Chapter 4
System Stability and Control Technologies after Large-Scale Wind Power Integration
Dezhi Chen, and Kun Ding
Abstract
This chapter focuses on the system stability control and relay protection of power system with large-scale wind power integrated and proposes corresponding measures. In addition, an automatic voltage control method is introduced.
Keywords
Automatic voltage control; FACTS; Relay protection; Short-circuit current characteristics; Small-disturbance stability; Synchronous stability; System stability and control; Voltage stability4.1. Impact of Large-Scale Wind Power Integration on Grid Protection
4.1.1. Analysis of Short-Circuit Current Characteristics of WTGs
4.1.1.1. Short-circuit current characteristics of FSIGs
The rotor winding of the fixed speed induction generator (FSIG) is generally 3-phase symmetric, which can be closed via external resistor or directly short connected. Compared with the synchronous generator, the asynchronous generator is not provided with independent field winding so that the voltage at the generator end will reduce close to zero when a 3-phase short circuit is present at the generator end, and since the generator is free from external excitation, the stator current will gradually attenuate and the stable short-circuit current will finally attenuate to zero. The short-circuit current of the asynchronous generator set consists of the attenuated direct current (DC) component and the attenuated alternating current (AC) component. The former attenuates by the time constant on the stator side, and the latter attenuates by the time constant on the rotor side.
Figure 4.1 shows the variation of the stator/rotor attenuation time constant with increase of the impedance from the short-circuit location to the generator end where the line resistance and reactance adopt the per-unit parameters of 110-kV overhead line (certain models).
Figure 4.1 shows that when the fault location is far from the point of common coupling (PCC) of the wind farm, the attenuation time constant of the generator stator will become smaller, which makes the aperiodic component of the fault current attenuate more rapidly; and the attenuation time constant of the generator rotor will increase, which makes the periodic component of the fault current attenuate more slowly.
Build a 49.5 MW wind farm model composed of 33 × 1.5 MW FSIGs, and connect it to an infinite voltage source and then conduct electromagnetic transient simulation. Suppose a 3-phase short-circuit fault occurs at the low voltage (LV) outlet of the generator-end transformer of the FSIG at moment of 0.5 s. See Figure 4.2 for the 3-phase short-circuit current of the stator after the fault. After the fault, since the short-circuit current of the FSIG contains the attenuated DC/AC component current, the short-circuit current will finally attenuate to zero if the fault is not cleared and the protection does not act, which is consistent with the analysis results.
4.1.1.2. Short-circuit current characteristics of double-fed induction generators
The short-circuit current of double-fed induction generators (DFIGs) are not only the characteristics of the wind turbine generator (WTG) itself but also affected by the control system, especially the status of the rotor overcurrent protection crowbar circuit, which has significant impact on the short-circuit current. First, we introduce the principle of the crowbar protection.
4.1.1.2.1. Rotor overcurrent protection crowbar of DFIGs
The rotor side of the DFIGs is integrated to the grid via the back-to-back voltage source frequency converter. Here is a typical problem—the frequency converter is an electric and electronic device, and it is easily damaged by overcurrent. Besides, the stator is directly integrated to the grid, and thus the failure of the grid will result in strong current on the stator side. Based on the electromagnetic coupling and flux linkage conservation between the stator and the rotor, the fault on the stator side will be transmitted to the rotor side. High voltage will be generated on the rotor side by induction and thus the large current will be generated. As a result, the frequency converter must be protected to prevent large current and uncontrollable energies from passing the converter. The simple and common practice is to make the rotor short-circuit via circuit when the large current occurs on the rotor side or overvoltage occurs on the DC link, which is generally called the crowbar. See Figure 4.3 for its structure.
The rotor side, composed of the thyristor and the external resistor Rcr controlled by the thyristor, is integrated to the rotor winding via the slip ring with the aim to restrain the rotor current. The basic principle is as follows: When overcurrent occurs on the rotor, the thyristor will be connected, the crowbar will be enabled, and the large current will be released via flowing Rcr, and at the same time, the rotor shall be short-circuited to protect the frequency converter from damage; and when the large current reduces below the limits, the thyristor will be disconnected, the crowbar will be disabled, and the control system of the frequency converter will be enabled again to recover the control functions of the DFIGs.
Once the crowbar is triggered, the rotor will be short-circuited, and the frequency converter on the rotor side will be out of control. In this case, the DFIG will run in the status of the squirrel-cage asynchronous generator provided with additional resistor. In case the frequency converter on the rotor side is out of control, the active/reactive output of the stator will be out of control during the fault period, and the field on the rotor side will be simultaneously lost. Since the frequency converter on the grid side is not directly connected to the generator winding, it is unnecessary to disable it in case the transient large current occurs. In this case, the frequency converter on the grid side can serve as a static var compensator to generate some reactive power. Obviously, the DFIG will run in different statuses when the crowbar is in various statuses. As a result, the short-circuit current characteristics of DFIGs shall be analyzed in the cases where the crowbar is enabled and disabled. The cases are as follows: (1) After the fault, the crowbar will be immediately put into operation and be in service during the fault period; (2) The crowbar is not put into operation during the whole fault period.
4.1.1.2.2. Crowbar enabled
After the crowbar is enabled, the operation status of the DFIG is similar to that of the ordinary asynchronous generator, and the short-circuit current is similar to that of the asynchronous generator, i.e., both of them contain the attenuated DC/AC components and will attenuate finally to zero. The difference between them lies in that (1) the initial value of the fault current is smaller and (2) and the attenuation time constant of the rotor is smaller and the fundamental-frequency AC component of the short-circuit current attenuates more rapidly.
The following simulation will be used to check the above analysis. Build a 49.5 MW wind farm model composed of 33 × 1.5 MW DFIGs, where the DFIGs are controlled by constant power factor and the power factor is controlled at 1. See Table 4.1 for the parameters of the DFIG.
Suppose that the 3-phase short-circuit fault occurs on the LV side of the step-up transformer of the WTG with duration of 0.5 s, and that the crowbar is enabled at the moment of fault occurrence and is not disabled. Figure 4.4 shows the variation curve of the generator-end voltage of the double-fed WTGs and Phase A short-circuit current of the stator as well as the comparison with the ordinary asynchronous WTGs at the same fault. Figure 4.4 shows that the simulation result is the same as that of the theoretical analysis.
4.1.1.2.3. Crowbar not enabled
If the crowbar is always not enabled after the fault, the frequency converter and the control system of the DFIGs will always work, and the frequency converter will control the active/reactive output of the DFIGs via control of the rotor current. In this case, the short-circuit current of the DFIGs will be affected by the control system. On one hand, the DFIG is designed with the asynchronous generator except that the frequency converter is installed on the rotor side so that the short-circuit current will take on the characteristics of the asynchronous generator; on the other hand, the DFIG can control the external voltage on the rotor side to regulate the power output and make it constant so that it will also take on the characteristics of the synchronous generator.
The analysis adopts the wind farm model and fault identical to that when the crowbar is used. Suppose the crowbar does not act before the fault is cleared. Figure 4.5 shows the stator 3-phase current variation curve of one of the DFIGs. Figure 4.5 shows that the short-circuit current of the DFIGs has the following three characteristics: (1) The variation trend of the current is to increase transiently after the fault and reduce rapidly, and then increase again. The reduction results from attenuation of the transient flux linkage, and the increase process is the action of the control system of the DFIGs. The short-circuit fault results in reduction of the generator-end voltage and the output electromagnetic power. In this case, the frequency converter on the rotor side will increase its reference value, resulting in increase of power output and simultaneously the stator current. (2) The stator current attenuates rapidly as shown in Figure 4.5, and the short-circuit current attenuates to half the impulse value of the short-circuit current in one period after the fault (t = 0.52 s). (3) If the fault is not cleared, the DFIGs will transmit the short-circuit current continuously, whose value is not larger than the rated current during stable operation.
Table 4.1
| Rated power P (MW) | 1.5 | Stator resistance Rs (p.u.) | 0.004 |
| Output voltage U (kV) | 0.69 | Stator reactance Xs (p.u.) | 0.1 |
| Field reactance Xm (p.u.) | 3.5 | Rotor resistance Rr (p.u.) | 0.01 |
| Rotor reactance Xr (p.u.) | 0.1 |
Figure 4.6 shows the stator 3-phase current variation curve of DFIGs when the generator-end voltage of DFIGs reduces to different levels due to fault and the crowbar is not enabled. In Figure 4.6, the generator-end voltage of DFIGs reduces to 0.15 p.u., 0.25 p.u., 0.75 p.u., and 0.9 p.u., respectively, from top to bottom. Figure 4.6 shows that the short-circuit current of the DFIGs has the following characteristics if the crowbar is not enabled after the fault: (1) The DFIGs will transmit the short-circuit current continuously if the generator-end voltage does not reduce to zero; (2) The stable short-circuit current varies when the generator-end voltage reduces to different levels, and the stable short-circuit current increases with growth of generator-end voltage, but the maximum value does not exceed the rated current.
4.1.1.3. Short-circuit current characteristics of DDPMSG
Build a 49.5 MW wind farm model composed of 33 × 1.5 MW direct driven permanent magnetic synchronous generators (DDPMSGs) where the DDPMSGs are controlled in the constant power factor mode. When the 3-phase short-circuit fault occurs in the grid at the moment of 0.5 s, the short-circuit current variation curve of the DDPMSGs is shown in Figure 4.7. The simulation results show that the current of the DDPMSG stays constant in case of grid faults, which is because the DDPMSG is designed with permanent magnet for excitation and the full-power frequency converter has some isolation function for the WTGs from the grid fault; the short-circuit current of the DDPMSGs increases at the moment of fault and then keeps constant in several periods after the fault, which is related to the control strategy of the control system of the DDPMSGs.
The control strategy of DDPMSGs on the grid side consists of two types: DC bus voltage/reactive power and DC bus voltage/AC voltage of frequency converter. For the whole WTG, the two control strategies are generally called the constant reactive power control mode and the constant voltage control mode of DDPMSGs. Below analyzes the short-circuit current characteristics at PCC of the wind farm in case of grid faults when the DDPMSG runs in the above two control modes.
Suppose the 3-phase symmetric short-circuit fault occurs in the middle of the transmission line of the wind farm at the simulation time of 0.1 s when the DDPMSGs are in constant power factor and constant voltage control modes where the grounding impedance is 4 Ω and the duration of fault is 0.3 s, the simulation curve of the short-circuit current at PCC of the wind farm is shown in Figure 4.8.
The calculation results of Figure 4.8 show that in the two control modes the current at the PCC of the wind farm increases rapidly at the moment of fault and the short-circuit current reaches the peak in the second period after the fault. The short-circuit current at the PCC in the constant voltage control mode is larger than that in the constant power factor mode, which is because the DDPMSG is excited by the permanent magnet and it can offer constant field current during the fault, and the full-power frequency converter has some isolation function for the grid fault so that the fault has smaller impact on the generator. In addition, the constant voltage control mode can offer some reactive power during the fault and the system can rapidly resume to the previous stable status after the fault is cleared.
4.1.2. Calculations of Short-Circuit Current after the WTGs are Integrated to the Gansu Grid
The short-circuit current at the main nodes of the grid shall be analyzed in the winter maximum operation mode of the Gansu Grid, 2010, when the wind farm is integrated to the grid in two cases—it is integrated and not integrated to Xinjiang. During the analysis, the WTGs in the wind farms are supposed to be all in service but their output is limited not to exceed the stable limits.
4.1.2.1. Impact of the wind farms in Jiuquan on the grid short-circuit current when the Northwest Grid is integrated to Xinjiang
When the Northwest Grid was integrated to Xinjiang in 2010, the Hexi 750 kV channel was provided with serial compensating capacitors and controllable high voltage (HV) reactors and the WTGs were in constant voltage control mode with limit transmission capacity of 3400 MW, and the WTGs in Jiuquan were supposed to be all in service, the 3-phase short-circuit current calculations of the main buses in the grid were given in Table 4.2 when the wind farms stopped operation and when they fully generated 3400 MW.
The calculation results shown that the integration of the planned wind farms in Jiuquan resulted in increase of the short-circuit current of the nearby nodes where the short-circuit current of Dunhuang 750 kV Substation bus, Jiuquan 750 kV Substation bus, Dunhuang 330 kV Substation bus, Yumenzhen 330 kV Substation bus, and Jiayuguan 330 kV Substation bus suffered from large impact, and the short-circuit current of Dunhuang 330 kV Substation bus and Yumenzhen 330 kV Substation bus even saw growth by 75%. Since the probability of start-up of all WTGs was low in actual operation and the short-circuit current had relations with the start-up mode of the WTGs in the system, the equipment verification such as the switchgear based on the results could ensure the reliability. If the existing switchgear could meet the requirements, no replacement was necessary.
4.1.2.2. Impact of the wind farms in Jiuquan Region on the grid short-circuit current when the Northwest Grid is not integrated to Xinjiang
When the Northwest Grid was not integrated to Xinjiang in 2010, the Hexi 750 kV channel was not provided with serial compensating capacitors and controllable HV reactors and the WTGs were in constant voltage control mode with limit transmission capacity of 1693 MW, and the WTGs in Jiuquan Region were supposed to be all in service, the 3-phase short-circuit current calculations of the main buses in the grid were given in Table 4.3 when the wind farms stopped operation and when they fully generated 1693 MW.
The calculation results shown that when the Northwest Grid was not integrated to Xinjiang, the integration of the planned wind farms in Jiuquan Region resulted in decrease of the short-circuit current of the nearby nodes compared to that of integration, but the short-circuit current when all WTGs were in service increased when all WTGs were shut down whereas the 330 kV bus of Dunhuang Substation rose by 112%.
Table 4.2
Calculated 3-Phase Short-Circuit Current of the Main Buses in the Grid When the Northwest Grid was Integrated to Xinjiang and the Wind Farms Stopped Operation and Fully Generated the Power
| No. | Name of the Short-Circuit Bus | Short-Circuit Current (kA) | Growth Rate (%) | |
| Shut Down of WTGs | Start-Up of All WTGs | |||
| 1 | Dunhuang 750 kV substation bus | 15.41 | 21.27 | 38.03 |
| 2 | Jiuquan 750 kV substation bus | 17.59 | 21.23 | 20.69 |
| 3 | Hexi 750 kV substation bus | 22.39 | 23.96 | 7.01 |
| 4 | Dunhuang 330 kV substation bus | 22.96 | 40.23 | 75.22 |
| 5 | Jiuquan 330 kV substation bus | 27.42 | 31.55 | 15.06 |
| 6 | Hexi 330 kV substation bus | 27.69 | 28.39 | 2.53 |
| 7 | Yumenzhen 330 kV substation bus | 9.45 | 16.54 | 75.03 |
| 8 | Jiayuguan 330 kV substation bus | 20.84 | 24.43 | 17.23 |
Table 4.3
Calculated 3-Phase Short-Circuit Current of the Main Buses in the Grid, the Northwest Grid was Not Integrated to Xinjiang, the Wind Farms Stopped Operation, and They Fully Generated the Power
| No. | Name of the Short-Circuit Bus | Short-Circuit Current (kA) | Growth Rate (%) | |
| Shut Down of WTGs | Start-Up of All WTGs | |||
| 1 | Dunhuang 750 kV substation bus | 8.73 | 14.60 | 67.24 |
| 2 | Jiuquan 750 kV substation bus | 12.46 | 17.63 | 41.49 |
| 3 | Hexi 750 kV substation bus | 19.38 | 22.44 | 15.79 |
| 4 | Dunhuang 330 kV substation bus | 15.62 | 33.14 | 112.16 |
| 5 | Jiuquan 330 kV substation bus | 23.51 | 29.67 | 26.20 |
| 6 | Hexi 330 kV substation bus | 26.25 | 27.76 | 5.75 |
| 7 | Yumenzhen 330 kV substation bus | 8.61 | 16.23 | 88.50 |
| 8 | Jiayuguan 330 kV substation bus | 18.43 | 23.39 | 26.91 |
4.1.3. Analysis of Protection Devices of WTGs
The typical wiring method of the wind farm is that the output voltage of one WTG (generally 0.69 kV) is boosted to the middle voltage (generally 10 kV or 35 kV) via one transformer, and then several WTGs are integrated to one MV bus (10 kV or 35 kV) and then boosted to another higher voltage level via the step-up transformer of the wind farm and integrated to the system via the transmission line of the wind farm (generally 110 kV or 330 kV bus). Figure 4.9 shows the typical wiring system of a wind farm.
The protection system of the wind farm consists of different protection zones, generally three zones: the WTGs and the generator-end transformer, the collecting system of the wind farm, and the step-up transformer of the wind farm. The protection of the WTGs protects the WTGs mainly via the control system, including high/low voltage, high/low frequency, WTG winding temperature control, etc. The step-up transformer of the wind farm and the bus are provided with a multifunctional digital protection system, mainly including the differential protection, the transformer overcurrent backup protection, the bus differential protection, under/overfrequency protection, under/overvoltage protection, and circuit breaker failure protection.
To realize coordination of the wind farm protection and the protection of the integrated system, the special wind farm integrated control system (Supervisory Control and Data Acquisition, SCADA) is generally for consistent monitoring and control over the wind farm.
Refer to Figure 4.10 for the typical protection configuration of the WTG integrated to the grid.
The functions of the protections in Figure 4.10 are described below:
1. The differential protection and the current quick-break protection serve as the main protection; they can disconnect the internal fault of the generator and the step-up transformer as well as the fault at the integration point of the WTG and the transformer. Since the capacity of the WTG is smaller than that of the traditional generator set, the longitudinal differential protection with percentage restraint characteristics is used to reflect the variation of the generator-end current, and after action, it will trip the WTG.
2. Zero-sequence overcurrent protection: It can reflect the earth fault between the generator and the step-up transformer and act to clear the fault. The WTG unit step-up transformer is generally designed with Y/△ wiring, and the zero-sequence current protection shall act to trip in case of stator winding or outlet earth fault. The principle of setting calculation of the protection action current Idz is to shunt the 3-phase imbalance current due to incomplete symmetry of the 3-phase current during start-up. The value of Idz based on the principle is small, and the zero-sequence current protection can be added with certain delay if it is difficult to shunt the imbalance zero-sequence current due to phase–phase short-circuit fault. Two sections of zero-sequence current protection can be provided where section Ⅰ can act to trip and section Ⅱ can act to alarm.
3. Zero-sequence overvoltage protection: It can reflect the earth fault on the HV side of the step-up transformer and the stator single-phase earth fault. The WTG step-up transformer is not provided with earth point on its HV side, and the earth capacitance current is too small to make the traditional earth protection act after a single-phase earth fault but it may result in intermittent arc. The neutral voltage deviation can be used to detect the voltage of the earth circuit and judge whether an earth fault is present where the zero-sequence voltage for protection is taken from the generator-end potential transformer (PT).
4. Low/high-frequency protection: It can reflect the abnormal system frequency and act to disconnect the generator to protect the equipment from damage. The WTG protection shall measure the frequency continuously and the measured frequency shall be subject first to mean value algorithm and then compared with the low/high limits of the grid frequency; if it exceeds the setting, the WTG will be disintegrated from the grid.
5. Low/high-voltage protection: The low-voltage judgment criterion is that all three line voltages are smaller than the setting of the undervoltage protection. In this case, the protection will act—it will act when the 3-phase voltage suddenly is lost during operation of the WTGs and the operating current is present by judgment. The action setting of the high-voltage protection can be based on the impulse voltage that the WTG can withstand. In case of voltage faults, the WTG must be disintegrated from the grid; the general measure is to shut down normally and then the associated measures shall be taken.
6. Overcurrent protection: This can reflect the internal fault of the WTGs and the phase–phase fault between the WTG and the step-up transformer. It can also serve as the backup protection of the fault on HV side of the step-up transformer. Heating is one of the main problems affecting the service life of the generator. Long-time overload or continuous starts in a short time will result in generator overheating and insulation aging, which is the common reason for generator damage. Since the overload of generators will result in overheating and the slight overload permits some time lag, the overload characteristics of the generator overload are inverse time lag. As a result, the generator shall be provided with inverse time lag protection to reflect the mean heating status of the stator winding and field winding, whose action time will decrease with growth of overcurrent, i.e., the action time lag is short when the current is large and the action time lag will be automatically prolonged when the current is small. Appropriate regulation shall be carried out to make the inverse time lag trip characteristics fit with the generator permissive overcurrent curve and then prevent the generator from overcurrent damages.
7. Directional power protection: This can reflect the abnormal status of the output power, prevent the equipment from damage, and ensure the normal generation operation of the WTGs. In some abnormal cases, the WTG may absorb the power and run as a motor, which may result in damage to the WTG. As a result, it shall be provided with directional power protection. The active power setting of the directional power protection is −0.2 ∼ −0.01 (per unit), and the time to disintegrate the WTG is based on the limit disintegrating time in case of generator-end 3-phase short-circuit fault.
4.1.4. Analysis of Impact of Wind Power Integration on Grid Protection
4.1.4.1. Impact of short-circuit current of wind farms on the grid
Take the actual grid of Gansu, 2010. Build the simulation model in DIgSILENT/PowerFactory as shown in Figure 4.11. The rated capacity of Changma Wind Farm is 200 MW and consists of 133 × 1.5 MW WTGs and is integrated to Yumenzhen 330 kV Substation via the single-circuit 110 kV line.
Suppose the Jiayuguan-Jiuquan 330 kV line has 3-phase short-circuit fault and the line is disintegrated after the fault holds on for 0.5 s. The following three cases are taken into account: (1) The wind farm is based on the DFIGs and the crowbar does not act during the fault; (2) The wind farm is based on the DFIGs and the crowbar is always in service during the fault, and the resistance of the bypass is 0.1 p.u.; (3) The wind farm is based on the FSIGs. Figure 4.12 shows the short-circuit current variation curve of the fault line in the above three cases.
Figure 4.12 shows that when the wind farm based on ordinary FSIGs is integrated, the short-circuit current on the fault line will increase rapidly at the moment of fault and then attenuate continuously and finally attenuate approximately to the current level before the wind power is integrated. Since the FSIG is not designed with independent excitation system, the 3-phase short-circuit fault of the system line will result in significant reduction of the WTG generator-end voltage, and the short-circuit current supplied by the WTG will attenuate continuously and finally attenuate to zero after the protection at the fault point trips the other circuit breakers. Besides, since no rapid electromagnetic control is available for the FSIG, the significant growth of current at the moment of short-circuit fault will have huge impact on the grid.
Enable the crowbar at the moment of fault and keep it in service all the time. In this case, the short-circuit current on the fault line is smaller than that when the crowbar is not enabled. This is because after the crowbar acts, the frequency converter on the rotor side will be short-circuited and its control system will not work simultaneously, i.e., the d, q components of the rotor current cannot be changed via regulation of the rotor voltage to control the active/reactive components of the stator current. Figure 4.13 shows the stator current variation curve of DFIGs when the crowbar is enabled and not enabled.
Obviously, if the crowbar does not act in case of the fault, the stator short-circuit current will be large and it will always provide current to the grid during the whole fault. If the crowbar is enabled in case of the fault, the stator current will rapidly attenuate approximately to zero, and the short-circuit current provided to the system during the whole fault period will be close to zero. In this case, the short-circuit current curve of the line is approximately overlapped with the short-circuit current curve when the WTGs are not integrated.
In addition, compare the short-circuit current of the DFIGs and the current of the ordinary FSIGs when the crowbar is enabled, and the results show that the current variation trend is similar to that of the ordinary FSIGs when the crowbar is enabled; it is approximately overlapped with the current curve when the ordinary FSIGs are integrated at the moment the fault is cleared, identical to the theoretical analysis.
The above analysis shows that the integration operation based on various types of WTGs has different impacts on the system short-circuit current. It shall consider the fault impulse current during equipment check and also the attenuation characteristics of the short-circuit current of the wind farms with various types of WTGs during protection setting and calibration.
It is recommended to work out the current of the FSIG at each action moment based on the attenuation time constant of the WTG stator/rotor during protection setting and calibration for integrating the wind farms with FSIGs; it shall first judge whether the crowbar acts during protection setting and calibration for integrating the wind farms with DFIG. If the crowbar is not enabled, the DFIG will provide continuous short-circuit current, and if the crowbar is enabled, the current provided during the fault period will reduce significantly.
4.1.4.2. Impact of harmonic generated by the wind farm on the grid protection
The root cause of harmonic generated in the power system lies in the nonlinear load. When the current flows through the load and is nonlinear to the voltage applied, the nonsinusoidal current will be generated, i.e., the harmonic will be generated in the circuit. The harmonic may reduce the power generation, transmission and utilization efficiency, overheat the electrical equipment, and result in vibration and noise, insulation aging and short service life, and even faults or burnout. In addition, the harmonic may result in local parallel or series resonance of the power system, amplifying the harmonic content and burning out the capacitor, etc. The harmonic may also result in unwanted actions of the relay protection and automation devices, confusing the energy metering system. As for the external of the power system, the harmonic may produce serious interference on the communication equipment and the electronic equipment.
At present, the leading WTGs in the world are provided with high-capacity power electronics frequency conversion equipment, which may input harmonic current to the grid. After the harmonic current is injected, the harmonic voltage of the grid bus has relations to the grid structural strength and the short-circuit capacity, and the harmonic current generated by the WTGs will be amplified at the node with weak grid structure and small short-circuit capacity, which may affect the service life and efficiency of the equipment in the grid and the relay protection equipment in the grid.
Figure 4.14 Voltage waveform and n-order harmonic ratio for voltage of the bus in Yumen 110 kV Substation. (a) Voltage waveform; (b) Harmonic ratio for voltage.
After the harmonic current generated by the two wind farms that are integrated to Yumen Substation is injected, the voltage waveform and n-order harmonic ratio for voltage of the bus in Yumen 110 kV Substation are shown in Figure 4.14, and the associated voltage total harmonic distortion (THD) is 9.45%, respectively. Figure 4.14 shows that the harmonic current injected by the wind farm to the grid and the harmonic voltage generated by the grid bus are very large in some cases, which, if not filtered, will put the grid security at risk.
The impact of harmonic on the relay protection is mainly distortion or reduced effect of the relay action characteristics and rejected or unwanted actions of the protection devices. Moreover, for the relays of various types, design performance and working principles, the impact of harmonic is also different.
The harmonic has little impact on the electromagnetic relay, and the setting error will be less than 10% when the harmonic content is less than 40%. Since the movable part of the inductive relay has large inertia and its action speed is low, the harmonic torque has little impact on it. For the rectifier relay protection device based on integral phase detector (e.g., the high-frequency phase differential protection and the differential protection), error is apt to occur during zero detection when the waveform has harmonic distortion, resulting in incorrect actions of the protection. The computer-based relay protection devices are usually provided with a filter unit in the hardware and digital filtration technology in the software to weaken the impact of harmonic. Nevertheless, for some high-speed protection with fast algorithm, the existence of harmonic will still have some adverse impact.
The differential protection of transformers generally adopts the break variable as the start criterion, which, however, cannot completely eliminate the impact of harmonic and may result in abnormal starts of protection. For the differential current quick-break and percentage differential protection, the harmonic component has little impact on the protection because the computer-based protection is provided with filter circuit and the one-cycle Fourier algorithm is used in the software computation, and thus good filtration performance is available.
When the system has large harmonic content, the line protection with phase current break power-frequency variable as the start component will face problems similar to the transformer protection, which may result in incorrect starts of the protection and the signal receiver/transmitter. In addition, since the distance measuring component of the distance protection is set according to the fundamental wave impedance of the line or the transformer, the measured impedance may be in error with the fundamental wave impedance when a fault occurs and the harmonic current is present, which may result in some error on the edge of the action zone.
4.2. Impact of Large-Scale Wind Power Integration on Stability of Power System
4.2.1. Impact of Wind Power Integration on Grid Voltage Stability
For the research on voltage stability of traditional power systems, the out-of-stability voltage or voltage collapse starts from the load point of the receiving-end system. Since the load demand exceeds the transmission power limits of the power grid and the system cannot maintain the balance of reactive power, the voltage becomes unstable. For the regional grid integrated with wind farms, the receiving-end system may be changed to the sending-end system when the wind farms runs with high output. The actual operation experience of the wind farms across the globe shows that the reduction of voltage stability still exists, which results from the reactive power characteristics of the wind farms. In essence, the voltage stability reduction or voltage collapse resulting from the wind farm has the same principle with the out-of-stability voltage in the traditional power system. As shown in the research on traditional power systems, the system disturbance may be load growth while in the research on the voltage stability of the power system containing wind farms, the system disturbance may be the output variation of the wind farm caused by wind velocity change and even the large disturbance fault present in the grid.
The wind farms in operation and those under planning in China are mostly provided with FSIGs, DFIGs, and the DDPMSGs. When the active power output by the asynchronous generator increases, the reactive power absorbed will increase, too. At the same time, the increase of the reactive power transmitted by the line will also result in increase of the reactive power consumed by the line reactance, and the reactive power consumption is in direct proportion to the square of the line current. As a result, when the wind power output is large, the total reactive load of the subsystem, including the wind farms and the equivalent lines, will be large, which reduces the reactive power reserve of the system and the grid voltage stability.
Since the DFIG and the DDPMSG can realize active/reactive decoupling control, the reactive characteristics of the wind farm based on the two types of generators depend on the control of the WTGs. Generally, the wind farm composed of DFIGs and DDPMSGs can control its output not to exchange the reactive power with the grid, i.e., the whole wind farm will not generate or consume reactive power. As a result, in the wind farm and the equivalent line, only the reactive losses of the line are the reactive load of the system; and compared with the wind farm composed of FSIGs, the reactive power consumption is smaller and the voltage stability is obviously better. At present, the DFIGs and the DDPMSGs in China mostly run in the control mode of constant power factor, and thus online regulation of the reactive power generated is unavailable, and the control over the grid reactive voltage is less than that of the synchronous generator. As a result, the voltage stability of the grid will be to some extent reduced with reduction grade dependent on the reactive power reserve of the thermal units in the grid when some thermal power is replaced by wind power in the grid.
4.2.2. Impact of Wind Power Integration on Stability of Synchronous Grid Power Angle
A lot of research has been carried out concerning the transient stability of synchronous generators. The imbalance between the mechanical torque and the electromagnetic torque during grid fault will speed up or down the rotor of the synchronous generator, directly resulting in swing of the power angle. Because the generator electromagnetic power is approximately in sinusoidal relation with the power angle, it will result in swing of the system voltage and power. If the fault is not eliminated in time, it will further result in being out of step with the generator set, destroying the stability.
The DFIG is based on the ordinary asynchronous induction generator with additional rotor converter and control system, and thus it is the AC excitation asynchronization-oriented synchronous generator. The angular speed of the synchronous rotating magnetic field ω0 is the sum of the electrical angular speed of the rotor and the angular speed of the rotating magnetic field generated by the additional excitation power supply of the rotor. The DFIG runs in the asynchronous manner, and the rotor speed can be regulated by changing the frequency of the AC excitation power supply, and thus it overcomes the requirement that the traditional synchronous generator must be strictly synchronous and the rigid connections between the mechanical part and the electrical part of the generator have been changed to flexible connections.
4.2.2.1. Mechanical torque characteristics of DFIGs
For every specific wind velocity, an associated speed-mechanical torque curve is given for the DFIG, as shown in Figure 4.15, where the associated wind velocity is v1 > v2 > v3 > v4. In normal operation status, the prime mechanical torque lies at the climax of the speed-torque characteristics curve, i.e., the working point of the optimum speed (e.g., points a, b, c, d).
In case the grid has fault, the electromagnetic power of the DFIG will decrease, which may result in speeding up of the WTG, and the operating point of the WTG will change in the direction of speed rise from the climax of the speed-torque characteristics curve (e.g., point a at wind velocity v1), and the mechanical torque will fall down during the fault process. According to the equal area criterion, the decrease of the mechanical torque will reduce the acceleration area of the power angle characteristics curve and relatively increase the speeding down area until the WTG returns to the normal operation status. Obviously, the mechanical torque characteristics of the DFIG can improve its transient characteristics.
4.2.2.2. Control of the rotor field current
The DFIG can change the speed of the generator rotor by means of controlling the frequency of the rotor field current. When a fault occurs on the grid side and the rotor is accelerated, it can keep the speed of the rotor rotating magnetic field at the synchronous speed via reducing the frequency of the rotor field current. As a result, the power angle δ has little change during the fault process and Pe will not have serious swing. In addition, if the control strategy of forced excitation similar to the synchronous generator is available for the DFIG during the fault process, it can raise the electromagnetic power via increasing the e.m.f. 
(the transient e.m.f. E′ for the transient process), which can also reduce the acceleration area and increase the speeding-down area. The rotor excitation control system of the WTG is good for improvement of the transient stability.
(the transient e.m.f. E′ for the transient process), which can also reduce the acceleration area and increase the speeding-down area. The rotor excitation control system of the WTG is good for improvement of the transient stability.4.2.2.3. Flywheel effect
From the view of energy balance, the acceleration characteristic of the DFIG during the fault process may convert some imbalanced transient energy to the kinetic energy for rotor rotation, alleviating the impulse on the grid. After the fault is cleared, it can be gradually released to the grid by means of regulation of the WTG control system. The variable speeds of the WTG make it equivalent to a charged flywheel in the transient process and provides the DFIG with better transient characteristics than the synchronous generator set. The above theoretical analysis shows that the DFIG has better transient characteristics than the synchronous generator set thanks to its variable-speed capacity, unique speed-prime mechanical torque characteristics, and advanced excitation control system.
4.2.3. Impact of Wind Power Integration on Grid Frequency Stability
The active power balance in the power system is the precondition for frequency stability of the power system. When a disturbance occurs in the system (e.g., short-circuit fault, trip, tie-line breaking, system disintegration, etc.), it may result in imbalance between the total generation power and the total load power. If the total generation power is more than the total load power (including the grid losses), the system frequency will rise; otherwise, if the total generation power is less than the total load power, the system frequency will fall. Based on the various frequency fluctuations and the actual operation status of the system, the associated measures, mainly including regulating the generator active output, disintegrating the generator, disintegrating the load, and so on, shall be taken. The frequency control is indispensable for stable operation and security of the power system. As the wind power penetration rises, it has become one of the key subjects in wind power research about how to ensure the frequency security and frequency stability during continuous operation of the power system after the wind power is integrated to the grid.
In case of the fault where the grid frequency is significantly reduced, the system inertia will play a crucial role in the variation rate of the system frequency—the lower the inertia, the more rapid the system frequency reduction. For any serious frequency accident in the power system, any reduction of inertia response is dangerous. For the WTGs of various generator technologies, the inertia frequency response characteristics are different. For the constant-speed WTGs based on ordinary asynchronous generators, since the coupling action between the rotor speed and the system frequency is very strong, the constant-speed WTGs can consequently reduce the speed to release some rotation kinetic energy and provide inertia response when the frequency of the power system reduces. In this case, the amplitude of inertia response is dependent on the energy charged in the WTG blades and rotors and the generator rotors as well as the grid frequency variation rate. Compared with the constant-speed WTGs, the variable-speed WTGs based on DFIGs have improved control capacity and can carry out decoupling control over the active/reactive power, respectively. The inertia response of WTGs is a key factor determining the impact of the increased wind power on the power system.
When a large amount of installed wind power is available at high or low wind velocity, the WTGs may replace some traditional generator sets of the system. If the WTG fails to provide inertia response, the effective inertia of the whole system will reduce, and as the proportion of the wind power grows, the adverse impact on the inertia of the power system will gradually increase. As a result, when a large number of WTGs in operation have replaced the traditional power plants and the power system based on DFIGs operates at low load and small system capacity, the reduction of system inertia is very adverse.
For the large-sized power plant, the typical inertia time constant of the synchronous generator set is usually 2 ∼ 9 s. And the typical inertia time constant of WTGs is 2 ∼ 6 s. This indicates that it does not really reduce the total rotation kinetic energy after the wind power is integrated to the grid and that for the DFIG, the inherent inertia in the action of the control system is shown as a “hidden inertia” for the grid, and consequently it cannot increase the inertia of the grid during the transient process of frequency variation.
4.2.4. Analysis of Impact of Wind Farm Integration on the System Small Disturbance Stability in Jiuquan Region
Since the Hexi transmission channel sees the maximum power transmitted in the summer minimum mode, which has the maximum impact on the system small disturbance stability, the calculations adopt the reference data of the Northwest Grid in the summer minimum mode, 2010. Since the small disturbance analysis algorithm used is a full-dimensional partial characteristics algorithm, it does not simplify the units and network in the power system.
In the analysis, it considers whether the Hexi transmission channel is provided with serial compensating capacitors/controllable HV reactors and whether the Gansu Grid is integrated with Xinjiang. Based on it, the research consists of three cases: (1) Not provided with serial compensating capacitors/controllable HV reactors, and not integrated with Xinjiang; (2) Not provided with serial compensating capacitors/controllable HV reactors, and integrated with Xinjiang to transmit out 1000 MW from Xinjiang; and (3) Provided with serial compensating capacitors/controllable HV reactors, and integrated with Xinjiang to transmit out 1000 MW from Xinjiang. Based on the calculated system stability limit level of the three cases, check the small disturbance stability, respectively, and work out the system swing and damping characteristic variation of the wind farms in Jiuquan Region at various output levels in the typical system electromechanical swing mode for each case.
4.2.5. Small Disturbance Stability Analysis and Damping Judgment Criteria
4.2.5.1. Analysis of characteristic values
Based on the judgment criteria of small disturbance stability, the system is stable only when the real parts of all the system characteristic roots are negative, and it is unstable if any characteristic root has positive real part. The focuses of the analysis on characteristic roots shall fall on the negative damping root of the system and the weak damping root with damping ratio less than 0.03, especially the characteristic root with low frequency. Based on the real and imaginary parts of the characteristic root, the swing frequency and damping ratio in the swing mode can be worked out.
4.2.5.2. Analysis of characteristic vectors
For each characteristic root, the generator has an associated characteristic vector, and the angle of the vector represents the relative position of the generator in the electromagnetic swing. Those generators with identical or similar angles are synchronous in the swing; otherwise, those with angle difference of 180° or approximate are swung between them. The value of the characteristic vector module shows the intensity of the generator swing in the swing mode where the amplitude of the generator swing will become larger with module growth.
4.2.5.3. Analysis of participation factors
For the swing mode with negative or weak damping, it shall determine the generators having the strongest correlation with the mode so that the power system stabilizer (PSS) can be installed at the most proper location. Generally, the generator with the maximum participation factor is the one having the strongest correlation with the swing mode. Compared with the generators of the strongest correlation, those generators with the participation factor of smaller magnitude order have little correlation with the swing mode, and the generator with smaller participation factor has smaller impact on the conduct characteristics of the swing mode. Generally, the generators with large participation factor are provided with the power system stabilizer.
4.2.5.4. Damping judgment criteria
According to “Specification on security and stability calculations of power systems, SGCC,” the judgment criteria for dynamic stability from the frequency domain solution are that the damping ratio in each swing mode shall be larger than zero. To ensure that the system has appropriate small disturbance dynamic stability, the system damping ratio shall meet the following criteria:
1. It is deemed as negative when the damping ratio is less than 0, and in this case, the system cannot run stably;
2. It is deemed as weak when the damping ratio falls in a range of 0 ∼ 0.02;
3. It is deemed as very weak when the damping ratio falls in a range of 0.02 ∼ 0.03, and in normal modes, the damping ratio of the regional swing mode and the swing modes having strong correlation with the large power plants and units shall be generally larger than 0.03;
4. It is deemed as appropriate when the damping ratio falls in a range of 0.04 ∼ 0.05, and the system will have good dynamic characteristics when the damping ratio is larger than 0.05;
5. The damping ratio shall fall in a range of 0.01 ∼ 0.015 as the minimum requirement in the special operation mode after faults.
4.2.6. Analysis of System Small Disturbance Stability and Damping Characteristics in Various Cases
4.2.6.1. Not provided with serial compensating capacitors/controllable HV reactors, and not integrated with Xinjiang
The analysis is based on the conditions that the Hexi 750 kV transmission channel is not provided with serial compensating capacitors/controllable HV reactors, and the Northwest Grid is not integrated with Xinjiang. First of all, work out all the characteristic values of the system before the wind farms in Jiuquan Region are integrated, sort out the swing mode between or among the weak damping zones of strong correlation with the large-capacity units in the grid, and carry out modal analysis; then, analyze and calculate the impact of the various output levels of the wind power in Jiuquan Region on the swing and damping characteristics of these modes. Figure 4.16 shows the characteristic value distribution of the system corresponding to all the swing modes before the wind farms in Jiuquan Region are integrated and when the system runs in the summer minimum mode of 2010.
Figure 4.16 shows that the real parts of the characteristic roots corresponding to all swing modes in the system are all negative before the wind farms in Jiuquan Region are integrated, i.e., the system is stable in case of small disturbance before the wind power is integrated, but the damping ratio corresponding to several characteristic values is low so that once these modes are triggered by some faults or operation modes of the system, the grid may generate low-frequency continuous power swing due to low damping. Based on the calculated characteristic roots of all swing modes, sort out the swing mode between or among the weak damping zones of strong correlation and participation with the units in Gansu Grid, and work out the characteristic root, swing frequency, modal and unit participation corresponding to these modes (see Table 4.4) where in the modal analysis, it shows the units with maximum module value in the right characteristics vector.
Table 4.4
Relevant Swing Mode Characteristics in Summer Minimum Mode, 2010, before the Wind Farms in Jiuquan Region Are Integrated (Not Provided with Serial Compensating Capacitors, and Not Integrated with Xinjiang)
| Mode | Characteristic Root | Damping Frequency (Hz) | Damping Ratio (%) | Modal Analysis | Participant Factor |
| 1 | −0.206 ±j4.439 | 0.706 | 4.643 | Swing between the units of Gansu, Qinghai and the units of Shaanxi and Ningxia Units #1, #2 of Liujiaxia, Gansu = 0.00019 Unit #1, technical improved, Ningxiashi = 0.00017 | Units #1, #2 of Liujiaxia, Gansu = 1 Unit #1, technical improved, Ningxiashi = 0.47 |
| 2 | −0.245 ±j5.355 | 0.852 | 4.565 | Swing between some units of Ningxia and the whole grid Units #1, #2 of Zhongning, Ningxia = 0.0003 Units #1, #2 of Liujiaxia = 0.00016 | Units #1, #2 of Zhongning, Ningxia = 1 Units #1, #2 of Liujiaxia, Gansu = 0.65 Units #1, #3 of Aluminum Manufacturer, Gansu = 0.598 |
| 3 | −0.211 ±j6.566 | 1.045 | 3.208 | Swing between some units of Gansu and Ningxia and the whole grid Units #1, #2 of the thermal power plants in western Ningxia = 0.0003 Unit #4, Dam Power Plant, Ningxia = 0.00041 | Units #1, #3 of Aluminum Manufacturer, Gansu = 1 Unit #4, Dam Power Plant, Ningxia = 0.87 |
| 4 | −0.446 ±j6.568 | 1.045 | 6.777 | Swing between the units in the radiation range of Hexi and the whole grid Units #1, #4 of Xiliushui, Gansu = 0.0029, Unit #1 = 0.0028 | Unit #2 of Xiliushui, Gansu = 1, Units #3, #4 = 0.945 |
| 5 | −0.053 ±j6.990 | 1.112 | 0.761 | Swing among the unit groups of the Ningxia grid, the radiation network of Gansu Hexi, and the ring network of Hexi Units #1, #2 of Jingyuan Power Plant, Gansu = 0.00021 | Units #1, #3 of aluminum manufacturer, Gansu = 1; Units #1, #4 of Jingyuan Power Plant, Gansu = 0.761 |
| 6 | −0.068 ±j7.122 | 1.133 | 0.949 | Swing among the unit groups of the radiation network of Gansu Hexi and the Ningxia grid Units #1, #2 of Jingyuan, Gansu = 0.0021 | Units #1, #2 of Lanzhou Thermal Power Plant, Gansu = 1; Units #1, #2 of Jingyuan, Gansu = 0.48 |
| 7 | −0.425 ±j8.314 | 1.323 | 5.104 | Swing between the units of the radiation network of Hexi, Gansu, and the units of the Hexi ring network and Ningxia Unit #1 of 803 Power Plant, Gansu = 0.00096; Unit #2 of Xiliushui = 0.00099; Zhangye Power Plant = 0.00091 | Unit #1, 803 Power Plant = 1 Units #1, #2 of Zhangye Power Plant = 0.903 |
| 8 | −0.545 ±j9.574 | 1.524 | 5.684 | Swing among the units in the radiation network of Hexi, Gansu Unit #1, 803 Power Plant = 0.012; Units #1, #2 of Zhangye Power Plant = 0.0069 | Unit #1, 803 Power Plant = 1 Units #1, #2 of Zhangye Power Plant = 0.335 |
The eight modes shown in Table 4.4 are the swing modes existing between two or more large unit groups in the whole grid where the number of participating units is large and the swing frequency and damping ratio are low and the units in the Gansu Grid; see high participation degree. Modes 1–7 are the swing modes between zones, and mode 8 is the internal swing mode in the radiation network of Hexi, Gansu. Obviously, modes 5 and 6 are the swing modes between zones participated in by several unit groups where the damping ratio is the minimum. The above sections show, in this case, the limit output of the WTGs in Jiuquan Region is 1693 MW when the system reaches transient stability limits. The variation of damping characteristics in the swing modes given in Table 4.4 will be analyzed for various output levels of the WTGs in Jiuquan (from zero to limit output). The variation of the associated damping characteristics in the swing modes given in Table 4.4 is analyzed in the four output cases of the WTGs in Jiuquan Region: zero output, 30% limit output, 60% limit output, and 100% limit output. See Figure 4.17.
The above variation of damping ratio in the swing modes shows that when the WTGs in Jiuquan Region have limit output (1693 MW), each swing mode still has positive damping, i.e., the system is still stable in case of small disturbance. After the WTGs are integrated, the damping ratio of modes 2, 3, 5, 6, and 8 is reduced. As the output level of the WTGs grows, the damping ratio of modes 1, 5, and 8 is basically in a rising trend while the damping of mode 6 gradually reduces. As a result, the large-scale wind power integration in Jiuquan Region will reduce the damping ratio in the swing modes between or in some weak damping zones, deteriorating the damping level of the system.
4.2.6.2. Not provided with serial compensating capacitors/controllable HV reactors, and integrated with Xinjiang to transmit out 1000 MW from Xinjiang
Figure 4.18 shows the distribution of all characteristic values of the system in the summer minimum mode, 2010, before the WTGs in Jiuquan are integrated when the Hexi 750 kV transmission lines are not provided with serial compensating capacitors/controllable HV reactors and Gansu is not integrated with Xinjiang.
Before the WTGs in Jiuquan are integrated, the real parts of the characteristic roots corresponding to all swing modes in the system are all negative, i.e., the system is stable in case of small disturbance but the damping ratio corresponding to several characteristic values is low. In case of corresponding fault or disturbance, it is apt to have low-frequency swing in the system due to low damping ratio in some modes. Based on the calculated characteristic roots of all swing modes, sort out the swing mode between or among the weak damping zones of strong correlation and participation with the units in Gansu Grid, and work out the characteristic root, swing frequency, and modal and unit participation corresponding to these modes (see Table 4.5) where in the modal analysis, it shows the units with maximum module value in the right characteristics vector.
Table 4.5
Relevant Swing Mode Characteristics in Summer Minimum Mode, 2010, before the Wind Farms in Jiuquan Region Are Integrated (Not Provided with Serial Compensating Capacitors, Integrated with Xinjiang)
| Mode | Characteristic Root | Damping Frequency (Hz) | Damping Ratio (%) | Modal Analysis | Participant Factor |
| 1 | −0.062 ±j3.948 | 0.706 | 1.563 | Swing between the units of Gansu and the radiation network of Hexi, Gansu, and the units of Ningxia Unit #1, technical improved, Ningxiashi = 0.00062 Units #1, #4 of Hadebute, Xinjiang = 0.00043 | Unit #1, technical improved, Ningxiashi = 1; Units #1, #2 of Luntai Thermal Power Plant, Xinjiang = 0.36 |
| 2 | −0.173 ±j4.370 | 0.852 | 3.966 | Swing between the units of Hexing Ring Network, Gansu, and Qinghai, and the units of Xinjiang Units #1, #2 of Liujiaxia = 0.00019 Unit #1, technical improved, Ningxiashi = 0.00016 | Units #1, #2 of Liujiaxia, Gansu = 1 Unit #1, technical improved, Ningxiashi = 0.454 |
| 3 | −0.216 ±j4.989 | 1.045 | 4.332 | Swing among some units of Ningxia, the units of Xinjiang, and the units of Qinghai Unit #1, technical improved, Ningxiashi = 0.00037 Units #1, #2 of Zhongning Power Plant = 0.00023 | Units #1, #2 of Liujiaxia, Gansu = 0.35 Unit #1, technical improved, Ningxiashi = 1 |
| 4 | −0.221 ±j5.250 | 1.045 | 4.203 | Swing between the units of Xinjiang and Ningxia and the units of Gansu and Qinghai Units #1, #2 of Zhongning Power Plant, Ningxia = 0.00033 Units #1, #2 of Liujiaxia = 0.00015 | Units #1, #2 of Zhongning Power Plant, Ningxia = 1 Units #1, #3 of Aluminum Manufacturer, Gansu = 0.548 |
| 5 | −0.389 ±j6.269 | 1.112 | 6.189 | Swing among the units in the radiation network of Hexi, Gansu, the units of the ring network of Hexi, and some units of Qinghai Units #3, #4 of Xiliushui, Gansu = 0.0036; Units #1, #2 = 0.0031 | Units #3, #4 of Xiliushui, Gansu = 1; Units #1, #2 = 0.75 |
| 6 | −0.022 ±j8.674 | 1.133 | 0.253 | Swing among the units in the radiation network of Hexi, Gansu, the units of the ring network of Hexi, and some units of Qinghai Units #1, #4 of Jingyuan, Gansu = 0.00029 | Units #1, #4 of Jingyuan, Gansu = 1 |
| 7 | −0.173 ±j9.088 | 1.323 | 1.902 | Swing between some units in Qinghai and the ring network of Hexi, Gansu Units #3, #10 of Hexi Salt Works, Gansu = 0.0014; Unit #1 of Tiecheng = 0.00097 | Units #3–10 of Hexi Salt Works, Gansu = 1 |
| 8 | −0.428 ±j9.174 | 1.524 | 4.659 | Swing among the units in the radiation network of Hexi Unit #1 of 803 Power Plant, Gansu = 0.0137; Unit #1 of Erlongshan = 0.0075; Unit #1 of Dagushan = 0.0073 | Unit #1 of 803 Power Plant, Gansu = 1 Units #1, #2 of Jiure #3 Power Plant = 0.41 |
The eight modes showed in Table 4.5 are the swing modes existing between two or more large unit groups in the whole grid where both the swing frequency and the damping ratio are low and the units in the Gansu Grid see high degree of participation. Modes 1–7 are the swing modes between zones and mode 8 is the internal swing mode in the radiation network of Hexi, Gansu. Obviously, the swing frequencies of these modes all fall in the frequency range of low-frequency swing where the damping ratio in mode 6 sees the minimum. The above sections show, in this case, the limit output of the WTGs in Jiuquan Region is 1736 MW when the system reaches transient stability limits. The variation of damping characteristics in the swing modes given in Table 4.5 will be analyzed for various output levels of the WTGs in Jiuquan (from zero to limit output).
The variation of the associated damping characteristics in the swing modes given in Table 4.5 is analyzed in the four output cases of the WTGs in Jiuquan Region: zero output, 30% limit output, 60% limit output, and 100% limit output. See Table 4.6. When the WTGs in Jiuquan Region have limit output (1736 MW), each swing mode still has positive damping, i.e., the system is still stable in case of small disturbance when it reaches the transient stability limits. After the WTGs in Jiuquan Region are integrated, the damping level of modes 3–6 is increased while the damping ratio of modes 2, 7, and 8 is decreased. As the output level of the WTGs grows, the damping ratio of the modes other than modes 1, 3, 4, and 7 is basically in a rising trend. When it reaches the system transient stability limits, modes 1 and 6 see a very low damping ratio.
4.2.6.3. Provided with serial compensating capacitors/controllable HV reactors, and integrated with Xinjiang to transmit out 1000 MW from Xinjiang
The analysis is based on the conditions that the Hexi 750 kV transmission line is provided with serial compensating capacitors/controllable HV reactors, and Gansu is not integrated with Xinjiang, and Figure 4.19 shows the distribution of all the characteristic values in the summer minimum mode, 2010, before the WTGs in Jiuquan Region are integrated.
The real parts of the characteristic roots corresponding to all swing modes in the system are all negative before the wind farms in Jiuquan Region are integrated, i.e., the system is stable in case of small disturbance but the damping corresponding to several characteristic values is low so that it is apt to generate low-frequency swing due to insufficient system damping. Based on the calculated characteristic roots of all swing modes, sort out the swing mode between or among the weak damping zones of strong correlation and participation with the units in Gansu Grid, and work out the characteristic root, swing frequency, and modal and unit participation corresponding to these modes (see Table 4.7), where in the modal analysis, it shows the units with maximum module value in the right characteristics vector.
Table 4.6
Damping Variation of Relevant Modes at Various Output Levels of WTGs
| Mode | Damping Ratio at Various Modes (%) | |||
| Zero Output of Wind Farms | 30% Limit Output | 60% Limit Output | 100% Limit Output | |
| 1 | 1.563 | 3.912 | 2.718 | 0.264 |
| 2 | 3.966 | 3.357 | 3.107 | 3.891 |
| 3 | 4.332 | 4.362 | 4.346 | 4.338 |
| 4 | 4.203 | 4.281 | 4.265 | 4.247 |
| 5 | 6.189 | 6.600 | 6.718 | 6.657 |
| 6 | 0.253 | 0.26 | 0.264 | 0.267 |
| 7 | 1.902 | 1.900 | 1.897 | 1.895 |
| 8 | 4.659 | 4.434 | 4.580 | 4.607 |
Figure 4.19 Characteristic values of system in summer minimum mode, 2010, before the WTGs in Jiuquan Region are integrated (provided with serial compensating capacitors/controllable HV reactors, and integrated with Xinjiang).
The eight modes shown in Table 4.7 are the swing modes existing between two or more large unit groups in the whole grid where both the frequency and the damping ratio are low and the units in the Gansu Grid see high participation degree. Modes 1–7 are the swing modes between zones and mode 8 is the internal swing mode in the radiation network of Hexi, Gansu. Obviously, the swing frequencies of these modes all fall in the frequency range of low-frequency swing where the damping ratio in mode 6 sees the minimum. The above sections show, in this case, the limit output of the WTGs in Jiuquan Region is 3400 MW when the system reaches transient stability limits. The variation of damping characteristics in the swing modes given in Table 4.7 will be analyzed for various output levels of the WTGs in Jiuquan (from zero to maximum output). The variation of the associated damping characteristics in the swing modes given in Table 4.7 is analyzed in the four output cases of the WTGs in Jiuquan Region: zero output, 30% limit output, 60% limit output, and 100% limit output. See Table 4.8.
The damping ratio variation in each swing mode given in Table 4.8 indicates that when the WTGs in Jiuquan Region have maximum output (3400 MW), each swing mode still has positive damping, i.e., the system is still stable in case of small disturbance when it reaches the transient stability limits. After the WTGs in Jiuquan Region are integrated, the damping of modes 1, 2, and 7 gradually reduces while the damping ratio of modes 3 and 4 gradually rises with the growth of WTG output level; and when the output of the WTGs reaches the limit level, the damping level of modes 5 and 6 is equivalent to that before the wind farms are integrated. At the maximum output of WTGs, modes 1 and 6 see a low damping ratio and the system has small stability margin in case of small disturbance, which is adverse for the system dynamic stability.
Table 4.7
Relevant Swing Mode Characteristics in Summer Minimum Mode, 2010, before the Wind Farms in Jiuquan Region Are Integrated (Provided with Serial Compensating Capacitors, and Integrated with Xinjiang)
| Mode | Characteristic Root | Damping Frequency (Hz) | Damping Ratio (%) | Modal Analysis | Participant Factor |
| 1 | −0.051 ±j4.161 | 0.662 | 1.229 | Swing between the units of Gansu and the radiation network of Hexi, Gansu, and the units of Ningxia Unit #1, technical improved, Ningxiashi = 0.0017 Units #1, #4 of Hadebute, Xinjiang = 0.0009 | Unit #1, technical improved, Ningxiashi = 1 Units #1, #2 of Luntai Thermal Power Plant, Xinjiang = 0.52 |
| 2 | −0.174 ±j4.415 | 0.703 | 3.942 | Swing between the units of Hexing Ring Network, Gansu, and Qinghai, and the units of Xinjiang Units #1, #2 of Liujiaxia = 0.00023 Unit #1, technical improved, Ningxiashi = 0.0002 | Units #1, #2 of Liujiaxia, Gansu = 1 Unit #1, technical improved, Ningxiashi = 0.491 |
| 3 | −0.202 ±j4.996 | 0.795 | 4.048 | Swing among some units of Ningxia, the units of Xinjiang, and the units of Qinghai Unit #1, technical improved, Ningxiashi = 0.00037 Units #1, #2 of Zhongning Power Plant = 0.00023 | Units #1, #2 of Liujiaxia, Gansu = 0.58 Unit #1, technical improved, Ningxiashi = 1 |
| 4 | −0.213 ±j5.269 | 0.839 | 4.037 | Swing between the units of Xinjiang, Ningxia and the units of Gansu, Qinghai Units #1, #2 of Zhongning Power Plant = 0.00052 Units #1, #2 of Liujiaxia = 0.00035 | Units #1, #2 of Zhongning Power Plant, Ningxia = 1 Units #1, #3 of Aluminum Manufacturer, Gansu = 0.593 |
| 5 | −0.417 ±j6.331 | 1.008 | 6.571 | Swing among the units in the radiation network of Hexi, Gansu, the units of the ring network of Hexi, and some units of Qinghai Units #3, #4 of Xiliushui, Gansu = 0.0039; Units #1, #2 = 0.0027 | Units #3, #4 of Xiliushui, Gansu = 1; Units #1, #2 = 0.71 |
| 6 | −0.022 ±j8.705 | 1.385 | 0.252 | Swing among the units in the radiation network of Hexi, Gansu, the units of the ring network of Hexi, and some units of Qinghai Units #1, #4 of Jingyuan, Gansu = 0.00025 | Units #1, #4 of Jingyuan, Gansu = 1 |
| 7 | −0.175 ±j9.109 | 1.450 | 1.920 | Swing between the units of Qinghai and some units of the ring network of Hexi, Gansu Units #3, #10 of Hexi Salt Works = 0.0031 Unit #1 of Tiecheng = 0.0012 | Units #3, #10 of Hexi Salt Works, Gansu = 1 |
| 8 | −0.448 ±j9.236 | 1.470 | 4.850 | Swing among the units in the radiation network of Hexi Unit #1 of 803 Power Plant, Gansu = 0.0072; Unit #1 of Erlongshan = 0.0043 Unit #1 of Dagushan = 0.0042 | Unit #1 of 803 Power Plant, Gansu = 1 Units #1, #2 of Jiure #3 Power Plant = 0.556 |
Table 4.8
Damping Variation of Relevant Modes at Various Output Levels of WTGs
| Mode | Damping Ratio at Various Modes (%) | |||
| Zero Output of Wind Farms | 30% Limit Output | 60% Limit Output | 100% Limit Output | |
| 1 | 1.229 | 1.091 | 0.674 | 0.143 |
| 2 | 3.942 | 3.532 | 3.215 | 3.064 |
| 3 | 4.048 | 4.089 | 4.122 | 4.189 |
| 4 | 4.037 | 4.144 | 4.159 | 4.237 |
| 5 | 6.571 | 6.970 | 6.858 | 6.523 |
| 6 | 0.252 | 0.261 | 0.270 | 0.258 |
| 7 | 1.920 | 1.914 | 1.911 | 1.908 |
| 8 | 4.850 | 4.761 | 4.721 | 4.554 |
4.3. Impact of Large-Scale WTG Disintegrations on Grid Stability and Prevention and Control Measures
4.3.1. Impact of Large-Scale WTG Disintegrations on Grid Stability
4.3.1.1. Analysis on grid voltage characteristics after large-scale disintegrations of WTGs
The typical mode shall be taken as the reference mode to study the reactive voltage characteristics after the large-scale WTGs are disintegrated. The boundary conditions for calculations are as follows: Xinjiang transmits 1000 MW to the Northwest Grid, and the full output of WTGs is 2150 MW. In the reference mode, the power transmitted at the key sections in the Hexi transmission channel is given below: Hadun Section: 1000 MW; Dunquan Section: 3014 MW; Quanhe Section: (including the Zhangshan 330 kV double-circuit lines) 3286 MW; Hewu Section: (including the Heliang 330 kV double-fed lines) 2773 MW.
See Table 4.9 for the voltage levels of the bus of the key substations in Hexi in the reference mode where the voltage levels of all buses fall in a reasonable range.
Table 4.9
Voltage Levels of Hexi Buses in Typical Mode (kV)
| Bus | Hami S/S | Dunhuang S/S | Jiuquan S/S | Hexi S/S | Wusheng S/S |
| 750 kV bus | 750/330 kV bus | 750/330 kV bus | 750/330 kV bus | 750/330 kV bus | |
| Voltage | 774.2 | 771.2/356.8 | 766.1/361.2 | 760.5/350.6 | 763.5/348.3 |
| Bus | Qiaoxi wind power step-up S/S | Yumen S/S | Jiayuguan S/S | Zhangye S/S | Liangzhou S/S |
| 330 kV bus | 330 kV bus | 330 kV bus | 330 kV bus | 330 kV bus | |
| Voltage | 357.1 | 361.4 | 360.4 | 356.2 | 348.5 |
Figure 4.20 shows the voltage dynamic variation curve of the bus of the key 750 kV substations in the Hexi transmission channel when a certain disturbance occurs in the grid or in the wind farm, resulting in the WTGs of 530 MW in Jiuquan (mainly Guazhou Region) disintegrated from the grid.
Figure 4.21 shows the voltage variation curve of the bus of Dunhuang 750 kV Substation in the Hexi grid when the capacity of WTGs in Jiuquan disintegrated from the grid rises to 956 MW. And Figure 4.22 shows the voltage variation curve of the bus of the rest 750 kV substations in the Hexi transmission channel.
In this case, the power angle variation curve and the frequency deviation curve of the system are shown in Figures 4.23 and 4.24, respectively.
Figures 4.23 and 4.24 show that both the system power angle and the frequency can still remain stable when the generators are disintegrated from the grid. The power flow of the Hexi transmission channel designed for wind power transmission, however, becomes smaller due to disintegration of WTGs, which results in rapid rise of the bus voltage in the 750 kV substations in the Hexi transmission channel with the bus of Dunhuang 750 kV Substation seeing the maximum variation; see Figures 4.20–4.22. The comparisons on the voltage rise after WTGs of 530 and 956 MW disintegrated from the grid show, the voltage rise will see larger amplitude as the capacity of WTGs disintegrated from the grid grows, and the bus voltage of Dunhuang Substation will be higher than the per-unit value 1.0 when the capacity of WTGs disintegrated from the grid reaches 956 MW, i.e., the actual value of the voltage is higher than the security voltage 800 kV, which will put the security of the Hexi transmission channel grid at risk.
4.3.1.2. Analysis of grid frequency characteristics after large-scale disintegrations of WTGs
The system spinning reserve has a significant impact on the system frequency characteristics after large-scale WTGs are disintegrated. Based on the operation specification, a spinning capacity of 2 ∼ 5% maximum generation capacity shall be reserved during normal operation of the grid. Since the Northwest Grid is likely to operate disintegrated from the grid in case of faults after it is integrated to Xinjiang, the spinning reserve of the Northwest Grid (excluding Xinjiang) shall be larger than the maximum capacity of a single unit, i.e., 1000 MW. In this case, the Northwest Grid (excluding Xinjiang) shall be designed with 1120 MW spinning reserve, which is approximately 2.37% maximum generation capacity. The impact of large-scale WTG disintegrations on the grid frequency characteristics is studied in the following conditions: the power flow transmitted out by Xinjiang is 1000 MW, the power flow at Quanhe Section is 3630 MW, and the wind power integrated to the grid is 3000 MW.
After WTGs of 3000 MW are disintegrated from the grid without any fault, the power shortage will result in reduction of system frequency, which may fall down to the minimum value—48.95 Hz, less than 49 Hz. In this case, the frequency will become unstable and trigger the third protection line of the grid—the low-frequency load-shedding protection—to act. The protection will first act to disintegrate 2800 MW loads, and then the frequency will recover to the stable frequency (49.56 Hz), and the system frequency will remain stable after the low-frequency load-shedding protection acts. See Figure 4.25 for the frequency simulation characteristic curve after WTGs are disintegrated from the grid.
4.3.2. Measures to Suppress the Impact of Large-Scale WTG Disintegration on the Grid
4.3.2.1. Prevention and control measures on system voltage stability after large-scale WTGs are disintegrated from the grid
4.3.2.1.1. The controllable HV reactor on the bus of Dunhuang Substation is enabled
The controllable HV reactor, a flexible alternative current transmission system (FACTS), is designed with rapid regulation and control capacity. It can follow the bus voltage fluctuation caused by random fluctuation of WTG output and control the voltage in a reasonable range. In addition, thanks to the rapid response capacity, it can suppress the bus voltage rise at the moment of fault occurrence when the WTG is disintegrated from the grid.
After WTGs of 956 MW are disintegrated from the grid, the controllable HV reactor installed on the bus of Dunhuang Substation can rapidly respond to the fault and carry out rapid regulation and control to output inductive reactive power to suppress the rapid rise of bus voltage. See Figures 4.26 and 4.27 for the simulation results.
Figure 4.27 shows that the controllable HV reactor on the bus of Dunhuang Substation will act to output inductive reactive power, which will rise rapidly from zero to 225 Mvar to further suppress the rise of the bus voltage of Dunhuang Substation after the WTGs are disintegrated from the grid. Figure 4.26 illustrates the control effect—when it does not consider the action of the controllable HV reactor, the bus voltage of Dunhuang Substation is approximately 1 p.u. (800 kV) at the dynamic climax and approximately 0.99 p.u. (792.32 kV) for the stable voltage; and when the controllable HV reactor is considered, the bus voltage of Dunhuang Substation is approximately 0.994 p.u. (795.2 kV) at the dynamic climax and approximately 0.976 p.u. (780.8 kV) for the stable voltage. Obviously, with the control of the controllable HV reactor, the bus voltage of Dunhuang Substation will not rise up to the security limit (800 kV) and the stable voltage level is more reasonable after the WTGs are disintegrated from the grid.
4.3.2.1.2. Disintegration of the 750 kV line
The 750 kV line can be disintegrated to suppress the rapid rise of the bus voltage after the WTGs are disintegrated from the grid. The principle lies in that after the line is disintegrated, it will increase the equivalent impedance of the regional grid and further raise the reactive losses of the line so that the equivalent charge reactive power increased to the Hexi transmission channel by power flow reduction will become smaller, suppressing the rise of bus voltage.
Based on the above example for calculation of WTGs disintegration, one circuit of the Dunhuang-Jiuquan 750 kV double-circuit lines can be rapidly disintegrated simultaneously at the moment the WTGs of 956 MW are disintegrated from the grid. See Figures 4.28 and 4.29 for the dynamic variation of the bus voltage of Dunhuang Substation.
Figure 4.28 shows that the bus voltage of Dunhuang Substation does not see the dynamic variation process of rapid rise after the line is disintegrated; instead, it falls from the initial value 0.964 p.u. (771.2 kV) directly to the minimum point 0.924 p.u. (739.2 kV). After a transient transitional process, the stable voltage is 744 kV. Although the voltage is low after the measure to disintegrate the line is taken, it is not too low to affect the system security. Figure 4.29 shows that after one circuit of the Dunhuang-Jiuquan line is disintegrated, the power transmitted by the other circuit is 2370 MW, which has large margin compared with the thermal stability limit (5000 MW). As a result, disintegrating the 750 kV line can be used as a regulation and control approach to handle the impact on the bus voltage of the Hexi 750 kV channel after large-scale WTGs are disintegrated from the grid.
4.3.2.1.3. Regulation and control via the SVC in the wind farm
The wind farms in Jiuquan are provided with static var compensator (SVC) of certain capacity, which can be used to regulate and control the reactive exchange between the wind farm and the main grid in case of wind power output and the impact of the random fluctuation of the wind power output on the PCC voltage and even the voltage of the main grid. The SVC can be used to rapidly regulate and control the bus voltage rise caused by the WTGs disintegrated from the grid. Simulation is carried out to analyze the suppression effect of the SVC provided in the wind farms of Jiuquan on the bus voltage rise of Dunhuang Substation after the WTGs are disintegrated from the grid. See Figure 4.30 for the analysis results.
When SVC is used for regulation and control, the bus voltage maximum of Dunhuang Substation is 0.993 p.u. (794.4 kV), and the stable voltage is 0.983 p.u. (786.4 kV). Obviously, the SVC regulation and control can suppress to some degree the bus voltage rise caused by large-scale WTGs disintegrated from the grid.
4.3.2.2. Prevention and control measures for system frequency stability after large-scale WTG disintegrations
Based on the rapid response characteristics of the DC system, the DC modulation can be used as the control measurement to improve the system frequency stability after large-scale WTGs are disintegrated from the grid. Generally, it shall follow three principles:
1. It shall effectively prevent the system from falling below the setting of the low-frequency load-shedding protection in case of ordinary overloading.
2. After DC modulation, the grid frequency shall not have overmodulation on the local side and have no large fluctuation on the opposite side.
3. The DC modulation shall not act in case of normal fluctuation of system frequency.
Generally, the rapid reduction of power should not exceed 50% the DC power when the DC modulation is used. In addition, the DC power rapid reduction is different from DC single/double-pole blocking, and it does not disintegrate the filter during modulation, which may result in excessively high voltage in the near zone, and thus special attention shall be paid, especially in case of large modulation capacity.
The typical mode for characteristic analysis shall be used again. The application of various DC modulation schemes shall be studied in case large-scale WTGs are disintegrated from the grid. See Table 4.10 for the calculation results and Figure 4.31 for the corresponding curve.
Table 4.10 and Figure 4.31 show that after the large-scale WTGs are disintegrated, the system can maintain stable power angle and voltage but see large power shortage. In this case, if no actions are taken, the low-frequency load-shedding system of the Northwest Grid will act in the first round, which may result in 2800 MW load losses and the system stable frequency is 49.56 Hz.
Since the integration points of Lingbao DC, Debao DC, and Ningdong DC systems in the Northwest all fall in the Central China and North China Grids, which are interconnected via ultra-high voltage (UHV) AC systems, the DC modulation approach shall be carried out only on the Debao DC system to avoid excessively large power shortage on the opposite side, which may affect the safe and smooth system operation. The simulation results, however, show that if the power rapidly reduced keeps below 1500 MW, the low-frequency load-shedding system of the Northwest Grid will still act for the first round, which may result in 2800 MW load losses; and if the rapidly reduced DC is 1200 MW, the system frequency will reduce to the minimum degree 49.95 Hz; if the rapidly reduced DC is 1500 MW, the system stable frequency is 50 Hz and overmodulation may occur during frequency stable process; if the rapidly reduced DC is 1800 MW, the low-frequency load-shedding system will not act and the system stable frequency is 49.17 Hz and, in this case, it is difficult to meet the requirement on system stable frequency (no less than 49.5 Hz) after fault if no other actions are taken, and at the same time, the bus voltage of Baoji 750 kV Substation will reach 800 kV, failing to meet the requirement of safe and smooth operation.
If it does not consider the bearing capacity of the opposite grid and modulation is carried out simultaneously on Debao DC and Ningdong DC systems, the system will see good frequency characteristics. If the power rapidly reduced is 1200 MW in the Debao DC system and 1000 MW in the Ningdong DC system, the low-frequency load-shedding system will not act and the system stable frequency is 49.4 Hz; in this case, it is difficult to meet the requirement of system stable frequency (no less than 49.5 Hz) after fault if no other actions are taken. If the power rapidly reduced is 1500 MW in the Debao DC system and 1000 MW in the Ningdong DC system, the low-frequency load-shedding system will not act and the system stable frequency is 49.56 Hz, which can meet the requirement on system operation. Simultaneous modulation of two DC transmission lines, however, will result in large power shortages in the Central China and North China grids. As a result, detailed study shall be conducted for the stability of the receiving-end grid before the measure is taken.
Table 4.10
Frequency Stability after Large-Scale WTGs Are Disintegrated from the System
| Scheme | Parameter | |||
| Type of Fault | DC Modulation Scheme | Load Disintegrated by Low-Frequency Load-Shedding Protection Action | System Frequency Variation after Fault | |
| I | WTGs of 3000 MW tripped without fault | 49.5 Hz: DC modulation starts; 49.3 Hz: 1 s delay; DC power rapidly reduced in Debao: 1200 MW | Low-frequency load-shedding protection act to disintegrate 2800 MW loads in the first round | Minimum frequency: 48.98 Hz; stable frequency: 49.95 Hz |
| II | 49.5 Hz: DC modulation starts; 49.3 Hz: 1 s delay; DC power rapidly reduced in Debao: 1500 MW | Low-frequency load-shedding protection act to disintegrate 2800 MW loads in the first round | Minimum frequency: 48.99 Hz; stable frequency: 50 Hz | |
| III | 49.5 Hz: DC modulation starts; 49.3 Hz: 1 s delay; DC power rapidly reduced in Debao: 1800 MW | Not act | Minimum frequency: 49.12 Hz; stable frequency: 49.17 Hz | |
| IV | 49.5 Hz: DC modulation starts; 49.3 Hz: 1 s delay; DC power rapidly reduced in Debao: 1200 MW; DC power rapidly reduced in Ningdong: 1000 MW | Not act | Minimum frequency: 49.18 Hz; stable frequency: 49.4 Hz | |
| V | 49.5 Hz: DC modulation starts; 49.3 Hz: 1 s delay; DC power rapidly reduced in Debao: 1500 MW; DC power rapidly reduced in Ningdong: 1000 MW | Not act | Minimum frequency: 49.19 Hz; stable frequency: 49.53 Hz | |
Based on the DC modulation effect, device-bearing capacity, load losses of the Northwest Grid, and the impact on stability of the receiving-end grid etc., it is difficult to settle the system frequency stability problem in case of large-scale WTG disintegrations only by the DC modulation approach if the Northwest Grid has low spinning reserve.
Figure 4.31 System frequency variation curve in different DC modulation schemes after WTGs of 3000 MW are disintegrated. (a) System frequency curve after the low-frequency load-shedding system acts when WTGs of 3000 MW are disintegrated and Debao DC rapidly reduced power of 1200 MW; (b) System frequency curve after the low-frequency load-shedding system acts when WTGs of 3000 MW are disintegrated and Debao DC rapidly reduced power of 1500 MW; (c) System frequency curve after WTGs of 3000 MW are disintegrated and Debao DC rapidly reduced power of 1800 MW; (d) System frequency curve after WTGs of 3000 MW are disintegrated and Debao DC rapidly reduced power of 1200 MW and Ningdong DC rapidly reduced power of 1000 MW; (e) System frequency curve after WTGs of 3000 MW are disintegrated and Debao DC power rapidly reduced power of 1500 MW and Ningdong DC rapidly reduced power of 1000 MW.
4.4. FACTS-Based Automatic Voltage Control of Hexi Transmission Channel
Jiuquan Wind Power Base is developed with the feature of large-scale concentrated integrations, by which the power can be transmitted to the Northwest Grid hundreds of kilometers away via the EHV transmission lines. Since the wind power is intermittent and random, the power output by the wind power base may have significant and frequent fluctuations in a day. To maintain the voltage of the whole grid at a reasonable level, the voltage regulation devices in the Hexi Region and even in the whole Northwest Grid shall act frequently in a coordinated manner. The whole process for voltage-reactive power control (VQC) and regulation is complicated.
The existing voltage controls in the Gansu Grid consist of distributed local controls. It is impossible to coordinate and optimize from the overall view. The limitations are as follows:
1. The rate of qualified voltage is not high, and it cannot meet the high requirements of the user on power quality.
2. Too many voltage points shall be monitored in the grid, resulting in excessively huge daily burden for the dispatcher.
3. The reactive voltage has strong nonlinear relations and the voltage control equipment has various characteristics, resulting in huge difficulty for manual dispatch.
4. The unreasonable flow of the reactive power will affect the safe operation of the grid and result in large grid losses, which is adverse to economical operation of the grid.
As the Gansu Grid develops rapidly and the wind power bases of thousands of MW are built in Jiuquan, it is difficult for the existing voltage control mechanism to meet the requirements on safe, quality, and economical operation of the grid. The FACTS-based automatic voltage-reactive power control system shall be built in the Hexi transmission channel, which is the general development direction for grid operation dispatch after Jiuquan Wind Power Base is integrated. It is recommended to build the automatic voltage control (AVC) system for the Hexi transmission channel in two stages: open-loop operation and closed-loop operation with the aim to reduce the burden of the planning, dispatch, and operation personnel, improve the system capacity to meet the requirements on power quality, safe and economic operation of the grid, and raise the integrated decision, dispatch, and management level of voltage-reactive power controls in the Gansu Grid.
4.4.1. Application of Automatic Reactive Voltage Control Technologies
The voltage-reactive power control shall focus on the following: to reasonably arrange and make full use of the reactive compensation capacity and regulation capacity of the grid, avoid long-distance large-capacity transmission of reactive power, and keep the voltage of each pilot node in the grid at the normal level during normal operation and after fault, and make sure of safe and smooth grid operation.
The traditional reactive power and voltage is generally provided with distributed controls. In this case, the voltage control equipment (the generator, the capacitor bank, on-load tap changer (OLTC) transformer, etc.) can only acquire the local information and control the local voltage in an independent manner, featuring high speed and independence from the control center. The controllers are not coordinated, and they can only control the local reactive power and voltage in the upper/lower limits, which, however, may result in adverse impact on the reactive distribution and voltage level of the main grid. Corresponding to the distributed control is the centralized voltage control. Its advantage lie in that it can carry out optimal control over the reactive voltage from the whole grid. The control method, however, must acquire the voltage information of each point in the system, and, thus, it has high requirements for reactive power measurement accuracy and data communications, which are difficult to implement in practice.
The leveled voltage control, a compromise of the above two control methods, is a good voltage control method. The three-level voltage control mode has been used in France and some other European countries where the voltage control functions of the power system are divided by time and space and the control structure is leveled and hierarchical.
The third-level voltage controls, built at the top, can control the whole system. They shall be executed by the system control center with response time of only tens of minutes. The main functions are to coordinate the second-level control system and instruct the interference behavior of the dispatcher. In addition to security monitoring, the economic issue is the main concern of the control level. And the economical dispatch serves as the daily work of the control level. The tasks of the third-level voltage controls are, based on the system information, to determine the amplitude of the voltage at the pilot nodes that can meet the limits on grid security and make the system run economically.
The second-level control system, the intermediate level, can control a region. It shall be executed by the regional control center with response time from tens of seconds to several minutes. The goals of the controls are to ensure the bus voltage at the pilot node equal to the setting. If any deviation occurs on the bus voltage amplitude at the pilot node, the second-level controller will change the setting reference value of the first-level controller based on the preset control rule.
The first-level controls, built at the bottom, are installed in the power plant, and the user and the power supply point where the closed-loop controls of rapid response are generally used to control the local voltage with response time usually from one to several seconds.
The main advantage of the three-level voltage control mode lies in the “time-space decoupling.” The voltage-reactive power control zones are divided by “space,” and the pilot nodes in each region are controlled to realize control of the whole region. In this way, the reactive resource in the region can be fully used, avoiding large-scale flow of reactive power. The control levels are decoupled by time, which can ensure the control effect of reactive power and voltage. The mode of grid zone division and pilot node control, of course, has some disadvantages. As for the zone division, the coupling among the control zones cannot be too strong; otherwise, the second-level voltage control in one control zone may affect the adjacent control zones, and the second-level voltage control cannot achieve good control effect and, in serious cases, it may result in unstable voltage in the adjacent control zones. As for selection of pilot nodes, the pilot node must have short electrical distance with other nodes in the zone and sufficiently long electrical distance with the adjacent control zones with the aim to avoid unnecessary mutual impact among the zones. As a result, zone division and selection of pilot nodes are sensitive to the grid structure. In case the grid structure has large changes, the original zones and pilot nodes shall be recalculated and adjusted. In addition, the second-level voltage control of each control zone in the scheme is executed by the regional control center, which is inappropriate to the developing grid.
Previously, most of the voltage-reactive power control devices based on online operation in China are basically designed with local voltage-reactive power control as the goal. The control principle is based on the nine-zone diagram and can only ensure that the local reactive power and voltage are controlled within the upper/lower limits, which may exert adverse impact on reactive power distribution and voltage level of the main grid. In addition, these devices are not designed with network joint-commissioning function, and optimal control cannot be realized for the grid reactive power and voltage.
At present, the study of voltage-reactive power automatic control in the system range is at a stage of rapid development in China. Many provincial grids, including Fujian, Henan, Jiangsu, Anhui, Liaoning, Hebei, etc., have achieved optimal control of reactive power and voltage. Each system has its own advantages and disadvantages, but such problems as unpractical dynamic zones and low robustness of three-level controls generally exist, which shall be settled in the follow-up AVC studies. Nevertheless, the practice of the AVC systems in the provincial grids indicates that the application and promotion of AVC systems are good for improving the system voltage quality and safe and smooth operation, reducing the grid losses and the burden of the operators for frequent reactive power regulation.
4.4.2. Primary Framework of Automatic Voltage Control System for Hexi Transmission Channel
For the Northwest Main Grid, the Hexi region is a relatively independent transmitting-end system. Based on the concept of voltage-reactive power leveled control previously described, the Hexi transmission channel is deemed as the first/second-level controls to build an AVC system that covers the FACTS devices, the power plants, and the voltage-reactive power controls (VQC) of the major substations in the region. The AVC system is designed with grid security and qualified voltage as the main goals and economical operation as the auxiliary goal. It can monitor the bus voltage of the pilot nodes and the output status of the wind farms and traditional power plants, carry out reactive optimization calculations, and work out the optimized control objectives of each subsystem. The communication system and the network equipment shall transmit the optimized objectives to each control subsystem, which shall realize these objectives, achieving centralized decision, multilevel coordination, and leveled controls.
Based on the matured experience of AVC operation both at home and abroad, the overall design framework of the AVC system is given in Figure 4.32 for the Hexi transmission channel. In Figure 4.32, EMS refers to the energy management system.
The AVC system of the Hexi transmission channel shall mainly include the following parts:
1. As for the current system status, the AVC system of the Hexi transmission channel shall use the voltage prevention and control software to judge whether the stability margin is sufficient. If the margin is insufficient, it will call the prevention and control module to regulate; and if the margin is sufficient, it will enter the optimization control model.
2. The reactive power optimization software shall carry out periodic calculations based on the real-time grid information at an interval of 5 ∼ 15 min. If the voltage off-limit is present, it will call the voltage verification calculation model to make it qualified; and if no voltage off-limit is present, it will call the optimization control model to carry out optimization control with grid loss reduction as the objective.
3. The target voltage of the control points in the power plant shall be directly delivered by the EMS/SCADA system of the Hexi transmission channel to the local VQC devices, which will automatically set the generator-end target voltage or reactive power for each unit based on the target voltage delivered and the preset distribution principles, and the unit distributed control system or the automatic voltage regulator will automatically regulate the voltage.
4. The target voltage of the control points in the key 330 kV substations shall be directly delivered by the EMS/SCADA system of the Hexi transmission channel to the local VQC devices, which shall regulate the settings of the dynamic reactive compensators or control the tap positions of the transformer taps and enable or disable the LV capacitors and the LV reactors based on the target voltage delivered. The VQC devices can also receive commands from the upstream system and, based on them, control the equipment.
5. The adjustment of the controllable HV reactor of the 750 kV substation and enabling/disabling of the LV capacitor/LV reactor can be directly controlled by the command of the AVC system of the Hexi transmission channel.
The key technologies and difficulties of the AVC system lie in the accuracy of real-time data, calculations of security monitoring indexes, real-time reactive power optimization method, selection of pilot nodes, coordination of control strategies, static voltage prevention and control, etc. The most important technologies lie in the real-time reactive power optimization method and static voltage control. The static voltage prevention and control covers modal analysis, load margin index calculation method based on continuous load flow, fault screening and sorting method, etc.
4.4.3. Coordination Control of FACTS Equipment in Hexi Transmission Channel in Stable Status
To handle the frequent system adjustment caused by wind power integration, the associated voltage-reactive power automatic control system shall be built. The system, however, is very complicated, technically difficult, and relatively weak in reliability. It is very difficult to build a matured and complete AVC system and put it into operation in a short time. There is still a long way to go even just to build the first stage of the AVC system, i.e., the AVC system running in open loop. As a result, it is necessary to discuss how to coordinate the FACTS equipment and ensure normal grid operation without the AVC system.
4.4.3.1. Coordination and control factors of FACTS in stable status
The Hexi Region has many dynamic reactive compensators, which aim to improve system stability and settle the system voltage regulation problem caused by wind power fluctuation. It has two means to make the system voltage compliant with requirements: (1) to settle the problem of stable voltage regulation; (2) to maintain the system voltage in a reasonable range and make certain that the system is stable.
4.4.3.1.1. Coordination and control principles of FACTS equipment
1. The system stable voltage must conform to the relevant regulations in any mode.
2. The coordinated FACTS equipment shall have good adaptability to the variation of system modes, and the system control demands shall be to the maximum degree satisfied by regulating FACTS equipment with minimal enabling/disabling of LV capacitors/LV reactors.
3. The coordinated FACTS equipment shall be designed with adequate sensibility to offer timely response to variation of system operation status.
4. The coordinated FACTS equipment shall be designed with adequately slow response, and the associated equipment cannot move in case of system mode changes. The action of one device will make the system status change toward a certain target, which shall not result in interlocking reaction of other equipment, no matter if the reaction is positive (making the system status approach the same target, an effect similar to overmodulation) or negative (making the system status away from the set target, the action of the two devices are at odds).
4.4.3.1.2. Function positioning of FACTS equipment in stable voltage regulation
To regulate the system voltage fluctuation caused by wind power fluctuation, the dynamic reactive compensators installed on the LV side of the step-up substation of the wind farm, including thyristor controlled reactor (TCR), SVC, magnetically controlled reactor (MCR), and STATCOM (also known as SVG, Static Var Generator), or the controllable HV reactor installed on the 750 kV HV bus can be regulated. Table 4.11 and Figure 4.33 show the voltage variation of some nodes in the Hexi transmission channel when Jiuquan Wind Power Base has 3200 MW output in the summer maximum mode in the following two cases: the dynamic reactive compensator of the wind farm integrated to Dunhuang 750 kV Substation has 75 Mvar inductive reactive power in service; and the controllable HV reactor of Dunhuang 750 kV Substation has 75 Mvar inductive reactive power in service. The results show that when the same inductive reactive power is put into service, the controllable HV reactor has better effect on system voltage regulation than that of the dynamic reactive compensators. As a result, the system voltage shall be regulated mostly by the controllable HV reactor with the dynamic reactive compensator as the alternative.
4.4.3.2. Long-time robustness control strategy of FACTS equipment
The long-time robustness control strategy of FACTS equipment (“robustness strategy” for short) stresses that the control strategy of FACTS equipment shall have the maximum adaptability to variation of operation modes. Based on it, the following control strategies are established:
Table 4.11
System Voltage When the Inductive Reactive Power Generated by the Dynamic Reactive Compensator of the Wind Farm is Put into Service and When the Controllable HV Reactor of Dunhuang S/S is Put into Service
| Bus | Mode | ||||
| Reference | 75 Mvar inductive reactive power by the dynamic reactive compensator is put into service | 75 Mvar inductive reactive power by the controllable HV reactor is put into service | |||
| Voltage amplitude (kV) | Voltage amplitude (kV) | Voltage fluctuation (%) | Voltage amplitude (kV) | Voltage fluctuation (%) | |
| 750 kV side of Dunhuang S/S | 778.8 | 778.3 | −0.07 | 776.4 | −0.32 |
| 330 kV side of Dunhuang S/S | 351.6 | 351.3 | −0.09 | 350.9 | −0.21 |
| 750 kV side of Jiuquan S/S | 778.7 | 778.3 | −0.12 | 777.1 | −0.48 |
| 330 kV side of Jiuquan S/S | 351.5 | 351.4 | −0.03 | 351.1 | −0.12 |
| 330 kV side of Beidaqiao East S/S | 355.3 | 355 | −0.09 | 354.8 | −0.15 |
| 330 kV side of Beidaqiao West S/S | 352.6 | 352.3 | −0.09 | 352.00 | −0.18 |
| 330 kV side of Ganhekou West S/S | 355.3 | 354.9 | −0.12 | 354.8 | −0.15 |
| 330 kV side of Ganhekou East S/S | 355.1 | 354.6 | −0.15 | 354.5 | −0.18 |
| 330 kV side of Ganhekou North S/S | 354.5 | 354 | −0.15 | 353.9 | −0.18 |
| 330 kV side of Qiaowan S/S | 356.9 | 356.8 | −0.03 | 356.6 | −0.09 |
| 330 kV side of Yumen S/S | 355.9 | 355.8 | −0.03 | 355.6 | −0.09 |
| 330 kV side of Jiayuguan S/S | 351.1 | 351 | −0.03 | 350.70 | −0.12 |
| 330 kV side of Guazhou S/S | 347.7 | 347.4 | −0.09 | 347.10 | −0.18 |
Figure 4.33 System voltage variation when the inductive reactive power generated by the dynamic reactive compensator of the wind farm is put into service and when the controllable HV reactor of Dunhuang S/S is put into service.
2. For the dynamic reactive compensator controlling the 330 kV voltage, the control strategy is based on that it shall act when the voltage at the given node is 3 kV above/below the target voltage; and for the dynamic reactive compensator controlling the 110 kV voltage, the control strategy is based on that it shall act when the voltage at the given node is 2 kV above/below the target voltage.
3. For the traditional LV capacitor and LV reactor, the control strategy is based on that it shall act when the voltage at the 750 kV node is above 795 kV and below 750 kV. When the voltage is above the upper limit, it shall act in the order of “LV capacitor disabled first, and then LV reactor enabled”; and when the voltage is below the lower limit, it shall act in the order of “LV reactor disabled first, and then LV capacitor enabled.”
4. For initial arrangement of the traditional LV capacitor and LV reactor, it shall be based on that “the controllable HV reactors shall be all put into service and the LV capacitors and LV reactors shall be enabled and regulated to make the voltage of the 750 kV node possibly high in the precondition of below 790 kV when the wind power output is zero.”
The control strategy is clear and simple, and the action strategy of equipment has little to do with the system mode variation. Accordingly, it can be used for a long time, and it can reduce the burden of the grid operator and dispatcher. The disadvantage is that it has large system voltage fluctuation.
4.4.3.3. Short-time fine-control strategy of FACTS equipment
The short-time fine-control strategy of FACTS equipment (“fine strategy,” for short) stresses that the FACTS can be regulated to achieve voltage microregulation in the system voltage fluctuation range when the operation mode fluctuates in a certain range. The LV capacitor and LV reactor can be enabled or disabled to regulate the voltage when the system mode has large changes. Based on it, the following control strategies are established:
1. For the controllable HV reactor, the control strategy is based on that it shall act when the voltage at the given 750 kV node has the specified voltage deviation from the target voltage.
2. For the dynamic reactive compensator controlling the 330 kV voltage, the control strategy is based on that it shall act when the voltage at the given node has the specified voltage deviation from the target voltage; and for the dynamic reactive compensator controlling the 110 kV voltage, the control strategy is based on that it shall act when the voltage at the given node has the specified voltage deviation from the target voltage.
3. For the LV capacitor and LV reactor, the control strategy is based on that it shall act when the voltage at the 750 kV node is above 795 kV or below 750 kV; when the voltage is above the upper limit, it shall act in the order of “LV capacitor disabled first, and then LV reactor enabled”; and when the voltage is below the lower limit, it shall act in the order of “LV reactor disabled first, and then LV capacitor enabled.”
4. For initial arrangement of the LV capacitor and LV reactor, it shall be based on that “the controllable HV reactors shall be all put into service and the LV capacitors and LV reactors shall be regulated to make the voltage of the 750 kV node possibly high in the precondition of below 790 kV when the wind power output is zero.”
The control strategy has close relations to the grid operation mode, and thus it has good regulation effect only in a certain period. The grid dispatcher shall regulate the associated settings as demanded or at a certain interval. The advantage lies in that it can reduce the voltage fluctuation amplitude and the system operation status is relatively smooth in most cases.
4.4.3.4. Comparisons and analysis on control effect of the two strategies
Take the Gansu Grid as an example. When it runs in the summer maximum load mode in 2010, compare the actual effect of the two control strategies.
4.4.3.4.1. Parameter setting for long-time robust control strategy
Table 4.12 shows the enabling/disabling of the LV capacitor and LV reactor of the 750 kV substation and the arrangement of the controllable HV reactor taps after the summer maximum typical mode is adjusted. Table 4.13 shows the voltage settings of FACTS control nodes, and Figure 4.34 shows the voltage of the associated buses in Hexi.
4.4.3.4.2. Parameter setting for short-time fine-control strategy
The short-time fine-control strategy shall determine the voltage setting and voltage permissive deviation of the control node for the FACTS equipment (including the controllable HV reactor). These settings have close relations to the system operation mode, and they can be worked out in the following steps:
1. Determine the boundary conditions in the most probable system operation mode. The control objective of the fine strategy is to keep the system smooth in a certain period with regulation of FACTS equipment. This needs to master and control the possible conditions in an upcoming period. Suppose the grid dispatcher prepares the operation plan every 6 h. Based on the wind power prediction system or the historical experience, the overall output of Jiuquan Wind Power Base most probably falls in a range of 0 ∼ 1000 MW in the interval of 6 h, followed by 1000 ∼ 2000 MW and the range above 2000 MW has the minimum probability. As a result, 1000 MW wind power output can be used as one of the conditions for most probable system operation mode in the period.
Table 4.12
Enable/Disable of the LV Capacitor and LV Reactor of the 750 kV S/S and Arrangement of the Controllable HV Reactor Taps (Mvar)
| Dunhuang S/S | Jiuquan S/S | ||||
| LV capacitor | LV reactor | Controllable HV reactor | LV capacitor | LV reactor | Controllable HV reactor |
| 0 × 60 | 1 × 60 | 4 × 75 | 0 × 90 | 0 × 90 | 2 × 4 × 52.5 |
| Hexi S/S | Wusheng S/S | ||||
| LV capacitor | LV reactor | Controllable HV reactor | LV capacitor | LV reactor | |
| 0 × 90 | 0 × 90 | 2 × 4 × 52.5 | 0 × 120 | 0 × 120 | |
Table 4.13
Voltage Setting of FACTS Control Nodes (kV)
| Node | Beidaqiao East S/S | Beidaqiao West S/S | Ganhekou West S/S | Ganhekou East S/S | Ganhekou North S/S | Qiaowan S/S |
| Voltage setting | 357 | 356 | 357 | 357 | 357 | 358 |
| Node | Changma West S/S | Xiangyang S/S | Guazhou Huajing | Liuyuan S/S | Diwopu S/S | |
| Voltage setting | 358 | 119 | 119 | 119 | 119 |
Figure 4.34 Voltage of some buses in Hexi in the robust strategy, the summer maximum load mode, 2010, zero output of wind power.
Wind power output in a range of 0 ∼ 1000 MW has the maximum probability, which indicates that the system will run at light loads in many cases. When the wind power output is 1000 MW, the stable voltage of the major nodes in the Hexi region shall be regulated at low level to prevent excessively high voltage in case of light loads. For example, the 750 kV bus voltage of the 750 kV substations shall be regulated at about 760 kV; the controllable HV reactor shall be set at the third tap so that there are still two taps to be disabled when the wind power output rises, and there is still one tap to be enabled when the wind power output is reduced; and the dynamic reactive compensator shall be set based on zero reactive output.
2. Voltage setting calculations and traditional LV capacitor/LV reactor enable/disable setting. Based on the boundary conditions given in Step (1), the system mostly probable operation mode can be worked out by analysis, trial calculations, and adjustment, and the subsequent results will indicate the voltage settings of the FACTS equipment and the enable/disable arrangement of the LV capacitor/reactor.
3. Calculations of permissive voltage deviation. It is necessary to set certain permissive voltage deviation in order to ensure the smooth operation of the FACTS equipment. (1) When the control HV reactor is changed by one tap, the resultant node voltage variation shall be less than the permissive deviation of the node voltage control; (2) And when the dynamic reactive power compensator has some capacity change (e.g., 10%), the resultant node voltage variation shall be less than the permissive deviation of the node voltage control. For example, suppose the controlled target voltage of Dunhuang Substation is 765 kV with ±1 kV deviation permitted, and when the controllable HV reactor of Dunhuang Substation is changed by one tap, it results in the variation of 4 kV. In this case, the voltage of Dunhuang reduces to 763 kV, and the controllable HV reactor can be disabled by one tap to raise the voltage up to 767 kV, which exceeds the upper limit. And then the controllable HV reactor shall be enabled by one tap. In this way, it will generate vibration.
The voltage of the FACTS control nodes in Hexi Region (see Figure 4.35) is worked out in the following conditions: The five sets of LV reactors of Dunhuang 750 kV Substation are all enabled, and the controllable HV reactor is enabled by three taps, the four sets of LV reactors of Jiuquan 750 kV Substation are all enabled, the line controllable HV reactor is enabled by three taps, the controllable HV reactor of the Hexi 750 kV line is enabled by three taps, and the LV capacitor and LV reactor of Wusheng 750 kV Substation are not enabled. Based on this, the voltage settings of the FACTS control nodes can be obtained (see Table 4.14). In the course of mode arrangement, it actually has shown the enable/disable of the LV capacitors and the LV reactors in each 750 kV substation. See Table 4.15. In this case, all the dynamic reactive power compensators in Hexi Region are in a status of zero reactive output.
Table 4.16 shows the node voltage fluctuation resulting from up/down by one tap of the controllable HV reactors, and 10% capacity variation of the dynamic reactive power compensators in Hexi Region. Based on the reasonable margin, the precondition of system voltage and minimal times of control HV reactor actions, it is recommended to take permitted voltage deviation of ±5 kV for the controllable HV reactor in the three 750 kV substations, Dunhuang, Jiuquan, and Hexi; to take permitted voltage deviation of ±2 kV for the dynamic reactive power compensators on the 330 kV control bus side in the Hexi region; and to take permitted voltage deviation of ±1 kV for the dynamic reactive power compensators on the 110 kV control bus side in the Hexi region. It is further worth noting that the voltage fluctuation showed in Table 4.16 has some relations with the mode but they are not close. As a result, the permissive voltage deviations of the controllable HV reactor and the dynamic reactive power compensator given in Table 4.16 have good adaptability to various modes.
Figure 4.35 Node voltage of Hexi in the fine strategy, the summer maximum load mode, 2010, WTGs of 1000 MW.
Table 4.14
Voltage Setting of FACTS Control Nodes (kV)
| Node | Dunhuang S/S | Jiuquan S/S | Hexi S/S | Beidaqiao East S/S | Beidaqiao West S/S | Ganhekou West S/S | Ganhekou East S/S |
| Voltage setting | 779 | 781 | 771 | 355 | 353 | 355 | 355 |
| Node | Ganhekou North S/S | Qiaowan S/S | Changma West S/S | Xiangyang S/S | Guazhou Huajing S/S | Liuyuan S/S | Diwopu S/S |
| Voltage setting | 354 | 357 | 357 | 119 | 119 | 118 | 119 |
Table 4.15
750 kV Enable/Disable of LV Capacitors/LV Reactors and Taps of Controllable HV Reactors in 750 kV S/S (Mvar)
| Dunhuang S/S | Jiuquan S/S | ||||
| LV capacitor | LV reactor | Controllable HV reactor | LV capacitor | LV reactor | Controllable HV reactor |
| 0 × 60 | 5 × 60 | 3 × 75 | 0 × 90 | 4 × 90 | 2 × 3 × 52.5 |
| Hexi S/S | Wusheng S/S | ||||
| LV capacitor | LV reactor | Controllable HV reactor | LV capacitor | LV reactor | |
| 0 × 90 | 0 × 90 | 2 × 3 × 52.5 | 0 × 120 | 0 × 120 | |
Table 4.16
Node Voltage Fluctuation Resulting from Up/Down by One Tap of the Controllable HV Reactors, and 10% Capacity Variation of the Dynamic Reactive Power Compensators in Hexi Region
| Node | Initial Mode (kV) | Controllable HV Reactor of Dunhuang S/S | Controllable HV Reactor of Jiuquan S/S | Controllable HV Reactor of Hexi S/S | Capacitive Capacity of Dynamic Reactive Power Compensators in Hexi Region | ||||
| Down by One Tap | Up by One Tap | Down by One Tap | Up by One Tap | Down by One Tap | Up by One Tap | Up by 10% | Down by 10% | ||
| 750 kV side of Dunhuang S/S | 779.9 | 2.3 | −2.2 | 1.1 | −1.1 | 0.4 | −0.3 | 0.4 | −0.4 |
| 750 kV side of Jiuquan S/S | 781.3 | 1.4 | −1.3 | 1.9 | −1.8 | 0.6 | −0.5 | 0.3 | −0.3 |
| 750 kV side of Hexi S/S | 771.6 | 0.8 | −0.8 | 0.6 | −0.6 | 1.7 | −1.7 | 0.2 | −0.2 |
| Beidaqiao East S/S | 355.7 | 0.5 | −0.5 | 0.3 | −0.2 | 0.1 | 0 | 0.3 | −0.2 |
| Beidaqiao West S/S | 353.2 | 0.6 | −0.7 | 0.3 | −0.4 | 0 | −0.1 | 0.2 | −0.3 |
| Ganhekou West S/S | 355.7 | 0.5 | −0.5 | 0.3 | −0.2 | 0.1 | 0 | 0.3 | −0.3 |
| Ganhekou East S/S | 355.4 | 0.5 | −0.5 | 0.3 | −0.2 | 0.1 | −0.1 | 0.3 | −0.3 |
| Ganhekou North S/S | 354.9 | 0.5 | −0.6 | 0.3 | −0.3 | 0.1 | −0.1 | 0.3 | −0.3 |
| Qiaowan S/S | 357.3 | 0.3 | −0.4 | 0.2 | −0.3 | 0 | −0.1 | 0.3 | −0.4 |
| Changma West S/S | 357.3 | 0.2 | −0.2 | 0.2 | −0.1 | 0.1 | 0 | 0.4 | −0.3 |
| Diwopu S/S | 119.8 | 0 | −0.1 | 0 | −0.1 | 0 | 0 | 0.2 | −0.2 |
| Xiangyang S/S | 119.3 | 0.1 | −0.1 | 0.1 | 0 | 0 | 0 | 0.1 | −0.1 |
| Guazhou Huajing S/S | 119.4 | 0.1 | −0.1 | 0 | −0.1 | 0 | 0 | 0.1 | −0.2 |
| Liuyuan S/S | 118.7 | 0.1 | −0.1 | 0.1 | −0.1 | 0 | 0 | 0.1 | −0.1 |
4.4.3.4.3. Simulation of control effect of the two strategies
The voltage fluctuation and the actions of the associated LV capacity/LV reactor and the controllable HV reactor are studied for the robust and fine-control strategies in the summer maximum mode, 2010, with wind power output varying in a range 0 ∼ 3188 MW. The results are shown in Tables 4.17 and 4.18. For easy observation, Figures 4.36–4.42 show the node voltage fluctuation and the actions of the controllable HV reactor. In the fine strategy, the system voltage fluctuation sees smaller amplitude. Take the voltage fluctuation of the 750 kV bus in Jiuquan Substation as an example. When the wind power output varies in a range of 0 ∼ 2000 MW, the robust strategy sees a fluctuation amplitude of 21.7 kV, and the fine strategy sees a fluctuation amplitude of 8.5 kV. The fine regulation capacity is realized by the relatively frequent regulations of the controllable HV reactor. Figures 4.40–4.42 show that the times of the controllable HV reactor are larger than that of the robust strategy.
Table 4.17
Voltage Fluctuation and Reactive Power Compensator Actions in the Robust Strategy and the Summer Maximum Load Mode, 2010, in Case of Wind Power Output Variation
| Wind Power (MW) | 0 | 500 | 1000 | 1500 | 2000 | 2500 | 3188 | |
| 750 kV side of Dunhuang S/S (kV) | 787 | 785.4 | 782.2 | 777.4 | 771.2 | 764.5 | 760.7 | |
| 750 kV side of Jiuquan S/S (kV) | 790 | 787.4 | 782.9 | 776.5 | 768.3 | 760 | 759.6 | |
| 750 kV side of Hexi S/S (kV) | 777.7 | 775.1 | 770.6 | 763.9 | 756.7 | 753.6 | 750.2 | |
| 750 kV side of Wusheng S/S (kV) | 772.3 | 771 | 767.9 | 763.2 | 757.5 | 753.5 | 750.7 | |
| Beidaqiao East S/S (kV) | 358.6 | 358.3 | 357.6 | 356.3 | 354.6 | 352.6 | 350.8 | |
| Beidaqiao West S/S (kV) | 357.1 | 356.6 | 355.5 | 353.8 | 351.5 | 349 | 346.8 | |
| Ganhekou West S/S (kV) | 358.6 | 358.3 | 357.6 | 356.3 | 354.5 | 352.5 | 350.8 | |
| Ganhekou East S/S (kV) | 358.4 | 358.1 | 357.3 | 356 | 354.2 | 352.1 | 350.3 | |
| Ganhekou North S/S (kV) | 358.2 | 357.8 | 356.9 | 355.5 | 353.6 | 351.4 | 349.5 | |
| Qiaowan S/S (kV) | 359.7 | 359.6 | 359.1 | 358.1 | 356.7 | 355 | 353.6 | |
| Changma West S/S (kV) | 359.5 | 359.5 | 359.2 | 358.4 | 357.2 | 355.8 | 354.6 | |
| Diwopu S/S (kV) | 120.5 | 120.5 | 120.3 | 119.8 | 119.2 | 118.4 | 117.2 | |
| Xiangyang S/S (kV) | 119.7 | 119.8 | 119.7 | 119.3 | 118.8 | 118.2 | 117.2 | |
| Guazhou Huajing S/S (kV) | 119.8 | 119.8 | 119.7 | 119.4 | 118.9 | 118.2 | 117.3 | |
| Liuyuan S/S (kV) | 119.5 | 119.4 | 119.1 | 118.6 | 117.9 | 117 | 115.8 | |
| Table Continued | ||||||||
| Wind Power (MW) | 0 | 500 | 1000 | 1500 | 2000 | 2500 | 3188 | |
| Dunhuang S/S | Tap position of controllable HV reactor | 4 | 4 | 4 | 4 | 4 | 4 | 4 |
| Sets of LV capacitors | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
| Sets of LV reactors | 1 | 1 | 1 | 1 | 1 | 1 | 1 | |
| Jiuquan S/S | Tap position of controllable HV reactor | (4,4) | (4,4) | (4,4) | (4,4) | (4,4) | (4,4) | (1,1) |
| Sets of LV capacitors | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
| Sets of LV reactors | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
| Hexi S/S | Tap position of controllable HV reactor | (4,4) | (4,4) | (4,4) | (4,4) | (4,3) | (1,1) | (1,1) |
| Sets of LV capacitors | 0 | 0 | 0 | 0 | 0 | 0 | 3 | |
| Sets of LV reactors | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
| Wusheng S/S | Sets of LV capacitors | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Sets of LV reactors | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
Table 4.18
Voltage Fluctuation and Reactive Power Compensator Actions in the Fine Strategy and the Summer Maximum Mode, 2010, in Case of Wind Power Output Variation
| Wind Power (MW) | 0 | 500 | 1000 | 1500 | 2000 | 2500 | 3188 | |
| 750 kV side of Dunhuang S/S (kV) | 782.3 | 783.1 | 779.9 | 776.7 | 775.3 | 771.8 | 761.4 | |
| 750 kV side of Jiuquan S/S (kV) | 784.2 | 785.8 | 781.3 | 777.4 | 777.3 | 769.8 | 758.2 | |
| 750 kV side of Hexi S/S (kV) | 775.8 | 776 | 771.6 | 766.6 | 767.5 | 759 | 752.1 | |
| 750 kV side of Wusheng S/S (kV) | 771.2 | 771.4 | 768.8 | 765.5 | 765 | 758.8 | 753.1 | |
| Beidaqiao East S/S (kV) | 356.2 | 356.5 | 355.7 | 354.8 | 354.2 | 352.9 | 349.6 | |
| Beidaqiao West S/S (kV) | 354.1 | 354.3 | 353.2 | 351.9 | 351.1 | 349.5 | 345.4 | |
| Ganhekou West S/S (kV) | 356.2 | 356.5 | 355.7 | 354.8 | 354.2 | 352.9 | 349.6 | |
| Ganhekou East S/S (kV) | 356 | 356.2 | 355.4 | 354.5 | 353.8 | 352.5 | 349.2 | |
| Ganhekou North S/S (kV) | 355.6 | 355.8 | 354.9 | 353.9 | 353.2 | 351.8 | 348.2 | |
| Qiaowan S/S (kV) | 357.4 | 357.8 | 357.3 | 356.6 | 356.1 | 355 | 352.2 | |
| Changma West S/S (kV) | 357.2 | 357.7 | 357.3 | 356.8 | 356.5 | 355.3 | 352.9 | |
| Table Continued | ||||||||
| Wind Power (MW) | 0 | 500 | 1000 | 1500 | 2000 | 2500 | 3188 | |
| Diwopu S/S (kV) | 119.9 | 120 | 119.8 | 119.4 | 119 | 118.2 | 116.8 | |
| Xiangyang S/S (kV) | 119.2 | 119.4 | 119.3 | 119.1 | 118.8 | 118.2 | 117 | |
| Guangzhou Huajing S/S (kV) | 119.3 | 119.5 | 119.4 | 119.1 | 118.8 | 118.3 | 117.1 | |
| Liuyuan S/S (kV) | 118.9 | 119 | 118.7 | 118.3 | 117.8 | 117.1 | 115.5 | |
| Dunhuang S/S | Tap position of controllable HV reactor | 3 | 3 | 3 | 3 | 3 | 1 | 1 |
| Sets of LV capacitors | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
| Sets of LV reactors | 5 | 5 | 5 | 5 | 5 | 5 | 5 | |
| Jiuquan S/S | Tap position of controllable HV reactor | (4,4) | (3,3) | (3,3) | (3,2) | (1,1) | (1,1) | (1,1) |
| Sets of LV capacitors | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
| Sets of LV reactors | 4 | 4 | 4 | 4 | 4 | 4 | 4 | |
| Hexi S/S | Tap position of controllable HV reactor | (4,3) | (3,3) | (3,3) | (3,3) | (1,1) | (1,1) | (1,1) |
| Sets of LV capacitors | 0 | 0 | 0 | 0 | 0 | 0 | 4 | |
| Sets of LV reactors | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
| Wusheng S/S | Sets of LV capacitors | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
| Sets of LV reactors | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
When the wind power output has significant changes, e.g., from 0 MW up to 3188 MW, in order to make the system voltage qualified, the LV capacitors of Hexi shall be enabled in both of the two strategies, and one more set of LV capacitors shall be enabled in the fine strategy, i.e., one more enabling operation is needed. The reason is that the objective of the fine strategy is to achieve minimal system fluctuation with the wind power output varying in a certain range. When the wind power output has significant changes, since the FACTS equipment has used all its control capacity during the course, the system shall use additional enabling/disabling operations of the LV capacitors/LV reactors to regulate the voltage.
4.4.4. Coordination Control of FACTS Equipment of the Hexi Transmission Channel in the Transient Status
4.4.4.1. Impact of dynamic reactive power compensation parameter variation on system stability in transient status
4.4.4.1.1. Dynamic reactive power compensator model
See Figure 4.43 for the SVC model used in the simulation analysis.
In Figure 4.43, VT is the bus voltage to be controlled, Ts1 is the time constant of the filter, VERR is the input voltage deviation signal, VSCS is the auxiliary control signal, VREF is the reference voltage, VEMAX is the upper limit of the maximum voltage deviation, VEMIN AX is the lower limit of the maximum voltage deviation, TS2 is the first-order leading time constant, TS3 is the first-order lagging time constant, TS4 is the second-order leading time constant, TS5 is the second-order lagging time constant, KSVS is the continuously controller gain, KSD is the intermittently controlled gain, TS6 is the thyristor triggered delay, and DV is the voltage deviation. If DV = 0, DVLO = B′MAX/KSVS, DVHI = B′MIN/KSVS; and if DV > 0, DVLO = DV, DVHI = –DV. BR is the intermediate variable, BMAX is the maximum admittance, B′MAX is the continuously controlled maximum admittance, BMIN is the minimum admittance, B′MIN is the continuously controlled minimum admittance, and BSVS is the equivalent admittance.
See Figure 4.44 for the STATCOM model used in the simulation analysis.
In Figure 4.44, V is the bus voltage to be controlled, VREF is the reference voltage, VSCS is the auxiliary control signal, T1 is the time constant of the filter and the measuring circuit, T2 is the first-order leading time constant, T3 is the first-order lagging time constant, T4 is the second-order leading time constant, T5 is the second-order lagging time constant, VMAX is the upper limit of the voltage amplitude-limiting link, VMIN is the lower limit of the voltage amplitude-limiting link, KP is the amplification factor of the proportion link, KI is the amplification factor of the integration link, TP is the time constant of the proportion link, VSMAX and VSMIN are the upper/lower limits of the proportional-integral controller output, VT is the control voltage, KD is the slope of STATCOM V–I curve, XT is the equivalent reactance between STATCOM and the system, TS is the response delay of STATCOM, ICMAX is the maximum capacitive current, and ILMAX is the maximum inductive current.
4.4.4.1.2. Impact of SVC action delay variation on system stability
1. WTGs in constant voltage control. When the Hexi transmission channel is not provided with series compensation capacitors and controllable HV reactors, the Jiuquan-Hexi 750 kV double-circuit transmission line transmits about 3300 MW power, and the impact of the SVC action delay variation on system stability is shown in Table 4.19.
Figure 4.45 shows the associated 750 kV bus voltage variation in Jiuquan Substation. When a “three-permanent” fault occurs on the Jiuquan side of the Jiuquan-Hexi 750 kV line, the thermal generators in Jiuquan Region and Xinjiang will lose stability of the power angle with respect to the Northwest Grid. During operation, it should be noted to make the SVC action delay compliant with the requirement.
2. WTGs in constant power factor control. When the Hexi transmission channel is not provided with series compensation capacitors and controllable HV reactors, the Jiuquan-Hexi 750 kV double-circuit transmission line transmits about 2750 MW power, and the impact of the SVC action delay variation on system stability is shown in Table 4.20.
Figure 4.46 shows the associated 750 kV bus voltage variation in Jiuquan Substation. When the SVC action delay is larger than 0.11 s and a “three-permanent” fault occurs on the Jiuquan side of the Jiuquan-Hexi 750 kV line, the thermal generators in Jiuquan Region and Xinjiang will lose stability of the power angle with respect to the Northwest Grid. During operation, it should be noted to make the SVC action delay compliant with the requirement.
Table 4.19
Impact of SVC Action Delay Variation on System Stability
| Operating Mode | SVC Action Delay (s) | Limitation Fault | Stability |
| Summer maximum mode, not provided with series compensation capacitors and controllable HV reactors | 0.1 (calculated initial value) | “Three-permanent” fault on Jiuquan side of the Jiuquan-Hexi 750 kV transmission line | Stable |
| 0.11 (critical stable value) | Stable | ||
| 0.12 (critical out-of-stability value) | The thermal power units in Jiuquan Region and Xinjiang lose stability of the power angle with respect to the Northwest Main Grid. |
Table 4.20
Impact of SVC Action Delay Variation on System Stability
| Operating Mode | SVC Action Delay (s) | Limitation Fault | Stability |
| Summer maximum mode, not provided with series compensation capacitors and controllable HV reactors | 0.1 (calculated initial value) | “Three-permanent” fault on Jiuquan side of the Jiuquan-Hexi 750 kV transmission line | Stable |
| 0.11 (critical stable value) | Stable | ||
| 0.12 (critical out-of-stability value) | The voltage of Dunhuang and Jiuquan Substations are excessively low, and the thermal power units in Jiuquan Region and Xinjiang lose stability of the power angle with respect to the Northwest Main Grid. |
4.4.4.1.3. Impact of SVC gain variation on system stability
1. WTGs in constant voltage control. When the Hexi transmission channel is not provided with series compensation capacitors and controllable HV reactors, the Jiuquan-Hexi 750 kV double-circuit transmission line transmits about 3300 MW power, and the impact of the SVC gain variation on system stability is shown in Table 4.21.
Figure 4.47 shows the associated 750 kV bus voltage variation in Jiuquan Substation. When the SVC gain is less than 23 and a “three-permanent” fault occurs on the Jiuquan side of the Jiuquan-Hexi 750 kV line, the thermal generators in Jiuquan Region and Xinjiang will lose stability of the power angle with respect to the Northwest Grid. During operation, it should be noted to make the SVC gain compliant with the requirement.
Table 4.21
Impact of SVC Gain Variation on System Stability
| Operating Mode | SVC Gain | Limitation Fault | Stability |
| Summer maximum mode, not provided with series compensation capacitors and controllable HV reactors | 200 | “Three-permanent” fault on Jiuquan side of the Jiuquan-Hexi 750 kV transmission line | Stable |
| 50 (calculated initial value) | Stable | ||
| 23 (critical stable value) | Stable | ||
| 22 (critical out-of-stability value) | The thermal power units in Jiuquan Region and Xinjiang lose stability of the power angle with respect to the Northwest Main Grid. |
2. WTGs in constant power factor control. When the Hexi transmission channel is not provided with series compensation capacitors and controllable HV reactors, the Jiuquan-Hexi 750 kV double-circuit transmission line can transmit about 2750 MW power, and the impact of the SVC gain variation on system stability is shown in Table 4.22.
Table 4.22
Impact of SVC Gain Variation on System Stability
| Operating Mode | SVC Gain | Limitation Fault | Stability |
| Summer maximum load mode, not provided with series compensation capacitors and controllable HV reactors | 200 | “Three-permanent” fault on Jiuquan side of the Jiuquan-Hexi 750 kV transmission line | Stable |
| 50 (calculated initial value) | Stable | ||
| 44 (critical stable value) | Stable | ||
| 43 (critical out-of-stability value) | The voltage of Dunhuang and Jiuquan Substations are excessively low, and the thermal power units in Jiuquan Region and Xinjiang lose stability of the power angle with respect to the Northwest Main Grid. |
Figure 4.48 shows the associated 750 kV bus voltage variation in Jiuquan Substation. When the SVC gain is less than 44 and a “three-permanent” fault occurs on the Jiuquan side of the Jiuquan-Hexi 750 kV line, the thermal generators in Jiuquan Region and Xinjiang will lose stability of the power angle with respect to the Northwest Grid. During operation, it should be noted to make the SVC gain compliant with the requirement.
4.4.4.2. Impact of controllable HV reactor parameter variation on system stability in transient status
4.4.4.2.1. Basic principle and model of controllable HV reactors
The shunt reactor, mainly applied to the EHV grid of 330 kV or above and the grids with more cable lines, is designed to absorb the surplus capacitive reactive power in the grid. The controllable HV reactor is one of the effective approaches to settling the limited overvoltage and the reactive phase modulation and voltage regulation. When the controllable reactor is in operation, the reactive power can be regulated in a certain range, which can prevent to some extent excessively high voltage in case of low loads and excessively high voltage in case of high or large loads. In addition, it can regulate the capacity to the maximum at the moment of fault occurrence, limiting the power-frequency overvoltage resulting from the fault. Besides, after the controllable reactor is put into operation, the dynamic reactive power compensation can be realized based on its maximum regulation range, improving the system voltage characteristics. In the end, it can also suppress the system voltage or power vibration in various disturbances, improving the dynamic stability.
Based on the composition principles, the controllable HV reactors are mainly composed of two types: those based on magnetically control principle and those on the high-impedance transformer (HIT).
The HIT-based staged controllable reactor is developed on the basis of thyristor controlled transformer (TCT) SVC. The double windings are used, and the transformer impedance is designed as 100%. Based on it, three sets of reactors are integrated in series to the LV side of the transformer, and then the thyristor and circuit breaker (CB) are used to achieve staged regulation and control the inductive reactive power. The LV side of the high-impedance transformer is integrated in series with three sets of reactors and three sets of taps are integrated. Based on different combinations, four output capacities can be obtained. See Figure 4.49 for the single-phase structure of the 500 kV controllable reactor. The thyristor on the secondary side and the bypass circuit breaker can be integrated or disintegrated to control the capacity of the controllable reactor. The typical staged capacity is 25, 50, 75, and 100%.
For the controllable reactor based on the magnetic control principle, in the whole capacity regulation range, the thyristor control system can change the core saturation to change the capacity of the reactor only in the linear area where the core is saturated and the yoke is not saturated. As a result, it is generally known as a magnetically controlled reactor (MCR). It consists of two parts: the main body of the reactor and the control system. When the DC field current is zero, the core limb will be unsaturated in the whole power-frequency cycle. In this case, the reactor is at no-load status with minimum capacity; as the control current rises, the core limb will see growing saturation time and the reactor capacity will rise, too. When the core is all saturated in a power-frequency cycle, the reactor capacity will also reach the maximum value Qm. See Figure 4.50 for the operation principle of MCR. See Figure 4.51 for the model of the controllable HV reactor.
In Figure 4.51, Vt is input voltage, TR is the time constant in the voltage measurement link, Vref is the reference voltage, DB is the dead zone, K1 is the transfer function coefficient from DC to AC, K2 is the transfer function coefficient from DC to AC, DELT is time interval, DELI is the step of current variation, I2MAX is the amplitude-limiting maximum of the field reference current IDREF, I2MIN is the amplitude-limiting minimum of the field reference current IDREF, KP1 is the factor of the proportion link in PI control of DC side, KI1 is the factor of the integration link in PI control of DC side, K3 is transfer function coefficient on DC side, K4 is transfer function coefficient on DC side, KF1 is the feedback coefficient on DC side, TF1 is the feedback time constant on DC side, IDMAX is the amplitude-limiting maximum of the field reference current, IDMIN is the amplitude-limiting minimum of the field reference current, I1MAX is the amplitude-limiting maximum of PI control on DC side, I1MIN is the amplitude-limiting minimum of PI control on DC side, and ISCR is output current.
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.