8
Distributed Generation
Abstract
This chapter begins with a description of different distributed generation (DG) technologies. These are presented from the point of view of their working principle and their way of connecting to the network. An introduction is also given to the load flow problem in DG systems. The different types of nodes are defined and the problem equations are formulated. The network connection of DG involves a change in the control of the system and the quality of the electrical power. In some cases, this connection leads to a deterioration of the quality indices of electric power. In other cases, it may serve to improve these indices. So, this chapter includes a section in which the impact of DG on the quality of electric power is analyzed. The last section is devoted to present and analyze different control strategies of the inverters that are used as interface for the connection between DG and network.
Keywords
distributed generation
power flow
power quality
renewable energy
microgrid
smart grid
This chapter begins with a description of different distributed generation (DG) technologies. Thermal solar plants, wind power plants, cogeneration, photovoltaic (PV) plants, small hydrogenation, fuel cell, flywheel, superconducting coils, and supercapacitors are some of technologies currently most widely used, or it is expected that they may be commonly connected to electrical networks in the near future. These are presented from the point of view of their working principles and how they connect to the network.
In addition, an introduction is made to the load flow problem in DG systems. The different types of nodes are defined and the equations of the problem are formulated.
Connecting DG systems to the network of involves a change in the control of the system and the quality of the electrical power. In some cases, this connection leads to a deterioration of the quality indices of electric power. In other cases, it may serve to improve these indices. Therefore, this chapter includes a section, which analyzes the impact of DG on the quality of electric power.
The last section is devoted to presenting and analyzing different control strategies of the inverters that are used as interface for the connection between DG and network. The main target of these strategies is to deliver the power generated by the source to the network. However, they are also designed to allow these interface devices to improve the electrical power quality.
8.1. Introduction
Until some years ago electrical systems were based on a centralized model. It was characterized by a system with large power plants, with high power generators connected to the transmission network. And the transmission system is used to transport the energy generated from these large power plants to the consumers (Figure 8.1). Most of the times this transmission is made over long distances, therefore with high voltage values. In Europe, 400 kV is usual but in other places, as North America or China, the transmission voltage can reach 750 kV.
Figure 8.1 Conventional electric system.
Once the energy has been transported to nearby places of the consumption centers, it is necessary to distribute it through a greater number of lines, with shorter length and lower rated power. From a technical and economical point of view it is desirable to reduce the voltage levels. All these networks and substations together form what is called the distribution system. Traditionally, this distribution network did not have generators units connected, so it was considered as a passive network. In this condition the power flow is one-way, from the transmission to the distribution network.
Approximately in the 1990s, this conception of the electrical system begins to change. There was an increased interest in harnessing energy resources that require small power plants, usually located dispersedly. These resources are often related to renewable energies. Thus, the concept of DG, distributed energy, or dispersed generation is introduced (Figure 8.2). While it was called initially in one way or another, it seems that in recent years the term, distributed generation is the most used in the technical literature, falling in disuse the remaining two. The concept of DG is not defined in a formal and unique way [1]. In some countries, this definition is based on the rating of the power plant, or the voltage level to which it is connected, as it is usually associated with the technical documents used to specify the connection conditions and the operation of this type of generation.
Figure 8.2 Electric system with DG.
Despite there is no complete agreement on the definition of DG, considering the different proposals, the main features of this kind of generation can be established as follows:
• It is connected to the distribution network.
• It is usual that part of this generation is consumed by the same facility and the rest is exported to the distribution network (e.g., cogeneration).
• There is no unified planning of this generation and is not usually dispatched centrally.
• The power rating of these units used to be lower than 50 MW.
The DG units can be connected to the system at different voltage levels, from low to high voltage. Traditional systems were designed to feed the loads with power flows from the higher voltage levels to the lower ones, hence the distribution network was considered as a passive system. With the integration of DG, the distribution system is no longer passive and becomes active, where the power flow can be in the opposite direction to conventional systems [2]. These changes bring as a result, the overload of networks and transformers, an increased risk of overvoltages, and a deterioration of the power quality.
On the other hand, the installation of the backup units close to the demand center avoids the cost of transmitting the power and any associated transmission losses. DG is foreseen to fit well in the new restructured energy market with all stockholders enjoying major benefits:
• For electricity distribution companies, DG will increase transmission and distribution capacity and, therefore, limit market influence on the energy cost increase.
• For large or small electricity consumers, DG uses a locally available mix of prime fuel sources, thus decreasing the dependency on importing. It can also be used for emergency backup and, moreover, it can be considered as an income-generating vehicle if properly interconnected with a hosting grid.
DG is closely related to the use of renewable energies. Among them, energy from the Sun and the wind are the ones that are having a greater development. The introduction of this type of energy in the electricity system marks another change in the safely and efficient operation and control of the network. This change can be partially solved by microgrids, which are entities that coordinate distributed energy resources in a consistently more decentralized way, thereby reducing the control burden on the grid and allowing them to provide their full benefits [3].
Microgrids can be ac or dc depending on the type of load and generation type that is installed, as shown in Figure 8.3 [4]. Each microgrid can be efficiently connected to the public network through an interface that controls the power flow between them, in optimal quality conditions.
Figure 8.3 Microgrids.
Recently, the concept of “smart grid” has appeared [5]. It defines a type of self-managing network with dynamic optimization techniques that uses real-time measurements to minimize losses, to maintain the voltage levels, to increase security, and to improve the system management. The data collected by the smart grid and its subsystems will enable system operators to quickly identify the best strategy to ensure proper system operation.
8.2. Different Technologies for Distributed Generation
In the last years, a wide variety of small power plants that are connected to the distribution network have proliferated [6]. Some of them use technologies already developed to generate electricity as wind turbines, PV solar, small hydro, or cogeneration units. Other techniques are more recent as fuel cells, solar thermal, microturbines, biomass, and marine renewable technologies. On the other hand, new energy storage technologies are also being developed such as flywheel, batteries, and supercapacitors. This section is intended to give an overview of those technologies that are presently taking more impact.
8.2.1. Thermal Solar Plant
Energy from the Sun is the primary energy source for life in our planet. Solar energy is really the main renewable energy available. This energy reaches the earth in the form of electromagnetic waves. The amount of this radiation that reaches a given area depends on many factors, such as the season, humidity, air mass, time of the day, etc. Most of the energy incident on the surface is used for the warming cycle, the weather cycle, wind, and waves. A small fraction is used for the photosynthesis in plants and the rest is emitted into space. The energy provided by the Sun can be approximately equal to 10,000 times the energy needs of the world [7]. Mankind has always tried to take advantage of solar energy and today this approach may be the solution to the energy problem of our societies.
One way of using this type of energy is through thermoelectric plants. Such plants use solar energy to heat a boiler, in order to produce the steam needed to run a turbine generator. A range of different technologies are used depending on the way that the solar energy is concentrated onto a receiver in which a working fluid is housed.
The solar concentrating systems (concentrating solar power (CSP) system) consist of a set of lenses designed to focus the energy on a receptor that acts as a boiler to generate steam. The system usually includes a tracking control system to maximize its efficiency. There are different types of reflectors, but the most widely used are parabolic mirrors, as shown in Figure 8.4.
Figure 8.4 CSP system with parabolic mirror.
Another approach concentrates the solar energy on a tower in which the receiver is placed, Figure 8.5. The use of a large number of reflectors, called heliostats, is needed, whose amount will depend on the system capacity. Water/steam is used as the heat-transfer fluid. Most recent designs use molten nitrate salt.
Figure 8.5 CSP system of tower.
Another technology used is based on the use of a Fresnel lens system. This type of lens design allows the construction of large aperture lens and a short focal length without the weight and volume of material that is needed in a lens of conventional design. The compact linear Fresnel reflector (CLFR) uses Fresnel lenses located along a single axis to concentrate solar energy and generate steam, Figure 8.6. It has the great advantage that the mirrors are cheaper to manufacture than the parabolic ones.
Figure 8.6 Compact Linear Fresnel Reflector, CLFR.
All the technologies described use solar energy to heat a fluid and produce steam to power a steam turbine. The technology to generate electricity from steam is the same as that used in conventional thermal power plants.
Finally, another technology that uses the heat of the Sun, based on the Stirling engine is presented (Figure 8.7). This thermal machine operates by cyclic compression and expansion of air or other gas, called working fluid, at different temperatures such that a net conversion of heat energy to mechanical energy occurs. The alternator is coupled to the machine shaft, to transform the mechanical energy into electric energy.
Figure 8.7 Motor stirling.
8.2.2. Photovoltaic Plant
PV plants convert solar radiation into electrical energy directly. Today it is a sufficiently mature technology; since its beginnings, it has gone from being a way to feed small loads that operate in isolation in remote areas to PV plants of certain power that connect to the electricity distribution network.
The amount of energy produced by a PV module is directly proportional to the area of the module (the terms PV module and PV panel are used interchangeably).
There are many models of “PV cells or bank” for calculating the output power due to the numerous technologies currently available. One alternative equation to approximate the output power of PV panels is
(8.1)where G is the incident irradiance, Pmp is the maximum power output, Pmp, ref is the maximum power output under standard testing conditions, T is the temperature, Tref is the temperature for standard testing conditions reference (25°C), Gref = 1000 W/m2, and γ is the maximum power correction for temperature.
The rated power generated by a module is limited. The maximum output voltage is corresponding to open circuit, which is usually reduced with respect to the mains voltage. Therefore, to increase the voltage of the plant, several PV modules are connected in series, composing strings. Furthermore, when an increase in power plant is required, several strings are connected in parallel to form an array.
Figure 8.8 shows a general wiring diagram of a PV plant to be connected to a distribution network. The output of a panel string can be connected to the dc bus through a dc/dc converter in order to adjust to the proper voltage levels. A dc/ac inverter converts the dc voltage to the appropriate ac voltage level of the distribution network.
Figure 8.8 PV plant.
In this basic scheme, a circuit of maximum power point tracking (MPPT) can be added to make the PV modules work in the maximum power point, with energy storage elements placed at the MPPT circuit output that is connected to the dc/dc converter. At the output of the inverter it usually has an isolating transformer and a filter to remove the harmonic components. Figure 8.9 shows the hardware structure to connect a PV system to the network.
Figure 8.9 Grid connection scheme of a PV plant.
PV inverters normally operate with unity power factor, not generating reactive power, so they do not take part in the voltage control of the system.
8.2.3. Wind Power Plants
Wind energy is one of the most abundant resources and is the renewable technology that is growing most rapidly worldwide. The improvement in the turbines and the design of power converters has led to a significant drop in the cost of this power generation technology, which places it as the most competitive renewable energy from the economical point of view. This has caused it to become the second largest energy renewable resource behind the hydroelectrics.
A wind turbine operates by drawing kinetic energy from the wind passing through the rotor. The power output of a wind turbine is given by
(8.2)where CP is the power coefficient, a measure of the effectivity of the rotor aerodynamics; P is the power in W; V is the wind speed in m/s; A is the swept area of rotor disk in m2; and ρ is the air density (1.25 kg/m3).
The power depends on the wind speed cubed, so it is important to install such facilities in places with high annual average wind speeds. This requires a detailed study to establish the most suitable location.
The generator is coupled to the turbine, usually through a reduction gear. There are several configurations of these, depending on system speed [8].
• Constant speed wind turbine. This type of turbine is coupled via a multiplier to a squirrel cage induction generator (Figure 8.10a). The generator is connected directly to the network or through a soft starter. A capacitor bank is necessary in addition, to compensate the reactive power of the machine.
• Variable speed wind turbine with variable rotor resistance. These use a wound rotor induction generator (Figure 8.10b). As in the previous configuration a soft starter and the capacitor bank are included.
• Variable speed wind turbine with partial-scale frequency converter. A double-fed induction generator (Figure 8.10c) is used to generate the energy. The machine control is performed with power electronics. In this way, the turbine is operated at the maximum power point according to the wind conditions to control the active and reactive power of the machine.
• Variable speed wind turbine with full-scale frequency converter. It may include a synchronous or induction generator (Figure 8.10d). The network connection is made with ac/dc and dc/ac converters. The power control is performed with power electronics.
Figure 8.10 (a–d) Wind turbine architectures.
Variable speed turbines with permanent magnet synchronous generators are often used in small wind turbines. They do not usually include reduction gear, and their speed varies widely. The output is ac of variable frequency, so it includes a ac/dc converter and an inverter as coupling interface to the voltage and frequency of the distribution network.
Nowadays there is a growing interest in offshore wind farm projects. They have advantages over those projected on land: reduced visual impact, better wind conditions for speed and reduced turbulence, and low wind shear leading to lower towers. The main disadvantage is their high cost.
8.2.4. Cogeneration Plant
Cogeneration is the simultaneous production of electricity and heat for industrial or residential use. The electricity produced is often consumed by the same industrial plant, although the surplus energy can be exchanged with the public distribution system. On the other hand, the heat generated is used in industrial processes or heating in a local area or district. This makes possible to achieve an overall system performance up to 67%. This means a 35% less fuel consumption than at a power plant in which the steam generated is only used to move the turbine.
Although a number of configurations can be given, Figure 8.11 shows a simple topology for a steam turbine. The generator is generally synchronous, connected directly to the network.
Figure 8.11 Cogeneration scheme with steam turbine.
8.2.5. Fuel Cell
Hydrogen is an important source of renewable and clean energy, and there is plenty of it in the universe. Among its highlight qualities are that it is nontoxic, colorless, and odorless. Hydrogen can be used as a transportable energy; it is storable and can be developed in places where it is needed. When used there are no carbon emissions since it only produces water.
To utilize the energy of hydrogen, a fuel cell can be taken (Figure 8.12). Its operation is as follows: the hydrogen that arrives at the anode side is dissociated in protons and electrons. Protons are driven to the cathode through the membrane, but the electrons are forced to go through an external circuit (producing energy) because the membrane is electrically isolated. In the catalyst of the cathode, the oxygen molecules react with electrons (conducted through the external circuit) and protons to form water. In this case, the only waste is water vapor or liquid water.
Figure 8.12 Fuel cell scheme.
It is important to establish the fundamental differences between conventional batteries and fuel cells. Conventional batteries are energy storage devices, that is, the fuel is therein and it produces energy until it is consumed. However, in the fuel cell, the reagents are supplied as a continuous flow from the outside, allowing uninterrupted power generation.
The output voltage of a fuel cell is constant, so its connection to an ac network requires a dc/ac converter. Thus, for connecting the fuel cell to a power grid, a similar circuit to that used in PV plants is often used (Figure 8.9).
8.2.6. Small Hydro Generation
The generation of electrical energy from the potential energy of water is a relatively mature technology that has been developed in Europe and North America. Most of the large water resources have already been developed. So, in recent years the trend is to give value to the small power availabilities. From the point of view of technology, research continues on more efficient turbines that are also capable to exploit variable flow rates and heads.
The output power of a hydraulic turbine is given by
(8.3)where P is the output power (W), Q is the flow rate (m3/s), H is the effective head (m), η is the overall efficiency, ρ is the density of water (1000 kg/m3), and g is the acceleration due to gravity (m/s2).
The feasibility study of a small hydroelectric plant requires knowledge of the mean annual flow that can be used and the height of the waterfall. For the first point, the most useful is to express the resource as a flow duration curve, which shows the percentage time that a given flow is equaled or exceeded.
The small hydroelectric plant can use induction or synchronous generators although it tends to use squirrel cage induction generators for its robustness and simplicity. The network connection of such machines is performed similarly to that shown in Figure 8.10.
8.2.7. Energy Storage
A major problem is that renewable energies are not manageable, that is, electricity production does not always coincide with the moment that it is needed, so it is practically impossible to achieve generation–load balance with this type of energy source. Although today there are sufficiently accurate models that predict the environmental conditions and the amount of energy of some sort that can be generated within a day in advance, is necessary to have additional plants that are manageable. As an example, solar energy is available only during daylight and is reduced significantly on a cloudy day.
To mitigate this problem, energy storage is used. There is no energy storage process that has an acceptable efficiency and that can store energy in large enough quantities. In many cases, projects to achieve these aims are in the experimental stage. In the case of solar thermal energy, to increase production in times when there is no Sun, molten nitrate salts have been used. This allows the heat generated during the day to be stored, so that the generator can keep injecting energy during periods of time when the energy from the Sun is not available.
Another method of energy storage uses a flywheel. The flywheel is a metal disc, which starts spinning when a torque is applied, see Figure 8.13. Once it is rotating, it decelerates when subjected to a resisting torque. The equation of the stored kinetic energy is as follows
Figure 8.13 Flywheel scheme.
(8.4)where E is the energy stored; I, the moment of inertia that is a function of the rotating mass and its distance to the axis of rotation; and ω is the angular speed.
A motor/generator absorbs energy from the grid and uses it to push/brake the rotor mass to reach a certain angular velocity. Under ideal conditions, once the flywheel is rotating, its speed will remain constant until a resisting torque is applied. In practice, we look for the friction losses to be minimal. This is achieved by installing the wheel in vacuum-sealed containers and using magnetic bearings where there is no mechanical friction.
There are facilities, which use flywheels of various megawatts. They can provide an amount of energy in a relatively short time, so they can play an important role in the primary regulation of the power–frequency control.
Another method of energy storage is the use of superconducting coils. The energy stored in a coil is given by the expression
(8.5)where L is the inductance (H) of the coil and I is the circulating current (A).
A constant electrical current is made to flow through a superconducting coil. If the conductor has no energy losses, that energy is stored permanently until the circuit changes. The coil will behave as a current source. An electronic conditioning system (inverter/rectifier) will be needed to allow the power flow in both directions and to prevent overvoltages in the system (Figure 8.14).
Figure 8.14 Superconducting coil.
Another way of storing energy is the use of supercapacitors. These are devices that have an operation principle similar to a traditional condenser. However, their capacity and discharge current is much higher and therefore they can be used as storage devices in power systems. The energy stored is given by
(8.6)where C is the capacity (F) and V is the terminal voltage (V).
It should be noted that the capacity is not constant and depends on the voltage across its terminals. Supercapacitors have the ability to be loaded and unloaded in very short periods of time, in the order of seconds or less, which makes them particularly suitable for responding to supply interruptions of short duration. With superconducting coils, a power converter for the network connection is also necessary.
Batteries are other devices that are used to store energy. Batteries are electrochemical devices that convert electrical energy (in the form of direct or constant current) into chemical energy as the battery charges, and convert chemical energy back into electrical energy when it is discharged. In energy storage systems, only rechargeable batteries can be used. The cost, duration, efficiency, battery life, and energy that can provide per unit volume are some of the most important features to consider before selecting a type of battery.
Pumped-storage hydroelectricity facilities are another efficient way of storing energy. Here, water is pumped to a reservoir located at a certain height to store potential energy; which can be exploited later using the gravitational field of the Earth on the water to drive a turbine that is coupled to a generator to produce electricity. The main problem with this approach is that two reservoirs located at different elevations are usually required, which often implies high investment costs.
Finally, the energy storage facilities of compressed air (CAES) could be mentioned. They function like big batteries. Powerful electric compressors compress the air into an underground geological formation (abandoned mines, cavities filled with mineral solutions or aquifers) during the periods of time when the use of electricity is lower, for example at night. And when the maximum power is needed during periods of high demand, the precompressed air is used in modified combustion turbines to generate electricity. Natural gas or other fossil fuels are still needed to run the turbines, but the process is more efficient. This method uses up to 50% less natural gas than a normal system power production.
8.3. Power Flow Control of a Distributed Generation
The integration of renewable energy sources as a part of the grid must also be examined through the power flow analysis, to ensure acceptable values of voltages and currents for a certain generation and distribution of the electrical energy throughout the network. The load flow problem calculates the voltage in each of the nodes in steady state and balanced conditions, as well as the active and reactive power flows through the system elements [9], for a given set of generation and loads. Knowledge of these parameters is necessary for the safe and effective operation and planning of the electric system.
The classical methods of analysis, by mesh currents or nodal voltages, do not apply directly when the load data are given in terms of power. The problem statement will lead to a system of nonlinear equations for whose solution is necessary to apply iterative techniques
Let “i” be a generic node with a power source that supplies to the node a current Ii and a complex power Si, with Ui as node voltage, Figure 8.15 (hereinafter the complex quantities will be represented in bold). The complex power can be expressed as
Figure 8.15 Generic bus.
(8.7)In the load flow analysis only balanced three-phase systems are considered. Thus, any system can be represented by their single-phase equivalent. Applying node analysis to this equivalent circuit, for node “i” the equation can be expressed as
(8.8)where n is the number of nodes of the system, Yik for k≠i is the mutual admittance between nodes i and k with the sign changed, Yii is the self admittance of the node i, and Uk is the voltage at node k.
(8.9)In this expression, phasors and complex numbers can be represented in exponential form, that is
(8.10)where δ is the phase angle of the voltages and γ is the argument of the admittances.
Thus equation (8.9) can be rewritten in the form
(8.11)where δik is the phase difference of the node voltages i and k, that is, δik = δi−δk.
Equation (8.11) can be decomposed into the following two real equations
(8.12)
(8.13)In equation (8.12), Pi is the real part of the apparent power Si, that is, the incoming active power in node i. In equation (8.13), Qi is the imaginary part of Si, incoming reactive power in node i.
Each node has two equations and four variables associated with it: the magnitude of the voltage Ui, the phase angle of the voltage δi, the injected active power Pi, and the injected reactive power Qi. To solve the resulting system of equations, it is necessary to use iterative methods such as Newton–Raphson or Gauss–Seidel.
On the other hand, the nodes of the system can be classified according to three basic categories:
• Generator nodes or PV nodes. In this node, the active power P and the voltage magnitude U can be taken as system data as these two variables are easily adjusted at the power plant (Figure 8.16a).
• Load nodes or PQ nodes. They are the nodes where the consumption of active and reactive power are connected and where either no generation exists or it is fixed and is taken as given (Figure 8.16b).
• Swing node or slack node (Figure 8.16c). This is a generation node (usually designated as node 1) in which instead of the active power the argument of the voltage is specified and fixed at δ1 = 0. This node is used as reference for the arguments of the voltages for the rest of nodes. It uses to be the node of an important generator of the system. The function of a slack bus is to balance power consumption and power loss with the net-injected generated power.
Figure 8.16 Bus type: (a) P–V bus; (b) PQ or load bus; (c) swing or slack bus; and (d) P constant active power.
PVs, wind, or other types of generation coupled to the network through inverters [10], can be considered as a new node category, as these kind of power sources used to work with unity power factor, therefore producing only active power. So, when the unit is connected to a local power network, it can be specified according to Figure 8.16d.
Example 8.1
The Figure 8.17 shows the single line diagram of a four-node system with a rated voltage of 66 kV. The generator connected to node 1 is set to a voltage of 1.04 per unit. A PV generator is connected at node 3, whose inverter operates at unity power factor. Two loads are connected at nodes 2 and 4, whose complex apparent powers are those shown in Figure 8.17. The resistances and inductances of the distribution lines are indicated in Table 8.1. Determine the solution of the power flow when the PV generator injects 1.5 MW.
Table 8.1
Impedances Line, Example 8.1
| Line 1-2 | Line 2-3 | Line 3-4 | Line 4-1 | Line 2-4 | |
| Resistance (Ω) | 0.8 | 0.5 | 0.6 | 0.8 | 0.4 |
| Inductance (mH) | 5 | 4 | 3.5 | 5.5 | 3 |
To solve the load flow problem a tool included in MATLAB–Simulink has been used, specifically in the SimPowerSystems library. It is necessary to build the Simulink model of the network shown in Figure 8.17 to use this tool. Each of the nodes must be defined by type. Thus node 1 will be the swing node, nodes 2 and 4 are PQ type and node 3 is PV type.
The iterative method of Newton–Raphson is used, where a tolerance of 10-4 has been set. Table 8.2 presents the results obtained with the node voltages and the net active and reactive powers of each one. The base voltage is 66 kV. To solve the power flow, Simulink allows to the load nodes to choose between having constant active and reactive powers or to maintain a constant impedance. In this case, the constant power option has been selected.
Table 8.2
Load Flow Results, Example 8.1
| Bus ID | Type | U (pu) | U_angle (deg) | P (MW) | Q (Mvar) |
| Bus_1 | Swing | 1.04 | 0 | 99.43 | 31.93 |
| Bus_2 | PQ | 1.0261 | −0.818 | 40 | 10 |
| Bus_3 | PV | 1.0259 | −0.846 | 1.5 | 0 |
| Bus_4 | PQ | 1.0249 | −0.884 | 60 | 20 |
As most remarkable results, the active power losses of the system are 0.93 MW and 1.93 Mvar of reactive power. The powers defined in the statement remain constant. In node 1 the active and reactive powers are 99.43 MW and 31.93 Mvar.
8.4. Distributed Generation Impact in Power Quality
Inclusion of DG on the electrical system may result in a worsening of the electric power quality [11,12]. The connection of a large number of generators may have effects on the distribution network locally or globally, may even affect the transmission system [7].
The grid connection of DG is performed basically with three technologies [13]: synchronous generators, asynchronous generators, and power electronic inverters. The impact on the network depends on the technology used [14], although in general, DG can impact positively or negatively on the power quality in the following aspects:
• Voltage fluctuations. Such disturbances can be produced due to the variability of the power fed into the grid of some energy sources. Wind power generation may be a clear example, as the wind speed is usually not constant and rapid changes may appear even in windy conditions, so that the voltage at the connection point undergoes random variations, which may result in the phenomenon of flicker. This phenomenon is more evident in the case of weak networks.
• Harmonics. Some connected interfaces emit high frequency harmonics. However, the problem of harmonics is not referenced in the literature, perhaps because it has not yet had sufficient penetration of DG in the system.
• Voltage unbalances. Many small generators are usually single-phase and cannot be connected to a three-phase distribution network. That can lead to unbalances in the system voltage.
• Voltage dips. In general, it can be said that DG improves the behavior bus voltage where it is connected. Voltage dips are less severe, although this depends on the size of DG, improving when the power is greater.
The presence of DG does not always have a negative impact on the system [15]. It is in fact possible to use an intentionally DG to improve the power quality. On the other hand, it is clear that there is also an impact from the point of view of control and operation of the electric system that is not covered in this section.
The following subsections will discuss how DG affects each of these aspects.
8.4.1. Voltage Fluctuations
Voltage fluctuations can be produced by sources whose available power varies strongly with time [16]. Figure 8.18 shows the equivalent circuit when a DG is connected to the utility grid. According to this model the current can be obtained as
Figure 8.18 Equivalent circuit with DG connected.
(8.14)where I is the line current, UPCC is the voltage at the point of common coupling (PCC), P is the output active power of DG, and Q is the output reactive power of DG.
Voltage between the connection points of the DG the utility grid is the voltage in the Z impedance, which is given by
(8.15)where Ug is the voltage at the grid; R is the line resistance; and X, the line reactance. When (8.14) is replaced in (8.15), the result is
(8.16)As can be deduced, variations in the active power or reactive power injected into the network cause voltage fluctuations in the grid. DG produced by solar energy and wind are two clear examples of systems with output powers that vary randomly.
In the case of wind power, wind is unstable in nature, which causes fluctuations in the output power of wind turbines. When it comes to large turbines in a wind farm operating at the same time, significant impacts may occur in the system. The impact on the voltage level of the wind turbines depends on the generator type. For induction generator systems connected directly to the network, fixed speed systems depend primarily on the X/R relation of the network [17]. A relation of two or three times generates a low impact, provided that the system is equipped with capacitor banks to compensate the reactive power. When the wind turbine system includes power converters connected to the grid, the inverter control usually includes a loop to control the voltage level and reactive power. Moreover, it usually includes a smooth control to avoid switching transients that occur when capacitor banks are included.
Wind turbulence together with the turbine dynamic itself creates power variations in the region of 0.01–10 Hz. The use of converters provides the possibility of reducing the dynamic voltage fluctuations. This is because incoming power fluctuations can be offset by slight changes in rotor speed.
For PVs, solar radiation varies randomly, causing a change in the system power flow [18] and hence variations in the output voltage. These variations also depend on the functioning of the whole PV system, PV modules, inverter, filters, control mechanisms etc. Some studies have shown that short fluctuations in irradiance and cloud over play an important role in distribution networks of low voltage with high penetration of PV.
8.4.2. Harmonics
When PV systems, fuel cells, micro-turbines connect with the grid or supply power for ac load, they need an inverter device to carry out the dc-ac conversion. If a constant speed, constant frequency, asynchronous wind generator connects to grid directly, its soft start-up phase needs power electronic devices such as thyristor, and variable speed constant frequency wind generator systems also need a rectifier and an inverter device to connect with grid. Therefore, from a harmonic modeling and simulation standpoint, a distributed generator is usually a converter-inverter type unit and can therefore be treated as a nonlinear load injecting harmonics into the distribution feeder. Under the present framework of IEEE 519-2014, the supplier of electricity is responsible for the quality of the voltage supplied. The end-user is responsible for limiting harmonic current injections based on the size of the end-use load relative to the capacity of the system. Distributed resources are small relative to system capacity, but the smaller sizes are much more likely to achieve significant penetration levels for economic reasons.
The harmonic emission from DG is smaller than that emitted by modern loads, whereby the voltage distortion is rarely a problem [7]. By contrast, DG may cause increases in harmonic components that traditionally in a power system were small as the even harmonics, interharmonics and high frequency harmonics.
The injected current in the grid should not have a total harmonic distortion larger than 5%. A detailed image of the harmonic distortion with regard to each harmonic is given in Table 8.3 for six-pulse converters, according to IEEE Std 519 [19].
Table 8.3
Distortion Limits as Recommended in IEEE Std 519 for Six-Pulse Converters
| Odd Harmonics | Distortion Limit (%) |
| 3rd–9th | <4.0 |
| 11th–15th | <2.0 |
| 17th–21st | <1.5 |
| 23rd–33rd | <0.6 |
On the other hand, the IEC 61000-3-15 standard [20] establishes emission limits and immunity requirements for DG systems in low voltage.
8.4.3. Voltage Unbalances
In an electrical system the main sources of voltage unbalance are mainly due to transmission networks nontransposed or poorly transposed and connection of large single-phase loads. For the transmission networks, the system unbalance depends on the transport power. In the second case, the impact will depend heavily on how large installations with power unbalances are connected. In this sense, the IEC 61000-3-13 [21] standard establishes recommendations for connecting unbalanced network facilities.
DG can generate voltage imbalances in different ways. Single-phase units can produce an increase of these imbalances, although these units are limited to low power systems and connections in low voltage, so that the incidence is very low. In the case of many single-phase generator connected to the system, they have resulted in increased levels of voltage unbalance. Balance is very complicated when the system has a large number of generators, so there is a high probability that the system voltage is unbalanced.
In the case of three-phase generators, they limit the voltage imbalance provided that currents injected to the system are balanced. Generally the presence of these generators connected to the distribution system reduces the negative sequence components of the voltage.
8.4.4. Voltage Dips
DG does not affect the increase of voltage sags to the system directly, but there may be indirect effects as follows [22]:
• The replacement of large power plants by DG systems leads to a weakening of the network, resulting in increased voltage dips.
• The high penetration of DG requires new lines and strengthening others. This is most obvious in the case of large wind farms. This increase in the line number causes an increase in number of sags to the consumers connected to nearby lines.
• The weakening of the transmission system leads to an increase in the timing to remove faults. This affects less to the distance and differential protection but it does affect overcurrent protection that is sometimes used in these lines.
• After clearing the fault, rotating machines absorb a large reactive power. In weak networks, this causes decreasing voltage, which generates a voltage dip.
• When the network connection of DG is done through power converters, these can help maintain the load voltage locally.
More directly, the impact on voltage sags may be caused by wind generation due to its high penetration in the present electrical system. When a wind turbine is connected to the grid, high currents occur that may cause voltage dips [23]. This may be aggravated in the case of turbines that use a capacitor as compensation device, since the value of reactive power is proportional to the square of the output voltage. If at the moment of connection the value of the voltage is low, the current consumption is greater and thus the dip depth.
In the past, [16] wind generators were not allowed to stay connected when the voltage at the common connection point was below 85% of rated voltage, forcing disconnection [2]. This caused stability problems in the system. The present wind energy penetration into the electrical network has forced system operators to adapt their grid codes to this new generation, preventing an unacceptable effect on the system safety and reliability.
8.5. Distribution Line Compensation
Much of distributed generation uses converters as the interface for connection to the grid. The inverter control determines the operation and synchronization with the network. For injection of a given active power can use different control strategies, although all generally include two control loops [24–27]. A current loop that regulates the current injected to the network and a voltage control loop that controls the voltage at the dc-link.
The current control loop is responsible for the quality of the power and dc-link voltage controller is designed to control the power flow in the system.
Under normal operating conditions, DG operates at unity power factor. Mainly there are two control strategies that enable this objective:
• Instantaneous unity power factor control strategy. The aim of this strategy is to maintain constant active power and reactive power zero. By contrast, the network currents may be distorted.
• Positive sequence control strategy. In calculating the reference current, the positive sequence component of the fundamental harmonic of the voltage at the PCC is used. With this strategy injected currents are sinusoidal and balanced.
In addition to these strategies, some authors propose other aims to improve the above when a fault occurs in the system. In this network situation, oscillations appear in the active power and reactive power of double frequency of the mains frequency. These oscillations cause voltage oscillations at the dc side of the inverter, making it more difficult to control network in such conditions [28], leading to imbalance in the network current.
8.5.1. Instantaneous Unity Power Factor Control Strategy
Figure 8.19 shows a DG system connected to the network through three-phase dc/ac converter, which injects a power P at the PCC. In a three-phase system as shown, an instantaneous equivalent conductance can be defined by
Figure 8.19 System with DG and utility grid.
(8.17)where |v|2 is instant norm module of the phase voltage vector at the inverter output.
The instantaneous current vector, i* can be obtained from the product of the equivalent conductance g by voltage vector v,
(8.18)This vector current allows injecting a P power of DG at the PCC, to the grid voltage whose phase values are the vector components, v.
Current vector components, i* form a balanced and sinusoidal system when the network voltage vector is balanced and sinusoidal too. If the voltage vector is unbalanced, the current vector is unbalanced. If any of the components of v has any harmonic, the current system will also be distorted.
Moreover, the current vector, i* is proportional to the voltage vector; therefore, it has no orthogonal component with respect to the mains voltage; hence, there is no reactive power transfer between the network and DG.
Example 8.2
A DG system is connected to a distribution network via a dc/ac converter, with a configuration as shown in Figure 8.19. The system injects an active power of 15 kW with unity power factor. The supply voltage is 400 V and 50 Hz frequency. At the PCC, a load has been connected. It transfers an active power of 1 MW and a reactive power of 0.75 Mvar to the rated voltage.
The instantaneous unity power factor control strategy is applied to the converter. Its performance will be verified in two different situations voltage:
• Sinusoidal, balanced voltage system.
• Unbalanced network voltage system with a negative sequence component of 20% and nonsinusoidal with a 5th order harmonic of 15%.
To verify the operation of this strategy, the system is simulated in MATLAB–Simulink. The DG system is connected to the network by means of an LC filter formed by an inductance of 50 mH and a capacity of 0.7 μF.
The distribution network is modeled by means of a programmable three-phase source of infinite power. This source allows include imbalances and harmonics in the voltage.
The dc-link controller has been omitted in order to not influence the creation of the current references. As a consequence, dc power sources are used to supply the necessary voltage in dc-link. The dc-link voltage is set to 1500 V.
The control strategy is implemented according to the block diagram shown in Figure 8.20, which is the development of equation (8.18). The input signals to the control scheme are the phase voltage vector, v, and active power of DG, P.
Figure 8.20 Instantaneous unity power factor control strategy, block diagram.
In the first case, the voltage is sinusoidal with an rms value of 400 V and 50 Hz. Figure 8.21a shows the waveform of the system voltage in the PCC. The rms value of the phase voltage is 230.9 V. The currents injected by the DG system have the waveform shown in Figure 8.21b. As noted, it is a balanced and sinusoidal system. The rms value of the current is 21.5 A.
Figure 8.21 Instantaneous unity power factor control strategy. Balance and sinusoidal voltage: (a) voltage grid and (b) grid current.
With respect to the power balance, DG system provides 14,884 W and distribution network 985,116 W, thus the active power consumed by the load is 1 MW.
In the second case, the voltage of the distribution network has a negative sequence component of 20% and a 5th order harmonic of 15%. The three-phase source that model the distribution network is programmed with these values. The voltages waveforms are shown in Figure 8.22a. The voltage rms values of each phase are 279, 214, and 214 V.
Figure 8.22 Instantaneous unity power factor control strategy. Unbalance and nonsinusoidal voltage: (a) voltage grid and (b) grid current.
Current waveforms injected into the system by the DG system are shown in Figure 8.22b. It is noted that these currents are not balanced and sinusoidal. Current rms values of each phase are: 23, 20, and 21 A.
Respect to powers, in this situation DG supplies 14,657 W, distribution network supplies 1,027,676 W, and load consumes 1,042,333 W.
8.5.2. Positive Sequence Control Strategy
In situations where the voltage vector is in balance and/or presents harmonics, the above strategy, instantaneous unity power factor, can cause a deterioration in the waveform of the current and thereby, loss of power quality. In this situation, for the circuit of Figure 8.19, an instantaneous equivalent conductance can be defined by means of
(8.19)In this equation, 
represents the magnitude of the instantaneous norm of the positive sequence voltage vector at the fundamental frequency [29]. The reference current vector is given by
represents the magnitude of the instantaneous norm of the positive sequence voltage vector at the fundamental frequency [29]. The reference current vector is given by
(8.20)This strategy requires an algorithm for detecting positive sequence components of the voltage.
When the voltage is balanced and sinusoidal, instantaneous active power coincides with the active power, P, and reactive power is zero. When the voltage is unbalanced the instantaneous active power is different from the power P.
When this strategy is applied, the instantaneous active power can be expressed in terms of the positive and negative sequence components by means of
(8.21)The first term in equation (8.21) is the power P and the second is 
, which represents a power oscillation of double frequency of the fundamental.
, which represents a power oscillation of double frequency of the fundamental.Similarly, the instantaneous reactive power can be calculated as
(8.22)In this case, the first term is reactive power, Q with a zero value and the second is an oscillatory term, 
, double frequency of the fundamental that appears in the reactive power injected into the network.
, double frequency of the fundamental that appears in the reactive power injected into the network.Example 8.3
Verification of the performance of positive sequence control strategy for the DG system is described in Example 8.2.
Figure 8.23 shows the block diagram of the control circuit designed. The positive sequence component of the phase voltage vector, v is obtained using the block va+. This component is calculated by means of
Figure 8.23 Positive sequence control strategy, block diagram.
The operator a can be obtained from two all pass filters connected in cascade and a2 inverting input signal in the all pass filter. For the design of this filter, it is used as a Simulink block that generates the transfer function
Figure 8.24 shows the block diagram that calculates the positive sequence component.
Figure 8.24 Block diagram to determine the positive sequence component.
The direct sequence component of the fundamental harmonic is obtained with the block 
. This block is similar to that used in Chapter 5 when the fundamental component is determined. After obtaining the fundamental harmonic of the positive sequence component of the phase a, the b and c phase components are determined. In Figure 8.23, this calculation is performed by the block 
. Here, the inverse transformation of Fortescue is applied, this is
. This block is similar to that used in Chapter 5 when the fundamental component is determined. After obtaining the fundamental harmonic of the positive sequence component of the phase a, the b and c phase components are determined. In Figure 8.23, this calculation is performed by the block . Here, the inverse transformation of Fortescue is applied, this is
Figure 8.25 Fortescue inverse transformation.
The rest of the diagram in Figure 8.23 is the same as that of Example 8.2.
When the voltage system is balanced and sinusoidal, positive sequence control strategy and instantaneous unity power factor control strategy show a similar behavior.
In the case of nonsinusoidal and unbalanced voltage system, network currents in DG have a sinusoidal waveform, as shown in Figure 8.26. For each of the phases, the current rms values are 21.15, 21.2, and 21.2 A. Thus, these form a balanced three-phase system.
Figure 8.26 Positive sequence control strategy. Unbalance and nonsinusoidal voltage: (a) grid voltage and (b) grid current.
Finally, DG supplies to the grid an active power of 14,647 W, the distribution network supplies 1,027,686 W, and the load consumes 1,042,333 W.
8.6. Power Quality Improvement in Distributed Environment
The increased penetration of DG can generate problems in power quality if appropriate measures are not taken. However, if the appropriate control strategies are taken, DG can help improve that quality. In this section, different strategies to achieve those improvements are analyzed.
In the distribution network, the voltage at different points differs from its rated value. The voltage profile of the network can be improved by installing DG in the appropriate places [30]. This takes system planning, so that the optimal locations and proper size of DG are chosen.
Figure 8.27a shows a network that feeds a set of loads distributed throughout the network. Usual voltages profile in an actual facility is also shown. Voltage in the feeding point nearest to the loads is greater than that of the most distant loads. When a DG system is connected to the network modifies the voltage profile with an increase in the tension of the farthest loads (Figure 8.27b). Connecting these units at different points in the network can generate a flat profile, which would be ideal from the point of view of system operation. The voltages profile obtained depends on the situation of DG and its size.
Figure 8.27 Voltage profile of an electric line. (a) Without DG and (b) with DG.
The presence of current harmonics in the network due to nonlinear loads can be mitigated by DG systems whose connection interface is an inverter. For this, it is necessary to include in its current control strategy a new control loop that allows compensating harmonics be injected [31,32]. In this case, the DG system inverter has a dual function, to inject active power and to act as an active filter to compensate harmonics.
The strategies used in this new control loop are varied, thus, in [33] the inverter operates as an active inductor to absorb harmonic currents. Another method is that the inverter acts as an active conductance to dampen the harmonics of the distribution network [34]. In Ref. [35] a control strategy is presented for renewable energy interface based on p–q theory.
In this work, a strategy that integrates two control loops is described. On the one hand, a control loop based on positive sequence control strategy to inject the active power, and on the other hand, another control loop is added to compensate the reactive power and harmonics transferred to the set of loads. For this, instantaneous active power vectorial theory, which has been described in previous chapters is applied. According to this theory, the compensation current is given by the expression
(8.23)where PL is the transferred power to the load, iL is the load current vector. The current vector, iC, is the reference current of the control loop that compensates reactive power and current harmonics.
DG transfers the active power, PDG, when the inverter generates an instantaneous current, iP, defined according to (8.20). The sum of iP and iC is the reference current for the inverter control, this is shown in Figure 8.28.
(8.24)The above equation can be rewritten in the form
(8.25)Figure 8.28 shows the block diagram that implements the equation (8.25). The blocks are the same as described in Example 8.3 with the exception of block “LFP” consisting of a low-pass filter to obtain the average value of the output voltage by load current, resulting in active power transferred to the load, PL.
Example 8.4
Figure 8.29 shows a power system with DG connected to the distribution network of 400 V and 50 Hz. In a system bus, a linear load of 30 kW active power and reactive power of 20 kvar and a nonlinear load are connected. The nonlinear load is formed by uncontrolled rectifier with a resistor of 15 Ω connected in series with an inductance of 100 mH at the dc side.
DG should inject an active power of 10 kW, compensate reactive power of all loads and mitigate current harmonics of the distribution network.
For the rated voltage, the set of linear and nonlinear loads transfer an active power of 46,812 W and reactive power of 22,413 var. The load power factor is 0.74 inductive. The current presents a waveform clearly nonsinusoidal, as shown in Figure 8.30a. Its rms value is 75.2 A and has a THD of 7.89%. The most significant harmonics are of order 5 and 7. When the DG system is not connected, this will be the waveform and power transferred from the distribution network.
The DG system inverter is controlled with a strategy based on positive sequence strategy and instantaneous active power vectorial theory. For this, the block diagram shown in Figure 8.28 is implemented in MATLAB–Simulink. When the DG system is connected, the waveform of the current supplied by the distribution network is shown in Figure 8.30b. The rms value of this current is 54.0 A and has a THD of 2.26%. From the viewpoint of the quality of power is a significant improvement in the current waveform as there is a large decrease THD. Furthermore, regarding the power balance, the active power transferred by the distribution network is 37,377 W with a power factor of 1.00.
Respect to the power provided by the DG, it is 9435 W. This value is somewhat lower than the expected power of 10,000 W. This is due to losses occurring in the inverter output LC filter, since the resistive effect of the inductance has been considered. Regarding the current waveform generated by the DG, it is shown in Figure 8.31, the rms value of the current is 35.8 A.
On the other hand, microgrids are proposed to isolate the problems associated with DG. Different solutions have been proposed to mitigate these problems of lack of quality in microgrids. Some of these are based on improving the functioning of the converters that connect equipment of DG to the network [36]. Other proposals are based on the use of a power quality compensator [37,38]. When the microgrid is connected to the public network, it should not cause disturbances to the rest of the system, so that a connection interface between the microgrid and the main grid is necessary to avoid these issues. These interfaces are based on a configuration of series-parallel active power filter [39].
The interface consists of an inverter connected in series with the system between main grid and microgrid. Figure 8.32 shows the general scheme, where the main network is represented by its Thevenin equivalent and the microgrid by a set of source and nonlinear load.
Figure 8.32 Scheme of interface between utility grid and microgrid.
The series APF consists of an inverter bridge of IGBTs. At the ac side a ripple filter is connected, which allow the high frequency components to be eliminated. Connection to the system is done with coupling transformers.
The objective of the proposed control strategy is to avoid transferring voltage unbalances and current harmonics from the main grid to microgrid or from microgrid to main grid [40]. Thus, the inverter must generate a compensation voltage such that at the network side, voltage unbalances are not transmitted to the microgrid and vice versa. Furthermore, the generated harmonic currents should not be transferred between two networks. Therefore, two control loops will be designed: one to compensate voltage unbalances and another to mitigate current harmonics.
Respect to loop control to compensate voltage unbalances, the direct sequence component of voltage vector of the “a” phase can be calculated by means of the following expression
(8.26)where va, vb, and vc are the components of the voltage vector and a, operator is defined as 
, which supposes a 120° phase shift. This operator can be implemented by an all pass filter [41].
, which supposes a 120° phase shift. This operator can be implemented by an all pass filter [41].Once (8.26) is applied, the Fortescue inverse transformation allows voltage vector of direct sequence component to be obtained. It is calculated by means of the expression
(8.27)where, va+, vb+, and vc+ are the components of the direct sequence voltage vector for each phase.
These components are obtained at the main network side (vPC) and at microgrid side (vm). So, the compensation voltage vector (vc) will be calculated by
(8.28)To mitigate the current harmonics, the inverter is controlled to present zero impedance at the fundamental frequency and high impedance at the frequencies of the load harmonics. For it, the compensation voltage will be proportional to the current harmonics.
(8.29)where k is the proportionality constant.
Figure 8.33 shows the single-phase equivalent circuit of Figure 8.32 for a harmonic of h order. Here, the main grid and the microgrid are represented by its Thevenin equivalent and the inverter by a controlled voltage source of vC value. By applying Kirchhoff’s voltage law and the equation (8.29), the current is given by
(8.30)According (8.30), the h order harmonic of the current can be mitigated when k has a high value, ideally k should be infinite, however, in the practice it is enough when k > > ZPCh + Zmh to the most significant harmonics.
Example 8.5
Figure 8.34 shows a serial interface connection between a microgrid and DG system. The values of the passive components are shown in Table 8.4. The main grid impedance is considered less than microgrid impedance. This is intended to consider the microgrid weaker than the main grid.
Figure 8.34 Simulation circuit for the series interface.
Table 8.4
Values Passive Elements
| Main grid | RPC = 1 Ω | LPC = 1 mH |
| Microgrid | Rm = 3 Ω | Lm = 3 mH |
| Ripple filter | Crf = 50 μF | Lrf = 10 mH |
The proposed control strategy will be applied and the system will be subjected to disturbances at the network side and microgrid side.
To verify the performance the system will be subjected to two different situations. In one, the power flow is from the main grid to the microgrid, and in the other, the microgrid is what gives power to the main grid. So, it will be subjected two different network situations:
1. The voltages at the main network side have an unbalance of 25% and further have a 5th order harmonic of 12% of the fundamental harmonic. In this case, the power flow is from the main grid to microgrid.
2. The voltages at the microgrid side have an unbalance of 25% and contain a 5th order harmonic of 12% of the fundamental harmonic. In this test, power flow is considered from the microgrid to the main grid.
In the first test, the main grid voltage contains a 5th order harmonic of 12% and a negative sequence component of 25%. Figure 8.35 shows the voltage waveforms at the PCC and the current before connecting the inverter.
Figure 8.35 Test 1, waveform voltage and current, before the interface is connected. (a) Voltage vPC and (b) current.
When the interface is connected, the voltage waveform at the main grid side and at the microgrid side shown in Figure 8.36 is obtained. The proposed control strategy was applied to the inverter. The proportionality constant defined in (8.29) was fixed in 50. The unbalance factor was calculated to be 5.2%. This represents a significant reduction of voltage unbalance but does not eliminate it completely. The remainder is due to the current system being unbalanced, resulting in different voltage drops across the equivalent impedances in each phase. These current unbalances cannot be compensated with a series topology, so this slight unbalance voltage due to the current unbalance will always remain.
Figure 8.36 Test 1, waveform voltage and current, after the interface is connected. (a) Voltage vPC, (b) voltage vm, and (c) current.
On the other hand, taking into account the “a” phase, the current THD is reduced from 24.6% to 4.9%. This reduction is even higher in the other phases. Therefore, it is possible to state that the interface “isolates” the microgrid from the main grid harmonics.
A similar analysis can be performed with respect to the harmonics of the voltage. The results shown in Figure 8.36a and b illustrate how the interface “isolates” the microgrid of the voltage harmonics from the main grid.
In the second test, the voltages at the microgrid side have an unbalance of 25% and contain a 5th order harmonic of 12% of the fundamental harmonic. At the main grid side, the voltages are balanced and sinusoidal.
Figure 8.37 shows the waveform before connection to the interface. Voltages and currents present a waveform that is not sinusoidal, with the 5th harmonic being the most significant.
Figure 8.37 Test 2, waveform voltage and current, before the interface is connected. (a) Voltage vPC and (b) current.
After, the interface is connected to the system with the control strategy proposed. The proportionality constant was fixed at 50. The voltage and current waveforms are shown in Figure 8.38. Voltage at the main grid and current are almost sinusoidal. However, the voltages at the microgrid side are strongly distorted. Therefore, the interface isolates the main grid of harmonics generated in the microgrid side.
Figure 8.38 Test 2, waveform voltage and current, after the interface is connected. (a) Voltage vPC, (b) voltage vm, and (c) current.
With respect to the voltage unbalance, with the proposed control is possible to reduce it. The calculated value of unbalance factor is of 3.8%. It is an important reduction however, as it occurred in test 1, its value is not null. This is justified by the current unbalance that causes an unbalance in the voltages due to drops in system impedances. These cannot be avoided with an interface topology of series connection.
In the technical literature, other control strategies and different circuit topologies have been proposed for the improvement of electrical power in environments with high penetration of DG [42]. It was not possible, given the scope of this chapter, to discuss them all, so the authors have limited the discussion to those proposals that they consider clearer for its simplicity.
8.7. Summary
Nowadays DG is a reality in the power system. The increased penetration of these technologies means that solutions must be found to the problems that this produces. In this chapter different technologies have been introduced, highlighting how they couple to the grid. In addition, the power flow problem is reformulated, defining a new category node to characterize this technology kind.
The inclusion of DG in the power system generates an impact on the distribution network. In some cases, it can lead to a worsening of power quality and in others, it can be used to improve power quality. The impact on the grid is analyzed according to the following topics:
• Voltage fluctuations. Such disturbances can be produced due to the variability of the power fed into the grid by some energy sources.
• Harmonics. Some connecting interfaces emit high harmonics of the main frequency.
• Voltage unbalances. Many small generators are usually single-phase and are connected to a three-phase distribution network. This can lead to unbalances in the system voltage.
• Voltage dips. In general, it can be said that DG improves the behavior bus voltage where it is connected.
Much of DG uses a power converter (dc/ac) as an interface for connection to the network. When the aim is to inject active power to the grid, there are two basic control strategies: instantaneous unity power factor and positive sequence. In the first strategy, generation system works with unity power factor. When the voltage system is balanced and sinusoidal, the injected current is balanced and sinusoidal. However, when the voltage system is unbalanced and/or nonsinusoidal, current is unbalanced and/or nonsinusoidal, which may contribute to imbalance system and/or current harmonic circulation to the mains. In the second strategy, positive sequence control strategy, when the supply voltage is unbalanced and/or nonsinusoidal, the currents are sinusoidal and balanced, which is an improvement from this viewpoint. However, it has the disadvantage that in fault situations in the network, the latter strategy produces reactive power fluctuations and active power fluctuations in the system.
Distributed technologies can be used to improve the voltage profile of an electrical network. For this it is necessary to connect these systems to the grid in the right place according to their size. This results in a flatter voltage profiles, which means an improvement in the supply voltage.
In DG systems that use inverter as coupling devices to the network, the electrical power quality can be improved with the right strategy. For this application a new loop is provided to the inverter control, so as to determine the waveform of the current compensation according to the proposed compensation objective. As an example, generation system has been simulated and its inverter has been applied a control strategy based on positive sequence control strategy and instantaneous active power vectorial theory. This mitigates current harmonics of a nonlinear load, compensates load reactive power and feed into the grid a specified active power.
Finally, another application of active power filters is based on its use as an interface device between the utility grid and a microgrid. With proper control strategy, this interface can isolate the microgrid of generated disturbances in the utility grid and isolate the utility grid of those generated disturbances in the microgrid.
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.