Computational Tools and Techniques

Although some would credit the abacus, the first computational machine was probably a mechanical calculator designed by the French mathematician Blaise Pascal in 1642. In 1804, Joseph Marie Jacquard designed the Jacquard loom, a pattern-making tool that could produce a variety of patterns by simply changing a template. This loom was the first (re)programmable machine to make use of punchcards as a way of encoding data.

Known as a “difference engine” and consisting of sets of gears, levers and pulleys made from cast bronze, the first fully (re)programmable computation machine was designed by Charles Babbage in 1830, but it wasn’t until the 1940s that the modern computer was conceived. During the Second World War, Alan Turing, in his search for ways of cracking the German secret code, conceived the idea of an endless tape loop being fed into a machine capable of performing an infinite number of calculations – the “universal computer.”

The often used acronym CAD originally stood for computer-aided-drafting. “Drafting” was replaced by “design” at some later stage, and although graphic computing has proven a useful aid to the draftsman, the ability of the computer to help with design is largely a product of its ability to produce rapid mathematical calculations (largely using ones and zeros – the binary system) and hence to (1) organize vast amounts of information and (2) simulate (model) the predicted behavior of both natural and artificial objects and systems.

This section is subdivided into three main categories:

Building Information Modeling

The ways in which computers can be used as interoperable libraries and databases and how building information can be modeled in a computer.

Structural Analysis

The ways in which computers can be used to predict the behavior of materials and structures under load and to explore structural forms within these parameters: form finding and finite element analysis.

Environmental Analysis

The ways in which computers can be used to predict environmental behavior and to explore the design of climatic enclosures within such parameters, viz.: thermal, daylight, air movement, and acoustic.

Building Information Modeling

Building Information Modeling (BIM)

Also known as BIMM – Building Information Modeling and Management – BIM is a generic term that covers a multitude of modern, computer applications. BIM relies entirely on software design and the ability of the different programs to become integrated within a single computer model.

A building information model represents the building as “an integrated database of co-ordinated information.” Standard computer aided drafting software is adapted to include object oriented systems. That is to say that since any line on a drawing is able to be encoded as an object, it is then a logical step to link objects in an architectural drawing to non-graphical data using standardized computer file formats. Any construction element in the drawing can also carry other (hidden) attributes such as material properties, costs, suppliers, etc. This non-graphical data, held in a series of spread sheets, can then be extracted for use in written schedules and specifications.

The following range of programs may contribute to the overall model:

Site surveys: 3D geo-technical surveys, thermal imaging, noise mapping, wind mapping.

Design: parametric modeling – an important part of an efficient building information model, parametric modeling is computer software that uses a relational database within a dynamic model of the whole project. The software takes account of the linked behavior of the different elements of a building – its parametric components – both graphically and informationally: it maintains consistent relationships between elements as the model is manipulated. The changes (to a floor plan, a section, a schedule) made in a project are reflected throughout the project and all necessary adjustments and alterations are made automatically.

Structural analysis: materials selection software; finite element analysis (FEA).

Environmental analysis: solar geometry, daylight penetration, thermal capacities, acoustic properties, air movement, and crowd control – the latter two using Computational Fluid Dynamics (CFD).

Optimization and clash detection: the parametric model is also able to identify all building elements on the 3D drawing and use them as a database for schedules, spreadsheets, and specifications. It can therefore help calculate detailed material quantities and track material quantities in cost estimates.

Construction: off-site (pre-) fabrication of construction elements using Computer-Integrated-Manufacturing where machine codes are converted directly from the designer’s graphic files; the contractor’s program, critical paths, sourcing, and costs etc. are incorporated into the optimization model. (The industry is establishing a standard specification format for BIM.)

Clients: BIM for clients is viewed on line as non-executable files that do not require the original software application; Facilities Management (FM) is able to use the model after hand-over to locate, repair, or replace any element in the building. Bar-coding of elements is also used for this purpose.

1 Centralized computer model.

2 Parametric design using Revit software.

2.1 The 3D nature of BIM means one change is a change everywhere; plans, sections, details, schedules, and quantities will all automatically update.

2.2 “Families” are the 3D components used in the model and can include cabinetry, equipment, building parts, walls, and columns. These components are completely parametric, allowing an increased level of flexibility and reducing the time spent updating traditional plans, sections, and details.

2.3 BIM allows for the project team members including consultants, client, specialists, suppliers, and contractor to work on a single shared 3D model which encourages collaborative working relationships and reduces the risk of construction conflicts and the cost and time impacts of redesign.

2.4 Projects can be visualized at an early stage giving clients a clear idea of the design intent. Before works on site commence BIM allows the project team to work through the “build” in a virtual environment to optimize the construction period.

2.5 BIM enables much faster and more efficient project delivery with the 3D model able to produce quantities, schedules, and fabrication and construction drawings.

2.6 BIM models can contain product information that assists with the ongoing operation and maintenance of a building once completed.

Responsive Systems

Computers may be used to enable buildings to adapt and respond to external conditions through electronic sensing and control and the mechanical actuation of their parts; the capacity for the built environment to respond and adapt in a controlled manner depends upon automotive systems.

Automation in buildings requires combined mechanical and electrical systems (M & E) and these range from programming the pumps and fans of heating and ventilation systems through to the control of lifts and the monitoring of fail-safe mechanisms. A modern office building will have a series of computer-controlled routines to monitor and regulate the levels of temperature, humidity and light, as well as air quality and other hazards. Modern, long span/high traffic bridges have built-in vibration and seismic sensors which are monitored 24 hours a day.

Sensors and Actuators

Sensors can monitor the environment in a variety of ways. They can detect motion, temperature, humidity, wind speed and direction, light levels, noise levels, air quality, pressure, vibrations, etc. By turning analogue waves into digital bytes, they can provide a constant stream of data that can either be stored for analytical purposes or fed directly into a program that will search for certain parameters within which to react. These programs may be considered as “if – then” routines: if the sensor detects a certain type of data, then it can instruct a motor to actuate in a certain way. Electronic stepper motors can be computer-controlled to operate in digital “steps” to very fine levels of accuracy.

A to D and Binary Logic

Generally, the type of input that might be required in order to modify an environment may be classified as either a simple switch, such as a light switch, or data that comes in the form of an electromagnetic wavelength such as sound, light or temperature. Sensors that are designed to read this type of environmental data must then transform the analogue waves into a digital format – A to D – so that electronic logic systems can understand and act on the information.

Digitization has developed using binary logic; all data can be encoded using only two symbols (e.g. one and zero) that can then be transformed into a voltage that is in turn directed through logic gates in order to control the desired output.

Expert Systems and Statistical Analysis

In the Case Studies section at the end of this book, two projects illustrate the use of responsive systems in architecture. Foster Associate’s HSBC building in Hong Kong (see p168) employs mirrors to track the sun in order to reflect light into the atrium; this is known as an expert system, as the computer knows where the sun is at all times and can adjust the mirrors accordingly. The case study of the “D” Tower by Nox Architecture (see p178) illustrates the use of a regularly updated statistical database to control the external appearance (or “mood”) of the building.

1 Binary logic tables are also known as truth tables.

2 Sound wave described using a 4-bit A to D convertor. The analogue wave will be digitally encoded as 1001 1011 1101 1100 1011 1000 0101 0010 0001, etc.

3 Logic gates: graphic symbols and truth tables.

4 Arab Institute, Paris, France. This building by Jean Nouvel employs the same mechanism as that used for camera shutters to control light entering the library. Sensors monitor light levels and automatically open and close the shutters to maintain the optimum levels.

5 Kingsdale School, London, UK (dRMM Architects). The ETFE (ethylene tetrafluoroethylene), inflated “pillows” that cover the open forecourt have an intermediate layer that is printed with the inverse, checkerboard pattern to that of the upper skin. When sensors detect overheating the pressure within the pillows is regulated to bring the two patterns closer together, thus decreasing solar gain.

6 Dynamic yacht mast. Rather than engineering for the worst case scenario, a lighter mast can react to increased loading through sensors and control systems that actuate tension wires within the mast.

Structural Analysis / Form Finding, Finite Element Analysis

Form Finding

Historically, finding and creating new structural forms was accomplished by extracting geometric information from physical models, in particular three-dimensional compressive surfaces (shells) or three-dimensional tensile surfaces (membranes). With the advent of computer-aided-design (CAD) along with an increased knowledge of the behavior of materials, a variety of approaches to form finding can now be pursued using computer programs to calculate optimum structural solutions for given geometric parameters.

Finite Element Analysis

The first step in using finite element analysis (FEA) is constructing a finite element model of the structure, to be analysed. Two- or three-dimensional CAD models are imported into an FEA environment and a “meshing” procedure is used to define and break the model up into a geometric arrangement of small elements and nodes. Nodes represent points at which features such as displacements are calculated. Elements are bounded by sets of nodes and define the localized mass and stiffness properties of the model. Elements are also defined by mesh numbers which allow reference to be made to corresponding deflections or stresses at specific model locations. Knowing the properties of the materials used, the software conducts a series of computational procedures to determine effects such as deformations, strains and stresses that are caused by applied structural loads. The results can then be studied using visualization tools within the FEA environment to view, and identify the implications of, the analysis. Numerical and graphical tools allow the precise location of data such as stresses and deflections to be identified.

1 3D “Nurbs” model of the canopy for a bandstand, with computergenerated section lines highlighted in yellow.

2 The complex roof geometry for a new roof on an existing tower was rationalized using three-dimensional models, and built with a simple steel ring beam and curved steel cross beams that support rafters and a double curved plywood deck.

3–5 Form-finding software used for the design of membrane structures. Control points (CPs) are used to create space and the program operates in such a way that when a force is applied to one point the load of the force is distributed homogeneously so that the membrane is always under tension to produce a smooth transition between points.

6 View of the mesh used to define the geometry of the bandstand canopy.

7 Exploded drawing of the structural components for the canopy.

8 FEA model of buckling in a proposed 230 ft high fiber-reinforced plastic (FRP) mast.

9 In this project for a 32 ft 9 in high tower (designed as a stack of solid acrylic blocks), FEA was applied in order to predict the behavior of the towers under wind load.

Environmental Analysis / Thermal, Daylight

Computer programs can enable architects to carry out environmental analysis for thermal comfort, daylighting, air movement, and acoustics. A three-dimensional graphic model is given material attributes and software produces both numerical and graphical results.

Thermal Comfort – Shading Design and Solar Analysis

Excessive solar exposure is one of the main causes of overheating in buildings, even in relatively cold climates. At the same time, the sun is one of the most effective sources of natural energy available. Thus, shading systems and the analysis of solar gains are inextricably linked.

Analytical computer software gives the architect the ability to calculate and visualize incident solar radiation on the windows and surfaces of buildings over any period of time. The computer model can display overshadowing from adjacent buildings and make it possible to compare incident gains during different seasons, showing variations in the solar resources available at times when heating or cooling is required. This enables architects to quantify the effects of different solar shading approaches, including building orientation and the size and location of window openings, at the design stage.

Sample Thermal Analysis Model

The effect of thermal mass is modeled by placing a building in Athens and applying the inbuilt weather files. The single-story building is 20 by 33 by 8 ft high and the envelope can be constructed from either dense, high thermal capacity materials such as brick and concrete or lightweight, low thermal capacity materials such as timber frames and suspended timber floors. Although the outside temperature reaches 103ºF at midday, the heavyweight building maintains a 9.7ºF temperature difference at this hottest time of the day. The maximum internal temperature of 94.8ºF occurs at 4 pm (diagram 1). The lightweight building shows the effects of instantaneous heat gains with a maximum temperature occurring nearer to midday, at 2 pm, with a higher internal temperature of 97.3ºF and a maximum temperature difference of only 6.8ºF. When comparing heavyweight and lightweight buildings, the concrete and brick envelope reduces the August cooling power required to 45 per cent of that of the timber-frame structure. If cooling were to be achieved by air-conditioning with refrigeration, the August running costs and carbon emissions for the concrete structure would be 70 per cent of those for the timber frame. By making the Z axis value on the original drawing –8 ft, the effect is to bury the entire zone in the earth. The resulting monthly load graph clearly shows that the maximum cooling requirement has reduced to 592W (diagram 2): there is therefore no requirement to heat the building.

Daylighting

Because analytical computer software can automatically generate sun path diagrams to show overshadowing periods for the entire year, it is possible to calculate daylight factors and illuminance levels at any point within a model (see diagrams 36).

1, 2 Thermal analysis bar charts.

3 Daylighting analysis; location London, shadows at 4 pm.

4 Daylighting analysis; view from sun’s position.

5, 6 Daylighting analysis; daylight values and penetration into interior.

Environmental Analysis / Air Movement, Acoustic

Air Movement

Using inbuilt weather data, analytical computer software can overlay annual wind speed, frequency and direction directly on top of a design model, making it appropriate for natural ventilation and wind shelter strategies, particularly when combined with computational fluid dynamics software.

Computational Fluid Dynamics (CFD)

The Navier-Stokes equations, named after Claude-Louis Navier and George Gabriel Stokes, are a set of equations that describe the motion of fluid substances such as liquids and gases. The equations are a dynamical statement of the balance of forces acting at any given region of the fluid. The various numerical approaches to solving the Navier-Stokes equations are collectively called computational fluid dynamics. When translated into a graphical format, the motion of the fluids can be seen as particles moving through space. CFD can then be used to simulate wind dynamics – speed and direction – in and around buildings. The architect can explore variations in design that can, for example, improve natural ventilation or minimize excessive downdrafts from tall buildings.

Acoustic Modeling

Analytical computer software offers a number of acoustic analysis options. These range from simple statistical reverberation times through to sophisticated particle analysis and ray tracing techniques.

Sample Acoustic Analysis Model

The effect of different materials and surfaces on reverberation time is explored within a computer model. The single-story building is 20 by 33 by 8 ft high and two different types of envelope are analyzed. In the first, the envelope is constructed from masonry cavity walls with a plaster joist ceiling and concrete slab floor. This gives reverberation time data of 2.3 seconds at 500 Hz and 1.14 seconds at 1 kHz (diagram 2). In the second, the envelope is constructed from masonry cavity walls with an acoustic tile ceiling and a carpeted suspended timber floor. This gives reverberation time data of 0.9 seconds at 500 Hz and between 0.31 and 0.75 seconds at 1 kHz (diagram 3). The graphical outputs clearly show the impact of absorbent surfaces on sound reflections within the room.

1 CFD applied to assess wind impact on a façade.

2, 3 Acoustic reverberation graphs.

Montage of the case studies.

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset