Beyond building simulation
A 30% reduction in carbon emissions from buildings is suggested for a software that can compare numerous variations of a building model to optimise design. Ken Sharpe has been finding out more.
Building simulation has long been a powerful method of assessing the performance of buildings in terms of energy consumption and the internal environment. And the more simulations that are performed, the better the outcome that can be expected. However, changing the various parameters takes time, limiting the number of simulations that are performed.
That problem has been addressed by a collaborative research programme led by IES that has developed an optimisation tool that will run many, even very many, simulations for various scenarios set by the designer. As well as predicting energy consumption, this tool can, for the first time, produce cost data, so energy and construction costs can be considered in the final design decision.
A key headline benefit is the production of designs with up to 30% lower carbon emissions than traditional design methods.
The starting point is subjecting the architect’s concept design to detailed engineering analysis.
In looking at energy performance, this tool could, for example, consider the orientation of the building at 15° steps over a complete 360°. There are 11 more optimisation variables.
• Orientation
• Construction (light, heavy or medium)
• Percentage window area
• Mechanical ventilation rate
• Natural ventilation rate
• Electric lighting
• HVAC strategy
• Heating set point
• Cooling set point
• Solar PV
• Wind-turbine power
Anyone with experience of simulation will realise that such a range of variables will lead to huge numbers of simulations — running into thousands or tens of thousands.
That is where the skill and experience of the designer comes in to choose which variables to assess and to what extent. For example, the number of orientations could be constrained.
 |
| Using Optimise to compare numerous variations of parameters for a building produces a graph showing the best solutions in green for further consideration by the design team. Purple solutions are sub-optimum design options. |
The number of simulations can be set, and John Gleeson, business development manager, suggests 2000 over a year, but at least a thousand to enable solutions to converge towards an optimum. The algorithm which runs these simulations was developed by Loughborough University and selects which to run, based on a evolutionary concept that can be expressed as ‘survival of the fittest’.
Huge time savings are possible. John Gleeson suggests setting up and running 2000 simulations manually might take 60 hours. In contrast, Niall Gibson business-development manager, told the audience at a launch event that a thousand simulations he carried took three hours.
The crucial point that Ruth Kerrigan, associate director of research and development, stresses is that Optimise is a tool, not a self-operating program. She said, ‘The concept is very different to what you have done before. An idea of the outcome is critical. In selecting the variables to be analysed and the constraints, a lot of judgment and knowledge is required to get your output.’
According to John Gleeson, ‘The existing design process may not ultimately arrive at the optimum configuration due to limitations in the design budget, so opportunities might be lost. Optimise enables more simulations to be carried out so that the design can be optimised.’
All the data from an Optimise analysis can be assessed against three criteria from a choice of five. Charts comparing two criteria, such as annual energy consumption and capital cost, show clearly how the various simulations compare. As the example shows, the program highlights sub-optimal designs in red for discarding. The best solutions are highlighted in yellow on the Pareto front for further consideration by the design team.
The five assessment criteria, are listed below.
• Performance, to minimise energy use and CO2 emissions.
• Cost, to minimise capital cost.
• Daylight, to maximise daylight and daylight uniformity.
• Comfort, to minimise PPD (percentage of people dissatisfied) and summer temperature.
• Air quality, to minimise CO2 ppm.
The cost tool was developed by Davis Langdon and uses Spon data. IES led the 2-year project, which was part funded by the Technology Strategy Board, to develop the dynamic simulation that integrated the optimisation algorithm and cost tool.
There were three industrial partners for steering the project and testing the outcome. They were Mott MacDonald, AECOM and Archial. CIBSE is involved in dissemination.
Optimise is due to become available in November 2012. At present, IES is using it in house to offer a consulting service.
Because Optimise represents a radically different way of thinking, it comes packaged with training to understand its concepts and how to use it to maximum benefit.
Don McLean, founder and managing director of IES summarises, ‘I am delighted that IES has played such a vital part in the successful completion of this project. All the partners have worked excellently together, and as a result we have managed to create a real ground-breaking tool. Right at the cutting edge, Optimise is a huge advancement in the sustainable building industry and a great example of the UK leading the way with this type of technology.’
