Modelling of building performance under the UK climate change projections and the prediction of future heating and cooling design loads in building spaces

New climate change projections for the UK were published by the United Kingdom Climate Impacts Programme in 2009. They form the 5th and most comprehensive set of predictions of climate change developed for the UK to date. As one of main products of UK Climate Projections 2009 (UKCP09), the Weather Generator, can generate a set of daily and hourly future weather variables at different time periods (2020s to 2080s) and carbon emission scenarios (low, medium and high) on a 5 km grid scale. In a radical departure from previous methods, the 2009 Projections are statistical-probabilistic in nature. A tool has been developed in Matlab to generate future Test Reference Year (TRY) and Design Reference Years (DRY) weather files from these Projections and the results were verified against results from alternative tools produced by Manchester University and Exeter University as well as with CIBSE’s Future Weather Years (FWYs) which are based on earlier (4th generation) climate change scenarios and are currently used by practitioners. The Northumbria tool is computationally efficient and can extract a single Test Reference Year and 2 Design Reference Years from 3000 years of raw data in less than 6 minutes on a typical modern PC. It uses an established ISO method for generating Test Reference Year data and an alternative method of constructing future Design Reference Years data is proposed. Fifteen different buildings have been identified according to alternative usage, thermal insulation, user activity and construction details. Besides these variants, the buildings were chosen specifically because they either exist, or have received planning consent and so represent ‘real’ UK building examples. Two investigations were then carried out based on the 15 case study buildings. The first involved applying TRYs generated for London, Manchester and Edinburgh for a variety of carbon emission scenarios at time horizons of 2030, 2050 and 2080. The TRYs were developed into a weather data format readable by the EnergyPlus energy simulation program to simulate summertime internal comfort (operative) temperatures, cooling demands and winter heating demands. All results were compared with a control data set of nominally current weather data, together with the same results produced using the alternative weather data generators of Manchester University, Exeter University and the CIBSE FWYs. Results revealed a good agreement between the various methods and show that significant increases in internal summer operative temperatures in non-air-conditioned buildings can be expected as time advances through this century, as well as increased air conditioning cooling energy demands and small reductions in winter heating energy demand. The second investigation involved generating time series of design internal peak summertime operative temperatures, design cooling demands and design winter heating demands for the same conditions as the first investigation. The results were then used to develop a simplified estimation method to predict future design cooling loads using multiple regressions fitting to selected data from the DRY simulation inputs and outputs. The simplified estimation method forms a useful tool for estimating how future cooling design loads in buildings are likely to evolve over time. It also provides a basis for designers and practitioners to determine how buildings constructed today will need to be adapted through life to cope with climate change