Predicted vs. actual energy performance of non-domestic buildings: Using post-occupancy evaluation data to reduce the performance gap

Abstract

With the increasing demand for more energy efficient buildings, the construction industry is faced with\nthe challenge to ensure that the energy performance predicted during the design stage is achieved\nonce a building is in use. There is, however, significant evidence to suggest that buildings are not\nperforming as well as expected and initiatives such as PROBE and CarbonBuzz aim to illustrate the\nextent of this so called ‘performance gap’. This paper discusses the underlying causes of\ndiscrepancies between energy modelling predictions and in-use performance of occupied buildings\n(after the twelve month liability period). Many of the causal factors relate to the use of unrealistic input\nparameters regarding occupancy behaviour and facilities management in building energy models. In\nturn, this is associated with the lack of feedback to designers once a building has been constructed\nand occupied.\nThe paper aims to demonstrate how knowledge acquired from Post-Occupancy Evaluation (POE) can\nbe used to produce more accurate energy performance models. A case study focused specifically on\nlighting, small power and catering equipment in a high density office building is analysed and\npresented. Results show that by combining monitoring data with predictive energy modelling, it was\npossible to increase the accuracy of the model to within 3% of actual electricity consumption values.\nFuture work will seek to use detailed POE data to develop a set of evidence based benchmarks for\nenergy consumption in office buildings. It is envisioned that these benchmarks will inform designers\non the impact of occupancy and management on the actual energy consumption of buildings.\nMoreover, it should enable the use of more realistic input parameters in energy models, bringing the\npredicted figures closer to reality.