On the Reliability of Buildings' Energy Use Predictions Can we Close the Occupant-related Performance Gap?

Abstract

(no german abstract) Computational building models can support building design, for instance via provision of performance predictions. Specifically, energy use predictions are considered both as useful feedback for iterative design improvements and as useful orientation for future building users and operators. However, in many instances, predictions of building energy use are not confirmed by actual monitored energy use data. This circumstance is generally referred to as performance gap. As such, multiple factors can play a role in the divergence of predictions and observations, including differences between the computational building model and the actually constructed building, weather conditions, and user presence and behaviour. The latter aspect has recently suggested to be a major factor behind the performance gap. In this paper, we reconsider the problem of performance gap as related to building energy use predictions. Toward this end, we: i) formulate a general definition of the occupant-centric performance gap research premise, ii) provide a systematic depiction of the many ways people's presence and control-oriented actions can influence buildings' energy demand, iii) reassess the validity of identification of the human factor as the main contributor to performance gap, and iv) explore the postulated prospect of sophisticated occupancy modelling techniques and behavioural modification schemes.

Computational building models can support building design, for instance via provision of performance predictions. Specifically, energy use predictions are considered both as useful feedback for iterative design improvements and as useful orientation for future building users and operators. However, in many instances, predictions of building energy use are not confirmed by actual monitored energy use data. This circumstance is generally referred to as performance gap. As such, multiple factors can play a role in the divergence of predictions and observations, including differences between the computational building model and the actually constructed building, weather conditions, and user presence and behaviour. The latter aspect has recently suggested to be a major factor behind the performance gap. In this paper, we reconsider the problem of performance gap as related to building energy use predictions. Toward this end, we: i) formulate a general definition of the occupant-centric performance gap research premise, ii) provide a systematic depiction of the many ways people's presence and control-oriented actions can influence buildings' energy demand, iii) reassess the validity of identification of the human factor as the main contributor to performance gap, and iv) explore the postulated prospect of sophisticated occupancy modelling techniques and behavioural modification schemes.