Thermal comfort, occupant control behaviour and performance gap - A study of office buildings in north-east China using data mining

Simulation techniques have been increasingly applied to building performance evaluation and building environmental design. However, uncertain and random factors, such as occupant behaviour, can generate a performance gap between the results from computer simulations and real buildings. This study involved a longitudinal questionnaire survey conducted for one year, along with a continuous recording of environmental parameters and behaviour state changes, in ten offices located in the severe cold region of north-east China. The offices varied from private rooms to open-plan spaces. The thermal comfort experiences of the office workers and their environmental control behaviours were tracked and analysed during summer and winter seasons. The interaction of the thermal comfort experiences of the occupants and behaviour changes were analysed, and window-opening behaviour patterns were defined by applying data mining techniques. The results also generated window-opening behaviour working profiles to link to building performance simulation software. The aim was to apply these profiles to further study the discrepancies between simulation and monitored results that arise from real-world occupant behaviour patterns.