A new approach to evaluate various thermal environments

Predicted mean vote (PMV) has been recognized as one internationally adopted measurement criterion of thermal comfort. It has been applied in an air-conditioned environment but may be less applicable to naturally ventilated (NV) buildings. In this study, a new approach was developed to evaluate the impact of thermal environment on human comfort. An effective algorithm has been devised to enable a selection of the steady physiological response most appropriate in a required situation. Three physiological parameters-mean skin temperature, skin wettedness, and skin evaporative heat loss-are computed for correlation with the actual thermal comfort vote based on published data. With previous research findings, the correlation is derived in an expert system. Human thermal adaptation, which is overlooked in PMV but actually affects thermal comfort in NV environments, has been incorporated. This approach may offer a more accurate estimation of values of thermal mean vote, especially when the evaporative heat loss is used in the evaluation. In the study, the estimation error in NV buildings was reduced by 61%, leading to a better prediction of thermal acceptability. As a conclusion, this approach may yield a better evaluation of the thermal environment than the conventional PMV in NV buildings.