The predicted mean vote (PMV) model predicts the mean thermal response of a large group of people, according to the ASHRAE thermal sensation scale, and has been proven to agree well with many field studies. However, the model may not predict well the mean response of a small number of people or individuals because it does not take into account the physiological variations of people due to changing thermal environments. In this paper, the thermal comfort sensations of twenty subjects (ten males and ten females) aged 21 to 30 were investigated by Fanger's PMV model. This experimental study found that the PMV model failed to predict the subjects' thermal sensations. The reason may be that the mean skin temperatures (MSTs) of the subjects were fixed uniformly, regardless of the variations of the thermal environments, which results in errors using the PMV model, especially for a small number of subjects. To obtain the MSTs of the subjects under different thermal environments, fourteen measurement methods were employed. Afterwards, the PMV model was used to predict the thermal response of the twenty subjects based on their measured MSTs. The results prove that the prediction accuracy of the PMV model is improved greatly when the MSTs are obtained through measurements instead of theoretical calculations. In addition, this study concludes that the Burton 3-point Method may be the best to measure MSTs for the PMV model.
Read full abstract