In order to assess the differences between the human body thermal sensation in naturally ventilated space and that in air-conditioned space,the fuzzy evaluation model was adopted in the research of thermal sensation in naturally ventilated space.Based on the questionnaires and field measurements,the membership functions were presented by the statistic of the covering frequency to the fuzzy subset.Dry-bulb temperature was taken as the only independent variable for membership functions.The maximum values of membership grades are all at 0.5 or so,which is a distinction character on thermal comfort of naturally ventilated space.By the calculating results of membership grades value to different fuzzy evaluation subsets,the Predicted Mean Vote(PMV)was obtained.Furthermore,energy coefficient(Ea)was introduced to calculate the energy consumption,and the prediction methods of residential building energy consumption were also discussed.Finally,the importance of evaluation model of thermal sense is shown through the energy consumption prediction in a specific residential building.