Modeling The Thermal Prediction Using The Fuzzy Rule Classifier
Chujun Zhong, Tingzhang Liu, Jianfei Zhao · 2019
The comfort of the indoor environment greatly affects the work efficiency of the personnel. Although “sensation” is a fuzzy conceptualization of `satisfaction with the thermal environment', it is frequently used as a measure for evaluating thermal comfort. Therefore, this is an innovation that the fuzzy rule classifier is used for modeling the thermal prediction based on the ambiguity of the data itself. This paper applies the fuzzy rule classifier to model the thermal prediction based on the PR-821 dataset. The simulation results show that the model has over-fitting but it can be solved by processing the data by SMOTEENN algorithm. The prediction accuracy of the improved model is greatly improved. Compared with other algorithms (CART, C4.5, Adaboost), the results show that the fuzzy rule classifier has a better predictive effect on thermal sensation.