Research On Apparent Temperature Modeling Based On Improved GEP Algorithm

With the improvement of life quality, people pay more and more attentions to indoor comfort control. Comfort is the body's subjective perception of surrounding environment and apparent temperature is the quantitative analysis of comfort, they are all affected by multiple environment variables. So, determining the relationship between apparent temperature and environmental variables is an important research context. For this context, Gene expression programming(GEP) algorithm will be a good solution, because it is widely used to solve various problems, such as Function Finding, Parameter Optimization, Data Mining and so on.Combined genetic algorithms and genetic programming, gene expression programming is a new evolutionary computation method, and its outstanding feature is that it can solve complex problems with simple coding. In this paper, for the situation of traditional GEP lowering efficiency in terms of function finding, there will be improvement about mutation rate and population diversity. Then adopting the improved GEP algorithm to get Apparent Temperature mathematical model. Next proposing a comfort control method based on the mathematical model,the method indicates Apparent Temperature to be the controlled variable, controllable environmental factors to be the control variables. Finally, an experiment platform based on MATLAB and FRDM-KL25 Z is built to test the comfort control method.There are two innovation points as following:1) Improving the GEP algorithm, then using the algorithm to get apparent temperature mathematical model;2) Proposing a comfort control method based on the Apparent Temperature mathematical model.