Research on a Visual Comfort Model Based on Individual Preference in China through Machine Learning Algorithm

Recently, decreasing energy consumption under the premise of building comfort has become a popular topic, especially visual comfort. Existing research on lacks standard how to select indicators. Moreover, studies individual preference considering interaction between internal and external environment are few. In this paper, we ranked common indicators by cloud model combined with failure mode effect analysis (FMEA) hierarchical technique for order similarity ideal solution (TOPSIS). Unsatisfied vertical illuminance, daylight glare index, luminance ratio, shadow position top four Based these indicators, also built through five categories personalized data obtained from experiment, which was trained machine learning algorithms. The results show that random forest best prediction performance support vector is second. Gaussian mixed classification tree have worst stability accuracy. addition, study programmed BIM plug-in integrating environmental personal predict appropriate illuminance specific occupant. Thus, managers can adjust intensity artificial light in office increasing or height table lamps, saving improving occupant This novel will serve as paradigm selecting make indoor space be tailored meet preferences.