Evaluating Indoor Air Quality by Fuzzy Logic with Statistical Pattern Recognition

This paper presents a new method for evaluating the indoor air quality based on fuzzy grade mathematical operation with frequency pattern of indoor environment parameters. A statistical pattern recognition approach was applied in this study for the generation of patterns. The fuzzy modeling methods were used to represent the indoor air quality value to increase the flexibility of the expression. The distribution frequency of each parameter was discussed in this article. Three different probability density functions, Gaussian distribution, Weibull distribution, and log-normal distributions, were used to calculate the pattern recognition of environmental factors. The model was used to express the measurement results in the building environment. Ten table of fuzzy representation of indoor air quality was created in this study, which was based on the suggested value of indoor air quality by Taiwan environmental Protection Administration. These tables include the carbon dioxide, carbon monoxide, PM10, PM2.5, bacteria, fungi, temperature, ozone, formaldehyde, and total volatile organic carbon etc. Model validation was done by a Monte-Carlo experiment with the statistical pattern recognition of different parameters. The proposed methods can overcome the drawback of the present method to express the condition of indoor air quality in a more simple way.

並列關鍵字

Fuzzy theory ； Indoor air quality ； membership function ； linguistic variable ； pattern recognition

國際替代計量

全文下載

主題瀏覽