Research On Building Energy Conservation Climate Zoning Based On Fuzzy Clustering Analysis

The boundaries between things, some are exact, some are fuzzy. Fuzzy clustering algorithm is a mathematics methods that categorize things according to certain requirements when the fuzzy boundaries between things.This thesis focuses on an application of fuzzy clustering analysis on building energy conservation climate zoning. In the process of studies, firstly the fuzzy c-means clustering algorithm and the weighted fuzzy c-means clustering algorithm on climate analysis of applicability are discussed in the thesis. Then the experimental results of two clustering algorithms are analyzed and inspected. Based on the analysis results, the weighted fuzzy c-means clustering algorithm is applied to building energy conservation climate zoning research.The weights is determined by the principal component analysis method, the Euclidean distance which is in the objective function of the fuzzy c-means clustering algorithm is defined by the weighted Euclidean distance, so all the indexs can be reflected the different proportion of contribution in building climate division, the fuzzy clustering results are more in line with the actual situation.The weighted fuzzy c-means clustering algorithm is an unsupervised fuzzy clustering algorithm, which needs to give a fuzzy clustering number in advance. This thesis give the value range of clustering number according to the actual situation of building climate division, and take these clustering numbers to fuzzy cluster. And use the cluster validity index which is put forward in the fifth chapter to analysis the clustering results. Practical results show that when the clustering number is 8, getting the optimal classification results. On this basis, the corresponding passive design strategies are put forward according to different climate area. In addition, our country building thermal zoning map is compared with the climate division of the thesis and some anomalies in the partitioning result are interpreted and analyzed. Starting from the national climate division of the overall, the climate division results of the thesis is relatively accurate.