Sampling for building energy consumption with fuzzy theory

The foundation of energy saving is knowing the real status of building energy consumption. For various kinds and a great number of building energy consumption data, the fuzzy theory is applied for sampling. It would make data representational. Firstly, a fuzzy clustering method is used to classify the data set and then the samples are extracted from the subclass. A modified clustering algorithm based on entropy weight method is proposed. It can determine the number of the classification of data set. The simulation results indicate that the new method can directly determine the optimal sample size. This method is suitably applied for dynamic energy consumption data and is more accurate compared with the statistical method.