A method of mining electricity consumption behaviour based on CC-DWT

Abstract

In this paper, a fusion clustering algorithm based on discrete wavelet transform called CC-DDWT is proposed. In this method, the preliminary feature construction of load curve is realized by multi-layer discrete wavelet transform, and the approximate component and detail component are formed, and then the clustering fusion is carried out. The experimental results show that the typical load patterns of a single user extracted by this method are more reliable than the traditional methods.