Abstract: A recognition method based on discrete wavelet transformation（DWT）and fuzzy K-means clustering is proposed to solve the problem that it is difficult to recognize the relationship between courts and users quickly and accurately. Firstly,the dimension of the daily voltage curve of each user is reduced by DWT,and the approximate curve set and the detailed curve set are obtained. The standardized and parallel translation methods are used to pretreat the approximate curve set. Then,the membership function is constructed based on the curve similarity,and the fuzzy K-means clustering method is used to cluster the two curve sets,and the clustering result and the comprehensive recognition method proposed are used to recognize the relationship between courts and users. The approximate curve and detail curve obtained by the discrete wavelet transform can retain more original curve information after the high-dimensional curve is reduced;The standardized and translation process makes the distance calculation between the curves completely dependent on the difference in curve shape and is not affected by the amplitude. The introduction of the membership function increases the influence of curve shape similarity on the recognition result,and greatly improves the correct rate of curve clustering. The simulation results show that the proposed method can effectively reduce the dimension of curves and improve the accuracy of curve clustering,and this method has strong applicability in curve clustering and so on.