A Method of Electricity Consumption Behavior Analysis Based on Rough Set Fuzzy Clustering

For data mining on big data of electricity consumption in smart grid, cluster analysis is used to extract information value and optimize strategy of demand response. Compulsively attributing electricity consumption behavior to the class whose center is closest to, existing methods have deficiencies when a behavior has characteristics of multiple classes. Therefore, a method of electricity consumption behavior analysis in smart grid based on rough set fuzzy clustering is proposed. By introducing rough set, the result of clustering is expressed as lower approximation set and upper approximation set to retain data information on the characteristics. Case study shows that the method is effective to improve clustering accuracy rate, and provide a precise description of target group for demand response.