In the present day retail market, there are several opportunities for load serving entities which are provided by large volumes of smart data for meters , which improves the knowledge of electricity consumption behaviors of customers by using load profiling instead of focusing on load curves. This paper proposes a unique approach for clustering the electricity consumption behavior dynamics such as transitions and relations between them in eventual periods. First, to downsize the scale of data a

more »

... symbolic aggregate approximation (SAX) is performed for each distinct customer and to model the electricity consumption dynamic, transforming the large data set of load curves to several state transition matrixes by using a time-based Markov model is applied. Second, to obtain the dynamics of consumption behavior a clustering technique by Fast Search and Find of Density Peaks (CFSFDP) is mainly carried out, with the distinction between any two consumption patterns measured by the Kullback-Liebler (K-L) distance, and to classify the customers into several clusters. To tackle the challenges of big data, the CFSFDP technique is integrated into a divide-and-conquer approach toward big data applications. A numerical case verifies the effectiveness of the proposed models and approaches. with the refinement between any two utilization designs measured by the Kullback-Liebler (K-L) remove, and to arrange the clients into a few bunches. To handle the difficulties of enormous information, the CFSFDP method is coordinated into a gap and-overcome approach toward huge information applications. A numerical case checks the adequacy of the proposed models and methodologies.