Line loss in power distribution system occupies one fifth to one quarter of the total power loss in power systems. Different from the line loss in transmission system, distribution line loss is closely related to the load characteristics and economic types of power users. And the proportion of line loss caused by management problems is significant. Hence the traditional physical model of line loss is not suitable for distribution line loss analysis. To analyze line loss of low-voltage distribution system accurately, a big-data method based on MSApriori association rule mining algorithm is proposed in this paper. Based on the operational data of more than 20,000 distribution stations and the detailed data of 1.8 million power users in Shanghai, a line loss analysis method based on big data is proposed in this paper. The relationship between the line loss and its influencing factors is analyzed from the perspectives of user type, user number, power supply volume, reactive power shortage, and transformer type. The association factors obtained from the analysis are helpful to further guide the reduction of line loss.
Keywords distribution line loss, big data, association rules