A Decomposition Method of Electricity Quantity Based on Typical Load Curve Simulation

In the settlement process of electricity spot market, it is crucial to obtain the real-time power of users. This paper designs a method to obtain the typical load curve of the users without smart meters in the spot market and then proposes a power decomposition strategy to determine the Time-of-Use consuming quantity of these users. Firstly, users are classified according to their industry characteristics, this paper assumes that the same type users will have similar daily load curve. Then, the history load data of the users with smart meters is collected, and the typical load curve is determined by an improved clustering algorithm based on these history data. Lastly, the Time-of-Use consuming quantity of the users without smart meters can be determined according to the typical load curve. In addition, an improved load curve clustering algorithm based on FCM is proposed, the improved XB clustering validity index is introduced to judge the effect of the clustering results. The actual power load data of Fujian province are used for case study, results verify the correctness and feasibility of the proposed method.