A Study Of Identification Method Of Wide-area Adjustable Load

With the rapid development of new energy in China,the problem of new energy consumption has become increasingly prominent.Involving adjustable loads in the grid dispatch is one of the effective measures to solve the problem of wind power consumption.The wide-area adjustable load has good regulation characteristics,and the development of the load has begun to take shape.Since the objects of power system load monitoring and statistics are mostly integrated loads,to achieve load control,the adjustable load must be identified from the integrated load.In view of the above situation,the paper studies the wide-area adjustable load identification method.Firstly,this paper analyzes the adjustable characteristics of typical adjustable industrial loads represented by electrolytic aluminum loads,ferroalloy loads and silicon carbide loads,and establishes a discretely adjustable and a continuously adjustable model.The adjustable characteristics of typical adjustable civil loads represented by a thermal storage electric boiler and an electric vehicle are also analyzed.Next,for the wide-area civilian loads represented by the thermal storage electric boiler and the electric vehicle,the adjustable characteristics after multi-load polymerization are analyzed,and the time-shiftable load adjustment model is established.Furthermore,based on the discretely adjustable,continuously adjustable and time-shiftable characteristics of wide-area adjustable load,five characteristic parameters characterizing various load regulation characteristics are extracted for wide-area adjustable load identification.Then,based on the five load characteristic parameters,a wide-area adjustable load identification method based on FCM and BP neural network is proposed to realize the identification of the proportion of wide-area adjustable load in the integrated load.Finally,based on the data of Gansu load energy management platform,the simulation analysis is carried out to verify the feasibility and effectiveness of the wide-area adjustable load identification method based on FCM and BP neural network.