1.Key Laboratory of the Ministry of Education on Smart Power Grids (Tianjin University), Tianjin 300072, China;2.Electric Power Research Institute of Guizhou Power Grid Co., Ltd., Guiyang 550007, China;3.China Electric Power Research Institute Co., Ltd., Beijing 100192, China;4.State Grid Tianjin Electric Power Company, Tianjin 300010, China
In the regional energy interconnection system, the demand response has changed the conventional power consumption habits of power users, and increased the uncertainties in forecasting environment. In view of this specific environment, this paper proposes a short-term load forecasting method of regional power grids considering demand response to meet the demand of enterprises for forecasting accuracy. This method improves the forecasting accuracy by constructing data processing model, load forecasting model and error forecasting model in turn. More specifically, for the historical data sample set, the gray relation analysis method is used to process the meteorological data to obtain the similar daily characteristic variables of the input forecasting model. For the power load forecasting, a long short-term memory network model is established. Its special gate structure is used to selectively control the influence of input variables on model parameters, and the forecasting performance of the model is improved. For the error data sample set, the dynamic modal decomposition technology is used to mine the potential value of the error data. Its data-driven characteristics are used to characterize the variation trend characteristics of the error series, and a good error prediction is achieved. Finally, combined with the actual power grid data, the effectiveness and superiority of the proposed method are verified.
This work is supported by National Key R&D Program of China (No. 2016YFB0901104).
|||LI Chuang, KONG Xiangyu, ZHU Shijian, et al. Short-term Load Forecasting of Regional Power Grid Considering Demand Response in Energy Interconnection Environment[J]. Automation of Electric Power Systems,2021,45(1):71-78. DOI:10.7500/AEPS20200428001|