Research on Short-term Load Forecasting under Demand Response of Multi-type Power Grid Connection Based on Dynamic Electricity Price

• Published in: 2021 IEEE International Conference on Power, Intelligent Computing and Systems (ICPICS) pp. 141 - 144
• Main Authors: Yuan, Li, Ying, Chen, Yongfu, Shi, Xuenan, Zhao, Yucai, Li, Yang, Lou
• Format: Conference Proceeding
• Language: English
• Published: IEEE 29.07.2021
• Subjects: Demand response; dynamic electricity price; grid connection of multiple power sources; Load forecasting; Power grids; Power supplies; Power system dynamics; Predictive models; radial basis function neural network; short-term load forecasting; Software; utility function
• DOI: 10.1109/ICPICS52425.2021.9524096

The demand response of multiple types of power supply is an effective means to alleviate the power supply pressure and optimize the allocation of resources in the power sector, and it also puts forward higher requirements for the traditional short-term load forecasting method of power system. In view of the problem that the traditional prediction model cannot accurately reflect the change of load curve after the implementation of multi-energy grid connection, firstly, according to the effect with theory, a demand response model based on dynamic electricity price is constructed to simulate the actual load curve of power users after implementing demand response. On this basis, the short-term load forecasting model of RBF neural network considering demand response factors is constructed. Finally, using the actual data of a certain region, the prediction performance of the model is simulated and verified by Matlab software. The results show that the average prediction error of this paper is 1.27 %, while the average prediction error of the traditional load forecasting model is 3.40 %. The average prediction error of the former is 2.13 % lower than that of the latter, which proves that the addition of demand response signal in the prediction model can effectively improve the prediction accuracy of short-term load, and accurately reflect the change of load curve due to the effect of demand response signal.