Research on Combined Algorithm for Medium and Long-term Electricity Load Forecasting

Load forecasting is one of the important work of the power sector. Reasonable and accurate load forecasting can effectively reduce the reserve capacity, improve the safety and stability of power grid operation, reduce the power generation cost, ensure the power demand, and provide a prerequisite for the development of the national economy. On this basis, this paper first discusses the purpose and significance of the research on medium and long-term load forecasting of power system, describes and compares the specific application methods of load forecasting, including multiple linear regression method, grey forecasting method and BP neural network method, and then evaluates the fitting effect of different medium and long-term forecasting algorithms through entropy weight method, determines the weight of different algorithms, and forms a combined forecasting algorithm, In order to improve the prediction accuracy of medium and long-term power load forecasting. Finally, taking the load data of Hebei Province from 2002 to 2021 as an example, the power load of Hebei Province in the next five years is predicted to verify the effectiveness of the prediction method used in this paper.