A Method for Predicting the Growth Trend of Load in Extreme Weather Zones Based on Wavelet - LSTM Algorithm

The proportion of renewable energy is constantly increasing, and new types of loads are developing rapidly, especially with a large number of distributed power sources, electric vehicles, etc. connected to the power grid, which poses huge challenges to the formulation of power generation plans and the arrangement of methods, affecting the safety and stability of the power grid. Therefore, the ability to accurately predict loads is becoming an important technical foundation. Under the new situation of “dual carbon”, it has become an inevitable choice to use technologies such as artificial intelligence and big data to accurately predict electricity load, integrate load measurement data, numerical weather forecasts, etc., and construct high-precision load forecasting models. An intelligent forecasting method based on Wavelet - LSTM algorithm to study the growth trend of regional load under extreme weather conditions is proposed in this paper. Various types of load data are collected through load partitioning and classification, and meteorological data is fused for normalization preprocessing. Through wavelet decomposition and LSTM modeling, the load and related parameters under extreme weather conditions are intelligently predicted, and the growth trend of each region is obtained by integrating the output results of various types of loads.