BP Neural Network Load Prediction For Low-Voltage Substations Considering Different Environmental Impacts

To reduce energy waste, improve energy utilization efficiency, and ensure power supply safety, short-term power load forecasting for low-voltage substations is essential. This article analyzes the load characteristics of typical low-voltage substations and establishes a short-term load forecasting model and solving algorithm for low-voltage substations based on them. By analyzing various factors that affect the load in the substation area, including weather, temperature, economy, time, etc., and obtaining the correlation between each factor and the load. Afterwards, the BP neural network was introduced and improved through the bacterial chemotaxis algorithm to obtain a predictive model. Finally, the accuracy and feasibility of the prediction model were verified through numerical examples, and it was proven that it can be used as a method and means for typical low-voltage substation load forecasting.