Short-term Load Forecasting of Power Grid Based on Multivariate Empirical Mode Decomposition and Genetic Algorithm Optimization BP Neural Network

In order to make short-term load forecasting of power grid more accurately, a new combination model of power system short-term load forecasting is proposed. This model combines multivariate empirical mode decomposition with the BP neural network optimized by genetic algorithm. First, it uses multivariate empirical mode decomposition to decompose the power load sequence and important influencing factors sequence into the same number of sub-sequences at the same time, and then import them into the GABP model for prediction and get the final prediction result. Taking the instantaneous load of a city power grid in Shandong Province in August 2019 as the research object, and considering the influence of instantaneous temperature, humidity, atmospheric pressure, precipitation, wind speed and other factors recorded by meteorological data, the established model is simulated and analyzed, and the load value of 24 hours on August 31, 2019 is compared with the predicted effect. The simulation results indicate that the prediction model has high precision, which proves its feasibility.