Short-term Power Load Forecasting Method Based on VMD-ARIMA-SVR

Aiming at the problem of strong randomness and low prediction accuracy of power load, a short-term prediction decomposition and reconstruction model of power load based on differential integral moving average autoregressive and support vector machine is proposed. The grey wolf algorithm is improved by using the nonlinear control factor and the random weight position update strategy. The improved algorithm is used to optimize the kernel parameters and penalty factors of the support vector machine, and the integrated moving average autoregressive model is used to predict the data after the variational mode decomposition, and the final prediction value is obtained after reconstruction. Combined with the actual data prediction analysis in a certain area, this method has better prediction accuracy and convergence speed than other prediction models.