Short‐term load forecasting based on wavelet neural network with adaptive mutation bat optimization algorithm

Abstract

To improve the accuracy of short‐term load forecasting of power systems, according to the nonlinearity and uncertainty of short‐term load sequence, a short‐term power load forecasting method combined with wavelet neural network (WNN) and adaptive mutation bat optimization algorithm (AMBA), which is based on the variance of the population's fitness, is proposed in this paper. The model determines the mutation probability of the current optimal individual based on the variance of the population's fitness and the current optimal solution, performs Gaussian mutation on the global optimal individual, and carries out the second optimization on the bat individuals after mutation. AMBA is employed to optimize the network parameters of WNN, improving the prediction accuracy of WNN and speeding up its training. Then the AMBA‐WNN forecasting model is built. The AMBA‐WNN model is used to predict short‐term load of a certain city in China as a case study. The results show that the model can effectively improve the accuracy of short‐term load forecasting and has good practical significance. © 2018 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.