Power load forecasting algorithm based on nonlinear inertial factor change pattern particle swarm optimization algorithm

College of Instrumentation & Electrical Engineering, Jilin University, Changchun, china

^* Corresponding author: iamwangyongzhi@126.com

Abstract

The common method of power load forecasting is the least squares support vector machine, but this method is very dependent on the selection of parameters. Particle swarm optimization algorithm is an algorithm suitable for optimizing the selection of support vector parameters, but it is easy to fall into the local optimum. In this paper, we propose a new particle swarm optimization algorithm, it uses non-linear inertial factor change that is used to optimize the algorithm least squares support vector machine to avoid falling into the local optimum. It aims to make the prediction accuracy of the algorithm reach the highest. The experimental results show this method is correct and effective.