Support Vector Machine Based on Chaos Particle Swarm Optimization for Lightning Prediction

The learn accuracy and generalization ability of support vector machine (SVM) depend on a proper setting of its parameters to a great extent. An optimal selection approach of support vector machine parameters is proposed based on chaos particle swarm optimization (CPSO) algorithm. Then a lightning prediction model for Shapingba district of Chongqing based on support vector machine is established, and the optimal parameters of the model are searched by CPSO. The upper air data and the ground data of the model are collected from the Micaps system of the national weather service and the actual thunderstorm data are collected from the ground station of Shapingba from year 1998 to 2008. The results show that the proposed prediction model has better prediction results than neural network trained by particle swarm optimization and least squares support vector machine.