舰船装备维修费用预测的PSO-LSSVM方法研究

Application of the PSO-LSSVM for the Weapon Equipment's Maintenance Cost Prediction

提出了一种基于粒子群优化的最小二乘支持向量机(PSO-LSSVM)模型的舰船装备维修费用预测方法,该方法利用PSO算法的收敛速度快和全局收敛能力,优化LSSVM模型的惩罚因子和核函数参数,避免了人为选择参数的盲目性,提高了LSSVM模型的预测精度。以某舰船装备维修费用为例进行实例验证,计算结果表明,这种方法比其他方法有更好的预测精度。

In order to improve the accuracy of weapon equipment＇s maintenance cost prediction, a least squares support vector machine （LSSVM） model optimized by the particle swarm optimization （PSO） is proposed in this paper. Optimizing two parameters of LSSVM model by PSO abilities of the fast convergence and whole optimization, thus avoiding the blindness of man-made choice, the LSSVM-PSO model can enhance the capability of forecasting. An example of the prediction of weapon equipment＇s maintenance costs is given, and the result shows that the method can bring less error and better precision compared with other methods.