基于双参数最小二乘支持向量机(TPA-LSSVM)的风电时间序列预测模型的优化研究  

Optimization of a wind power time series prediction model based on a two-parameter least squares support vector machine

作　　者：刘云[1] 易松 LIU Yun;YI Song(Faculty of Informalion Engineering and Automation,Kunming University of Science and Technology,Kunming 650500,China)

机构地区：[1]昆明理工大学信息工程与自动化学院,昆明650500

出　　处：《北京化工大学学报（自然科学版）》2019年第2期97-102,共6页Journal of Beijing University of Chemical Technology(Natural Science Edition)

基　　金：国家自然基金(61262040)

摘　　要：风电时间序列预测模型的优劣直接影响风功率的应用价值,最小二乘支持向量机(least squares support vector machine,LSSVM)在处理风电预测问题上具有明显优势。提出了一种双参数算法(two-parameter algorithm,TPA),从理论上证明了任意初始值均可线性收敛到全局最优值。调用TPA算法对LSSVM模型的惩罚因子和径向基宽度进行寻优赋值,并将训练好的TPA-LSSVM模型应用于风电预测中。仿真结果表明,与LSSVM模型、粒子群最小二乘支持向量机(PSO-LSSVM)模型、径向基函数神经网络(RBFNN)模型相比,TPA算法可以更好地实现LSSVM的参数寻优,TPA-LSSVM模型能有效提高预测精度。The advantages and disadvantages of the wind power time series prediction model directly affect the application value of wind power.The least squares support vector machine(LSSVM) has obvious advantages in dealing with wind power prediction.This paper proposes a two-parameter algorithm(TPA),which shows that any initial value can converge linearly to the global optimal value.The two-parameter(TPA) algorithm is employed to optimize the penalty factor and radial base width of the LSSVM model,and the trained TPA-LSSVM model is then applied to wind power prediction.The simulation results show that compared with the LSSVM model,the particle swarm least squares support vector machine(PSO-LSSVM) model and the radial basis function neural network(RBFNN) model,the TPA algorithm can better realize the parameter optimization of LSSVM,and thus the TPA-LSSVM model can effectively improve the prediction accuracy.

关 键 词：最小二乘支持向量机 时间序列预测 双参数算法 参数寻优 

分 类 号：TN929.5[电子电信—通信与信息系统]