一种新的粒子群算法优化支持向量机的短期负荷预测  

A new power load chaotic predicting based on support vector machine and particle swarm optimization algorithm

作　　者：陶琳[1] 岳小冰[1] 

机构地区：[1]河南工业职业技术学院电子信息工程系,河南南阳473000

出　　处：《电子设计工程》2016年第16期151-154,共4页Electronic Design Engineering

基　　金：河南省科技攻关项目(142102210368)

摘　　要：通过研究电力负荷预测中支持向量机的参数优化问题,将改进后新的粒子群算法导入支持向量机参数中,从而建立一种新的电力负荷预测模型(IPSO-SVM)。首先将支持向量机参数编码为粒子初始位置向量,然后通过对粒子个体之间信息交流、协作的分析找到支持向量机的最优参数,并针对标准粒子群算法的缺陷进行一定的改进,从而应用于电力负荷的建模与预测,最后通过仿真对比实验来测试它的性能。实验结果表明,这种新的电力负荷预测模型能够获得较高精度的电力负荷预测结果,大大减少了训练时间,能够满足电力负荷在线预测要求。By studying the parameter optimization of support vector machine in power load forecasting,the new particle swarm algorithm is introduced into the support vector machine parameters,and a new power load forecasting model （IPSO-SVM） is established. Firstly,support vector machine parameters encoding as the initial position vector,and then through the information exchange between particles and the collaborative analysis to find the optimal parameters of the support vector machine, and for the standard particle swarm algorithm to improve the defect of the standard particle swarm algorithm, and thus applied to the power of negative load modeling and forecasting, and finally to test its performance by simulation comparison experiments. Experimental results show that this new power load forecasting model can get high accuracy of the load forecasting results, greatly reducing the training time, can meet the requirements of power load online forecasting.

关 键 词：粒子群优化算法 电力负荷预测模型 支持向量机 混沌理论 

分 类 号：TN98[电子电信—信息与通信工程]