摘要
针对光伏组件中常用的最大功率跟踪方法存在的不足,将支持向量机用于预测光伏组件的最大功率点工作电压.支持向量机是一种新型的机器学习方法,该方法以结构风险最小化原则取代传统机器学习方法中的经验风险最小化原则,在小样本的机器学习中有着优异的性能.根据光伏组件的特点和最大功率点工作电压的影响因素,建立了支持向量机的最大功率点工作电压预测模型.实际仿真分析表明,与BP神经网络的模型相比,支持向量机的模型具有更高的预测精度.
As the traditional ways of Maximum Power Point Track(MPPT) suffers from some dissatisfaction,Support Vector Machines(SVM) is proposed to predict the output voltage of Maximum Power Point of PV module.SVM is a novel machine learning approach,which is based on the principle of structural risk minimization,unlike other traditional machine learning approach which is based on empirical risk minimization principle.SVM can perform well in Machine Learning with small sample.This paper presents a kind of SVM for the prediction and simulation of the voltage of the maximum power output in PV module.Compared with BP Neural Network,the simulation result of predict shows that the presented method based on SVM is more accurate.
出处
《西安工业大学学报》
CAS
2007年第4期371-375,共5页
Journal of Xi’an Technological University
关键词
支持向量机
光伏组件
最大功率点
最大功率跟踪
预测
support vector machine(SVM)
photovolatic module
maximum power point(MPP)
maximum power point track(MPPT)
predict