机构地区: 华南理工大学土木与交通学院
出 处: 《太阳能学报》 2010年第12期1668-1672,共5页
摘 要: 建立了基于开关磁阻发电机的波浪发电系统以及SRG的磁链特性和转矩特性模型,分别用Levenberg-Mar-quardt算法和弹性BP算法进行训练,比较了BP神经网络的几种常用训练算法的性能;模型的泛化性能用数据插值方法进行了检验。建模效果及模型泛化能力试验表明,BP神经网络建立的开关磁阻发电机模型性能优越。 A typical wave power system using SRG was set up, and the novel flux characteristics model and torque characteristics model of SRG was set up too. The models were trained by the Levenberg-Marquardt algorithm and the elastic BP algorithm. During the training course, the performance of several commonly-used BP neural networks training algorithm was compared. And the generalization ability of the SRG model was verified by an interpolation test. It have shown that the neural network is good for SRG modeling.
领 域: [电气工程]