RBF神经网络及其在电力谐波测量中的应用研究

【作者】 郑一鹏；

【导师】 陈昌；

【作者基本信息】 大连理工大学 ， 控制理论与工程， 2000， 硕士

【摘要】 本文采用一种基于人工神经网络的谐波测量方法，利用模拟并行谐波测量装置的的基本原理，构造一个径向基函数神经网络（RBFNN），给出网络的训练算法、步骤以及阈值参数的选取原则。并根据电力系统中谐波的主要特点，分析和研究训练样本的形成方法，建立一个多输入多输出的网络结构，与应用反向传播算法（BP）训练的网络进行比较研究。仿真结果说明这一方法的可行性和有效性。更多还原

【Abstract】 Based on Artificial Neural Network(ANN), an approach for measuring harmonics is used in this paper. Using the basic principle of analog parallel harmonics measurement device, a Radial Basis Function Neural Network(RBFN7N) is built. In addition, the algorithm and the procedure for training the network and the principle of selection for threshold parameters are given. According to the primary characters of harmonics in power system, the forming method of training samples is analyzed and studied and a network with multi inputs and multi outputs is built. And the comparison between it and the network, which is trained by Back Propagation (BP) algorithm is studied. The simulation results illustrate the effectiveness and the feasibility of the presented approach.更多还原