【摘要】 将模糊神经网络应用于单相自适应重合闸中,以模糊理论和人工神经网络理论为基础构造了一个多输入模糊神经网络,用于识别瞬时性故障与永久性故障。该网络以取大取小运算部分代替了乘积求和运算,并采用了从样本中获取模糊规则的方法。利用Matlab进行了大量仿真实验,验证了该方法的可行性与准确性;在仿真的基础上,将多输入模糊神经网络与BP神经网络进行了比较,证明了多输入模糊神经网络在单相自适应重合闸中应用的优越性。更多还原

【Abstract】 Applying fuzzy neural network to single phase adaptive auto-reclosure and based on fuzzy theory and artificial neural network (ANN) a multi-input fuzzy neural network is built to distinguish transient faults from permanent faults. In this fuzzy neural network the summation of products are partly replaced by the operations of fetching maximum or minimum and the method of gaining fuzzy rules from samples is used. A lot of simulation is carried out by use of Matlab and the feasibility and accuracy of the proposed method are verified. On the basis of the simulation results the multi-input fuzzy neural network is compared with BP neural network and the superiority of applying multi-input fuzzy neural network to single phase adaptive auto-reclosure is proved.更多还原