基于混沌神经网络的供配电系统故障诊断  

Fault Diagnosis of Power Supply Based on Chaotic Neural Network

作　　者：吴凡[1] 张志利[1] 颜宁[2] 

机构地区：[1]第二炮兵工程学院,陕西西安710025 [2]第二炮兵装备研究院,北京100085

出　　处：《兵工自动化》2005年第2期55-56,62,共3页Ordnance Industry Automation

摘　　要：基于混沌神经网络的供配电系统故障诊断,采用引入动量项和混沌映射的改进BP算法。先分析系统典型故障,建立典型网络模型。在BP算法中加入动量项和混沌映射,选择神经网络初值。再进行学习训练,分别给训练后的子网络输入现场采集的装备数据,通过网络直接获得故障诊断结果。The improved BP algorithm added momentum item and chaotic mapping was adopted in fault diagnosis of power supply system based on chaotic neural network. At first, the typical fault of the system was analyzed, and the typical network model was established. The momentum item and chaotic mapping was added into BP algorithm, and the initial value of the network was selected. The network was learned and trained using improved BP algorithm, the data of device gathered in field was inputted into trained sub-network respectively. Fault diagnosis results are directly achieved through the network.

关 键 词：混沌 神经网络 供配电系统 故障诊断 

分 类 号：TP274.5[自动化与计算机技术—检测技术与自动化装置] TP183[自动化与计算机技术—控制科学与工程]