| 基于群智能神经网络的线性直流电源故障诊断 |
| |
| 引用本文： | 王家林,吴正国,杨宣访.基于群智能神经网络的线性直流电源故障诊断[J].微电子学,2008,38(6). |
| |
| 作者姓名： | 王家林 吴正国 杨宣访 |
| |
| 作者单位： | 海军工程大学,电气与信息工程学院,武汉,430033 |
| |
| 基金项目： | 国家自然科学基金资助项目
|
| |
| 摘 要： | 反向传播算法是应用广泛的一种多层前馈神经网络模型,具有求解精度低、易于陷入局部极小值的缺点.群智能研究领域主要有粒子群优化算法和蚁群算法.粒子群优化算法有收敛速度快、算法参数简洁等特性;蚁群算法具有正反馈、启发性收敛等特性.将群智能神经网络的方法应用于线性直流电源的故障诊断:利用蚁群算法来约简故障特征参数;用粒子群优化算法来训练神经网络的权值.实验表明:此方法提高了网络训练效率和故障定位准确性.
|
| 关 键 词： | 群智能算法 神经网络 线性直流电源 故障诊断 |
Fault Diagnosis of Linear DC Power Supply Based on Swarm Intelligence Algorithm and Neural Network |
| |
| WANG Jia-lin,WU Zheng-guo,YANG Xuan-fang.Fault Diagnosis of Linear DC Power Supply Based on Swarm Intelligence Algorithm and Neural Network[J].Microelectronics,2008,38(6). |
| |
| Authors: | WANG Jia-lin WU Zheng-guo YANG Xuan-fang |
| |
| Affiliation: | WANG Jialin WU Zhengguo YANG Xuanfang (College of Electrical , Information Engineering,Naval University of Engineering,Wuhan 430033,P.R.China) |
| |
| Abstract: | Back propagation is a widely used feedback neural network,which has disadvantages such as low-precision solution and easy convergency to local minimum points.Swarm intelligence algorithm mainly includes particle swarm optimization(PSO)algorithm and ant colony algorithm.The PSO algorithm has the advantages of rapid convergency and terse parameters,while the ant colony algorithm features positive feedback and heuristic convergency. A method for fault diagnosis of linear DC power supply circuits based on swarm... |
| |
| Keywords: | Swarm intelligence algorithm Neural network Linear DC power supply Fault diagnosis |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
