摘要
针对传统BP神经网络在模拟电路故障诊断中存在的不足,提出遗传算法和BP神经网络相结合的遗传神经网络模拟电路故障诊断方法;充分利用遗传算法全局、并行寻优的能力对BP神经网络的学习过程进行优化,防止神经网络训练时出现收敛速度慢和陷入局部极小等缺陷;在MATLAB平台上编程实现模拟电路故障诊断的仿真实验;仿真结果表明,相对于传统的BP神经网络算法,遗传神经网络算法不仅提高了网络训练收敛速度,而且提高了模拟电路故障诊断平均正确率,为模拟电路智能化诊断提供一种新的思路。
The traditional BP neural network in fault diagnosis of analog circuit has some defect, this paper put forward a fault diagnosis of analog drcuits method based on genetic algorithm and BP neural network, The proposed method fully uses genetic algorithm's global search and parallel optimization ability to optimize BP neural network learning process and prevent appearing the slow convergence speed and local minimum defect. In the MATLAB platform, the analog circuit fault diagnosis simulation experiments are carried out . The simulation results show that, compared with the traditional BP neural network algorithm, the proposed method not only improves the network training speed, but also improves the average correct rate of analog circuit fault diagnosis, it provides a new thought for intelligent diagnosis of r ana- log circuit.
出处
《计算机测量与控制》
CSCD
北大核心
2012年第11期2926-2928,2931,共4页
Computer Measurement &Control
关键词
模拟电路
故障诊断
遗传算法
神经网络
analog eircuit
fault diagnosis
genetic algorithm
neural networks