摘要
为提高对模拟电路故障模式的准确分类和减少网络模型的训练时间,提出基于小波包变换(WPT)和果蝇算法(FOA)优化广义回归神经网络(GRNN)的模拟电路故障诊断方法。首先采用小波包变换提取电路优质故障特征,以减少网络训练时间,然后建立GRNN网络模型,选择FOA算法优化GRNN网络参数,构建最优模型对电路故障特征进行训练测试,最后采用仿真测试其性能。实验结果表明,FOA算法有效提高诊断模型训练效率,相比于其它电路故障诊断模型,FOAGRNN模型具有更高的诊断率和优越性。
With the purpose of improving the accurate classification of analog circuit failure mode and reduce the training time of network model,an analog circuit fault diagnosis method based on wavelet packet transform(WPT)and Fruit fly optimization algorithm(FOA)generalized regression neural network(GRNN)was proposed.Firstly,WPT was used to extract high quality fault features and reduce the network training time.Then,GRNN network mod-el was established,the FOA algorithm was selected to optimize the GRNN network parameters.amd the optimal mod-el was constructed to train and test the circuit fault characteristics.Finally,the simulation test was used to test its performance.The experimental results show that the FOA algorithm effectively improves the training efficiency of di-agnostic model,and compared with other circuit fault diagnosis models,the FOAGRNN model has higher diagnostic rate and superiority.
作者
郭庆
张文斌
苏海涛
GUO Qing;ZHANG Wen-bin;SU Hai-tao(School of Electronic Engineering and Automation,Guilin University of Electronic Technology,Guilin Guangxi 541004,China)
出处
《计算机仿真》
北大核心
2020年第1期355-359,共5页
Computer Simulation
基金
桂林市科学研究与技术开发计划项目(2016010404-3)
广西自然科学青年基金(.2016GXNSFBA380117)
厦门大学水声通信与海洋信息技术教育部重点实验室开放课题(201601)。
关键词
果蝇优化算法
广义回归神经网络
小波包变换
故障诊断
模拟电路
Fly algorithm optimization(FOA)
Generalized regression neural network(GRNN)
Wavelet packet transform(WPT)
Fault diagnosis
Analog circuit
