- 1
- 2
- 3
- 4
- 5
基于QGA优化Simple-MKL的模拟电路故障诊断方法
资料介绍
为解决模拟电路故障诊断中传统支持向量机(SVM)分类精度不高的问题,提出基于量子遗传算法(QGA)优化简单多核学习(Simple-MKL)支持向量机模型的模拟电路故障诊断方法。对电路进行瞬态分析,采用多分辨分析(MRA)提取电路故障特征,用量子遗传算法优化简单多核支持向量机中的正则化参数作为训练模型,用于模拟电路故障的诊断。仿真结果表明,本方法可实现模拟电路故障的精确分类。
In order to deal with the problem of accurate classification that exist in fault diagnosis of analog circuits by traditional support vector machine(SVM),a fault diagnosis method for analogy circuits is proposed based on the simple multiple kernel learning(Simple-MKL)SVM model which is optimized by using the quantum genetic algorithm(QGA).First of all,extract fault characteristics by using the multi-resolution analysis(MRA)based on the transient analysis of the circuits.Secondly,optimize the parameters of the Simple-MKL SVM model by using the QGA and train the model to diagnose the circuits.Simulation results show that the proposed method can effectively achieve the accurate classification of analog circuit faults.
部分文件列表
文件名 | 大小 |
基于QGA优化Simple-MKL的模拟电路故障诊断方法.pdf | 1M |
全部评论(0)