论文部分内容阅读
为了提高模拟电路故障诊断的识别率,提出一种基于人工鱼群优化支持向量机的模拟电路故障诊断方法。该方法用主成分分析对采集到的模拟电路故障信息进行降维,将提取后的故障特征输入支持向量机进行故障诊断,同时采用人工鱼群算法对支持向量机的惩罚参数和核函数参数进行优化,避免参数选择的盲目性,提高模型的诊断精度。通过对Sallen-Key带通滤波器电路的故障诊断仿真结果表明,该方法是有效的,与BP神经网络和传统SVM等方法的诊断结果相比较能够诊断更多的故障类别,并具有更高的故障诊断率。
In order to improve the recognition rate of analog circuit fault diagnosis, an artificial fish swarm optimization support vector machine based analog circuit fault diagnosis method is proposed. In this method, Principal Component Analysis (PCA) is used to reduce the fault information of the analog circuits. The extracted fault features are input to SVM for fault diagnosis. At the same time, the artificial fish swarm algorithm is used to calculate the penalty parameters and kernel function parameters of SVM Optimization, to avoid the blind selection of parameters, improve the diagnostic accuracy of the model. The simulation results of fault diagnosis of Sallen-Key band-pass filter show that this method is effective. Compared with the results of BP neural network and traditional SVM, it can diagnose more fault categories and have higher Fault diagnosis rate.