基于WPEE-RF的模拟电路故障诊断

为实现高效的模拟电路故障诊断,提出了基于小波包能量熵(WPEE)和随机森林(RF)的模拟电路故障诊断方法;选择合适的测试激励信号,监测电路收集数据,对模拟电路监测数据进行5层小波包分解,计算多频带WPEE向量表征故障特征,由RF分类器实现故障诊断;仿真实验结果表明,该方法在双二次滤波电路、Sallen-key滤波电路容差故障诊断以及对数放大器综合故障诊断中体现出良好的性能,故障诊断准确率达99%以上,且该方法具有参数鲁棒性,RF模型训练时间短;较支持向量机和BP网络方法相比,表现出更好的综合性能,更能贴近工程实践应用。

Analog Circuit Fault Diagnosis Based on WPEE-RF

In order to achieve high-performance fault diagnosis of analog circuit, an analog circuit fault diagnosis method based on Wavelet Packet Energy Entropy(WPEE) and Random Forest(RF) is proposed. Select the appropriate test excitation signal, perform five-layer wavelet packet decomposition on the monitoring data, calculate the multi-band WPEE vector to characterize the fault features, and implement the fault diagnosis by RF classifier. The experimental results show that this method has good performance in the tolerance fault diagnosis of the biquadratic filter circuit, the Sallen-key filter circuit and the integrated fault diagnosis of the logarithmic amplifier circuit. The fault diagnosis accuracy in the experiment is more than 99%. This method has good parameter robustness, and the RF model has short training time. Compare with the method based on support vector machine and BP network, it shows better comprehensive performance and is more applicable to practice applications.