In electronic equipment fault diagnosis, waveform identification is an important basis of fault diagnosis. According to the requirements of real time monitoring and diagnosis system, a test waveform recognition method based on Zemike moments and support vector machine) SVM) is proposed. The features of test waveform are extracted based on Zemike moments, and the feature set is constructed through the feature selection based on Relief algorithm. At last, test waveform is identified by SVM. Experimental results show that the method can realize automatic identification of waveform in time domain, and has a better performance.%在电子设备故障诊断中,波形识别是进行故障诊断的重要依据.根据电子设备实时检测与诊断系统的需要,提出了一种基于Zernike矩和支持向量机的测试波形识别方法.采用基于区域的Zernike矩描述子提取测试波形的特征,通过Relief算法选择特征,构造特征集,运用支持向量机完成测试波形的识别.实验结果表明,该方法能够实现时域波形的自动识别,具有较高的识别性能.
展开▼