【摘要】 针对现有高速列车轨道电路故障诊断方法大多采用电压电流信号有效值,没有充分利用其丰富的频域信息的问题,为提高轨道电路故障诊断系统的性能,提出基于电压电流实时信号时频分析的故障诊断方法.基于短时傅里叶变换(Short-Time Fourier Transform,STFT)生成频谱特征;根据可分性准则进行特征频带选择;采用支持向量机(Support Vector Machine,SVM)进行故障特征分类.基于现场数据的试验结果表明该方法有效.更多还原

【Abstract】 Most of the existing fault diagnosis methods for track circuits are developed using root-meansquare values of voltage and current signals,but the plenty of frequency domain information is often ignored.To improve the performance of fault diagnosis system of track circuits,a fault diagnosis method is proposed on the basis of time-frequency analysis of real-time voltage and current signals.The frequency spectrum features are generated by Short-Time Fourier Transform( STFT),and the characteristic frequency bands are classified by Support Vector Machine( SVM).The test results based on the field data show that the method is effective.更多还原