基于VMD的广义三次互相关管道泄漏定位检测OACSTPCD

Considering the problem of presence of relatively big error with second cross-correlation time delay estimation algorithm in the acoustic location technology for gas pipeline leak detection,a delay estimation algorithm based on variational mode decomposition(VMD)combined with generalized third cross-correlation was proposed.Firstly,VMD algorithm was used to decompose the two signals and reconstruct the signals.Secondly,a correlation was carried out again on the basis of the second cross-correlation,and Hilbert transform(HT)was introduced in the peak detection stage of the cross-correlation algorithm to sharpen the peak value,which becomes a new generalized third cross-correlation delay estimation method.Through the simulation experiment on the data collected by the oil and gas pipeline leakage detection system built for the platform,the accuracy of each algorithm was analyzed.Experimental results show that compared with second cross-correlation,the improved generalized third cross-correlation delay estimation algorithm has significantly improved positioning average accuracy,has higher accuracy and better anti-noise performance,and has a wider application prospect in gas pipeline leakage locating.