Pipeline signal feature extraction method based on multi-feature entropy fusion and local linear embedding

This paper considers the problem of effective feature extraction acoustic signals from oil and gas pipelines under different working conditions. A pipeline leakage detection method is proposed based on multi-feature entropy fusion local linear embedding (LLE). First, seven kinds commonly used which can reflect characteristics signal better are extracted through experiments, including permutation entropy, envelope approximate fuzzy energy sample dispersion entropy. The seven-dimensional vectors obtained by fusion. Second, LLE algorithm to reduce dimension vector complete secondary extraction. Finally, support machine (SVM) identify conditions pipeline. experimental results show that, compared with other dimensionality reduction methods, single-feature method, types effectively problems false negatives positives in detection.