A Pipeline Leak Localization Method Based on the EEMD-HT Algorithm and the Leakage-Grading Resolution Strategy

Pipeline transportation plays a key role in energy security and economic development, and the accurate localization of leak points significantly reduces environmental pollution and economic losses. This paper proposes a pipeline leak localization method based on the ensemble empirical modal decomposition algorithm with the Hilbert spectrum analysis (EEMD-HT) and the leakage-grading resolution strategy. The method classifies the pipeline leakage level by adaptive pressure threshold and uses the acoustic signal as the basis for leak localization. At the same time, the leakage-grading resolution strategy is proposed for the leak localization problem in different leak conditions of pipelines to use the most applicable leak localization method for different leak classes. The method can improve the accuracy of the localization results, using the complementary nature of acoustic pressure signals and the high sensitivity of acoustic signals. This paper uses the Hilbert spectral analysis combined with the EEMD as a noise reduction tool. Firstly, the acoustic signal is decomposed by the EEMD to obtain a series of intrinsic mode functions (IMFs) and the residual terms; then the Hilbert spectral analysis is performed on the above components and residual terms to obtain the instantaneous energy spectrum of each component; finally, the original signal is reconstructed by analyzing the instantaneous energy change of each IMF and the residual at the signal inflection point and screening the effective components to complete the signal noise reduction. The experimental results show that the average localization error of this method is 6.75% in the small leakage condition of the short transmission pipeline and 0.34% and 0.16% in the medium condition and large leakage condition, respectively.