The Study Of Fault Diagnosis Method For Information Fusion In Pipeline Leakage Based On The Acoustic Characteristics

Pipeline transport plays an important role in the transport of oil, gas and other fluid. Once the pipeline fault occurs, it not only affects the normal transport of oil and gas, but also causes an explosion or fire accidents. At the same time, it poses a threat to human life and the country’s economy. Therefore, it is important in theoretical meaning and application value to study the fault diagnosis method for information fusion in the pipeline leakage based on the acoustic characteristics.In the paper, the fault acoustic signals of oil and gas pipeline is taken as the research object, and extraction method of pipeline fault feature is analyzed. According to the characteristics of pipeline fault signal, time-frequency analysis method is given, and the Hilbert transform method is selected to analyze the fault signal. Then an open-closing and close-opening mixed morphological filtering method is proposed to filter out noise in the acoustic signal, which realizes the signal preprocessing function. Aiming at the modal aliasing phenomenon of empirical mode decomposition, an improved time-frequency analysis method of empirical mode decomposition is proposed to achieve pipeline leak detection of acoustic signal.Through the simulation experiment, it shows that mixed morphological filtering method can realize signal preprocessing function. The improved empirical mode decomposition method can eliminate the modal aliasing phenomenon of empirical mode decomposition effectively and extract characteristic in time-frequency domain accurately. Thus, the fault diagnosis method for information fusion in the pipeline leakage based on the acoustic characteristics is studied, which provides a new way to solve pipeline fault diagnosis problem.