Experimental study on leak location detection of directly buried hot water heating pipeline

In this paper, the Fourier filtering method and wavelet threshold method are used to make a comparative study of signal noise reduction and leak location. The leak detection experiment is carried out on the directly buried hot water heating system of the DN300 pipeline. The small diameter branch pipe and the regulating valve are used to simulate the leak point, and the acceleration sensor is used to measure the acoustic vibration signal. Two filtering methods are selected successively to reduce the noise of acoustic vibration signal of the directly buried hot water heating pipeline leak. The frequency domain characteristics and the location accuracy of the leak signal before and after denoising are discussed. Acoustic noise reduction leak detection engineering verification is carried out in a directly buried heating pipeline in a residential area of Beijing, and the positioning deviation is less than 1.0 m. The results show that the Fourier filtering method and the wavelet threshold method have a remarkable effect on acoustic vibration signal noise reduction for leak location of the directly buried hot water heating pipeline. The wavelet threshold method keeps the signal characteristics of the whole band and can obtain higher positioning accuracy.