Abstract
In this paper, we propose to combine the intercorrelation method and the Multiple Signal Classification algorithm (MUSIC) to locate a leak on a gas pipeline using acoustic sensors. We compare the Root Mean Square Error (RMSE) of the leak position estimate using hybrid localization based on a combination of MUSIC and intercorrelation method to the MUSIC algorithm and the intercorrelation method using two acoustic sensors and a pipeline of length 100m. At average Signal to Noise Ratio equal to 0 dB, the RMSE of the leak position estimate is equal to 0.88 m for the hybrid algorithm while it is 1.6 m for the MUSIC algorithm and 7.4 m for the intercorrelation method. The intercorrelation method, the hybrid, and the MUSIC algorithms are unbiased as the RMSE converges to zero at high SNR. (C) 2021 INT TRANS J ENG MANAG SCI TECH.