Localization of multiple leaks in a fluid pipeline based on ultrasound velocity and improved GWO

Abstract It is difficult to effectively locate the position of multiple leaks in fluid pipelines; therefore, a localization method for multiple leak positions is proposed based on ultrasound velocity and improved grey wolf optimization algorithm (GWO). First, the mathematical relationship between ultrasound velocity and the pressure signal inside a pipeline is established. Second, the ultrasound velocity is decomposed by local mean decomposition (LMD), and the inflection point of the ultrasound velocity by using a wavelet transform after denoising is extracted. Then, the simulated annealing GWO (SAGWO) method is proposed to improve the performance of GWO which is easy to converge to local optimum. Finally, the inflection time of the ultrasound velocity is obtained at the ends of a pipeline with multiple leaks, and an objective function is established to estimate the localization of the multiple leaks by SAGWO. The field experiment demonstrates that the proposed method can effectively and accurately locate multiple leak positions in a fluid pipeline.