Localization Method of Abnormal Sound Source in Substation Based on Maximum Likelihood Estimation

The audible signal of substation power equipment can reflect its operation status. The location and analysis of the audible signal is one of the effective means to realize the operation status evaluation and fault diagnosis of equipment. For this reason, based on the non-contact sound sensor array, from the perspective of statistical analysis, the maximum likelihood estimation method is proposed to locate the abnormal sound source of power equipment in the whole station. In addition, the mutation particle swarm optimization algorithm is used to reduce the amount of computation, and the equipment fault sound source is located efficiently and accurately in low signal-to-noise ratio environment. Simulation analysis and laboratory test show that the positioning error of the proposed method is about 30% lower than that of MUSIC method in low signal-to-noise ratio environment, which can quickly and effectively locate the abnormal sound source of equipment, and provide reference for accurate fault location and diagnosis of subsequent equipment.