MEMS Array Voiceprint Imaging Algorithm and System Design

As the key equipment of the electric power supply, safe and stable operation of power transformers have a significant impact on the operation of a large-scale electric power supply. The voiceprint monitoring based on MEMS sensor arrays can detect and diagnose the overall mechanical faults of power transformers which can be installed externally. It could be widely applied in online monitoring of transformer operation status. This article designs the MEMS sensor array algorithm, which is called Beam-Forming Algorithm, to detect the voiceprint directions and voice types. And further more, AI LSMT algorithm is used to analyze the signals of MEMS sensor arrays which is in time sequence. With the industrial voiceprint databases, LSMT algorithm can correctly distinct the power transformer sub-optimal health working status. The experiments show that with BFA and AI LSMT algorithm, abnormal working conditions can effectively be detected and monitored. Therefor, this MEMS array based voiceprint imaging equipment can provide early warning effectively and timely, which has much important application value for electric power supply.