Industrial equipment fault signal positioning and identification method based on microphone array

Abstract

The invention discloses an industrial equipment fault signal positioning and recognition method based on a microphone array, and the method comprises the steps: collecting sound signals, and dividingthe sound signals into a training set, a verification set and a test set; performing feature extraction on the sound signals in the training set, and extracting a phase spectrogram and an amplitude spectrogram of the spectrogram; sending the output of the feature extraction module to a convolutional neural network as an input, and learning translation invariance in the spectrogram by using a 2D CNN in each layer of CNN; between each layer of CNN, normalizing the output by using batch normalization, and reducing the dimension by using a maximum pooling layer along a frequency axis; transmittingthe output from the CNN layer to an RNN layer; using a linear activation function; and inputting the output of the full connection layer into two parallel full connection layer branches for fault identification and fault positioning respectively. According to the invention, a non-contact method based on audio signals is used, the limitation of environment and contact is smaller, fault positioningand fault identification are synchronously carried out, and the operation efficiency is improved.

Description

technical field [0001] The invention relates to a technology for locating and identifying industrial equipment fault signals, in particular to a method for locating and identifying industrial equipment fault signals based on a microphone array. Background technique [0002] With the increasing degree of modern automation and the rapid development of science and technology, all walks of life are increasingly trending toward intelligence, high speed and automation. In all kinds of mechanical equipment, not only the various parts of the equipment are closely related, but also different equipment are closely related to each other, eventually forming a series of complex production systems. In the production process, the advanced fault diagnosis technology is used to include the inspection and maintenance of mechanical equipment in the production process, so as to realize the predictability of mechanical equipment, which can not only prolong the service life of the equipment, but ...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the deficiencies in the prior art of industrial equipment fault signal location and identification of doped interference information leading to limited accuracy of fault signal location, the problem to be solved by the present invention is to provide a microphone array-based industrial equipment fault signal location and identification method, so that The location of the fault signal has a higher accuracy

Images