The use of artificial neural network models in the acoustic diagnostics of multi-shaft gear drives

Abstract

The article considers the possibility of diagnosing a multi-shaft gear mechanical system based on the analysis of an acoustic signal using artificial neural network models on the example of the speed box of the SN-501 lathe. A sufficiently high efficiency and accuracy of detecting a local defect of the gear wheel in conditions of high acoustic activity of all components of the drive when monitoring its condition is shown.