A Fault Diagnosis Method of Belt Conveyor Based on Sound Signal

Abstract

The invention provides a belt conveyor fault diagnosis method based on sound signals, which can reduce the labor intensity of inspection personnel and has the characteristics of fast detection speed, high real-time performance, strong safety, etc.; it includes the following steps: S1, Collect the sound signals of the belt conveyor; S2, perform improved wavelet threshold denoising processing on the collected sound signals; S3, perform MFCC and deep learning feature extraction on the noise-reduced sound signals of the belt conveyor; S4, establish support The vector machine classifies the model and forms the trained SVM model; S5. Put the extracted feature information data into the trained SVM model to obtain the posterior probability, then use D-S evidence theory to perform decision-level fusion, and finally use the fusion output results It matches the operating status of the belt conveyor under the known operating status of SVM, and the one with the highest matching degree with the fusion output result corresponds to the current operating status of the belt conveyor, thus completing the fault diagnosis of the belt conveyor.

Description

Technical field [0001] The invention relates to the technical field of belt conveyor fault diagnosis, specifically a belt conveyor fault diagnosis method based on sound signals. Background technique [0002] With the rapid development of science and technology, industrial production is increasingly modernized, and various highly intelligent and highly integrated large-scale mechanical equipment are gradually emerging. In industries such as ports, mining, and coal, the production and operation of belt conveyors has the characteristics of large usage, difficulty in inspection, and difficulty in predicting failures. After on-site investigation of the port production site, it was found that due to its large production and transportation throughput, the belt conveyor It requires long-term and high-load operation, and fault events that cannot be detected in time by manual inspections often occur. Based on this production pain point, research on belt conveyor fault diagnosis techno...

AI Technical Summary

Problems solved by technology

[0003] At present, the belt conveyor adopts the traditional manual inspection method. The inspectors need to carry heavy inspection tools to shuttle around the site, which greatly increases the labor risk of the inspectors. Moreover, there are many fault detection points on the belt conveyor. High precision requirements, this inspection method makes it difficult for the inspection work of the belt conveyor to achieve fault detection characteristics such as fast detection speed, high real-time performance, and strong safety

Images