Abnormal sound automatic detection method and device and computer readable storage medium

Abstract

The invention relates to an abnormal sound automatic detection method. The method comprises the following steps: S1, acquiring an original sound signal sent by a to-be-detected sample during normal work; s2, filtering the original sound signal to suppress noise interference and enhance a useful signal; s3, performing feature extraction on the useful signal to obtain an energy entropy, taking the energy entropy as an acoustic feature of the to-be-detected sample, and calculating the size of the energy entropy; and S4, matching the energy entropy of the to-be-detected sample with the threshold interval of the known abnormal sound type so as to judge the abnormal sound type of the to-be-detected sample. According to the method, the abnormal sound of the industrial product can be automatically detected, the abnormal sound type of the to-be-detected sample can be automatically judged, the defects of a traditional human ear detection method can be avoided, and the detection accuracy can reach a relatively high level.

Description

technical field [0001] The invention relates to the technical field of abnormal sound detection of industrial products, in particular to an automatic abnormal sound detection method, equipment and computer-readable storage medium. Background technique [0002] Products manufactured in the industry need to undergo various tests before leaving the factory to ensure that the product performance is qualified and to prevent unqualified products from flowing to customers. For some types of products, abnormal noise is one of the product quality evaluation indicators. Abnormal noise can reflect defects in product structure and help manufacturers analyze and solve product problems. Therefore, abnormal sound detection is of great significance. [0003] At present, the abnormal sound detection methods adopted by various manufacturers are basically the traditional method of listening directly to human ears. The method of listening with the human ear has the following disadvantages: (...

AI Technical Summary

Problems solved by technology

The method of listening with the human ear has the following disadvantages: (1) Human hearing is subjective and may affect judgment; (2) There is a lack of quantitative indicators for the definition of abnormal noise; different people have different judgment criteria , even for the same person at different times of the day; (3) long-term listening can lead to hearing loss or even deafness

Therefore, the existing method of detecting abnormal noise by human ear is difficult to meet the abnormal noise detection requirements of industrial products.

Images