Automatic Recognition of Machinery Noise in the Working Environment

Abstract

A necessity for the suitable recognition of different machinery and equipment based on the sound they generate is constantly present and will increase in the future. The main motivation for the discrimination between different types of machinery sounds is to develop algorithms that can be used not only for final quality inspection but for the monitoring of the whole production line. The objective of our study is to recognize the operation of the individual machine in a production hall, where background noise level is high and constantly changing. An experimental plan was designed and performed in order to confirm the hypothesis proposing that automatic speech recognition algorithms can be applied to automatic machine recognition. The design of the automatic machine recognition procedure used in our study was divided into three stages: feature extraction, training, and recognition (classification). Additionally, a traditional mel-frequency cepstral coefficient (MFCC) procedure was adjusted for machinery noise by using different filter compositions. Finally, two classifiers were compared, the k-NN classifier and the multivariate Gaussian distribution. The results of the experiment show that machinery noise features frequency cepstral coefficients (FCC) should be extracted by using linear filter compositions and processed with recognition algorithm based on the multivariate Gaussian distribution.

Automated identification and assessment of environmental noise sources

“…In recent years, we have done a lot of research on microphone arrays and environmental noise measurement. We have tried to mimic the human ability of acoustic spatial filtering as much as possible, [42,[78][79][80][81]126,127]. The use of spatial domain offers great potential for better identification of individual noise sources and to better understand their temporal and spatial dynamics.…”

“…State-of-the-art approach to the identification, evaluation, and classification of environmental noise sources is presented in this paper. To mimic the human ability to spatially filter the acoustic environment (i.e., the Cocktail-Party effect), we have done much research and development in recent years in the area of microphone arrays for measuring environmental noise, [42,[78][79][80][81]126,127]. The implementation of the spatial domain to environmental noise measurements offers exceptional potential for more efficient identification of individual sources and clearer representation of the temporal and spatial dynamics of environmental noise.…”

Condition classification of heating systems valves based on acoustic features and machine learning

“…Combined vibro-acoustic and electrical measurements have been used for the implementation of a system for automatic quality monitoring of vacuum-cleaner motors [12]. Modified MFCC, combined with the k-nearest neighbour classifier and multivariate Gaussian distribution, have been applied in order to automatically recognise machinery noise in a working environment [13]. Gaussian mixture model based classification methods have also been shown to be competitive and efficient methods in the field of fault detection and condition monitoring [14].…”