{"title":"Vibration Feature Extraction Using Audio Spectrum Analyzer Based Machine Learning","authors":"Jyun-Shun Liang, Kerwin Wang","doi":"10.1109/icice.2017.8479273","DOIUrl":null,"url":null,"abstract":"To develop a tool to identify different mechanical vibrations, this paper presents a novel instrument to extract the vibration features of rotating machinery. The instrument consists of an audio spectrum analyzer, a signal processing circuit and a single-board computer. This architecture offers a cost-effective solution for machine status monitoring and analyzing. The experiment results show that the training data collected from audio spectrum analyzer can work well with both KNN and SVM methods to construct accurate machine-learning models with the 95.8% and 97.2% accuracy respectively.","PeriodicalId":233396,"journal":{"name":"2017 International Conference on Information, Communication and Engineering (ICICE)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2017-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"7","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2017 International Conference on Information, Communication and Engineering (ICICE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/icice.2017.8479273","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 7
Abstract
To develop a tool to identify different mechanical vibrations, this paper presents a novel instrument to extract the vibration features of rotating machinery. The instrument consists of an audio spectrum analyzer, a signal processing circuit and a single-board computer. This architecture offers a cost-effective solution for machine status monitoring and analyzing. The experiment results show that the training data collected from audio spectrum analyzer can work well with both KNN and SVM methods to construct accurate machine-learning models with the 95.8% and 97.2% accuracy respectively.