基于HMM的轴承故障音频诊断方法研究-计算机应用技术专业论文.docxVIP

空匿丕堂亟±堂焦逾塞 空匿丕堂亟±堂焦逾塞 埴墨 于CGHMM算法。而使用本文的CGHMM改进算法，大大缩短了运 算时间，平均训练时间还不到DHM 方法的四分之一，诊断时间也 比DHMM方法少，诊断精度更是明显提高。 (5)在Visual C++7．0环境下，自主开发了基于HMM的轴承故 障音频诊断平台，并在此平台上完成实验研究，验证论文方法的有效 性，具有较好的应用推广价值。 关键词隐马尔可夫模型，轴承，故障诊断，音频信号 虫直盍堂亟±堂僮途塞 虫直盍堂亟±堂僮途塞 一——————————尘5迪盟 Abstract Rolling bearhlg is the most widespread application of。otatmg machinery，by the status of which，the normal performance of the wh01e machine is influenced directly．On the basis of combining theory with 0n Hidden Dractice，bearing fault diagnosis by acoustic signals based Markov Model(HMM)is researched systematically in this paper．The details are studied as follows： First，plentiful significant information about the operation status of bearings，which is potential for the fault diagnose after processed properly, b。 is contained in their acoustic signals．And acoustic signals can collected with non-contact sensors，SO as to be convenient and cheap· Therefore，taking acoustic signals as object，and according协e characteristics of MFCC with human auditory，the MeI-Frequ。ncy Cepstrum Coefficient(MFCC)is applied to the b。a。ing a。ou8‘1。819nal5 for the first time。 Second．because all parmneters are discrete values，Discrete Hidden Markov Model(DHMM)，simple model，fast speed，and high diagn08i8 accuracy is applied to the acoustic signals emitted by bearing in thl8 pap。。 Experiments resuIts prove that，with an average training time of 55·88， diagnosis time of 0．03 1 s，and diagnosis rate of 90％，the presented method iS effe‘ctive． Third．beeause output probability is described logically by Cornimlous Gaussian Mixture density function，Continuous Gaussian Mixture Hidden MarkoV Model(CGUMM)is applied tO the acoustic signals emitted by bearing in this paper．At the san2e time，the algorithms of training and diagnosis are improved by model parameter8 initialization based on based on clustering algorithm and forward-backward algorithm 生亩叁望亟±堂垃论塞