论文部分内容阅读
针对目前轴承在线式监测与故障检测系统价格昂贵,安装调试及维护困难而不适合中小型设备故障诊断的问题,该文提出了DSP环境下的轴承故障DHMM音频信号诊断方法,利用了音频信号采集方便、成本低廉,不会破坏设备原有机械结构的特点,而且首次将DHMM方法和音频信号方法和具有强大数据处理能力的DSPs相结合,应用于轴承故障检测与诊断,并对MFCC特征参数的提取和DHMM模型的训练方法作出了改进.实验结果表明效果良好,切实可行,具有很好的应用前景.“,”At present,the on line equipment fault diagnosis system is not suitable for fault detection and diagnosis of small and mediumsized equipment, because of its high cost and its inconvenience of installation and maintenance. The paper ,basing on audio signal and the DSP platform ,adopts DHMM method to reseach,utilizing the convience and low costs of audio signal information collected ,not destroying the equipment structures.And the article firstly applys the methods of DHMM,audio signal and DSPs to the sysytem,improving the abstract of MFCC and model training of DHMM.The experimental results prove that it is successful, feasible and has a great prospect.