低速重载齿轮箱声发射信号滤波消噪方法的研究-机械设计及理论专业论文.docxVIP

万方数据 万方数据 Abstract Gearbox is the core equipment in mechanical transmission, and the low-speed heavy-duty gearbox of which working speed is low, operation condition is typical intermittent and bears a larger impact load, whose condition monitoring and fault diagnosis has been a hotspot in research area. As a new real-time detection technology, the acoustic emission (AE) testing has the advantages of high sensitivity, and more easy to extract characteristic information of running status of equipment, application in the fault diagnosis of low-speed heavy-load gearbox. But the uncertainty of AE signals and the diversity of AE source about low-speed heavy-duty gearbox, the AE signals are susceptible to be interfered by noise and be embedded in background noise, so it is of great significance to do research on how to eliminate noise interference and extract useful information. This subject tries to study the noise eliminating technology on the low speed and heavy-duty gearbox about AE signals, and the results obtained are shown as follows: According to the structure and failure mechanism of the low speed and heavy-duty gearbox and characteristics of AE testing technology. Miter gate mechanical hoist horizontal reducer as a research object, was designed sensor layout and collected AE signals. According to the research results, the state detection and analysis technology was formed about the low speed and heavy-duty gearbox. Based on the theoretical analysis of multiple-scale variable step size LMS adaptive filtering and lifting wavelet, a modified lifting wavelet variable step size LMS adaptive filtering algorithm was proposed. Then collection AE signals were filtered and denoised respectively by multi-scale variable step size LMS adaptive filtering, lifting wavelet and the improved lifting wavelet- variable step size LMS adaptive filtering algorithm According to the characteristics of particle filter technology, AR model coefficients were seen as the state equation coeff