【作者】 杨俊；

【导师】 吴建华；

【作者基本信息】 浙江大学 ， 电机与电器， 2007， 硕士

【摘要】 随着电机在工业、农业等领域的广泛应用，如何测试、分析和抑制电机振动和噪声，越来越受到人们的广泛关注。虚拟仪器技术，相比于传统仪器拥有性能高、扩展性强等优点，在工程测试等领域得到越来越广泛的应用。因此，结合虚拟仪器技术，建立电机噪声和振动的测试分析系统是一种可行的解决途径。本文将虚拟仪器技术应用于电机的噪声和振动问题，建立了基于虚拟仪器的电机噪声振动测试分析系统。全文主要研究工作分为三部分：前两部分分别研究了系统的硬件和软件组成，建立了完整的硬件和软件系统；第三部分进行了噪声振动实验研究，验证了系统的正确性和有效性。本文的主要研究内容如下：1．硬件部分。探讨了系统的硬件组成，建立了以传感器、信号调理电路和数据采集卡为核心的测试系统。系统硬件部分是正确采集电机噪声和振动信号的关键，是测试分析的基础。2．软件部分。用LabVIEW虚拟仪器编程语言完成了软件部分的设计，实现了信号采集、显示、处理、诊断、打印报告等一系列功能。针对电机噪声振动的复杂性，建立了以快速傅里叶变换、功率谱函数分析、分数倍频谱分析、小波分析等信号处理方法为核心的信号分析处理功能，并用最小二乘支持向量机实现了电机故障诊断功能。3．实验研究。实验验证了系统的信号采集、信号分析和故障诊断的正确性。构造三类电机故障，实验研究了采用最小二乘支持向量机进行故障诊断的有效性。在总结全文的基础上，提出了该电机噪声和振动测试分析系统有待深入研究的若干问题。更多还原

【Abstract】 The requirement to measure, analyze and reduce the noise and vibration of electrical machines involves great concern around the world as the electrical machines are more and more widely used in industry, agriculture and other fields. And compared with traditional instrument technology, virtual instrument technology is widely used in engineering measurement because of many advantages, such as better performance and strong expansibility. Thus, it is an available solution to develop a system to measure and analysis noise and vibration of electrical machines using virtual instrument technology.In this dissertation a measurement and analysis system of noise and vibration of electrical machines is developed using virtual instrument technology. The main works can be divided into three parts. The components of the hardware and software of the system are discussed in the first and second part. And three experiment are designed to verify the correctness and validity of the system. Following are the main works of this dissertation.1. The hardware. The components of the hardware are discussed, and a system including sensors, signal conditioning circuit and data acquisition card is established. It is the foundation of the whole system, and is used to acquire the noise and vibration signals of the electrical machines.2. The software. A software is developed using the virtual instrument programming language LabVIEW, and several functions are realized including signal acquisition, signal display, signal processing, fault diagnosis, report generation. Four signal processing tools are studied, including fast fourier transformation, power spectrum analysis, fractional octave analysis, wavelet transform analysis, and their application in noise and vibration signal processing is conducted. SVM in electrical machines fault diagnosis is studied, And a Least Square-Support Vector Machine(LS-SVM) based on fault diagnosis system is designed.3. Experiments research. The correctness of the signal acquisition, signal processing and the fault diagnosis is verified,, and the validity to fault diagnosis is studied using LS-SVM.At last, the dissertation concludes with a summary and perspectives of future research of the measurement and analysis system.更多还原