基于HHT与神经网络的旋转机械故障诊断研究

发布时间：2018-01-10 10:32

本文关键词：基于HHT与神经网络的旋转机械故障诊断研究　出处：《南京航空航天大学》2012年硕士论文　论文类型：学位论文

【摘要】：旋转机械是航空、电力、化工等众多领域的关键设备，所以对其进行故障诊断研究具有重要的现实意义。而随着振动检测和信号处理等相关技术的不断发展，以振动信号检测、处理和分析为基础的故障诊断技术已成为故障诊断领域一个重要的研究方向，，同时神经网络、遗传算法等理论的发展也为故障诊断技术的研究和应用开辟了一条崭新的途径。本文详细介绍了Hilbert-Huang变换（简称HHT）方法以及神经网络等相关内容。一方面，介绍了HHT方法的基本原理和实现过程，并分析了该方法存在的端点效应和虚假模态问题；另一方面，介绍了BP神经网络和遗传算法的基本理论，并研究了遗传算法优化BP网络的过程，即针对BP网络的不足采用遗传算法进行优化。同时，为研究旋转机械的故障诊断问题，采用多功能转子试验台模拟旋转机械的常见故障，并运用HHT方法对各故障信号进行处理和分析，在此基础上，利用模糊熵能够表示信号复杂程度且具有相对稳定性等特点，将模糊熵理论引入到故障诊断领域，并提出了一种基于EMD和模糊熵相结合的特征向量提取方法，同时将它用于转子故障的特征提取中，证明了该特征提取方法的可行性和有效性。最后，综合地运用HHT方法和经遗传算法优化的BP神经网络进行转子系统的故障诊断研究。提取转子系统常见故障的特征量，再将该特征量输入到经遗传算法优化的BP网络模型中进行故障诊断，结果表明上述方法应用在转子系统故障诊断中能够取得较好的效果。
[Abstract]:Rotating machinery is the key equipment in many fields, such as aviation, electric power, chemical industry and so on. Therefore, it is of great practical significance to study the fault diagnosis of rotating machinery. However, with the development of vibration detection and signal processing and other related technologies. Fault diagnosis technology based on vibration signal detection, processing and analysis has become an important research direction in the field of fault diagnosis, and neural network. The development of genetic algorithm also opens a new way for the research and application of fault diagnosis technology. In this paper, the Hilbert-Huang transform method and neural network are introduced in detail. On the one hand, the basic principle and implementation process of HHT method are introduced. The endpoint effect and the false modal problem of the method are analyzed. On the other hand, the basic theory of BP neural network and genetic algorithm is introduced, and the process of optimizing BP network by genetic algorithm is studied. At the same time, in order to study the fault diagnosis of rotating machinery, the common faults of rotating machinery are simulated by multi-function rotor test-bed, and each fault signal is processed and analyzed by using HHT method. The fuzzy entropy theory is introduced into the field of fault diagnosis by using the characteristics of fuzzy entropy which can express the signal complexity and has relative stability. A feature vector extraction method based on the combination of EMD and fuzzy entropy is proposed and applied to the feature extraction of rotor faults. The feasibility and effectiveness of the feature extraction method are proved. Finally, the fault diagnosis of rotor system is studied by using HHT method and BP neural network optimized by genetic algorithm, and the characteristic quantity of common faults of rotor system is extracted. Then the feature is input into the BP neural network model optimized by genetic algorithm for fault diagnosis. The results show that the above method can achieve good results in rotor system fault diagnosis.
【学位授予单位】：南京航空航天大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH165.3

【参考文献】