基于约束独立分量分析的齿轮箱故障诊断研究

发布时间：2018-02-15 00:55

本文关键词： 独立分量分析约束独立分量分析故障诊断齿轮箱矿用提升机　出处：《河南理工大学》2015年硕士论文　论文类型：学位论文

【摘要】：随着现代工业科技的快速发展和自动化程度的提高,机械设备不仅具有大型、集成、高速及重载的特点,而且其工作环境十分复杂。机械设备的某一关键部件一旦发生故障,往往会影响整台设备甚至整个生产系统的运行,不仅会造成巨大的经济损失,而且可能会导致严重的人员伤亡后果。开展机械故障诊断技术是保障设备安全可靠运行的主要措施之一,它不仅能对机械设备故障的发生以及进一步的发展做出预测,而且对故障产生的原因作出判断,提出对策以减少事故的发生。因此,对机械设备开展故障诊断技术的研究具有重要的理论意义和现实的工程价值。故障信息特征的提取是设备故障诊断技术的核心内容之一,如何从较强的背景噪声下提取出隐含故障信息的信号特征并将其应用于工程实际是故障诊断领域研究的热点。新理论、新技术和新方法方面的研究层出不穷,丰富和完善了机械故障诊断技术。本论文以齿轮箱为研究对象,以约束独立分量分析(Constrained Independent Component Analysis,cICA)为信号分析工具,深入研究了基于约束独立分量分析的齿轮箱故障诊断技术方法,通过实验验证了该方法的有效性,并应用相应的工程案例予以说明,取得了一定的效果。论文主要包括以下内容:(1)介绍独立分量分析(Independent Component Analysis,ICA)的数学基础、基本原理和算法。首先介绍ICA的数学基础,然后叙述了ICA的数学模型、模型的不确定性和可辨识性及ICA算法相关原理。最后研究了约束独立分量分析(cICA)的优点、算法介绍以及参考信号的建立。(2)从齿轮箱的结构分别阐述了各振动源的特征,并对其振动机理进行分析。介绍了齿轮箱主要零、部件(齿轮、轴和轴承)的故障类型以及相应的故障信号特征。分析了齿轮箱振动源信号的混合机理并建立了齿轮箱振动信号的模型。(3)实验验证和分析。利用实验研究论证了约束独立分量分析方法在齿轮箱故障诊断中的有效性和优势。通过动力传动故障综合试验台模拟齿轮箱常见的齿轮局部故障,应用数据采集仪和相应软件拾取了断齿和局部断齿情况下齿轮箱振动信号。然后利用cICA方法对齿轮断齿和局部断齿故障进行了特征提取和分析,得出了相应的结论,进而验证了该算法的有效性及其特点。(4)工程案例分析。针对矿用提升机传动齿轮箱故障信号的微弱性,并结合其故障信息的先验知识,将cICA方法应用于矿用提升机齿轮箱故障信号特征提取中,消除了提升机齿轮箱故障特征中的冗余信息,提取出了其故障特征,实现了提升机齿轮箱的精密故障诊断。工程应用进一步验证了cICA方法应用于齿轮箱故障诊断中的有效性和实用性,并对保障煤矿提升机安全生产具有重要的现实意义。
[Abstract]:With the rapid development of modern industrial science and technology and the improvement of automation, mechanical equipment not only has the characteristics of large scale, integration, high speed and heavy load, but also its working environment is very complex. It will often affect the operation of the whole equipment and even the whole production system, which will not only cause huge economic losses, And it may lead to serious casualties. The development of mechanical fault diagnosis technology is one of the main measures to ensure the safe and reliable operation of equipment. It can not only predict the occurrence and further development of mechanical equipment failures, And to judge the cause of the fault, put forward countermeasures to reduce the occurrence of accidents. The research of fault diagnosis technology for mechanical equipment has important theoretical significance and practical engineering value. The feature extraction of fault information is one of the core contents of equipment fault diagnosis technology. How to extract the signal features of hidden fault information from strong background noise and apply it to engineering is a hot topic in the field of fault diagnosis. In this paper, the gearbox fault diagnosis method based on constrained independent component analysis (ICA) and constrained Independent Component analysis (ICA) is studied. The effectiveness of the method is verified by experiments, and the corresponding engineering cases are applied to illustrate the effectiveness. The paper mainly includes the following contents: 1) introducing the mathematical basis of Independent Component Analysis (ICA). This paper introduces the mathematical basis of ICA, then describes the mathematical model of ICA, the uncertainty and identifiability of the model, and the related principles of ICA algorithm. Finally, the advantages of constrained Independent component Analysis (ICA) are studied. From the structure of the gearbox, the characteristics of each vibration source are expounded, and its vibration mechanism is analyzed. The main parts of the gearbox (gear, gear, gear, gear, etc.) are introduced. The fault types of shaft and bearing and the corresponding fault signal characteristics are analyzed. The mixing mechanism of gear box vibration source signal is analyzed and the model of gear box vibration signal is established. The effectiveness and advantage of constrained Independent component Analysis (ICA) method in gear box fault diagnosis. The vibration signals of gear box in the case of broken teeth and local broken teeth are picked up by data acquisition instrument and corresponding software. Then the feature extraction and analysis of gear broken teeth and local broken teeth faults are carried out by using cICA method, and the corresponding conclusions are obtained. Furthermore, the validity and characteristics of the algorithm are verified. The engineering case analysis. Aiming at the weak fault signal of the transmission gearbox of mine hoist, and combining with the prior knowledge of fault information, The cICA method is applied to the fault signal extraction of mine hoist gearbox. The redundant information of hoist gearbox fault feature is eliminated and its fault feature is extracted. The precision fault diagnosis of hoist gearbox is realized. The engineering application further verifies the validity and practicability of cICA method applied in gearbox fault diagnosis, and has important practical significance to ensure the safety of coal mine hoist.
【学位授予单位】：河南理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TH132.41

【相似文献】