基于信息融合的液压缸故障模式识别方法研究

发布时间：2018-03-29 10:10

本文选题：液压缸　切入点：内泄漏　出处：《武汉理工大学》2012年硕士论文

【摘要】：液压系统因其结构轻巧、运行平稳、容易实现无极调速、远距离操纵等优势被广泛应用于各工程领域,在众多设备中起着关键的控制、驱动等作用。及时准确地对液压系统实施故障诊断,保证其正常运行,对提高生产效率,降低维护成本以及减少不必要的经济损失有着重要的意义。液压缸作为液压系统的执行元件,如果出现故障将直接影响液压系统甚至整机的正常运作。本文以液压缸为对象,在对国内外研究现状进行分析的基础上,将信息融合技术应用于液压缸的内泄漏故障诊断,充分利用各个传感器之间包含的冗余和互补信息,提高系统决策的准确性和鲁棒性。本文首先介绍了液压缸内泄漏的失效机理,结合专门设计的液压缸内泄漏故障模拟实验台,介绍了包括其中的硬件选择、数据采集程序的编写等在内的故障信息获取的具体实现过程。在获取所需的故障信息后,提出了基于BP神经网络的信息融合诊断方法,在液压缸内泄漏上进行了验证。通过对融合后的结果分析发现,由于不确定因素的影响,将BP神经网络信息融合应用于液压缸内泄漏故障诊断时存在一定的不足；而D-S证据理论在处理不确定知识方面存在的优势恰好能够弥补这一不足,但还不能实现完全诊断。论文在对以上两种方法有所探索的基础上提出了基于BP神经网络与D-S证据理论相结合的综合诊断方法,并验证了这种方法的有效性,为液压缸故障诊断确诊率的提高提供了一条有效途径。
[Abstract]:Hydraulic system is widely used in various engineering fields because of its advantages of light structure, stable operation, easy to realize pole free speed regulation and remote control, and plays a key role in the control of many equipments. Driving and other functions. Timely and accurate fault diagnosis of hydraulic system, to ensure its normal operation, to improve production efficiency, It is of great significance to reduce the maintenance cost and reduce the unnecessary economic loss. As an executive component of hydraulic system, the failure of hydraulic cylinder will directly affect the normal operation of hydraulic system or even the whole machine. Based on the analysis of domestic and foreign research status, this paper applies the information fusion technology to the diagnosis of internal leakage fault of hydraulic cylinder, and makes full use of the redundant and complementary information contained in each sensor. Firstly, the failure mechanism of hydraulic cylinder leakage is introduced in this paper. Combined with the specially designed hydraulic cylinder leakage simulation test bench, the hardware selection is introduced. After obtaining the needed fault information, the information fusion diagnosis method based on BP neural network is put forward. Through the analysis of the results after fusion, it is found that BP neural network information fusion in hydraulic cylinder leakage fault diagnosis has some shortcomings because of the influence of uncertain factors. The advantage of D-S evidence theory in dealing with uncertain knowledge can make up for this deficiency. On the basis of the above two methods, a comprehensive diagnosis method based on BP neural network and D-S evidence theory is put forward, and the validity of this method is verified. It provides an effective way to improve the diagnosis rate of hydraulic cylinder fault diagnosis.
【学位授予单位】：武汉理工大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TH137.51

【相似文献】