盾构刀盘驱动液压系统故障诊断研究

发布时间：2019-01-01 16:38

【摘要】：盾构机长期处于恶劣的环境下,负载随机多变、冲击强的工作特点使得其液压系统成为了故障高发部位。一旦发生故障,若不能及时有效地排除,将严重影响施工进度、造成无法挽回的经济损失,甚至可能造成重大伤亡事故。当前盾构机液压系统故障诊断大多通过单传感器检测、人工判断的方法,导致诊断效率低以及诊断精度不高。鉴于此,提出了改进型PCA和SVM相结合的故障诊断方法,以刀盘驱动液压系统为研究对象,以理论研究、仿真建模、试验分析为研究手段,旨在降低对维修人员的依赖性、实现盾构机液压系统自动化和智能化。主要完成了以下工作:(1)通过对刀盘驱动液压系统原理进行分析,总结了其常见故障机理及特性,并建立了其关键元件的数学模型,找到了模型参数与故障特性之间对应的关系,为故障模拟提供理论依据。(2)建立了刀盘驱动液压系统AMESim仿真模型,根据实际液压元件及系统对仿真模型进行了详细的参数设置,确定了数据测点并对典型的故障进行仿真模拟的同时采集相应数值为故障诊断研究提供样本数据。(3)通过改进型PCA对仿真模型模拟的样本数据进行故障检测,对其特征参数进行降维、去掉其中的冗余信息、解除特征间的相关性;然后将经过主元分析提取到的各类状态下的主元得分向量作为SVM的输入样本集;最后使用SVM分类器对其进行故障类型识别分类。(4)完成了实车试验,试验结果表明,本文所提出了故障诊断方法可行,其诊断准确率能达到95%以上,具有良好的工程应用价值。
[Abstract]:Shield machine is in a bad environment for a long time, the work characteristics of random load and strong impact make its hydraulic system become a high fault location. Once failure occurs, if it can not be eliminated in time and effectively, it will seriously affect the progress of construction, cause irreparable economic losses, and even cause serious casualties. At present, the fault diagnosis of shield machine hydraulic system is mostly through single sensor detection and manual judgment, which leads to low diagnosis efficiency and low diagnostic accuracy. In view of this, an improved fault diagnosis method combining PCA and SVM is put forward, which takes the hydraulic system driven by the cutter head as the research object, takes the theoretical research, the simulation modeling and the experimental analysis as the research means, in order to reduce the dependence on the maintainers. The automatic and intelligent hydraulic system of shield machine is realized. The main works are as follows: (1) through the analysis of the principle of the hydraulic system driven by the cutter head, the mechanism and characteristics of its common faults are summarized, and the mathematical model of its key components is established. The relationship between the model parameters and the fault characteristics is found, which provides a theoretical basis for fault simulation. (2) the AMESim simulation model of the hydraulic system driven by the cutter head is established. According to the actual hydraulic components and systems, the simulation model is set up in detail. At the same time, the corresponding data are collected to provide the sample data for fault diagnosis research. (3) the fault detection of the simulated sample data of the simulation model is carried out by improved PCA. The dimension of the feature parameters is reduced, the redundant information is removed, and the correlation between the features is removed. Then the principal component score vector extracted by principal component analysis (PCA) is used as the input sample set of SVM. Finally, SVM classifier is used to classify the fault type. (4) the test results show that the fault diagnosis method proposed in this paper is feasible, and the diagnostic accuracy can reach more than 95%. It has good engineering application value.
【学位授予单位】：湘潭大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U455.39

【参考文献】