基于有向图的数控车床故障传播机理研究

发布时间：2018-05-22 11:24

本文选题：数控车床 + 分层故障传播有向图　；参考：《吉林大学》2017年硕士论文

【摘要】：数控车床是集机、电、液于一体的复杂系统,系统结构的复杂性导致其故障存在着传播性、层次性、并发性等特征。机床运行故障是系统组件独立故障和级联故障共同作用的结果,若不能及时阻断级联故障传播,便会因系统的规模效应波及其余组件、以致整机瘫痪。级联故障的传播与组件间故障传播模型结构性指标、组件自身可靠性水平和环境载荷冲击等因素密切相关。故障传播机理分析是故障诊断与健康维护的根本保障,是设备关键智能基础共性技术之一,因此,进行数控车床故障传播机理研究对于实现制造装备智能化具有重要意义。本文以吉林省自然科学基金为依托,以国产某型数控车床为研究对象,依据故障机理分析,基于多指标定义故障传播强度以表征故障传播行为,结合分层故障传播有向图模型,提出一种关键组件节点、关键故障传播路径的识别方法,从微观到宏观、定性到定量研究数控车床故障传播扩散行为,主要内容如下:(1)数控车床系统故障传播模型构建。基于数控车床故障机理分析,应用有向图理论构造故障传播有向图模型,借助矩阵工具、集成DEMATEL(Decision Making Trial and Evaluation Laboratory)/ISM(Interpretative Structural Modeling)方法将故障传播有向图模型分层重构,得到初始分层故障传播有向图模型,实现了系统层级划分,描述了故障的分步扩散过程。增设虚节点确定数控车床系统分层故障传播有向图模型,用于最终关键路径识别。(2)故障传播模型结构性指标评估。首先,依据初始分层故障传播有向图模型,计算各边介数以衡量各有向边的负荷量;其次,基于故障机理分析获得的直接影响矩阵,计算修正的状态转移概率矩阵,引入PageRank算法评估系统组件节点故障影响度,为系统故障源定位提供依据。结合初始分层故障传播有向图模型,进而评估各有向边故障影响度,实现系统组件间故障影响能力的定量评估。有向边故障影响度、边介数影响故障传播行为,视为关键路径识别原则的重要组成部分。(3)数控车床系统关键故障传播路径识别。依据数控车床故障信息合理评估故障传播模型中组件节点故障概率,提出基于组件节点故障影响度与故障概率乘积值定位系统故障源的方法。同时,组件节点故障概率结合故障传播模型结构性指标定义故障传播强度作为关键路径识别原则,结合系统分层故障传播有向图模型,搜索所有存在的故障传播路径,模拟故障分步扩散过程,识别各层级关键节点并确定关键故障传播路径。本文所提方法克服了现有方法多基于单一指标模拟故障传播过程而产生的偏差,同时为系统故障诊断奠定基础。
[Abstract]:CNC lathe is a complex system which integrates machine, electricity and liquid. The complexity of system structure leads to the characteristics of transmission, hierarchy, concurrency and so on. Machine tool operation fault is the result of independent fault and cascade fault of system components. If the cascade fault propagation can not be interrupted in time, the whole machine will be paralyzed because of the scale effect of the system. The propagation of cascaded faults is closely related to the structural index of fault propagation model between components, the reliability level of components themselves and the impact of environmental load. The analysis of fault propagation mechanism is the fundamental guarantee of fault diagnosis and health maintenance and one of the basic common technologies of key intelligence of equipment. Therefore, it is of great significance to study the fault propagation mechanism of NC lathe for the realization of intelligent manufacturing equipment. In this paper, based on Jilin Province Natural Science Foundation, a kind of CNC lathe made in China is taken as the research object, according to the fault mechanism analysis, the fault propagation intensity is defined based on multiple indexes to characterize the fault propagation behavior, and the hierarchical fault propagation directed graph model is combined. In this paper, a key component node and identification method of critical fault propagation path are proposed. From microscopic to macroscopic, qualitative to quantitative research on the behavior of fault propagation and diffusion of CNC lathe is carried out. The main contents are as follows: 1) Construction of fault propagation model of CNC lathe system. Based on the fault mechanism analysis of NC lathe, the directed graph model of fault propagation is constructed by using directed graph theory, and the fault propagation directed graph model is reconstructed layer by layer with the help of matrix tool and integrated DEMATEL(Decision Making Trial and Evaluation Laboratory)/ISM(Interpretative Structural modeling method. The initial hierarchical fault propagation directed graph model is obtained, and the hierarchical partition of the system is realized, and the step-by-step diffusion process of the fault is described. Virtual nodes are added to determine the hierarchical fault propagation directed graph model of CNC lathe system, which is used to evaluate the structural index of the ultimate critical path identification. Firstly, according to the initial hierarchical fault propagation directed graph model, the mediums of each side are calculated to measure the load of each directed edge; secondly, based on the direct influence matrix obtained by fault mechanism analysis, the modified state transition probability matrix is calculated. PageRank algorithm is introduced to evaluate the impact of system component node fault, which provides the basis for system fault source location. Combined with the initial hierarchical fault propagation directed graph model, the influence degree of each directed edge fault is evaluated, and the quantitative evaluation of fault impact ability between components of the system is realized. The influence degree of directed edge fault and the number of edge mediums affect the behavior of fault propagation, which is regarded as an important part of the principle of critical path identification. 3) the identification of critical fault propagation path of CNC lathe system. According to the fault information of NC lathe, the fault probability of component node in the fault propagation model is reasonably evaluated, and the method of locating the fault source of the system based on the product value of fault influence degree and fault probability of component node is presented. At the same time, the component node fault probability combined with the structural index of the fault propagation model defines the fault propagation intensity as the key path identification principle, combined with the hierarchical fault propagation directed graph model to search all existing fault propagation paths. The process of fault diffusion is simulated, the critical nodes are identified and the critical fault propagation path is determined. The method proposed in this paper overcomes the deviation caused by most of the existing methods based on a single index to simulate the fault propagation process, and lays a foundation for system fault diagnosis at the same time.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TG519.1

【相似文献】