基于LabVIEW的圆盘式刀库可靠性台架试验中的状态监测及故障预警

发布时间：2018-02-27 01:10

本文关键词： 圆盘式刀库 LabVIEW 状态监测支持向量回归故障预警　出处：《吉林大学》2017年硕士论文　论文类型：学位论文

【摘要】：圆盘式刀库作为数控机床关键功能部件之一,其工作状态直接影响了整机的可靠性。对圆盘式刀库在可靠性台架试验中的状态监测及故障预警进行研究,有助于及时掌握其状态变化趋势、制定合理的维修策略以及提升可靠性,同时对提高圆盘式刀库可靠性试验系统自动化水平有着重要的实践意义,是数控机床可靠性方面的重要研究内容。本文以国产圆盘式刀库为研究对象,结合实验室搭建完成的圆盘式刀库可靠性试验系统,在分析了圆盘式刀库主要故障模式、状态监测系统设计方案、虚拟仪器和Lab VIEW功能特点的基础上,与相关的硬件系统相结合开发了基于Lab VIEW的圆盘式刀库状态监测系统;提出了信号降噪和电机堵转报警方法,同时,以机械手转位偏差故障为典型研究对象,使用基于遗传算法优化的支持向量回归实现预警,并通过多组实验验证了本文所研究方法的可行性。本文主要研究工作如下:1.基于NI Compact RIO的圆盘式刀库状态监测系统。根据加工中心现场可靠性试验数据统计结果,圆盘式刀库故障在整机中占有较高的比例,结合故障树分析确定出圆盘式刀库主要故障模式,并以此为依据建立状态监测系统框架。提出了传感器监测点确定原则和方法,选用合适的传感器并配置于圆盘式刀库可靠性试验系统中,以NI c RIO-9035作为状态监测系统硬件基础,选用合适的C系列模块组成状态监测系统,并对传感器及监测系统中各硬件的性能参数进行了介绍。2.软件系统的设计与实现。对虚拟仪器和Lab VIEW的基本概念以及主要特点进行了介绍,使用Lab VIEW开发了一套全面、实用的圆盘式刀库状态监测系统,实现了FPGA、RT与Windows间的数据传递、显示、存储以及故障报警功能,并给出了该系统的结构、功能、软件操作和程序框图。3.圆盘式刀库故障预警方法研究。因实际获取的时域信号噪声较多,为提高故障预警的准确性,本文提出一种改进的相邻系数法实现对圆盘式刀库状态监测系统获取到的振动信号进行降噪处理;在完成电机定子电流分析和故障分析的基础上,通过建立电流能量随时间的变化趋势,有效地区分了启动电流和堵转电流,实现电机堵转报警;在故障预警方法研究中,分析了圆盘式刀库振动机理,指出在圆盘式刀库换刀过程中振动信号时域特征参数更能明显地反应换刀状态变化,并以机械手转位偏差故障为典型研究对象,基于支持向量回归建立了机械手转位偏差预警模型,利用遗传算法对模型中回归误差惩罚因子和核宽度参数进行了优化。4.实验及数据分析。运用本文所研制的状态监测系统进行监测及预警实验,证明了所提出和选用的信号降噪、电机堵转故障报警以及机械手转位偏差预警方法的可行性。
[Abstract]:As one of the key functional components of CNC machine tool, the working state of disc cutter library has a direct impact on the reliability of the whole machine. It is helpful to grasp the trend of state change in time, to formulate reasonable maintenance strategy and to improve reliability. At the same time, it is of great practical significance to improve the automation level of reliability test system of disc type cutter storehouse. It is an important research content of NC machine tool reliability. In this paper, the main failure modes of disc cutter storehouse are analyzed by taking the domestic disc cutter library as the research object, and combining with the reliability test system of the disc cutter database built in the laboratory. Based on the design scheme of state monitoring system, virtual instrument and Lab VIEW, the paper develops a disc cutter database condition monitoring system based on Lab VIEW, and presents signal noise reduction and motor shutoff alarm method. At the same time, taking the fault of manipulator transposition deviation as the typical research object, the support vector regression based on genetic algorithm optimization is used to realize the early warning. The feasibility of this method is verified by many experiments. The main research work of this paper is as follows: 1. Based on NI Compact RIO, the status monitoring system of disc cutter storehouse is developed. According to the statistical results of field reliability test data of machining center, The failure of disc cutter storehouse occupies a high proportion in the whole machine. According to the fault tree analysis, the main failure mode of disc cutter storehouse is determined, and the frame of state monitoring system is established based on it. The principle and method of sensor monitoring point determination are put forward. The appropriate sensor is selected and configured in the reliability test system of disc cutter storehouse. The hardware of the system is based on NI c RIO-9035, and the appropriate C series modules are selected to constitute the status monitoring system. Secondly, the design and implementation of software system are introduced. The basic concepts and main features of virtual instrument and Lab VIEW are introduced, and a set of comprehensive development with Lab VIEW is developed. The practical disk cutter database condition monitoring system realizes the functions of data transfer, display, storage and fault alarm between FPGA RT and Windows, and gives the structure and function of the system. Software operation and program block diagram .3.Research on disc cutter storehouse fault warning method. In order to improve the accuracy of fault warning, the noise of time domain signal acquired in practice is much more. In this paper, an improved adjacent coefficient method is proposed to reduce the noise of vibration signals obtained from the disk cutter database condition monitoring system, and on the basis of the analysis of stator current and fault analysis of the motor, By establishing the trend of current energy changing with time, the starting current and the blocking current are effectively distinguished, and the motor shutoff alarm is realized. In the research of fault warning method, the vibration mechanism of disc knife storehouse is analyzed. It is pointed out that the time domain characteristic parameters of vibration signal can more obviously reflect the change of tool changing state in the process of changing the cutter in the disc cutter storehouse, and the typical research object is the fault of the displacement deviation of the manipulator. Based on support vector regression, an early warning model of manipulator transposition deviation is established. The regression error penalty factor and kernel width parameters in the model are optimized by genetic algorithm. Experiments and data analysis are carried out. The state monitoring system developed in this paper is used for monitoring and early warning experiments, and it is proved that the proposed and selected signals are de-noised. The feasibility of the alarm method of motor shutoff fault and the early warning method of manipulator transposition deviation.
【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】：TG659

【相似文献】