减速器故障诊断方法的应用研究

发布时间：2018-06-21 14:55

本文选题：减速器 + 故障诊断　；参考：《沈阳理工大学》2011年硕士论文

【摘要】：减速器是一种非常常见的传动设备,它被广泛地应用在各种各样的大型机械设备中,因此以减速器为研究对象进行故障诊断,不但具有代表性,而且具有十分重要的意义。减速器是一种依靠齿轮进行能量传递的传动装置。在运行过程中,会随着齿轮以及轴承的旋转而产生一系列的振动,而其运行状态的优劣则可以直接通过对振动信号的分析而得出。本文在对减速器的各种不同运行状态进行学习的基础上,深入研究了基于蚁群算法的减速器故障诊断方法,主要工作包括以下几方面: (1)简单介绍了蚁群算法的基本原理,在此基础上以旅行商问题(TSP)为例阐述了蚁群算法的参数选取、性能评价等问题; (2)对减速器的常见故障形式及其产生原因进行了分析,并以齿轮为研究对象,实现了对齿轮故障现象的初步诊断; (3)在完成了对采集到的数据的进行了特征选取和归一化处理,在此基础上,将蚁群算法(ACO)和BP神经网络应用到了减速器的故障诊断过程中,分别建立了基于蚁群算法的和基于BP神经网络的故障诊断模型,详细阐述了基于蚁群算法的故障诊断模型的参数选取,并对两个模型进行了训练学习、测试、性能分析和比较。 (4)设计完成了减速器故障诊断系统,其主要功能包括:初期诊断模块、精密诊断模块、人机交互模块等。该系统具有运算速度快,准确率高、易于维护等优点。
[Abstract]:Reducer is a very common transmission equipment, it is widely used in a variety of large mechanical equipment, so it is not only representative, but also very important to take the reducer as the research object for fault diagnosis. A reducer is a transmission device that relies on gears for energy transfer. In the process of operation, a series of vibration will be produced with the rotation of the gear and bearing, and the advantages and disadvantages of the running state can be directly obtained by the analysis of the vibration signal. In this paper, based on the study of various running states of reducer, the fault diagnosis method of reducer based on ant colony algorithm is studied. The main work includes the following aspects: (1) the basic principle of ant colony algorithm is introduced, and the parameter selection of ant colony algorithm is illustrated by TSPS (traveling Salesman problem) as an example. The common fault forms and their causes of reducer are analyzed, and the gear is taken as the research object. The primary diagnosis of gear fault phenomenon is realized, and the feature selection and normalization processing of the collected data are completed, on the basis of which, The ant colony algorithm (ACO) and BP neural network are applied to the fault diagnosis of reducer, and the fault diagnosis models based on ant colony algorithm and BP neural network are established, respectively. The parameter selection of fault diagnosis model based on ant colony algorithm is described in detail, and the two models are trained, tested, analyzed and compared. Its main functions include: initial diagnosis module, precision diagnosis module, man-machine interaction module and so on. The system has the advantages of fast operation, high accuracy and easy maintenance.
【学位授予单位】：沈阳理工大学
【学位级别】：硕士
【学位授予年份】：2011
【分类号】：TH165.3

【引证文献】