基于时序分析的齿轮故障预警仪器系统研究

发布时间：2018-02-25 10:16

本文关键词： 故障预测分布式故障局部故障 ARMA模型采集器在线监测　出处：《郑州大学》2016年硕士论文　论文类型：学位论文

【摘要】：设备故障预测与管理系统是近年来国内外研究的热点。设备故障预测是指根据设备当前时刻的运行状态特征以及此前一段时间内的运行状态特征判断设备下一时刻的状态特征,通过特征分析对设备的健康状况和剩余使用寿命等内容进行预测。齿轮系统是设备中关键的传动或传力部件,对齿轮设备进行状态监测和故障预测的研究对设备的良好运行有重要的意义。鉴于齿轮故障预测研究的重要性,本文主要研究了以下内容:1)ARMA模型构建的理论和过程;2)齿轮的故障机理和频谱特征以及齿轮故障的诊断标准和故障分析中的频带划分问题;3)设备振动信号采集器和在线监测系统。研究内容围绕齿轮振动信号采集,信号的分析处理,齿轮状态特征提取、状态识别和状态预测进行。首先,本文通过理论研究构建了应用于齿轮频谱结构预测的ARMA模型。然后将建立的ARMA模型应用于某水泥厂窑主减速机齿轮频谱的预测中,根据齿轮分布式故障和局部故障在频谱中表现的差异性,对减速机高速端齿轮发生分布式故障或局部故障的趋势进行了分析。从而得出,该齿轮处于较良好的运行状态,但是有发生分布式故障的趋势或可能,应当持续关注其趋势的发展。随后,本文对设备故障预测与管理系统中的分布式采集器和在线监测软件的实现进行研究。对于分布式采集器,其采用A8+DSP处理器和FPGA控制采样,拥有8个振动通道和1个转速通道,能够对设备的加速度、速度和位移信号进行采集;采集器对振动信号的处理包括隔直、滤波、放大、积分、A/D转换等。为了能够在通讯出现异常的情况下实现信号的本地存储采集器还配备了扩展内存,同时为了获得良好的通讯和信号传输效率其兼有USB/RS485接口和网络接口,可以实时地将信号传输到终端进行监测分析。对于在线监测软件,其由监测概貌、分析/评价、综合报告、系统管理功能模块构成,通过对从采集器实时接收的信号的分析处理和特征提取来实现对设备的状态监测、预报和管理。采集器和在线监测系统经过工程的实际应用检验,具有了良好的工程适用性。
[Abstract]:In recent years, equipment fault prediction and management system is a hot topic in domestic and international research. Equipment fault prediction refers to judging the state characteristics of equipment at the next moment according to the operating state characteristics of the current moment of the equipment and the operating state characteristics of the previous period of time. Through characteristic analysis, the health condition and remaining service life of the equipment are predicted. The gear system is the key transmission or force transfer component in the equipment. The research on condition monitoring and fault prediction of gear equipment is of great significance to the good operation of gear equipment. This paper mainly studies the following contents: the theory and process of constructing the 1 / 1 ARMA model) the fault mechanism and spectrum characteristics of gear, the diagnostic standard of gear fault and the problem of frequency band division in fault analysis. On-line monitoring system. Research contents around gear vibration signal collection, Signal analysis and processing, gear state feature extraction, state recognition and state prediction. First, In this paper, the ARMA model applied to the prediction of gear frequency spectrum structure is constructed through theoretical research, and then the established ARMA model is applied to the prediction of gear spectrum of main reducer in a cement factory. According to the difference of distributed fault and local fault of gear in frequency spectrum, the trend of distributed fault or local fault of gear at high speed end of reducer is analyzed, and it is concluded that the gear is in a good running state. However, there is a tendency or possibility of distributed failure, and the trend should be continuously followed. In this paper, the realization of distributed collector and on-line monitoring software in the equipment fault prediction and management system is studied. For the distributed collector, the A8 DSP processor and FPGA are used to control the sampling, and there are 8 vibration channels and 1 rotational speed channel. Can collect the acceleration, velocity and displacement signal of the equipment; the processing of the vibration signal by the collector includes straightening, filtering and amplifying, In order to realize the local storage collector in case of abnormal communication, the local storage collector is equipped with extended memory, and in order to obtain good communication and signal transmission efficiency, it has both USB/RS485 interface and network interface. For on-line monitoring software, it is composed of monitoring overview, analysis / evaluation, comprehensive report, system management function module. The state monitoring, prediction and management of the equipment are realized through the analysis, processing and feature extraction of the signals received in real time from the collector. The collector and the on-line monitoring system have been tested in engineering practice and have good engineering applicability.
【学位授予单位】：郑州大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TH132.41;TH17

【相似文献】