基于数据挖掘的物流设备隐性故障预警模型研究

发布时间：2018-03-03 09:36

本文选题：物流设备　切入点：故障诊断　出处：《燕山大学》2014年硕士论文　论文类型：学位论文

【摘要】：随着现代物流的快速发展，现代物流设备逐步走向自动化和智能化的道路，也使得维护其安全运转的检测与维修成为物流设备管理的重要方面。本文的研究目的是建立基于数据挖掘的物流设备隐性故障预警模型，估测隐性故障对显性故障发生的影响值，以此来实现监控隐性故障变化趋势的目的，并对由隐性故障引起的突发故障起到一定的预防作用。针对课题的特点，本文运用归纳分析法对国内外相关文献进行总结；运用比较分析法验证算法的有效性；运用例证法对本文提出的隐性故障预警模型进行实例分析，评价模型的可靠性。首先，从研究背景和意义入手，对数据挖掘技术、故障诊断技术、数据挖掘技术在故障诊断中的应用等的国内外研究现状进行综述，得出现有故障诊断技术的不足之处。并对物流设备、故障诊断技术和数据挖掘技术的相关基础知识进行详细阐述，奠定本文的理论基础。其次，为进一步提高显性故障诊断的效率，并根据故障诊断过程中不同故障因素具有不同的故障贡献度的实际情况，提出更适用于设备故障诊断现实需求的加权关联增量更新算法。再次，在对故障状态下与隐性故障相关的因素进行分析的基础上，提出隐性故障预警模型，通过引入神经网络的方法解决隐性故障这类不确定性因素较多的非线性问题。该模型不仅实现了对显性故障的诊断，而且通过对多种故障因素的综合分析达到对隐性故障进行监测的目的。最后，从结构特点、故障因素等方面对物流设备的液压系统进行简要分析，选择具有代表性的故障数据建立用于隐性故障预警的神经网络，并验证该模型在物流设备故障诊断中的可行性。
[Abstract]:With the rapid development of modern logistics, modern logistics equipment gradually moving towards automation and intelligent way, also makes the detection and maintenance of the safe operation of the equipment has become an important aspect of logistics management. The purpose of this study is to establish the data mining logistics equipment hidden fault early warning model based on the estimated effect of hidden failures occurred on dominant fault the value, in order to achieve the purpose of monitoring the hidden failure trend, and a preventive role for sudden failure caused by hidden failures. According to the characteristics of the subject, this paper uses the inductive analysis of the related literature at home and abroad were summarized; using the method of comparative analysis verify the effectiveness of the algorithm; analysis of hidden failure warning model the proposed method using examples, the reliability evaluation model.
First of all, starting from the research background and significance of the data mining technology, fault diagnosis technology at home and abroad the status quo of the research of data mining technology in the application of fault diagnosis of the existing shortcomings of fault diagnosis technology. And the logistics equipment, basic knowledge of fault diagnosis technology and data mining technology in detail this laid the theoretical basis of this paper.
Secondly, in order to further improve the efficiency of explicit fault diagnosis, and according to the actual situation of different fault factors with different fault contribution in the process of fault diagnosis, a weighted incremental updating algorithm is put forward which is more suitable for the actual demand of equipment fault diagnosis.
Again, based on the factors associated with the recessive fault condition analysis, put forward the hidden failure warning model and to solve nonlinear problems more hidden failures of this kind of uncertainty factors by introducing a method of neural network. The model not only realizes the diagnosis of dominant fault, and through a comprehensive analysis of the various factors of the fault achieve the purpose of monitoring of hidden failures.
Finally, we briefly analyze the hydraulic system of the logistics equipment from the aspects of structure characteristics and failure factors, select representative fault data, build a neural network for hidden fault early warning, and verify the feasibility of the model in the logistics equipment fault diagnosis.

【学位授予单位】：燕山大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：F252;TP311.13

【参考文献】