电梯故障诊断专家系统的研究与开发

发布时间：2018-01-06 17:05

本文关键词：电梯故障诊断专家系统的研究与开发　出处：《东北大学》2014年硕士论文　论文类型：学位论文

【摘要】：近几十年我国经济发展迅猛,楼房的平均高度也在不断增加,电梯越来越成为人们生活和工作中必不可少的工具。电梯的可靠性和安全性时刻在威胁着人们的人身和财产安全,因此必须做到及时发现和处理电梯故障。要保证电梯的可靠性需要做到以下两点：在电梯正式出厂安装使用之前对电梯进行可靠性测试；及时发现和处理电梯运行过程中出现的故障。根据当前对电梯故障诊断的需要,本文研究了专家系统和支持向量机用于故障诊断的相关理论,在专家系统的推理机中融入支持向量机,利用支持向量机的数据挖掘能力,从故障数据中分析出电梯故障原因。通过专家系统建立知识库并进行管理维护,使用规则对电梯故障进行推理诊断并用解释器做出相应的解释说明。系统发挥了两者的优势,提高诊断能力。本文的主要研究内容如下：首先本文介绍了故障诊断技术,对故障诊断方法进行了分析与比较,介绍了几种典型的故障诊断专家系统,并分析了电梯的常见故障。其次建立电梯故障诊断专家系统模型,选取产生式规则表示法、正向推理分别对知识进行表示和推理。再次研究了电梯故障诊断算法,包括支持向量机的基本原理及其关键点；选用径向基核函数,对其参数的选择采用交叉验证和网格搜索的方法,用“一对一”的方法来处理故障诊断的多分类问题,建立了支持向量机故障诊断多分类器模型。最后本文采用了专家系统与支持向量机相结合的方法,建立了电梯故障诊断专家系统。实现了专家系统知识的获取、知识的表示、知识库、推理机及解释器。
[Abstract]:In recent years China's rapid economic development, the average height of the building is also increasing, elevator has become more and more indispensable in people's work and life tools. Lift the reliability and safety of threaten people's personal and property safety, and therefore must be timely detection and treatment of elevator fault to ensure the reliability of the elevator. Need to do the following two points: testing the reliability of the elevator installation before officially lift the factory; failure to timely detection and treatment of elevator operation. According to the current needs of elevator fault diagnosis, this paper studies expert system and support vector machine theory for fault diagnosis, into the support vector machine in expert system in the use of support vector machine data mining ability, analysis of the reasons of elevator fault from fault data. Through the establishment of expert system knowledge Management and maintenance of knowledge base and reasoning rules, using the diagnosis of elevator fault and interpreter to make the appropriate explanation. The system can play the advantages of both, to improve the diagnosis ability. The main contents of this paper are as follows: firstly, this paper introduces the fault diagnosis technology, the fault diagnosis methods are compared and analyzed, introduced several the typical fault diagnosis expert system, and analyzes the common faults of the elevator. Secondly, the fault diagnosis expert system of elevator model, selection of generative rule representation and reasoning are representation and reasoning of knowledge. Another study of elevator fault diagnosis algorithm, including the basic principle of support vector machine and its key points; the radial basis kernel function of the parameter selection using cross validation method and grid search, use the multi classification problem of method "to deal with the fault diagnosis A support vector machine, fault diagnosis of multi classifier model. Finally this paper adopts the method of combining expert system with support vector machine, a fault diagnosis expert system of elevator. The knowledge acquisition of expert system, knowledge representation, knowledge base, inference engine and the interpreter.

【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TU857

【相似文献】