中国科学院光电技术研究所机构知识库
Advanced  
IOE OpenIR  > 光电技术研究所博硕士论文  > 学位论文
题名:
光电成像跟踪设备智能故障诊断技术研究
作者: 侯明亮
学位类别: 博士
答辩日期: 2008-06-06
授予单位: 中国科学院光电技术研究所
授予地点: 光电技术研究所
导师: 吴钦章
关键词: 智能故障诊断 ; 专家系统 ; 神经网络 ; 虚拟现实 ; 光电跟踪系统
其他题名: study of intelligent fault diagnosis technology for photoelectric tracking devices
学位专业: 信号与信息处理
中文摘要: 随着科学技术的持续发展与进步,光电跟踪测量设备朝着大型化、复杂化、高速化、集成化以及自动化的方向不断迈进。其结构层次日益复杂,功能更加强大,各种信息技术、智能技术广泛应用其中。现有的故障诊断理论和技术面临着新形势下外场测试任务不断扩展的性能需求以及相关测量设备的高可靠性、维修性、智能化、实时性需求的严峻挑战。在此背景下,本文针对光电成像跟踪系统的非线性、动态性、复杂性、层次性、相关性和不确定性等根本特点,在认真探讨并分析各种国内外先进的动态系统故障检测与诊断技术的基础上,采用人工智能相关理论和实现技术,结合日益成熟的虚拟现实技术,研究构建基于虚拟诊断环境的智能故障诊断系统。主要研究内容和成果包括以下几个方面。 (1) 基于产生式规则的智能故障诊断专家系统研究 论文对比分析国内外成熟的故障诊断技术、人工智能技术及虚拟现实技术,讨论了若干类故障检测和故障模式分类的诊断方法,重点研究比较应用于复杂动态系统的各种智能诊断技术。同时,运用故障层次化、专家知识分级处理等方法对设备故障从功能和结构上进行了全面分析,总结了设备的常见故障机理、特征及识别推理方法,同时分析比较跟踪设备各层级故障特点及关联耦合关系。在充分的数据和理论基础上,提出构建一种以虚拟诊断环境为平台,融合神经网络推理的诊断专家系统的技术方案和路线。 针对项目实际情况,研究了专家系统的组成结构、工作原理及相关实现技术,选择基于规则和框架的产生式诊断专家系统。其中,在知识表示方面,选择稳定性和表示能力强的产生式表示和框架表示法,同时结合多种故障树共同完成专家诊断知识的表示。而在推理机设计方面,采用了正反向两种推理策略,完成分阶段递进式诊断推理过程。 (2) 基于BP神经网络的诊断专家系统推理技术研究 在推理诊断过程中,引入了BP神经网络技术,实现由单一推理控制策略转换为混合智能推理策略,避免了诊断专家系统自适应能力差、缺乏灵感、知识领域狭窄等缺陷。在对比分析拟牛顿法、弹性BP方法、共轭梯度法等相关BP网络改进算法后,设计了基于Levenberg-Marquardt法的神经网络故障诊断推理方法,结合诊断专家系统的实现针对性地解决多故障并发和耦合关联难点,提高了系统鲁棒性、容错性、普适性及诊断推理效率。 (3) 光电成像跟踪测量设备虚拟诊断环境研究 在智能诊断系统中引入虚拟现实技术,阐述了基于OpenGL的虚拟现实开发技术,包括OpenGL成像原理、建模技术、虚拟诊断环境构建、Billboard技术、纹理映射技术等相关关键技术。智能故障诊断系统虚拟诊断环境的创建,提高了系统实用性、直观性及整体性能。论文最后给出了系统相应的软硬件实现,并介绍了诊断专家系统的知识库、数据库、接口设计及关键技术。  本文旨在通过结合人工智能技术、虚拟现实技术以及分布式网络技术,研究构建基于虚拟诊断环境下的智能故障诊断系统,以保证和适应新形势下外场测量任务不断扩展的性能需求及相关光电跟踪设备的高可靠性、维修性、智能化、实时性需求。实践证明,在智能诊断系统中融入虚拟现实技术,具有良好的实用价值和广阔的应用前景。
英文摘要: Along with the continued development and progress of science, the photoelectric tracking and measuring equipments are facing the change of large-scale integration, complication, high speed and automation. Its structure is becoming more and more complicated and powerful, meanwhile, various information technology and intellectual technology are widely used. The conventional fault diagnosis technology is confronted with the severe challenges of the ever-expanding performance requirements of the outfield test mission and the requirements of high reliability, maintenance, intelligence and timeliness. Against this background, according to the leading particulars of inherent complexity, nonlinearity, dynamics, relevance and nondeterminacy for the photoelectric tracking devices, on the basis of a thorough analysis of the international and domestic advanced fault diagnosis technology, an intelligent fault diagnosis system based on desktop virtual environment is proposed, which uses artificial intelligence theory and implementation technique , combined with the growing maturity of virtual reality technology.The main research contents and achievements can be summarized as follows. (1) Research on intelligent fault diagnosis expert system based on production rule The international and domestic sophisticated technologies of fault diagnosis, AI and virtual reality are elaborated In this paper, and attention is concentrated on the intelligent diagnosis technology applied to the Dynamic and complicated system. Meanwhile, this paper made an analysis of the functional configuration of the devices, which use stage treatment methods for expert knowledge and use fault layering methods. Also , the common fault mechanism, feature, symptom and decision-theoretic approach are reviewed and disccused. Finally, on the basis of sufficient data and theory, a technologies development solution of a diagnosis expert system is presented, which use neural network reasoning technique, based on a virtual diagnosis environment. According to the practical situation of the project, the composition, structure, principle and related implementation technique is studied and the fault diagnosis expert system based on the rule and frame knowledge representation is selected. Thereinto, as regards the knowledge representation, the frame knowledge representation, production and are used, which has the feature of high stability and is easy to express. Also, the fault trees are used in the system. For the design of inference engine, the forward and reverse inference strategy is adopted in the staged reasoning process. (2) Study of inference engine of diagnosis expert system based on BP neural network In the diagnosis reasoning process, the BP neural network is used, which made a convert from a single control strategy to a mixed strategy of intelligent reasoning and avoided the deficiencies of poor adaptive capacity, lack of inspiration and narrow domain knowledge of expert system. After making a comparative analysis of the related BP neural network algorithms, such as the quasi-Newton method, the stretch BP method and the conjugate gradient method, a neural network fault diagnosis reasoning method based on the Levenberg-Marquardt is designed, which combined the implementation of the diagnosis expert system according to the difficulty of multiple and coupling fault diagnosis. Practical applications and experiments demonstrate that the proposed approach is effective, robust and universal. (3) 3D visualization of fault diagnosis environment for photoelectric tracking and measuring equipments The virtual reality technology applied to the intelligent diagnosis system. Several essential implementation issues of the system, including the OpenGL image-forming principle, the modelling techniques, the billboard technique ,the texture mapping technique and so on, are been discussed. The practicability , intuitionism and overall performance of the system was improved greatly by the 3D visualization of the fault diagnosis environment. Finally, the realization of the hardware and software of the system was presented in this paper. And the rule base, the diagnosis database, the interface design and other related key techniques were also introduced. In order to meet and suit the ever-expanding performance requirement of the outfield test mission and the requirements of high reliability, maintenance, intelligence and timeliness of the related the photoelectric tracking devices under the new situation, the intelligent fault diagnosis system on the basis of the virtual diagnosis environment is developed, which combined with the artificial intelligence, the virtual reality and the distributed network technology. Practical applications demonstrate that using the virtual reality in the intelligent diagnosis system illustrates a good prospect of application and extension.
语种: 中文
内容类型: 学位论文
URI标识: http://ir.ioe.ac.cn/handle/181551/296
Appears in Collections:光电技术研究所博硕士论文_学位论文

Files in This Item:
File Name/ File Size Content Type Version Access License
10001_200518015127035侯明亮_paper.pdf(2288KB)----限制开放View 联系获取全文

Recommended Citation:
侯明亮. 光电成像跟踪设备智能故障诊断技术研究[D]. 光电技术研究所. 中国科学院光电技术研究所. 2008.
Service
Recommend this item
Sava as my favorate item
Show this item's statistics
Export Endnote File
Google Scholar
Similar articles in Google Scholar
[侯明亮]'s Articles
CSDL cross search
Similar articles in CSDL Cross Search
[侯明亮]‘s Articles
Related Copyright Policies
Null
Social Bookmarking
Add to CiteULike Add to Connotea Add to Del.icio.us Add to Digg Add to Reddit
文件名: 10001_200518015127035侯明亮_paper.pdf
格式: Adobe PDF
此文件暂不支持浏览
所有评论 (0)
暂无评论
 
评注功能仅针对注册用户开放,请您登录
您对该条目有什么异议,请填写以下表单,管理员会尽快联系您。
内 容:
Email:  *
单位:
验证码:   刷新
您在IR的使用过程中有什么好的想法或者建议可以反馈给我们。
标 题:
 *
内 容:
Email:  *
验证码:   刷新

Items in IR are protected by copyright, with all rights reserved, unless otherwise indicated.

 

 

Valid XHTML 1.0!
Copyright © 2007-2016  中国科学院光电技术研究所 - Feedback
Powered by CSpace