故障树计算机辅助分析优化算法研究与应用
详细信息    本馆镜像全文|  推荐本文 |  |   获取CNKI官网全文
摘要
系统故障树分析,简称FTA(Fault Tree Analysis),是系统可靠性工程学的重要分支,是目前国内外公认的评价复杂系统可靠性与安全性的一种实用方法。随着计算机技术的不断提高,FTA也得到了不断发展。借助计算机来研究快速大规模的FTA算法及软件成为可靠性分析人员的一个新的研究热点。但FTA中的组合爆炸困难(计算量随故障树规模指数增长)、可视化软件开发、应用领域拓宽等问题都还有待更进一步的研究。研究高效的FTA优化算法,对拓宽工程应用领域及开发大型的可靠性维修性安全性软件都有十分重要意义,本文就这些问题开展了分析研究工作。
     本文详细阐述了故障树计算机辅助分析基本原理,研究了故障树主要参数(最小割集、最小路集、不交化最小割集、不交化最小路集)间的相互转化规律,实现了采用割集、路集相互比较的故障树定性分析优化算法及采用直接不交化来实现故障树定量分析的新算法。这些新算法的合理应用能有效地降低FTA的NP困难,为FTA的简化提供新的途径。
     用C++语言在VC开发环境下研制了故障树定性定量分析计算程序组件,并开发了一套基于此组件的FTA分析软件。该软件应用了本文提出的优化算法理论,具有最小割集和最小路集计算、可用性计算、重要度计算、基于故障树的故障诊断等主要功能。其实现采用了高阶数组降维技术及动态数值方法,当某个动态数组完成计算功能后,立即释放所占内存空间,运算速度快,能够满足微机上进行大型故障树的分析要求。
     通过对一些故障树实例进行故障分析及与其他软件的比较计算,证明了本文所实现的优化算法和应用软件是科学有效的。最后通过对优化算法及软件在安全评定、故障诊断等领域中具体应用实例的进一步研究,表明优化算法及软件在工程中具有很好的实用价值。
Fault tree analysis (FTA), an important branch in systems reliability engineering, is a common practical technique used in evaluating reliability and safety for large-complex systems at home and abroad presently. With gradual advancing techniques in computer, FTA is in a state of sound development. It has become a new hot research for reliability analyzer to study rapid and large-scale FTA algorithm and software by using computer. But problems of combinatorial explosion that computational complexity increases with the size of fault tree by exponential growth, visual software developed and application domains widened are awaiting to be further researched. Studies on efficient FTA algorithm play a more important role in widening domains of application of FTA and developing large-scale software for reliability, maintainability and safety. This paper gives an elaborate study on these problems.
    Computer-aided FTA theory has been expounded in this paper. What's more, the reciprocal transformation rule for main parameters in FTA has been studied. Also, a new method using concepts of intercomparsion between numbers of cut set and path set for fault tree qualitative analysis algorithm and applying the boolean manipulation method and disjoint method for fault tree quantitative analysis is presented here. The proposed method applied properly can be developed to reduce the size of tree and simplify qualitative and quantitative analysis of FTA.
    By use of C++ language and VC6.0, a program module has been successfully developed for fault tree qualitative and quantitative analysis and also FTA software has been realized based on this module. The software applied the optimized algorithm proposed in this paper with main functions of easily calculating minimal cut sets, minimal path sets, reliability, probability of failure, importance factors and fault diagnosis based on fault tree. The implementation of software has been adopted techniques of high-level arrays reducing dimension and dynamic numerical method with characteristics of quick memory release when some dynamic arrays complete calculations, high computing speed and sound ability to meet the demand of microcomputer for large-scale fault tree analysis.
    From the comparison of analysizing an illustrative example between FTA software developed in this paper and other reliability software, the optimized algorithm and application software presented here proved scientific and efficient. Else, by further studying on some examples in the domain of safety evaluation and fault diagnosis, Results show it is a practical method for the application to engineering systems.
引文
[1] 曾声奎,赵延弟,张建国等.系统可靠性设计分析教程.北京:北京航天航空大学出版社,2001
    [2] 卿和民,邓林,李勇.可靠性维修性工程软件的现状分析与应用.电子产品可靠性与环境试验2001,8(4):20~26
    [3] 周广涛.计算机辅助可靠性工程.北京:宇航出版社,1990
    [4] 赵廷弟,曾声奎.计算机辅助可靠性设计分析系统研究[J].航空学报,2002,21(3):206~209
    [5] 郭波,武小悦等.系统可靠性分析.长沙:国防科技大学出版社,2002
    [6] 史定华,王松瑞.故障树分析技术方法和理论.北京:北京师范大学出版社,1993
    [7] 胡云昌,陈金水.求系统失效树最小割集的新方法.中国造船,1989,104:110~120
    [8] 李剑中,龚海里等.设备安全运行与维护维修智能管理系统研究.可靠性工程进展(2002年全国可靠性学术交流会会议论文集),辽宁大连,2002.北京:科学出版社,2002:176-178
    [9] Gerogíannis, V.C., Caragiannis, I.E. A general framework for applying safety analysis to safety critical real-time applications using fault trees. Real-Time Systems, 1997, Proceedings, Ninth Euromicro Workshop on, 1997: 168-175
    [10] Gabor Kocza, Andre Bossche. Application of the integrated reliability analysis system (IRAS). Reliability Engineering and System Safety. 1999, 64: 99~107
    [11] 曾慧娥等.基于3F装备可靠性设计与管理系统研究.计算机系统应用,2002,2:48~50
    [12] Leader S, Friend R. A Probabilistic diagnostic and prognostic system for engine health and usage management. Aerospace Conference Proceedings, IEEE 2000, Vol. 6: 185~192
    [13] 王永传等.基于模糊数的故障树分析方法研究.电子产品可靠性与环境试验,2000,1:10~15
    [14] 赵艳萍,贡文伟.模糊故障树分析及其应用研究.中国安全科学学报,2001,11(6):31~35
    [15] Bossche A. Computer-aided fault tree synthesis 1.Reliability Engineering and System Safety, 1991,33:217~241
    [16] 华小洋,胡宗武.模糊故障树分析方法.机械强度,1998,20(1):35~40
    [17] 朱大奇等.基于知识的故障诊断方法综述.安徽工业大学学报,2002,19(3):197~204
    [18] 周祖德,陈幼平.现代机械制造系统的监控与故障诊断.武汉:华中理工大学出版社,1999
    [19] 吴今培,肖健华.智能故障诊断与专家系统.北京:科学出版社,1997
    [20] 梅启智,廖炯生,孙惠中.系统可靠性工程基础.北京:科学出版社,1987
    [21] 黄祥瑞.可靠性工程.北京:清华大学出版社,1990
    [22] 廖炯生.不交型布尔代数及其应用..科学通报,1983,10
    [23] Kocza G. Models and algorithms for automatic reliability assessment of complex systems.. PhD thesis, Delft University of Technology, The Netherlands, 1997
    [24] 金星,洪延姬等.工程系统可靠性数值分析方法.北京:国防工业出版社,2002
    [25] 王少萍.工程可靠性.北京:北京航天航空大学出版社,2000
    [26] 闫剑平.一种故障树模块的划分方法.北方交通大学学报,2000,24(5):63~66
    
    
    [27] Dutuit Y. A Lineartime Algorithm to Find Modules of Fault Tree [J]. IEEE Transaction on Reliability, 1996, 45(3): 423~325
    [28] 刘永宾等.割集矩阵在早期不交化FTA中的应用.天津大学学报,2000,33(3):318~323
    [29] 方逵,罗强等.用矩阵法FTA进行非单调关联系统的早期不交化,国防科技大学学报,1997,19(2):85~89
    [30] 金星,沈怀荣.故障树定性分析的优化方法.指挥技术学院学报,2001,12(6):27~29
    [31] 金星等.直接不交化方法在故障树定量分析中的应用.中国空间科学技术,21(3):60~64
    [32] 郑裕国,张康达.故障树定性和定量分析的算法.浙江工业大学学报,1995,23(1):42~47
    [33] 潘爱民,王国印.Visual C++技术内幕.第四版.北京:清华大学出版社,2001
    [34] 候俊杰..深入浅出MFC.第二版.武汉:华中科技大学出版社,2001
    [35] 戴赛.电力系统实时图形系统的基于COM的设计与实现.研究与试验,2003,23(1):12~15
    [36] 钱能.C++程序设计教程.北京:清华大学出版社,2001
    [37] 候福均,肖贵平等.计算机辅助事故树分析研究,安全与环境学报,2001,1(2):33~36
    [38] 余浩章,聂万胜.FTA可视化计算平台中计算模块的设计与实现.装备指挥技术学院学报,2002,13.(5):36~40
    [39] Y.Wang, T.Teague. A new algorithm for computer-aided fault tree synthesis journal of loss prevention in the process industries. ELSEVIER 2002, (15): 265~277.
    [40] 金星,洪延姬等.基于故障树的智能型故障诊断系统.宇航学报,2001,22(3):111~113
    [41] 朱大奇,于盛林.基于故障树最小割集的故障诊断方法研究.数据采集与处理,2002,17(3) 341~344
    [42] 王魏,黄文虎等.基于故障树最小割集和最小路集的诊断方法研究[J].数据采集与处理,1999,14(1):26~29
    [43] 余浩章等.基于故障树的故障诊断推理新方法.上海海运学院学报,2001,22(3):65~67
    [44] 王占全.面向产品质量规划的故障分析专家系统研究.浙江大学硕士学位论文,2002
    [45] 韦家增.故障树分析和模糊理论在机械故障诊断中的应用研究.合肥工业大学硕士学位论文,2002
    [46] 许化东.基于故障树分析法的汽车故障诊断专家系统的研究.合肥工业大学硕士学位论文,2002
    [47] 尹朝庆,尹皓.人工智能与专家系统。北京:中国水利水电出版社,2002
    [48] Takehisa Kohda, Koichi Inoue. Probability Evaluation of System-Failure Occurrence Based on Minimal Cut-Sets. Reliability and Maintainability Symposium, 2002. Proceedings, Annual, 28-31 Jan. 2002:190~194
    [49] Hennings W, Kuznetsov N. FAMOCVTN&CUTQN: Programs for fast analysis of large fault tree with replicated & Negated Gates. IEEE TRANSACTIONS ON RELIABILITY. 1995, 44(3): 368~375
    [50] Hae Sang Lee, Chang Hoon Lie et al. A computation method for evaluating importance-measures of Gates in a fault tree. IEEE TRANSACTIONS ON RELIABILITY. 1997, 46 (3): 360~365