﻿ 基于可拓案例推理的故障诊断方法
 文章快速检索 高级检索

1. 海军航空工程学院科研部, 烟台 264001;
2. 海军航空工程学院研究生管理大队, 烟台 264001

Fault diagnosis method based on extension case-based reasoning
WEN Tianzhu1, XU Aiqiang1, SUN Weichao2
1. Department of Scientific Research, Naval Aeronautical and Astronautical University, Yantai 264001, China;
2. Graduate Student's Brigade, Naval Aeronautical and Astronautical University, Yantai 264001, China
Abstract: Extension case-based reasoning is a kind of knowledge reasoning method which is the combination of the extenics and case-based reasoning. Firstly, the representation of extension case using compound element was introduced, and the similarity computing method of fault symptom with interval value was given. Secondly, the reasoning process of extension case was analyzed; in the retrieve of extension cases, rough set method was adopted to confirm weight of fault symptom, and search strategy guided by rules was proposed; in the reuse of extension cases, conductive transformation was adopted to fulfill the reuse of extension case according to different search results; in the revise of extension cases, increasing and decreasing transform were adopted to add or delete fault symptoms of reused extension case and change their weight; in the retain of extension cases, three retain modes of revised extension case were discussed, including adding, deleting and replacing. Thirdly, the fault diagnosis process of this method was explained by an application case, and the time complexity of extension case retrieve was analyzed. By the comparison of the proposed search strategy and global search strategy, it is known that using the proposed search strategy can increase search efficiency and improve diagnosis speed.
Key words: extenics     case-based reasoning     case retrieve     fault diagnosis     extension transformation

1 可拓案例表示

1) 点与区间.

d(x,X)<0时,表示xX,即新案例的故障征兆值满足历史案例的故障征兆范围,两者的相似度1≥s(x,X)>0.

d(x,X)>0时,表示xX,即新案例的故障征兆值不满足历史案例的故障征兆范围,两者的相似度s(x,X)<0.

d(x,X)=0时,表示x=ax=b,即新案例的故障征兆值与历史案例的故障征兆边界值相同,两者的相似度s(x,X)=0.

2) 区间与区间.

X1X2=∅时:

X1X2≠∅时:

d(X1,X2)<0时,表示X1X2≠∅,即新案例的故障征兆范围与历史案例的故障征兆范围有交集,两者的相似度1≥s(X1,X2)>0.

d(X1,X2)>0时,表示X1X2=∅,即新案例的故障征兆范围与历史案例的故障征兆范围不相符,两者的相似度s(X1,X2)<0.

X1=∅或X2=∅时,两者的相似度s(X1,X2)=0.

3) 点与区间套.

d(x,X0)=d(x,X)且xX0时:

k(x)>0时,表示xX0,即新案例的故障征兆值满足历史案例的故障征兆范围,两者的相似度1≥s(x,X0,X)>0.

k(x)<-1时,表示xX,即新案例的故障征兆值不满足案例库中该故障征兆的最大取值范围,两者的相似度s(x,X0,X)<-1.

k(x)=-1时,表示x=cx=d,即新案例的故障征兆值与案例库中该故障征兆最大取值范围的边界值相同,两者的相似度s(x,X0,X)=-1.

2 可拓案例推理

 图 1 可拓案例推理的生命周期模型 Fig. 1 Life cycle model of extension case-based reasoning

2.1 可拓案例检索

2.1.1 相似度计算

2.1.2 确定权值

2.1.3 检索策略

2.2 可拓案例重用

2.3 可拓案例修改

2.4 可拓案例保存

3 应用案例

 故障征兆 离散化值 0 1 2 3 c1 断开 闭合 c2 (5.8,8.9) [8.9,11.6] (11.6,14.4) 其他 c3 (6.4,9.3) [9.3,13.4] (13.4,15.7) 其他 c4 (4.6,7.8) [7.8,10.9] (10.9,13.9) 其他 c5 (7.2,10.2) [10.2,12.1] (12.1,14.1) 其他 c6 (18.9,26.5) [26.5,28.5] (28.5,30.6) 其他 c7 (16.4,27.8) [27.8,29.7] (29.7,33.5) 其他 c8 (23.6,27.4) [27.4,30.2] (27.4,34.7) 其他 c9 (25.1,29.1) [29.1,31.4] (29.1,33.9) 其他

 故障征兆 依赖度 权重 c1 0.30 0.060 c2 0.69 0.138 c3 0.71 0.142 c4 0.80 0.160 c5 0.76 0.152 c6 0.42 0.084 c7 0.44 0.088 c8 0.50 0.100 c9 0.38 0.076

 检索策略 故障征兆计算次数 可拓案例计算个数 全局检索策略 900 100 本文检索策略 315 35

4 结 论

1) 与全局检索策略相比,本文提出的检索策略时间复杂度低、检索效率高.

2) 可拓案例推理中结合可拓变化实现的可拓案例重用和可拓案例修改更利于计算机实现.

3) 该方法可用于机载电子设备故障诊断,并具有多故障诊断能力.

 [1] Dendani-Hadiby N, Khadir M T.A fault diagnosis application based on a combination case-based reasoning and ontology approach[J].International Journal of Knowledge-based and Intelligent Engineering Systems, 2013, 17(4):305-317. Click to display the text [2] 张秦岭,郭宏,姜旭.大型无人机电源系统故障诊断专家系统[J].北京航空航天大学学报, 2013, 39(8):1026-1030. Zhang Q L, Guo H, Jiang X.Fault diagnosis expert system for electric power system of large-scale UAVs[J].Journal of Beijing University of Aeronautics and Astronautics, 2013, 39(8):1026-1030(in Chinese). Cited By in Cnki (1) [3] 李青,史雅琴,周扬.基于案例推理方法在飞机故障诊断中的应用[J].北京航空航天大学学报, 2007, 33(5):622-626. Li Q, Shi Y Q, Zhou Y.CBR methodology application in fault diagnosis of aircraft[J].Journal of Beijing University of Aeronautics and Astronautics, 2007, 33(5):622-626(in Chinese). Cited By in Cnki (53) [4] Im H K, Park S C.Case-based reasoning and neural network based expert system for personalization[J].Expert Systems with Applications, 2007, 32(1):77-85. Click to display the text [5] 贾艳华,莫蓉,杨海成,等.基于可拓理论的CBR系统实例检索模型[J].计算机工程与应用, 2013, 49(2):258-260. Jia Y H, Mo R, Yang H C, et al.Model for case retrieval in CBR system based on extenics[J].Computer Engineering and Applications, 2013, 49(2):258-260(in Chinese). Cited By in Cnki (2) [6] 韩丹.基于CBR及可拓学的汽车故障诊断系统研究[D].合肥:合肥工业大学, 2010. Han D.Research of automotive fault diagnosis system based on CBR and extenics[D].Hefei:Hefei University of Technology, 2010(in Chinese). Cited By in Cnki (3) [7] 豆增发.基于可拓规则和案例推理的混合专家系统[D].西安:西安电子科技大学, 2007. Dou Z F.A novel hybrid expert system based on extension rule and case-based reasoning[D].Xi'an:Xidian University, 2007(in Chinese). Cited By in Cnki (8) [8] Xu M, Yu H Y, Shen J.New algorithm for CBR-RBR fusion with robust thresholds[J].Chinese Journal of Mechanical Engineering, 2012, 25(6):1255-1263. Click to display the text [9] Yang C Y, Cai W.Extenics theory, method and application[M].Beijing:Science Press, 2013:49-68. [10] 柳玉,贲可荣.案例推理的故障诊断技术研究综述[J].计算机科学与探索, 2011, 5(10):865-879. Liu Y, Ben K R.Survey of fault diagnosis based on case reasoning[J].Journal of Frontiers of Computer Science and Technology, 2011, 5(10):865-879(in Chinese). Cited By in Cnki (11) [11] Lau A, Tsui E, Lee W B.An ontology-based similarity measurement for problem-based case reasoning[J].Expert Systems with Applications, 2009, 36(3):6574-6579. Click to display the text [12] 王国胤,姚一豫,于洪.粗糙集理论与应用研究综述[J].计算机学报, 2009, 32(7):1229-1247. Wang G Y, Yao Y Y, Yu H.A survey on rough set theory and applications[J].Chinese Journal of Computers, 2009, 32(7):1229-1247(in Chinese). Cited By in Cnki (5) [13] Geng Z Q, Zhu Q X.Rough set-based heuristic hybrid recognizer and its application in fault diagnosis[J].Expert Systems with Applications, 2009, 36(2):2711-2718. Click to display the text [14] Qi J, Hu J, Peng Y H, et al.A case retrieval method combined with similarity measurement and multi-criteria decision making for concurrent design[J].Expert Systems with Applications, 2009, 36(7):1036-1057. Click to display the text [15] 王波.基于CBR的飞机机载设备故障诊断系统[D].西安:西北工业大学, 2007. Wang B.Fault diagnosis system of on-board equipments of airplane by case-based reasoning[D].Xi'an:Northwestern Polytechnical University, 2007(in Chinese). Cited By in Cnki (15)

文章信息

WEN Tianzhu, XU Aiqiang, SUN Weichao

Fault diagnosis method based on extension case-based reasoning

Journal of Beijing University of Aeronautics and Astronsutics, 2015, 41(11): 2124-2130.
http://dx.doi.org/10.13700/j.bh.1001-5965.2014.0736