粗糙集理论与神经网络相结合的故障诊断方法研究
|
摘要
随着计算机技术与控制技术的发展与应用,系统设备的自动化程度大幅提高。故障原因、故障过程和故障现象错综复杂,计算智能领域的一些理论,如人工神经网络、粗糙集理论等已经在故障诊断中得到了广泛的应用。粗糙集理论是一种处理模糊和不确定知识的工具,它简化决策规则,提取有效的信息,能够消除知识冗余性的问题。但粗糙集的容错能力还不够理想,当核属性受噪声污染时有可能会出现误判的情况。因此为了提高粗糙集的容错能力,将粗糙集与神经网络相结合,构建粗糙集(RS)和神经网络(ANN)相结合的系统,充分利用粗糙集理论对知识的约简能力和神经网络的容错学习能力。本文利用二者的优势,将RS和ANN相结合,提出了一种基于粗糙集-神经网络的故障诊断模型。
首先对粗糙集理论的数据约简问题进行了深入探讨。数据约简是粗糙集理论的核心内容之一。主要研究了运用粗糙集理论进行属性约简的方法,由于运用粗糙集理论进行属性约简是一个典型的NP(多项式复杂程度的非确定性)问题,提出了一种基于改进的二进制可辨识矩阵与布尔代数的属性约简方法,从而大大简化了可辨识矩阵,使属性约简计算量大幅度减小,可以快速得到给定要求下的属性约简。
其次,对基于神经网络的连续数据离散化与故障诊断进行研究。粗糙集理论是一种基于离散数据进行处理的方法,连续数据的离散化直接影响到它的处理效果,为此提出了一种基于自组织映射神经网络的离散化算法,使得权值向量在输入向量空间中相互分离,形成各自代表的输入模式,实现特征自动识别的聚类分析功能。仿真结果验证了方法的有效性。
最后,采用基于粗糙集-神经网络故障诊断方法,对田纳西-依斯曼过程(TEP)过程进行故障诊断,取得了良好的故障诊断效果。
With the development and appliment of computer and control technical, the automation of system devices has being elevated sharply. The reason、process and phenomenon of fault is complicated. Some theory in computational intelligence fields, such as artificial neural network, Rough Set theory has been widely used in fault diagnosis. Rough Set theory is a tool for fuzzy and dubious knowledges. It can simplify decision rules and find useful information. It also can settle the problem caused by redundant data. But the fault-tolerance ability of rough sets is insufficiently ideal. When core’s attributes are polluted by noise, it may arouse error judgment. Therefore in order to enhance the fault-tolerance ability of Rough sets theory, unify Rough sets and neural network, construct an intelligent mixture system of Rough sets and neural networks, which fully develops the reduction ability of Rough sets and the classification ability of neural networks. In this paper, combining RS and ANN, using the both advantages, giving a fault diagnosis system model based on the rough set-neural network.
Firstly, studied the attributes reduction methods used in rough set theory. The knowledge reduction is the core part of Rough sets, but it is a NP(Non-deterministic Polynomical) problem on theory. And on this basis giving a method based on the improved binary matrix and boolean algebra for attribute reduction, reduce the amount of computation, the request of attribute reduction can be quickly given.
Secondly, researching the discrete method. Rough set theory is based on discrete data processing methods, discretization of continuous data directly impact on its effect. In this paper given a method based on SOM(Self-Organizing feature Map), transformated the input vector values which are detached from each other into a new input mode. The results of example demonstrated the effectiveness of this method.
Finally, we used the fault diagnosis system by rough-neural network on Tennessee-Eastman process(TEP) fault diagnosis, it achieved good diagnosis result.
首先对粗糙集理论的数据约简问题进行了深入探讨。数据约简是粗糙集理论的核心内容之一。主要研究了运用粗糙集理论进行属性约简的方法,由于运用粗糙集理论进行属性约简是一个典型的NP(多项式复杂程度的非确定性)问题,提出了一种基于改进的二进制可辨识矩阵与布尔代数的属性约简方法,从而大大简化了可辨识矩阵,使属性约简计算量大幅度减小,可以快速得到给定要求下的属性约简。
其次,对基于神经网络的连续数据离散化与故障诊断进行研究。粗糙集理论是一种基于离散数据进行处理的方法,连续数据的离散化直接影响到它的处理效果,为此提出了一种基于自组织映射神经网络的离散化算法,使得权值向量在输入向量空间中相互分离,形成各自代表的输入模式,实现特征自动识别的聚类分析功能。仿真结果验证了方法的有效性。
最后,采用基于粗糙集-神经网络故障诊断方法,对田纳西-依斯曼过程(TEP)过程进行故障诊断,取得了良好的故障诊断效果。
With the development and appliment of computer and control technical, the automation of system devices has being elevated sharply. The reason、process and phenomenon of fault is complicated. Some theory in computational intelligence fields, such as artificial neural network, Rough Set theory has been widely used in fault diagnosis. Rough Set theory is a tool for fuzzy and dubious knowledges. It can simplify decision rules and find useful information. It also can settle the problem caused by redundant data. But the fault-tolerance ability of rough sets is insufficiently ideal. When core’s attributes are polluted by noise, it may arouse error judgment. Therefore in order to enhance the fault-tolerance ability of Rough sets theory, unify Rough sets and neural network, construct an intelligent mixture system of Rough sets and neural networks, which fully develops the reduction ability of Rough sets and the classification ability of neural networks. In this paper, combining RS and ANN, using the both advantages, giving a fault diagnosis system model based on the rough set-neural network.
Firstly, studied the attributes reduction methods used in rough set theory. The knowledge reduction is the core part of Rough sets, but it is a NP(Non-deterministic Polynomical) problem on theory. And on this basis giving a method based on the improved binary matrix and boolean algebra for attribute reduction, reduce the amount of computation, the request of attribute reduction can be quickly given.
Secondly, researching the discrete method. Rough set theory is based on discrete data processing methods, discretization of continuous data directly impact on its effect. In this paper given a method based on SOM(Self-Organizing feature Map), transformated the input vector values which are detached from each other into a new input mode. The results of example demonstrated the effectiveness of this method.
Finally, we used the fault diagnosis system by rough-neural network on Tennessee-Eastman process(TEP) fault diagnosis, it achieved good diagnosis result.
引文
1. Shengwei Wang,Zhiming Jiang.Valve fault detection and diagnosis based on CMAC neural networks.Energy and Buildings[J].2004,53(36):599-610
2.周东华,叶银忠.现代故障诊断与容错控制[M].北京:清华大学出版社,2001.
3. B Liu.Selection of wavelet packer basis foe rotating machinery fault diagnosis. Journal of Sound and Vibration[J].2005,284(3):567-582.
4. Wang S, Xiao F.AHU sensor fault diagnosis using principal component analysis method.Energy and Buildings[J]. 2004,36(2):147-160.
5. Frank Kimmich,Anselm Schwarte,Rolf Isermann.Fault detection for modern Diesel engines using signal and process model-based methods.Control Engineering Practice [J].2005,13(1):189-203.
6. Bear R V.Failure accommodation in linear systems through self-reorganization.report, Information Sciences[J].1971,284(3):567-582.
7. Pawlak Z.Rough Sets.International Journal of Computer and Information Science[J]. 1982,11(1):341-356.
8. Mohamed Q.A-RST:A generalization of rough set theory.Information Sciences[J],2000,12(124):301-316.
9. Grzymala-Busse J W.A new version of the rule induction system LERS.Fundamenta informatics[J].1997,12(2):27-39.
10. Nigel Bevan.Measuring Usability as Quality of Use.Journal of Software Quality[J].1995,22(4):115-130.
11.郝丽娜,王伟.粗糙集-神经网络故障诊断方法研究[J].东北大学学报,2003,24,(3):252-255.
12. Tsumoto S.Automation extraction of mediacal expert system rules from clinical databases based on rough sets theroy.Journal of Information Sciences[J].1998,11(112):67-87
13. Pawlak Z,et al.Rough sets.Communications of ACM,1995,38(11)89-95.
14.郭小荟,马小平.基于粗糙集-神经网络集成的故障诊断[J].控制工程,2007, 14(1):53-56.
15. Stefanowski J. On rough set based approaches to induction of decision rules [C].In:L Polkowski A Skowron eds.Rough sets in knowledge discovery 1 methodology and applications Physica-Verlag Heideberg:1998.500-530.
16. Ziarko W T Arciszewski S Hoda. Urban rail corridor control through machine learning[J]. An Intelligent Vehicle-Highway System Approach,1994,45(2):91-97.
17. Czogala E,etc.Idea of a rough fuzzy controller and its application to the stabilization of a pendulum-car system[J].Fuzzy Sets and Systems,1995,72(1):61-73.
18. Peters J F K Ziaei S Ramanna.Approximate time rough control concepts and application to satellite attitude control[C].In.Proc of RSCTC 98.Warsaw Poland,1998.491-498.
19. Jackson A G M Ohmer H Ai-kamhawi.Rough sets analysis of chalcopyrite semiconductor band gap data[C].In.T Y Lin ed.The Third International Workshop on Rough Set and Soft Computing Proceedings(RSSC’94)San Jose State university. San Jose CA,1994.408-417.
20. Nowicki R,R Slowinski,J Stefanowski.Evaluation of vibroacoustic diagnostic systems by means of the rough set theory[J].Journal of Computers in Industry,1992,20(1),141-152.
21. Czyzewski A R Krolikowski.Applications of fuzzy logic rough sets to audio signal enhancement[C]. In. S L Pal A Skowron eds .Rough Fuzzy Hybrudization-A New Trend in Decision Making Springer-Verlag Singapore,1999.397-410.
22. Pawlak Z.Rough Sets[J].International Journal of Computer and Information Science,1982,341(4):341-356.
23. Sheng Fa,Yuan.Fu,Lei Chu.Support vector machines-based fault diagnosis for turbo-pump rotor[J].Mechanical Systems and Signal Processing,2006,20(9):39-42.
24.谭天乐,宋执环,李平.基于粗糙集的故障诊断方法[J].浙江大学学报(工学报),2003,37(1):47-50
25.张文修,吴志伟.粗糙集理论与方法[M].北京:科学出版社,2001.
26.陈文清.基于ANNES的故障诊断技术研究[D]:[硕士学位论文].武汉:武汉理工大学,2002..
27.付琼.智能故障诊断技术的研究与应用[D]:[硕士学位论文].大连:大连理工大学,2002.
28.文福栓,韩桢祥.基于遗传算法和模拟退火算法的电力系统的故障诊断[J].中国电机工程学报,1994,14(3):110-113.
29. Leger,R P,Garland,Wm,Poehlman W.F.S.Fault detection and diagnosis using statistical control charts and artificial neural networks[J].Artificial Intelligence in Engineering,1998,12(2):35-47.
30.郝丽娜,王伟,吴光宇.粗糙集-神经网络故障诊断方法研究[J].东北大学学报(自然科学版),2003,24(3):252-255.
31.曾黄霖.粗糙集理论与应用(修订版)[M].重庆:重庆大学出版社,1998
32.郑乐,黄湘林.离散数学[M].上海:华东师范大学出版社,1984
33. Lingras P L.Rough neural networks.Proc of the 6th Int Conf on Information Processingand Management of Uncertainty in Knowledge-based Systems[C],Granada,1996.1445-1450.
34.王庆东.基于粗糙集的数据挖掘方法研究[D]:[博士学位论文].浙江:浙江大学,2005.
35. Plonka L,Mrozek A.Rule-based stabilization of the inverted pendulum[J].Computational Intelligence,1995,11(2):348-356.
36. Czogola E.Idear of a rough fuzzy controller and its application to the stabilization of a pendulum-car system.Fuzzy sets and systems, 1995,72(1):61-73.
37. Mrozek A.Rough sets and dependency analysis among attributes in computers implementations of expert’s inference models[J].International Journal of Man-Machine Studies,1989,30(4):457-473.
38. Slowinski R.Rough set approach to decision analysis[J].AI expert,1995,18(5):19-25.
39.刘清.Rough集及Rough推理[M].北京:科学出版社,2001.
40. D.Dubois,Prade.Putting Rough Sets and Fuzzy Sets Together[J].Kluwer Academic Publishers,1992,45(3):23-232.
41.韩祯祥,张琦,文福栓.粗糙集理论及其应用综述[J].控制理论与应用,1999,16(2):153-157.
42. Slowinski R.A generalized definition of rough approxomations based on similarity[J]. Knowledge and data engineering,2000,12(2):331-336.
43.曾志民,江戈.粗糙集理论中析取范式生成算法[J].福建电脑,2008,30(2):70-71.
44. Pawlak Z, Peters J F,Skowron A,et al.Rough measures:Theory and Application.of Rough Set Theory[J].Granular Computing,2001,5(2):177-184.
45.庞彦军,刘开第,吴海玉,姚兰.粗糙集理论及其在故障诊断中的应用[J].河北建筑科技学院学报2006,6(1):103-105.
46.何明,冯博琴,马兆丰等.一种基于粗糙集的神经网络构造方法[J].西安交通大学学报,2004,38(12):1240-1242.
47.孙增圻,智能控制理论与技术[M].北京:清华大学出版社,1997。
48. Amitaba Roy,Sankar K.Pal.Fuzzy discretization of feature space for a rough set classifier[J].Pattern Recognition Letters,2003,24(2):921-933.
49.王国胤.Rough集理论与知识获取[M].西安:西安交通大学出版社,2001.
50.候利娟,王国胤,聂能.粗糙集理论中的离散化问题[J].计算机科学,2000,12(1):89-94.
51. Nguyen H S.Discretization of real value attributes:Boolean reasoning Approach[J].Ph D Dissertion.Warsaw University,1997,34(1):214-216.
52. Eberhart R C,Kennedy J.A new optimizer using Particle Swarm Theory[C].In:Proc of Sixth International Symposium on Micro Machine and HumanScience,Nogoya,Japan,1995,39-43.
53. J.M.Hyman,S.Steimberg.The Convergenc of Mimetic Discretization for Rough Grids[J].Computers and Mathematics with Applications,2004,47(2):1565-1610.
54. Richard Butterworth,Dan A.Simovici.A greedy algorithm for supervised discretization[J].Journal of Biomedical Informatics,2004,37(2):285-292.
55.石红.一种基于粗糙集的离散化方法[J].模式识别与人工智能,2006,3(2):412-416.
56.何亚群,胡寿松.具有区间数的粗糙神经网络故障认定方法[J].应用科学学报,2004,12(3):283-286.
57.朱大奇,史慧.人工神经网络原理集应用[M].北京:科学出版社,2006:120-138
58.杨叔子,丁洪,史铁林.基于知识的诊断推理[M].北京:清华大学出版社,1993.
59.陶志,许宝栋,汪定伟.一种基于粗糙集理论的连续属性离散化方法[J].东北大学报,2003,8(1):747-750.
60.凌维业,贾民平,许飞云.粗糙集神经网络故障诊断系统的优化方法研究[J].中国电机工程学报,2003,23(5):98-102.
61. Yahia M E,Mahmod R,Sulaiman N,et al.Rough nearul expert systems[J].Expert System with Application,2000,18(1):87-99.
62. Chen Zunde.Intelligent system combined rough set with aritificial neural network and its application[J].Pattern Reconition and Artificial Intelligence,1999,12(1):1-5.
63.叶斌,王存进,周庆敏。优化的粗糙神经网络在故障诊断中的应用[J].微计算机信息,2007,34(3):158-159.
64. E.L.Russell,L.H.Chiang,and R.D.Braatz.Data-driven Techniques for Fault Detection and Diagnosis in Chemical Processes[J].Springer Verlag,London,2000,3(1):213-216.
65.张文修,仇国芳.基于粗糙集的不确定决策[M].北京:清华大学出版社,2005.
2.周东华,叶银忠.现代故障诊断与容错控制[M].北京:清华大学出版社,2001.
3. B Liu.Selection of wavelet packer basis foe rotating machinery fault diagnosis. Journal of Sound and Vibration[J].2005,284(3):567-582.
4. Wang S, Xiao F.AHU sensor fault diagnosis using principal component analysis method.Energy and Buildings[J]. 2004,36(2):147-160.
5. Frank Kimmich,Anselm Schwarte,Rolf Isermann.Fault detection for modern Diesel engines using signal and process model-based methods.Control Engineering Practice [J].2005,13(1):189-203.
6. Bear R V.Failure accommodation in linear systems through self-reorganization.report, Information Sciences[J].1971,284(3):567-582.
7. Pawlak Z.Rough Sets.International Journal of Computer and Information Science[J]. 1982,11(1):341-356.
8. Mohamed Q.A-RST:A generalization of rough set theory.Information Sciences[J],2000,12(124):301-316.
9. Grzymala-Busse J W.A new version of the rule induction system LERS.Fundamenta informatics[J].1997,12(2):27-39.
10. Nigel Bevan.Measuring Usability as Quality of Use.Journal of Software Quality[J].1995,22(4):115-130.
11.郝丽娜,王伟.粗糙集-神经网络故障诊断方法研究[J].东北大学学报,2003,24,(3):252-255.
12. Tsumoto S.Automation extraction of mediacal expert system rules from clinical databases based on rough sets theroy.Journal of Information Sciences[J].1998,11(112):67-87
13. Pawlak Z,et al.Rough sets.Communications of ACM,1995,38(11)89-95.
14.郭小荟,马小平.基于粗糙集-神经网络集成的故障诊断[J].控制工程,2007, 14(1):53-56.
15. Stefanowski J. On rough set based approaches to induction of decision rules [C].In:L Polkowski A Skowron eds.Rough sets in knowledge discovery 1 methodology and applications Physica-Verlag Heideberg:1998.500-530.
16. Ziarko W T Arciszewski S Hoda. Urban rail corridor control through machine learning[J]. An Intelligent Vehicle-Highway System Approach,1994,45(2):91-97.
17. Czogala E,etc.Idea of a rough fuzzy controller and its application to the stabilization of a pendulum-car system[J].Fuzzy Sets and Systems,1995,72(1):61-73.
18. Peters J F K Ziaei S Ramanna.Approximate time rough control concepts and application to satellite attitude control[C].In.Proc of RSCTC 98.Warsaw Poland,1998.491-498.
19. Jackson A G M Ohmer H Ai-kamhawi.Rough sets analysis of chalcopyrite semiconductor band gap data[C].In.T Y Lin ed.The Third International Workshop on Rough Set and Soft Computing Proceedings(RSSC’94)San Jose State university. San Jose CA,1994.408-417.
20. Nowicki R,R Slowinski,J Stefanowski.Evaluation of vibroacoustic diagnostic systems by means of the rough set theory[J].Journal of Computers in Industry,1992,20(1),141-152.
21. Czyzewski A R Krolikowski.Applications of fuzzy logic rough sets to audio signal enhancement[C]. In. S L Pal A Skowron eds .Rough Fuzzy Hybrudization-A New Trend in Decision Making Springer-Verlag Singapore,1999.397-410.
22. Pawlak Z.Rough Sets[J].International Journal of Computer and Information Science,1982,341(4):341-356.
23. Sheng Fa,Yuan.Fu,Lei Chu.Support vector machines-based fault diagnosis for turbo-pump rotor[J].Mechanical Systems and Signal Processing,2006,20(9):39-42.
24.谭天乐,宋执环,李平.基于粗糙集的故障诊断方法[J].浙江大学学报(工学报),2003,37(1):47-50
25.张文修,吴志伟.粗糙集理论与方法[M].北京:科学出版社,2001.
26.陈文清.基于ANNES的故障诊断技术研究[D]:[硕士学位论文].武汉:武汉理工大学,2002..
27.付琼.智能故障诊断技术的研究与应用[D]:[硕士学位论文].大连:大连理工大学,2002.
28.文福栓,韩桢祥.基于遗传算法和模拟退火算法的电力系统的故障诊断[J].中国电机工程学报,1994,14(3):110-113.
29. Leger,R P,Garland,Wm,Poehlman W.F.S.Fault detection and diagnosis using statistical control charts and artificial neural networks[J].Artificial Intelligence in Engineering,1998,12(2):35-47.
30.郝丽娜,王伟,吴光宇.粗糙集-神经网络故障诊断方法研究[J].东北大学学报(自然科学版),2003,24(3):252-255.
31.曾黄霖.粗糙集理论与应用(修订版)[M].重庆:重庆大学出版社,1998
32.郑乐,黄湘林.离散数学[M].上海:华东师范大学出版社,1984
33. Lingras P L.Rough neural networks.Proc of the 6th Int Conf on Information Processingand Management of Uncertainty in Knowledge-based Systems[C],Granada,1996.1445-1450.
34.王庆东.基于粗糙集的数据挖掘方法研究[D]:[博士学位论文].浙江:浙江大学,2005.
35. Plonka L,Mrozek A.Rule-based stabilization of the inverted pendulum[J].Computational Intelligence,1995,11(2):348-356.
36. Czogola E.Idear of a rough fuzzy controller and its application to the stabilization of a pendulum-car system.Fuzzy sets and systems, 1995,72(1):61-73.
37. Mrozek A.Rough sets and dependency analysis among attributes in computers implementations of expert’s inference models[J].International Journal of Man-Machine Studies,1989,30(4):457-473.
38. Slowinski R.Rough set approach to decision analysis[J].AI expert,1995,18(5):19-25.
39.刘清.Rough集及Rough推理[M].北京:科学出版社,2001.
40. D.Dubois,Prade.Putting Rough Sets and Fuzzy Sets Together[J].Kluwer Academic Publishers,1992,45(3):23-232.
41.韩祯祥,张琦,文福栓.粗糙集理论及其应用综述[J].控制理论与应用,1999,16(2):153-157.
42. Slowinski R.A generalized definition of rough approxomations based on similarity[J]. Knowledge and data engineering,2000,12(2):331-336.
43.曾志民,江戈.粗糙集理论中析取范式生成算法[J].福建电脑,2008,30(2):70-71.
44. Pawlak Z, Peters J F,Skowron A,et al.Rough measures:Theory and Application.of Rough Set Theory[J].Granular Computing,2001,5(2):177-184.
45.庞彦军,刘开第,吴海玉,姚兰.粗糙集理论及其在故障诊断中的应用[J].河北建筑科技学院学报2006,6(1):103-105.
46.何明,冯博琴,马兆丰等.一种基于粗糙集的神经网络构造方法[J].西安交通大学学报,2004,38(12):1240-1242.
47.孙增圻,智能控制理论与技术[M].北京:清华大学出版社,1997。
48. Amitaba Roy,Sankar K.Pal.Fuzzy discretization of feature space for a rough set classifier[J].Pattern Recognition Letters,2003,24(2):921-933.
49.王国胤.Rough集理论与知识获取[M].西安:西安交通大学出版社,2001.
50.候利娟,王国胤,聂能.粗糙集理论中的离散化问题[J].计算机科学,2000,12(1):89-94.
51. Nguyen H S.Discretization of real value attributes:Boolean reasoning Approach[J].Ph D Dissertion.Warsaw University,1997,34(1):214-216.
52. Eberhart R C,Kennedy J.A new optimizer using Particle Swarm Theory[C].In:Proc of Sixth International Symposium on Micro Machine and HumanScience,Nogoya,Japan,1995,39-43.
53. J.M.Hyman,S.Steimberg.The Convergenc of Mimetic Discretization for Rough Grids[J].Computers and Mathematics with Applications,2004,47(2):1565-1610.
54. Richard Butterworth,Dan A.Simovici.A greedy algorithm for supervised discretization[J].Journal of Biomedical Informatics,2004,37(2):285-292.
55.石红.一种基于粗糙集的离散化方法[J].模式识别与人工智能,2006,3(2):412-416.
56.何亚群,胡寿松.具有区间数的粗糙神经网络故障认定方法[J].应用科学学报,2004,12(3):283-286.
57.朱大奇,史慧.人工神经网络原理集应用[M].北京:科学出版社,2006:120-138
58.杨叔子,丁洪,史铁林.基于知识的诊断推理[M].北京:清华大学出版社,1993.
59.陶志,许宝栋,汪定伟.一种基于粗糙集理论的连续属性离散化方法[J].东北大学报,2003,8(1):747-750.
60.凌维业,贾民平,许飞云.粗糙集神经网络故障诊断系统的优化方法研究[J].中国电机工程学报,2003,23(5):98-102.
61. Yahia M E,Mahmod R,Sulaiman N,et al.Rough nearul expert systems[J].Expert System with Application,2000,18(1):87-99.
62. Chen Zunde.Intelligent system combined rough set with aritificial neural network and its application[J].Pattern Reconition and Artificial Intelligence,1999,12(1):1-5.
63.叶斌,王存进,周庆敏。优化的粗糙神经网络在故障诊断中的应用[J].微计算机信息,2007,34(3):158-159.
64. E.L.Russell,L.H.Chiang,and R.D.Braatz.Data-driven Techniques for Fault Detection and Diagnosis in Chemical Processes[J].Springer Verlag,London,2000,3(1):213-216.
65.张文修,仇国芳.基于粗糙集的不确定决策[M].北京:清华大学出版社,2005.