图模型与学习算法结合的贝叶斯网络自动建模

沈琳; 于劲松; 唐荻音; 刘浩

doi:10.13700/j.bh.1001-5965.2015.0445

图模型与学习算法结合的贝叶斯网络自动建模

doi: 10.13700/j.bh.1001-5965.2015.0445

沈琳¹,
于劲松^1,2, ,,
唐荻音¹,
刘浩^1,3

1.
北京航空航天大学自动化科学与电气工程学院, 北京 100083
2. 北京航空航天大学先进航空发动机协同创新中心, 北京 10008
3. 解放军95809部队, 93分队, 沧州 061736

详细信息

作者简介:
沈琳女,硕士研究生。主要研究方向:故障预测与健康管理技术。Tel.: 010-82338693 E-mail: shenlin9177@163.com;于劲松男,副教授,硕士生导师。主要研究方向:预测与健康管理技术、自动测试系统。Tel.: 010-82338693 E-mail: yujs@buaa.edu.cn

通讯作者:
于劲松,Tel.: 010-82338693 E-mail: yujs@buaa.edu.cn

中图分类号: TP277
计量
- 文章访问数: 816
- HTML全文浏览量: 55
- PDF下载量: 563
- 被引次数: 0
出版历程
- 收稿日期: 2015-07-02
- 网络出版日期: 2016-07-20

Automatic learning of Bayesian network structure using graph model and learning algorithm

SHEN Lin¹,
YU Jinsong^{1,2
, ,},
TANG Diyin¹,
LIU Hao^1,3

1.
School of Automation Science and Electrical Engineering, Beijing University of Aeronautics and Astronautics, Beijing 100083, China
2. Collaborative Innovation Center of Advanced Aero-Engine, Beijing University of Aeronautics and Astronautics, Beijing 10008
3. Unit 93, Army 95809 of PLA, Cangzhou 061736, China

摘要

摘要: 针对纯数据驱动的贝叶斯网络结构学习算法的准确度和效率较低的问题，提出了一种融合多信号流图模型与K2学习算法的贝叶斯网络自动建模方法。该方法利用多信号流图模型能够描述信号之间传递与依赖关系的能力，结合K2学习算法在结构学习中的优势，实现了专家知识与数据驱动方法有效融合的贝叶斯网络结构自动学习算法。通过与常用网络结构学习算法的对比实验证明，该融合算法显著降低了结构学习对学习范围和训练数据规模的要求，具有更高的学习准确度和运算效率。采用真实系统实例阐述了该融合算法的应用过程，验证了算法的实用性。
- 贝叶斯网络 /
- 结构学习 /
- 多信号流图 /
- K2算法 /
- 故障诊断
Abstract: In order to improve the accuracy and efficiency of the data-driven approaches in learning Bayesian network structure, expert knowledge is usually implemented in the learning algorithm. To deal with the lack of effective ways to combine the expert knowledge and the data-driven learning approaches in the existing methods, this paper proposes an automatic learning method for Bayesian network structure learning, which combines multi-signal flow graphs and learning algorithm K2. The method inserts expert knowledge into data-driven learning methods, using the information of relationships between signals from multi-signal flow graphs and the structure learning algorithm K2, to achieve automatic learning of Bayesian network structure. Numerical analysis, compared with other typical network structure learning algorithms, proves that the proposed method significantly lowers the structure learning requirements for learning scale and training data size and provides a higher learning accuracy and computation efficiency. The application of the proposed method is illustrated using a real engineering system and verified the practicability of the algorithm at the same time.
- Bayesian networks /
- structure learning /
- multi-signal flow graphs /
- K2 algorithm /
- fault diagnosis

HTML全文

参考文献(21)

[1]	JENSEN F V.Bayesian networks and decision graphs[M].New York:Springer,2001:32-34.
[2]	EFRON B.Bayes' theorem in the 21st century[J].Science,2013,340(6137):1177-1178.
[3]	AOUAY S,JAMOUSSI S,BEN AYED Y.Particle swarm optimization based method for Bayesian network structure learning[C]//2013 5th International Conference on Modeling,Simulation and Applied Optimization (ICMSAO).Piscataway,NJ:IEEE Press,2013:1-6.
[4]	张振海,王晓明,党建武,等.基于专家知识融合的贝叶斯网络结构学习方法[J].计算机工程与应用,2014,50(2):1-4.ZHANG Z H,WANG X M,DANG J W,et al.Bayesian network structure learning method based on expert knowledge fusion[J].Computer Engineering and Applications,2014,50(2):1-4(in Chinese).
[5]	BALABAN E,SAXENA A,NARASIMHAN S,et al.Prognostic health-management system development for electromechanical actuators[J].Journal of Aerospace Information Systems,2015,12(3):329-344.
[6]	代京,于劲松,张平,等.基于多信号流图的诊断贝叶斯网络建模[J].北京航空航天大学学报,2009,35(4):472-475.DAI J,YU J S,ZHANG P,et al.Diagnostic Bayesian networks modeling based on multi-signal flow graphs[J].Journal of Beijing University of Aeronautics and Astronautics,2009,35(4):472-475(in Chinese).
[7]	陈希祥,邱静,刘冠军.测试不确定条件下基于贝叶斯网络的装备测试优化选择技术[J].中国机械工程,2011,22(4):379-384.CHEN X X,QIU J,LIU G J.Test selection of materiel based on Bayesian network under test uncertainty[J].China Mechanical Engineering,2011,22(4):379-384(in Chinese).
[8]	Qualtech Systems Inc.Testability,engineering and maintenance system(TEAMS)[EB/OL].[2010-07-26]. http://www.teamqsi.com/products/teams-designer/.
[9]	吴红,王维平,杨峰.融合先验信息的贝叶斯网络结构学习方法[J].系统工程与电子技术,2012,34(12):2585-2591.WU H,WANG W P,YANG F.Structure learning method of Bayesian network with prior information [J].Systems Engineering and Electronics,2012,34(12):2585-2591(in Chinese).
[10]	MASEGOSA A R,MORAL S.New skeleton-based approaches for Bayesian structure learning of Bayesian networks[J].Applied Soft Computing,2013,13(2):1110-1120.
[11]	BOUCHAALA L,MASMOUDI A,GARGOURI F,et al.Improving algorithms for structure learning in Bayesian networks using a new implicit score[J].Expert Systems with Applications,2010,37(7):5470-5475.
[12]	COOPER G F,HERSKOVITS E. A Bayesian method for the induction of probabilistic networks from data[J].Machine Learning,1992,9(4):309-347.
[13]	CHENG J,GREINER R,KELLY J,et al.Learning Bayesian networks from data: An information-theory based approach[J].Artificial Intelligence,2002,137(12):43-90.
[14]	SILANDER T,ROOS T,MYLLYMÄKI P.Learning locally minimax optimal Bayesian networks[J].International Journal of Approximate Reasoning,2010,51(5):544-557.
[15]	WELLMAN M P,BREESE J S,GOLDMAN R P.From knowledge bases to decision models[J].The Knowledge Engineering Review,1992,7(1):35-53.
[16]	DEB S,PATTIPATI K R,SHRESTHA R.QSI's integrated diagnostics toolset[C]//1997 IEEE Autotestcon Proceedings AUTOTESTCON,97.Piscataway,NJ:IEEE Press,1997:408-421.
[17]	YU K,WANG H,WU X.A parallel algorithm for learning Bayesian networks[M]//Advances in Knowledge Discovery and Data Mining. Berlin Heidelberg:Springer,2007:1055-1063.
[18]	曾安,李晓兵,杨海东,等.基于最小描述长度和K2的贝叶斯网络结构学习算法[J].东北师大学报(自然科学版),2014,46(3):53-58.ZENG A,LI X B,YANG H D,et al.Bayesian network structure learning based on minimum description length and K2 algorithm[J].Journal of Northeast Normal University(Natural Science Edition),2014,46(3):53-58(in Chinese).
[19]	吴永广,庞世春.K2 & HC结构学习算法[J].计算机与数字工程,2014,42(7):1137-1140.WU Y G,PANG S C.K2 & HC structure learning algorithm[J].Computer & Digital Engineering,2014,42(7):1137-1140(in Chinese).
[20]	TIAN J,PEARL J.Causal discovery from changes[C]//Proceedings of the Seventeenth Conference on Uncertainty in Artificial Intelligence. San Francisco: Morgan Kaufmann Publishers Inc.,2001:512-521.
[21]	TSAMARDINOS I,BROWN L E,ALIFERIS C F.The max-min hill-climbing Bayesian network structure learning algorithm[J].Machine Learning,2006,65(1):31-78.