置信规则库参数训练的布谷鸟搜索算法
|
摘要
置信规则库(belief rule base,BRB)中参数的选取直接影响着推理的精度.为了得到更加有效的参数训练方法,本文基于群智能算法中的布谷鸟搜索(Cuckoo Search,CS)算法进行改进拓展.针对布谷鸟搜索算法中Levy飞行后期出现的搜索速度慢和精度低的问题,引入自适应扰动函数进行优化,进而提出一种新的BRB参数训练方法.在多极值函数的拟合实验中,以均方误差与运行时间作为比较指标,验证本文改进方法的有效性.在输油管道泄漏检测的问题实例中,以平均绝对误差和运行时间作为比较指标,与其他现有的参数训练方法进行比较,实验结果表明,本文提出的算法具有更好的推理效率和准确度.
The selection of parameters in the belief rule base( BRB) is that directly affects the reasoning accuracy of the problem. In order to obtain a more effective approach of parameter training,this paper is based on the Cuckoo Search algorithm which is one of the swarm intelligence algorithms and then to improve and expand it. Aiming at the problem of slow searching and low precision in the later period of Levy flight in the Cuckoo Search algorithm,this paper introduces the self-adaptive disturbance function and then a new parameter training approach of BRB parameter training is proposed. In the fitting experiment of multi-extreme function,the mean square error and the run time are used as the comparison indexes to verify the effectiveness of the improved approach. Then,in the case of pipeline leak detection,the average absolute error and the run time are used as the comparison indexes to compare with other existing training approaches. The experimental results show that the algorithm proposed in this paper has better reasoning efficiency and accuracy.
The selection of parameters in the belief rule base( BRB) is that directly affects the reasoning accuracy of the problem. In order to obtain a more effective approach of parameter training,this paper is based on the Cuckoo Search algorithm which is one of the swarm intelligence algorithms and then to improve and expand it. Aiming at the problem of slow searching and low precision in the later period of Levy flight in the Cuckoo Search algorithm,this paper introduces the self-adaptive disturbance function and then a new parameter training approach of BRB parameter training is proposed. In the fitting experiment of multi-extreme function,the mean square error and the run time are used as the comparison indexes to verify the effectiveness of the improved approach. Then,in the case of pipeline leak detection,the average absolute error and the run time are used as the comparison indexes to compare with other existing training approaches. The experimental results show that the algorithm proposed in this paper has better reasoning efficiency and accuracy.
引文
[1]Dempster A P.A generalization of Bayesian inference[J].Journal of the Royal Statistical Society-Series B,1968,30(2):205-247.
[2]Shafer G.A mathematical theory of evidence[M].Princeton:Princeton University Press,1976.
[3]Huang C L,Yong K.Multiple attribute decision making methods and applications,a state-of art survey[M].Berlin:Springer-Verlag,1981.
[4]Zadeh L Z.Fuzzy sets[J].Information and Control,1965,8(3):338-353.
[5]Sun R.Robust reasoning:integrating rule-based and similaritybased reasoning[J].Artificial Intelligence,1995,75(2):241-295.
[6]Zhou Zhi-jie,Yang Jian-bo,Hu Chang-hua,et al.Belief rule based expert system and complex system modeling:1stedition[M].Beijing:Science Press,2011.
[7]Liu J,Yang J B,Ruan D,et al.Self-tuning of fuzzy belief rule bases for engineering system safety analysis[J].Annals of Operations Research,2008,163(1):143-168.
[8]Zhou Z J,Hu C H,Yang J B,et al.Online updating belief rule based system for pipeline leak detection under expert intervention[J].Expert Systems with Applications,2009,36(4):7700-7709.
[9]Yang J B,Liu J,Xu D L,et al.Optimization models for training belief-rule-based systems[J].IEEE Transactions on Systems,M an,and Cybernetics-part A:Systems and Humans,2007,37(4):569-585.
[10]Ma Chang-feng.Optimization method and Matlab programming[M].Bejing:Science Press,2010.
[11]Wu Wei-kun,Yang Long-hao,Fu Yang-geng,et al.Parameter training approach for belief rule base using the accelerating of gradient algorithm[J].Journal of Frontiers of Computer Science and Technology,2014,8(8):989-1001.
[12]Su Qun,Yang Long-hao,Fu Yang-geng,et al.Parameter training approach based on variable particle sw arm optimization for belief rule base[J].Journal of Computer Applications,2014,34(8):2161-2165.
[13]Wang Han-jie,Yang Long-hao,Fu Yang-geng,et al.Differential evolutionary algorithm for parameter training of belief rule base under expert intervention[J].Computer Science,2015,42(5):88-93.
[14]Yang Hui,Wu Pei-ze,Ni Ji-liang.Belief rule base parameter training approach based on improved particle sw arm optimization[J].Computer Engineering and Design,2017,38(2):400-404.
[15]Yang X S,Deb S.Cuckoo search via Levy flights[C].Proceedings of World Congress on Nature&Biologically Inspired Computing,India,USA:IEEE Publications,2009:210-214.
[16]Chang Rui,Wang Hong-wei,Yang Jian-bo.An algorithm for training parameters in belief rule-bases based on the gradient and dichotomy methods[J].Journal of Systems Engineering,2007,25(Sup.):287-291.
[17]Chang L,Zhou Z J,You Y,et al.Belief rule based expert system for classification problems w ith new rule activation and w eight calculation procedures[J].Information Sciences,2016,336(C):75-91.
[18]Yang X S,DEB S.Engineering optimization by cuckoo search[J].Int'l Journal of Mathematical Modeling and Numerical Optimization,2010,1(4):330-343.
[19]Chen Y W,Yang J B,Xu D L,et al.On the inference and approximation properties of belief rule based systems[J].Information Sciences,2013,234(11):121-135.
[20]Price K.Differential Evolution vs.the functions of the 2nd ICEO[C].Proceedings of the 1997 IEEE International Conference of Evolutionary Computation,1997:153-157.
[21]Xu D L,Liu J,Yang J B,et al.Inference and learning methodology of belief-rule-based expert system for pipeline leak detection[J].Expert Systems w ith Applications,2007,32(1):103-113.
[6]周志杰,杨剑波,胡昌华,等.置信规则库专家系统与复杂系统建模:第1版[M].北京:科学出版社,2011.
[10]马昌凤.最优化方法及其Matlab程序设计[M].北京:科学出版社,2010.
[11]吴伟昆,杨隆浩,傅仰耿,等.基于加速梯度求法的置信规则库参数训练方法[J].计算机科学与探索,2014,8(8):989-1001.
[12]苏群,杨隆浩,傅仰耿,等.基于变速粒子群优化的置信规则库参数训练方法[J].计算机应用,2014,34(8):2161-2165.
[13]王韩杰,杨隆浩,傅仰耿,等.专家干预下置信规则库参数训练的差分进化算法[J].计算机科学,2015,42(5):88-93.
[14]杨慧,吴沛泽,倪继良.基于改进粒子群置信规则库参数训练算法[J].计算机工程与设计,2017,38(2):400-404.
[16]常瑞,王红卫,杨剑波.基于梯度法和二分法的置信规则库参数训练方法[J].系统工程,2007,25(增刊):287-291.
[2]Shafer G.A mathematical theory of evidence[M].Princeton:Princeton University Press,1976.
[3]Huang C L,Yong K.Multiple attribute decision making methods and applications,a state-of art survey[M].Berlin:Springer-Verlag,1981.
[4]Zadeh L Z.Fuzzy sets[J].Information and Control,1965,8(3):338-353.
[5]Sun R.Robust reasoning:integrating rule-based and similaritybased reasoning[J].Artificial Intelligence,1995,75(2):241-295.
[6]Zhou Zhi-jie,Yang Jian-bo,Hu Chang-hua,et al.Belief rule based expert system and complex system modeling:1stedition[M].Beijing:Science Press,2011.
[7]Liu J,Yang J B,Ruan D,et al.Self-tuning of fuzzy belief rule bases for engineering system safety analysis[J].Annals of Operations Research,2008,163(1):143-168.
[8]Zhou Z J,Hu C H,Yang J B,et al.Online updating belief rule based system for pipeline leak detection under expert intervention[J].Expert Systems with Applications,2009,36(4):7700-7709.
[9]Yang J B,Liu J,Xu D L,et al.Optimization models for training belief-rule-based systems[J].IEEE Transactions on Systems,M an,and Cybernetics-part A:Systems and Humans,2007,37(4):569-585.
[10]Ma Chang-feng.Optimization method and Matlab programming[M].Bejing:Science Press,2010.
[11]Wu Wei-kun,Yang Long-hao,Fu Yang-geng,et al.Parameter training approach for belief rule base using the accelerating of gradient algorithm[J].Journal of Frontiers of Computer Science and Technology,2014,8(8):989-1001.
[12]Su Qun,Yang Long-hao,Fu Yang-geng,et al.Parameter training approach based on variable particle sw arm optimization for belief rule base[J].Journal of Computer Applications,2014,34(8):2161-2165.
[13]Wang Han-jie,Yang Long-hao,Fu Yang-geng,et al.Differential evolutionary algorithm for parameter training of belief rule base under expert intervention[J].Computer Science,2015,42(5):88-93.
[14]Yang Hui,Wu Pei-ze,Ni Ji-liang.Belief rule base parameter training approach based on improved particle sw arm optimization[J].Computer Engineering and Design,2017,38(2):400-404.
[15]Yang X S,Deb S.Cuckoo search via Levy flights[C].Proceedings of World Congress on Nature&Biologically Inspired Computing,India,USA:IEEE Publications,2009:210-214.
[16]Chang Rui,Wang Hong-wei,Yang Jian-bo.An algorithm for training parameters in belief rule-bases based on the gradient and dichotomy methods[J].Journal of Systems Engineering,2007,25(Sup.):287-291.
[17]Chang L,Zhou Z J,You Y,et al.Belief rule based expert system for classification problems w ith new rule activation and w eight calculation procedures[J].Information Sciences,2016,336(C):75-91.
[18]Yang X S,DEB S.Engineering optimization by cuckoo search[J].Int'l Journal of Mathematical Modeling and Numerical Optimization,2010,1(4):330-343.
[19]Chen Y W,Yang J B,Xu D L,et al.On the inference and approximation properties of belief rule based systems[J].Information Sciences,2013,234(11):121-135.
[20]Price K.Differential Evolution vs.the functions of the 2nd ICEO[C].Proceedings of the 1997 IEEE International Conference of Evolutionary Computation,1997:153-157.
[21]Xu D L,Liu J,Yang J B,et al.Inference and learning methodology of belief-rule-based expert system for pipeline leak detection[J].Expert Systems w ith Applications,2007,32(1):103-113.
[6]周志杰,杨剑波,胡昌华,等.置信规则库专家系统与复杂系统建模:第1版[M].北京:科学出版社,2011.
[10]马昌凤.最优化方法及其Matlab程序设计[M].北京:科学出版社,2010.
[11]吴伟昆,杨隆浩,傅仰耿,等.基于加速梯度求法的置信规则库参数训练方法[J].计算机科学与探索,2014,8(8):989-1001.
[12]苏群,杨隆浩,傅仰耿,等.基于变速粒子群优化的置信规则库参数训练方法[J].计算机应用,2014,34(8):2161-2165.
[13]王韩杰,杨隆浩,傅仰耿,等.专家干预下置信规则库参数训练的差分进化算法[J].计算机科学,2015,42(5):88-93.
[14]杨慧,吴沛泽,倪继良.基于改进粒子群置信规则库参数训练算法[J].计算机工程与设计,2017,38(2):400-404.
[16]常瑞,王红卫,杨剑波.基于梯度法和二分法的置信规则库参数训练方法[J].系统工程,2007,25(增刊):287-291.