基于递归神经网络的非线性系统辨识研究
|
摘要
非线性动态系统的辨识一直是控制领域研究的难点和热点。传统的前馈神经网络在处理非线性动态系统辨识中存在很多缺陷。而递归神经网络能够利用自身内部的状态反馈来描述系统的非线性特性,更适用于非线性系统的辨识。
递归神经网络有很多不同的结构,本文介绍了5种常用的递归网络：Hopfield、Jordan、Elman、DRNN、ESN网络。并将Elman、DRNN、ESN网络模型用于实际非线性系统的辨识。
对学习算法的研究一直是神经网络研究的核心。而递归神经网络常用的学习算法有BPTT、RTRL算法,它们都是基于梯度下降法的,因此避免不了误差梯度带来的诸多缺陷。因此本文在研究了3种常用智能优化算法的基础上,提出了2种改进的优化算法,对3个基准函数的优化比较了5种优化算法的性能,找到了性能最好的优化算法—IACA(Improved Ant Colony Algorithm),并将此算法用于优化神经网络的结构和参数。
本文以兰州石化某厂乙烯、丙烯精馏塔动态化工系统为辨识对象,首先通过实验研究了Elman、DRNN网络用于非线性系统辨识的优势及不足,针对传统递归网络训练算法的缺陷,将IACA用于网络结构和参数的设计,获得了较好的辨识性能；此外,文章比较细致地研究了新型递归神经网络——ESN用于乙烯、丙烯精馏塔的辨识问题,主要研究了ESN网络的3个重要参数(储备池规模N、谱半径γ、连接密度(?))对网络辨识性能的影响,并针对常规ESN网络学习算法易导致病态解的问题,将IACA用于优化ESN网络的输出权值,获得了相对最佳的辨识结果。
本文通过大量的仿真实验结果表明,递归神经网络相比于前馈网络更适合于非线性动态化工系统的辨识,将IACA算法引入到神经网络的结构及参数的设计中,有效地改善了网络的学习效率,提高了网络的训练速度,较好地克服了传统梯度下降法的局部极小问题以及ESN网络的病态解问题。
基于对乙烯、丙烯精馏塔的辨识研究,本文最后提出了基于IACA算法的ESN网络辨识模型的精馏塔产品质量实时预测方案。
文章最后对全文的工作进行了总结,并提出了一些今后研究工作的方向和重点。
The identification of nonlinear dynamic system is always the difficulty and the focus in the fields of control research.There are many defects of Traditional feedforward neural networks in dealing with nonlinear dynamic system identification.And the recurrent neural network can use its own internal system of state-feedback to reflect the non-linear characteristics,so it is more suitable for nonlinear system identification.
There are many different types of recurrent neural networks, and this paper has only described five common recurrent neural networks:Hopfield networks, Jordan network, Elman network, DRNN network and ESN network.In this paper,Elman, DRNN, ESN network models have used in the actual nonlinear system identification.
The research of learning algorithm has always been the core of the neural network research.The learning algorithm of the recurrent neural network commonly use BPTT algorithm, RTRL algorithm, which are based on gradient descent.Therefore the commonly algorithms can not prevent many defects to the error gradient.After the study of three kinds of intelligent optimization algorithm,this paper has proposed two kinds of improved optimization algorithms and has compared the optimization results of three benchmark functions to find the best performance algorithm in the five kinds of optimization algorithms.The IACA algorithm has the best performance,and the algorithm has used to optimize the neural network structure and parameters.
Taking the factory in Lanzhou,the dynamic chemical system petrochemical ethylene, propylene distillation column for the identification objects,First,It is advantages and disadvantages of the experimental study of the Elman,DRNN network for nonlinear system identification.There are many shortcomings of the traditional recurrent network training algorithm.So it has a better recognition performance for using the IACA algorithm to design the network structure and parameters.In addition,this paper has detailed studied a new type of recurrent neural networks which is ESN network,and has used it for the identification problem of ethylene, propylene distillation.Also this paper has studied the three important parameters (reserve pool size N, the spectral radiusγ,connection density (?)) of ESN network on the Network Identification properties.It is easily lead to pathological problems of ESN network regular learning algorithm,so this paper has used the IACA algorithm for optimizing the output weights of ESN network, and it has obtained the relative best results of identification.
Through a large number of simulation experiments,the results show that recurrent neural networks are more suitable for nonlinear system identification than feedforward neural networks.Using the IACA algorithm to design the network structure and parameters,it can Effectively improve the network efficiency,improve the training speed of network,also it can solve the local minimum problem of traditional gradient descent and pathological problem of ESN Network.
Finally,the distillation product quality real-time prediction program,which used the ESN network based on IACA algorithm,has proposed for the study identification of ethylene, propylene distillation.
In the final of this paper,it summarizes the work of this article and proposes some future research directions and priorities.
递归神经网络有很多不同的结构,本文介绍了5种常用的递归网络：Hopfield、Jordan、Elman、DRNN、ESN网络。并将Elman、DRNN、ESN网络模型用于实际非线性系统的辨识。
对学习算法的研究一直是神经网络研究的核心。而递归神经网络常用的学习算法有BPTT、RTRL算法,它们都是基于梯度下降法的,因此避免不了误差梯度带来的诸多缺陷。因此本文在研究了3种常用智能优化算法的基础上,提出了2种改进的优化算法,对3个基准函数的优化比较了5种优化算法的性能,找到了性能最好的优化算法—IACA(Improved Ant Colony Algorithm),并将此算法用于优化神经网络的结构和参数。
本文以兰州石化某厂乙烯、丙烯精馏塔动态化工系统为辨识对象,首先通过实验研究了Elman、DRNN网络用于非线性系统辨识的优势及不足,针对传统递归网络训练算法的缺陷,将IACA用于网络结构和参数的设计,获得了较好的辨识性能；此外,文章比较细致地研究了新型递归神经网络——ESN用于乙烯、丙烯精馏塔的辨识问题,主要研究了ESN网络的3个重要参数(储备池规模N、谱半径γ、连接密度(?))对网络辨识性能的影响,并针对常规ESN网络学习算法易导致病态解的问题,将IACA用于优化ESN网络的输出权值,获得了相对最佳的辨识结果。
本文通过大量的仿真实验结果表明,递归神经网络相比于前馈网络更适合于非线性动态化工系统的辨识,将IACA算法引入到神经网络的结构及参数的设计中,有效地改善了网络的学习效率,提高了网络的训练速度,较好地克服了传统梯度下降法的局部极小问题以及ESN网络的病态解问题。
基于对乙烯、丙烯精馏塔的辨识研究,本文最后提出了基于IACA算法的ESN网络辨识模型的精馏塔产品质量实时预测方案。
文章最后对全文的工作进行了总结,并提出了一些今后研究工作的方向和重点。
The identification of nonlinear dynamic system is always the difficulty and the focus in the fields of control research.There are many defects of Traditional feedforward neural networks in dealing with nonlinear dynamic system identification.And the recurrent neural network can use its own internal system of state-feedback to reflect the non-linear characteristics,so it is more suitable for nonlinear system identification.
There are many different types of recurrent neural networks, and this paper has only described five common recurrent neural networks:Hopfield networks, Jordan network, Elman network, DRNN network and ESN network.In this paper,Elman, DRNN, ESN network models have used in the actual nonlinear system identification.
The research of learning algorithm has always been the core of the neural network research.The learning algorithm of the recurrent neural network commonly use BPTT algorithm, RTRL algorithm, which are based on gradient descent.Therefore the commonly algorithms can not prevent many defects to the error gradient.After the study of three kinds of intelligent optimization algorithm,this paper has proposed two kinds of improved optimization algorithms and has compared the optimization results of three benchmark functions to find the best performance algorithm in the five kinds of optimization algorithms.The IACA algorithm has the best performance,and the algorithm has used to optimize the neural network structure and parameters.
Taking the factory in Lanzhou,the dynamic chemical system petrochemical ethylene, propylene distillation column for the identification objects,First,It is advantages and disadvantages of the experimental study of the Elman,DRNN network for nonlinear system identification.There are many shortcomings of the traditional recurrent network training algorithm.So it has a better recognition performance for using the IACA algorithm to design the network structure and parameters.In addition,this paper has detailed studied a new type of recurrent neural networks which is ESN network,and has used it for the identification problem of ethylene, propylene distillation.Also this paper has studied the three important parameters (reserve pool size N, the spectral radiusγ,connection density (?)) of ESN network on the Network Identification properties.It is easily lead to pathological problems of ESN network regular learning algorithm,so this paper has used the IACA algorithm for optimizing the output weights of ESN network, and it has obtained the relative best results of identification.
Through a large number of simulation experiments,the results show that recurrent neural networks are more suitable for nonlinear system identification than feedforward neural networks.Using the IACA algorithm to design the network structure and parameters,it can Effectively improve the network efficiency,improve the training speed of network,also it can solve the local minimum problem of traditional gradient descent and pathological problem of ESN Network.
Finally,the distillation product quality real-time prediction program,which used the ESN network based on IACA algorithm,has proposed for the study identification of ethylene, propylene distillation.
In the final of this paper,it summarizes the work of this article and proposes some future research directions and priorities.
引文
[1]魏东.非线性系统神经网络参数预测及控制[M].北京：机械工业出版社,2008.
[2]毕革新.递归神经网络的动态系统辨识及其在船舶运动控制中的应用研究[J].大连海事大学硕士学位论文,2009.
[3]Subudhi B,Jena D. Gupta M.M. Memetic Differential Evolution Trained Neural Networks For Nonlinear System Identification[J].IEEE Industrial and Information Systems,2008:1-6.
[4]Xinli Li,Yan Bai,Congzhi Huang. Nonlinear System Identification Using Dynamic Neural Networks Based on Genetic Algorithm[J].Intelligent Computation Technology and Automation,2008:213-217.
[5]Dongfeng,Wang Zijie,Wang Yu.The thermal process identification with radial basis function network based on quantum particle swarm optimization[J]. Sustainable Power Generation and Supply,2009:1-4.
[6]Ji Zhao,Wei Chen,Wenbo Xu. Nonlinear System Identification Based on Radial Basis Function Neural Network Using Improved Particle Swarm Optimization[J].Natural Computation,2009:409-413.
[7]王学厚,韩璞,李岩.基于PSO的RBF神经网络在热工系统辨识中的应用[J].华北电力大学学报,2008,35(1)：52-56.
[8]张蓉,冯斌.更有效的非线性系统辨识新方法[J].计算机工程与设计,2008,29(16)：4289-4292.
[9]于海波,马翠.基于对角递归神经网络系统辨识及应用[J].微计算机信息,2007,23(11)：216-217.
[10]高钦和,王孙安.基于Elman神经网络的非线性动态系统辨识[J].计算机工程与应用,2007,43(31)：87-89.
[11]彭靖波,谢寿生,孙东等.基于PSO算法的递归网络航空发动机随动系统辨识[J].系统仿真学报,2008,20(24)：6818-6820.
[12]Deng Xiaolong, Zhou Pingfang.Nonlinear Identification Based on Diagonal Recurrent Neural Network and Particle Filter[J].Natural Computation,2009:217-221.
[13]奚海蛟,张晓林.基于Elman网络的共轴式直升机动力学系统辨识[J].北京航空航天大学学报,2008,34(7)：861-864.
[14]H. Jaeger.The echo state approach to analysing and training recurrent neural networks[J]. German National Research Center for Information Technology, Tech. Report.148.2001.
[15]郭帅,韩敏.回声状态网络的非线性系统自适应辨识[J].大连理工大学硕士学位论文,2006.
[16]盛旺.基于回声状态网络的电子节气门复合智能控制[J].自动化仪表,2009,30(8)：64-66.
[17]韩敏,王亚楠.基于Kalman滤波的储备池多元时间序列在线预报器[J].自动化学报,2010,36(1)：169-173.
[18]刘颖,赵珺,王伟等.基于数据的改进回声状态网络在高炉煤气发生量预测中的应用[J].自动化学报,2009,35(6)：731-738.
[19]王玉昆,张勇.基于主元分析的转炉终点ESN静态预测模型[J].辽宁科技大学学报,2009,32(4)：360-364.
[20]王瑟.基于改进的回声状态神经网络的非线性预测[D].南京：南京工业大学硕士学位论文,2006.
[21]宋青松,冯祖仁,李人厚.用于混沌时间序列预测的多簇回响状态网络[J].物理学报,2009,58(7)：5057-5063.
[22]付琳娟,翟正军,郭阳明.基于回声状态网络的多变量预测模型的研究[J].计算机测量与控制,2009,17(7)：1356-1361.
[23]刘兴堂,梁炳成,刘力等.复杂系统建模理论、方法与技术[M].北京：科学出版社,2008.
[24]徐丽娜.神经网络控制[M].北京：电子工业出版社,2009.
[25]刘海峰,陈建军.基于神经网络的系统辨识研究[D].西安：西安电子科技大学硕士学位论文,2007.
[26]孙玮,韩璞.智能算法在系统辨识中的研究应用[D].华北电力大学硕士学位论文,2008.
[27]张良均,曹晶,蒋世忠等.神经网络实用教程[M].北京：机械工业出版社,2008.
[28]蒋宗礼.人工神经网络导论[M].北京：高等教育出版社,2005.
[29]侯媛彬,杜京义,汪梅.神经网络[M].西安：西安电子科技大学出版社,2007.
[30]丛爽,戴谊.递归神经网络的结构研究[J].计算机应用,2004(8)：18-20.
[31]丛爽,高雪鹏.几种递归神经网络及其在系统辨识中的应用[J].系统工程与电子技术,2003,25(2)：195-197.
[32]宋军,杨凌.基于递归神经网络的石化生产软测量建模方法研究[D].兰州：兰州大学硕士学位论文,2008.
[33]KU C C, Lee K Y. Diagonal Recurrent Neural Networks for Dynamic System Control[J].IEEE Trans Neural Networks,1995,6(1):144-156.
[34]段慧达,郑德玲,刘聪.基于对角递归神经网络的建模及应用[J].北京科技大学学报,2004,26(1)：103-105.
[35]Herbert Jaeger,Adaptive nonlinear system identification with echo state networks[J]. Advances in neural information processing systems.2002:593-600.
[36]Herbert Jaeger,Reservoir riddles:Suggestions for echo state network research.In Proceedings of the international joint conference on neural networks,2005:1460-1462.
[37]Herbert Jaeger,Harald Haas.Harnessing Nonlinearity:Predicting Chaotic Systems and Saving Energy in Wireless Communication[J]. Science,2004,304(5667):78-80.
[38]徐丽娜.神经网络控制(第二版)[M].北京：电子工业出版社,2003.
[39]Werbos,P.J.Backpropagation through time:what it does and how to do it[J].Proceedings of the IEEE, 1990,78(10):1550-1560.
[40]Herbert Jaeger.A tutorial on training recurrent neural networks,covering BPTT,RTRL,EKF and the "echo state network"approach[J].GMD Report 159Fraunhofer Institute,AIS,2002.
[41]丛爽,戴谊.递归神经网络逼近性能和稳定性能的研究[D].北京：中国科学技术大学硕士学位论文,2005.
[42]丁小帅.递归神经网络几个梯度学习算法的收敛性[D].辽宁：大连理工大学硕士学位论文,2007.
[43]李明.基于动态神经网络的非线性自适应逆控制研究[D].江苏：南京理工大学博士学位论文,2007.
[44]董正宏,王元钦.一种结构噪声抑制的非线性滤波器算法及模型[J].系统仿真学报,2008,20(1)：21-25.
[45]王雪.测试智能信息处理[M].北京：清华大学出版社,2008.
[46]马少平,朱小燕.人工智能[M].北京：清华大学出版社,2006.
[47]汪定伟,王俊伟,王洪峰.智能优化方法[M].北京：高等教育出版社,2007.
[48]王洪生.人工智能及其应用[M].北京：国防工业出版社,2006.
[49]阎平凡,张长水.人工神经网络与模拟进化计算[M].北京：清华大学出版社,2000.
[50]王莉.遗传算法的收敛性统一判据[J].控制理论与应用,2004,23(6)：16-19.
[51]郭观七,喻寿益.遗传算法收敛性分析的统一方法[J].控制理论与应用,2001(3)443-446.
[52]李成江.遗传算法收敛性分析及有效性初探[D].山东科技大学硕士学位论文,2001.
[53]Kennedy J,Eberhart R.Particle swarm optimization.Proceedings of IEEE International Conference on Neural Networks,Piscataway:IEEE Service Center,1995,1942-1948.
[54]纪震,廖惠连,吴青华.粒子群算法及应用[M].北京：科学出版社,2009.
[55]高尚,杨静宇.群智能算法及其应用[M].北京：中国水利水电出版社,2007.
[56]Ismail A,Engelbrecht AP.Training product units in feedforward neural networks using particle swarm optimization[A].Proceedings of the International Conference on Artificial Intelligence[C].Durban,South Africa,1999.36-40.
[57]Shi Y, Eberhart R. A modified particle swarm optimizer[J].IEEE World Congress on Computation Intelligence,1998,69-73.
[58]Dortgo M,Maniezzo V,Colomi A.Ant System:Optimization by a Colony Cooperating Agents,IEEE Trans.on Systems,Man,and Cybernetics Part B:Cybemetics,1996,26(1):29-41.
[59]Dortgo M,Gambardella L M,Ant Colony System:A Cooperative Learning Approach to the Traveling Salesman Problem,IEEE Trans,on Evolutionary Computation,1997,1(1):53-66.
[60]章国升,杨凌,肖修吉等.基于改进蚁群算法的DRNN网络动态系统辨识研究[J].中国科技在线,2009.
[61]Man Chun-tao,Li Xiao-xia.Radial Basis Function Neural Network Based on Colony Optimization[A]. 2007 International Conference on Computational Intelligence and Security Workshops[C].America:IEEE Computer Workshops,2007,59.
[62]Stutzle T,Hoos H H.Max-Min Ant System and Local Search for The Traveling Salesman Problem,In: IEEE IntConf.on Evolutionary Computation,Indianapolis:IEEE Press,1997:309-314.
[63]Zhu Qing-Bao,Yang Zhi-Jun.An Ant Colony Optimization Algorithm Based on Mutation and Dynamic Pheromone Updating,Journal of Software,2004,15(2):185-192.
[64]裴志利,王建,时小虎等,带个体差异的蚁群算法的应用[J].计算机应用研究,2008,25(4)：1036-1038.
[65]龚本灿,李腊元.基于启发式变异的蚁群算法[J].计算机工程,2008,34(8)：35-37.
[66]冀俊忠,黄振,刘椿年.基于变异和信息素扩散的多维背包问题的蚁群算法[J].计算机研究与发展,2009,46(4)：644-654.
[67]赵恒平,俞金寿.化工数据预处理及其在建模中的应用[J].华东理工大学学报,2005,31(4)：223-226.
[68]王玉涛,严其艳,杨钢等.基于主成分分析的动态神经网络预报方法及其应用[J].控制与决策, 2006,11(21)：1312-1315.
[2]毕革新.递归神经网络的动态系统辨识及其在船舶运动控制中的应用研究[J].大连海事大学硕士学位论文,2009.
[3]Subudhi B,Jena D. Gupta M.M. Memetic Differential Evolution Trained Neural Networks For Nonlinear System Identification[J].IEEE Industrial and Information Systems,2008:1-6.
[4]Xinli Li,Yan Bai,Congzhi Huang. Nonlinear System Identification Using Dynamic Neural Networks Based on Genetic Algorithm[J].Intelligent Computation Technology and Automation,2008:213-217.
[5]Dongfeng,Wang Zijie,Wang Yu.The thermal process identification with radial basis function network based on quantum particle swarm optimization[J]. Sustainable Power Generation and Supply,2009:1-4.
[6]Ji Zhao,Wei Chen,Wenbo Xu. Nonlinear System Identification Based on Radial Basis Function Neural Network Using Improved Particle Swarm Optimization[J].Natural Computation,2009:409-413.
[7]王学厚,韩璞,李岩.基于PSO的RBF神经网络在热工系统辨识中的应用[J].华北电力大学学报,2008,35(1)：52-56.
[8]张蓉,冯斌.更有效的非线性系统辨识新方法[J].计算机工程与设计,2008,29(16)：4289-4292.
[9]于海波,马翠.基于对角递归神经网络系统辨识及应用[J].微计算机信息,2007,23(11)：216-217.
[10]高钦和,王孙安.基于Elman神经网络的非线性动态系统辨识[J].计算机工程与应用,2007,43(31)：87-89.
[11]彭靖波,谢寿生,孙东等.基于PSO算法的递归网络航空发动机随动系统辨识[J].系统仿真学报,2008,20(24)：6818-6820.
[12]Deng Xiaolong, Zhou Pingfang.Nonlinear Identification Based on Diagonal Recurrent Neural Network and Particle Filter[J].Natural Computation,2009:217-221.
[13]奚海蛟,张晓林.基于Elman网络的共轴式直升机动力学系统辨识[J].北京航空航天大学学报,2008,34(7)：861-864.
[14]H. Jaeger.The echo state approach to analysing and training recurrent neural networks[J]. German National Research Center for Information Technology, Tech. Report.148.2001.
[15]郭帅,韩敏.回声状态网络的非线性系统自适应辨识[J].大连理工大学硕士学位论文,2006.
[16]盛旺.基于回声状态网络的电子节气门复合智能控制[J].自动化仪表,2009,30(8)：64-66.
[17]韩敏,王亚楠.基于Kalman滤波的储备池多元时间序列在线预报器[J].自动化学报,2010,36(1)：169-173.
[18]刘颖,赵珺,王伟等.基于数据的改进回声状态网络在高炉煤气发生量预测中的应用[J].自动化学报,2009,35(6)：731-738.
[19]王玉昆,张勇.基于主元分析的转炉终点ESN静态预测模型[J].辽宁科技大学学报,2009,32(4)：360-364.
[20]王瑟.基于改进的回声状态神经网络的非线性预测[D].南京：南京工业大学硕士学位论文,2006.
[21]宋青松,冯祖仁,李人厚.用于混沌时间序列预测的多簇回响状态网络[J].物理学报,2009,58(7)：5057-5063.
[22]付琳娟,翟正军,郭阳明.基于回声状态网络的多变量预测模型的研究[J].计算机测量与控制,2009,17(7)：1356-1361.
[23]刘兴堂,梁炳成,刘力等.复杂系统建模理论、方法与技术[M].北京：科学出版社,2008.
[24]徐丽娜.神经网络控制[M].北京：电子工业出版社,2009.
[25]刘海峰,陈建军.基于神经网络的系统辨识研究[D].西安：西安电子科技大学硕士学位论文,2007.
[26]孙玮,韩璞.智能算法在系统辨识中的研究应用[D].华北电力大学硕士学位论文,2008.
[27]张良均,曹晶,蒋世忠等.神经网络实用教程[M].北京：机械工业出版社,2008.
[28]蒋宗礼.人工神经网络导论[M].北京：高等教育出版社,2005.
[29]侯媛彬,杜京义,汪梅.神经网络[M].西安：西安电子科技大学出版社,2007.
[30]丛爽,戴谊.递归神经网络的结构研究[J].计算机应用,2004(8)：18-20.
[31]丛爽,高雪鹏.几种递归神经网络及其在系统辨识中的应用[J].系统工程与电子技术,2003,25(2)：195-197.
[32]宋军,杨凌.基于递归神经网络的石化生产软测量建模方法研究[D].兰州：兰州大学硕士学位论文,2008.
[33]KU C C, Lee K Y. Diagonal Recurrent Neural Networks for Dynamic System Control[J].IEEE Trans Neural Networks,1995,6(1):144-156.
[34]段慧达,郑德玲,刘聪.基于对角递归神经网络的建模及应用[J].北京科技大学学报,2004,26(1)：103-105.
[35]Herbert Jaeger,Adaptive nonlinear system identification with echo state networks[J]. Advances in neural information processing systems.2002:593-600.
[36]Herbert Jaeger,Reservoir riddles:Suggestions for echo state network research.In Proceedings of the international joint conference on neural networks,2005:1460-1462.
[37]Herbert Jaeger,Harald Haas.Harnessing Nonlinearity:Predicting Chaotic Systems and Saving Energy in Wireless Communication[J]. Science,2004,304(5667):78-80.
[38]徐丽娜.神经网络控制(第二版)[M].北京：电子工业出版社,2003.
[39]Werbos,P.J.Backpropagation through time:what it does and how to do it[J].Proceedings of the IEEE, 1990,78(10):1550-1560.
[40]Herbert Jaeger.A tutorial on training recurrent neural networks,covering BPTT,RTRL,EKF and the "echo state network"approach[J].GMD Report 159Fraunhofer Institute,AIS,2002.
[41]丛爽,戴谊.递归神经网络逼近性能和稳定性能的研究[D].北京：中国科学技术大学硕士学位论文,2005.
[42]丁小帅.递归神经网络几个梯度学习算法的收敛性[D].辽宁：大连理工大学硕士学位论文,2007.
[43]李明.基于动态神经网络的非线性自适应逆控制研究[D].江苏：南京理工大学博士学位论文,2007.
[44]董正宏,王元钦.一种结构噪声抑制的非线性滤波器算法及模型[J].系统仿真学报,2008,20(1)：21-25.
[45]王雪.测试智能信息处理[M].北京：清华大学出版社,2008.
[46]马少平,朱小燕.人工智能[M].北京：清华大学出版社,2006.
[47]汪定伟,王俊伟,王洪峰.智能优化方法[M].北京：高等教育出版社,2007.
[48]王洪生.人工智能及其应用[M].北京：国防工业出版社,2006.
[49]阎平凡,张长水.人工神经网络与模拟进化计算[M].北京：清华大学出版社,2000.
[50]王莉.遗传算法的收敛性统一判据[J].控制理论与应用,2004,23(6)：16-19.
[51]郭观七,喻寿益.遗传算法收敛性分析的统一方法[J].控制理论与应用,2001(3)443-446.
[52]李成江.遗传算法收敛性分析及有效性初探[D].山东科技大学硕士学位论文,2001.
[53]Kennedy J,Eberhart R.Particle swarm optimization.Proceedings of IEEE International Conference on Neural Networks,Piscataway:IEEE Service Center,1995,1942-1948.
[54]纪震,廖惠连,吴青华.粒子群算法及应用[M].北京：科学出版社,2009.
[55]高尚,杨静宇.群智能算法及其应用[M].北京：中国水利水电出版社,2007.
[56]Ismail A,Engelbrecht AP.Training product units in feedforward neural networks using particle swarm optimization[A].Proceedings of the International Conference on Artificial Intelligence[C].Durban,South Africa,1999.36-40.
[57]Shi Y, Eberhart R. A modified particle swarm optimizer[J].IEEE World Congress on Computation Intelligence,1998,69-73.
[58]Dortgo M,Maniezzo V,Colomi A.Ant System:Optimization by a Colony Cooperating Agents,IEEE Trans.on Systems,Man,and Cybernetics Part B:Cybemetics,1996,26(1):29-41.
[59]Dortgo M,Gambardella L M,Ant Colony System:A Cooperative Learning Approach to the Traveling Salesman Problem,IEEE Trans,on Evolutionary Computation,1997,1(1):53-66.
[60]章国升,杨凌,肖修吉等.基于改进蚁群算法的DRNN网络动态系统辨识研究[J].中国科技在线,2009.
[61]Man Chun-tao,Li Xiao-xia.Radial Basis Function Neural Network Based on Colony Optimization[A]. 2007 International Conference on Computational Intelligence and Security Workshops[C].America:IEEE Computer Workshops,2007,59.
[62]Stutzle T,Hoos H H.Max-Min Ant System and Local Search for The Traveling Salesman Problem,In: IEEE IntConf.on Evolutionary Computation,Indianapolis:IEEE Press,1997:309-314.
[63]Zhu Qing-Bao,Yang Zhi-Jun.An Ant Colony Optimization Algorithm Based on Mutation and Dynamic Pheromone Updating,Journal of Software,2004,15(2):185-192.
[64]裴志利,王建,时小虎等,带个体差异的蚁群算法的应用[J].计算机应用研究,2008,25(4)：1036-1038.
[65]龚本灿,李腊元.基于启发式变异的蚁群算法[J].计算机工程,2008,34(8)：35-37.
[66]冀俊忠,黄振,刘椿年.基于变异和信息素扩散的多维背包问题的蚁群算法[J].计算机研究与发展,2009,46(4)：644-654.
[67]赵恒平,俞金寿.化工数据预处理及其在建模中的应用[J].华东理工大学学报,2005,31(4)：223-226.
[68]王玉涛,严其艳,杨钢等.基于主成分分析的动态神经网络预报方法及其应用[J].控制与决策, 2006,11(21)：1312-1315.