基于神经网络和遗传算法的油田采油控制系统的研究
|
摘要
油田采油控制系统主要是用来测量油田液面高度并进行智能采油的系统。我国油田地质状况决定了部分油井在经过一段时间的开采后就会出现抽空现象，但电机仍在日夜不停的工作，这就造成电机大马拉小车现象，耗电量高，在节约能源方面造成了巨大的难题，所以该系统不仅可以满足油田采油智能控制的需要。同时，为新型采油控制系统的研制、开发和生产提供了有效的设计方案。
本系统可以根据用户的要求设计各种量程的采油控制系统。在本课题中设计的电机额定电流为50A。该类系统在国内的研究处于领先地位，因此在系统设计、设备的工艺结构设计、设备安装、VEGA测控软件开发等方面需要许多理论和技术上的创新。
本文根据实际工程背景，针对油田采油控制系统的工作原理、方案论证、硬件设计、软件算法设计、测控软件开发平台以及工艺结构设计等关键问题进行了深入的研究和讨论。
首先，按照系统提出的技术要求，依据高可靠性、高安全性、高效率、实用性强、操作方便的原则，合理设计了油田采油控制系统的总体结构，并阐述了油田采油控制系统的工作原理。
其次，根据油田抽油机采油控制系统的工作特点和技术要求，对系统的硬件电路进行了深入细致的设计研究。
然后，本文重点对油田采油控制系统的软件算法进行了深入研究。在
系统建模方面深入的研究了常规BP算法、原始训练数据的初始化方法，为了克服常规BP算法存在收敛速度慢、容易陷入局部极小点等弊病研究了同伦及非线性同伦BP算法，最终设计了采用了非线性化规范原始数据，非线同伦算法对液面和电流、蓄油曲线、电流和时间曲线等进行建模；针对优化停机时间，深入研究了常规遗传算法。并且将以上的算法都进行了仿真研究，证明了所提出的算法的有效性，对油田采油控制系统的优化具有理论指导意义和实际应用价值。
最后，本文对油田采油控制系统的外形、内部结构、设备的安装进行了设计，为系统的实施提高了强有力的保证；本文在最后还设计了计算机和控制系统间的可视化软件开发平台VEGA测控界面，给出了相应的控制功能，为系统的分析和论证奠定了基础。在本文中，我们通过软件开发平台可以深入分析研究实验数据，并对其进行了理论分析。论证了硬件方案和软件算法的高效性、准确性、实用性。上述的各种策略都是在硬件及VEGA软件的基础上实现的，最终建成了一个完整的油田采油控制系统，并达到了要求的技术指标。
Oil extraction control system in the oil field is mainly used to measure the level of the oil underground and extract oil intelligently. According to the geology of oil field in our country, most of the oil well appears empty after exploitation for some time. But when this happens, the electromotor still works on all day and all night that results in the phenomenon of “a big horse hauling a small garage”. So it makes the difficulties in retrenching the energy sources because of the large amount of the power energy used. Overall this system in the dissertation is essential that it can not only meet with the need of intelligent oil extraction, also offer the effective plan of design for the research, development and production of the new-type oil extraction control system.
This System can be designed in various scales, complying with the request of the user. It is concerned that in this dissertation the rating current of the electromotor is 50 Amperes. It is in the highest flight to set up such a system in China and is very significant to study on it. In order to build such a system, many theoretical and technological problems have to be solved in designing system, designing technics structure of the equipment, installing equipments, developing VEGA configuration and so on. Some of them are innovative work.
Based on the practices in engineering, in this dissertation some key technological problems are deeply studied and discussed on the principle of the oil extraction control system, in demonstrating the scheme, designing the
whole technics structures and hardware, designing software algorithms, and developing VEGA configuration for this system.
First, considering the technological requirements of the system and based on the study and absorption of relative theories and technologies for the key problems, the whole structure of this system is designed to be satisfied with the requirements of high reliability, safety, efficiency, practicality and convenience operations and the principle of the control system is demonstrated.
Secondly, considering the working characteristic of the control system for oil extraction and specific requirement, the research of the system’s hardware is carried on thoroughly and particularly.
Then, in this dissertation the key research is focused greatly on the software algorithms. On the one hand the general BP neural network is adopted on the system modeling, based on which the normalization method of the original training data is studied, in order to resolve the problems of the tardy convergency velocity and easily getting into the local minimum, homotopic and nonlinear homotopic BP algorithm are done deeply. In the end, nonlinear homotopic algorithm including nonlinear normalization method to original samples is used to model the relation between height and current, current and time, and the curve of the oil cumulation. On the other hand, in order to optimize the outwork time of the electromotor, the general GA is researched thoroughly. At last the simulations to the all algorithms are studied, and the validity of the algorithms is proved. All referred above are significant and valuable to the theoretical guidance and application.
Finally, the configuration, interior structure, and installation of the control system are designed, which offer an agressive potent to the actualization of the system. After that the visual interface of software developing platform named VEGA between control system and the computer is designed for the analysis and demonstration of the system, containing the corresponding control function. In the dissertation, developing platform may be used to analyze the experiment data and demonstrate the high efficiency, accuracy and practicality of the
hardware, software algorithms scheme. The above methods are all realized depending on the hardware, software and VEGA platform. Ultimately, the oil extraction control system is successfully built, and the requirement of technology of the system is satisfied.
本系统可以根据用户的要求设计各种量程的采油控制系统。在本课题中设计的电机额定电流为50A。该类系统在国内的研究处于领先地位，因此在系统设计、设备的工艺结构设计、设备安装、VEGA测控软件开发等方面需要许多理论和技术上的创新。
本文根据实际工程背景，针对油田采油控制系统的工作原理、方案论证、硬件设计、软件算法设计、测控软件开发平台以及工艺结构设计等关键问题进行了深入的研究和讨论。
首先，按照系统提出的技术要求，依据高可靠性、高安全性、高效率、实用性强、操作方便的原则，合理设计了油田采油控制系统的总体结构，并阐述了油田采油控制系统的工作原理。
其次，根据油田抽油机采油控制系统的工作特点和技术要求，对系统的硬件电路进行了深入细致的设计研究。
然后，本文重点对油田采油控制系统的软件算法进行了深入研究。在
系统建模方面深入的研究了常规BP算法、原始训练数据的初始化方法，为了克服常规BP算法存在收敛速度慢、容易陷入局部极小点等弊病研究了同伦及非线性同伦BP算法，最终设计了采用了非线性化规范原始数据，非线同伦算法对液面和电流、蓄油曲线、电流和时间曲线等进行建模；针对优化停机时间，深入研究了常规遗传算法。并且将以上的算法都进行了仿真研究，证明了所提出的算法的有效性，对油田采油控制系统的优化具有理论指导意义和实际应用价值。
最后，本文对油田采油控制系统的外形、内部结构、设备的安装进行了设计，为系统的实施提高了强有力的保证；本文在最后还设计了计算机和控制系统间的可视化软件开发平台VEGA测控界面，给出了相应的控制功能，为系统的分析和论证奠定了基础。在本文中，我们通过软件开发平台可以深入分析研究实验数据，并对其进行了理论分析。论证了硬件方案和软件算法的高效性、准确性、实用性。上述的各种策略都是在硬件及VEGA软件的基础上实现的，最终建成了一个完整的油田采油控制系统，并达到了要求的技术指标。
Oil extraction control system in the oil field is mainly used to measure the level of the oil underground and extract oil intelligently. According to the geology of oil field in our country, most of the oil well appears empty after exploitation for some time. But when this happens, the electromotor still works on all day and all night that results in the phenomenon of “a big horse hauling a small garage”. So it makes the difficulties in retrenching the energy sources because of the large amount of the power energy used. Overall this system in the dissertation is essential that it can not only meet with the need of intelligent oil extraction, also offer the effective plan of design for the research, development and production of the new-type oil extraction control system.
This System can be designed in various scales, complying with the request of the user. It is concerned that in this dissertation the rating current of the electromotor is 50 Amperes. It is in the highest flight to set up such a system in China and is very significant to study on it. In order to build such a system, many theoretical and technological problems have to be solved in designing system, designing technics structure of the equipment, installing equipments, developing VEGA configuration and so on. Some of them are innovative work.
Based on the practices in engineering, in this dissertation some key technological problems are deeply studied and discussed on the principle of the oil extraction control system, in demonstrating the scheme, designing the
whole technics structures and hardware, designing software algorithms, and developing VEGA configuration for this system.
First, considering the technological requirements of the system and based on the study and absorption of relative theories and technologies for the key problems, the whole structure of this system is designed to be satisfied with the requirements of high reliability, safety, efficiency, practicality and convenience operations and the principle of the control system is demonstrated.
Secondly, considering the working characteristic of the control system for oil extraction and specific requirement, the research of the system’s hardware is carried on thoroughly and particularly.
Then, in this dissertation the key research is focused greatly on the software algorithms. On the one hand the general BP neural network is adopted on the system modeling, based on which the normalization method of the original training data is studied, in order to resolve the problems of the tardy convergency velocity and easily getting into the local minimum, homotopic and nonlinear homotopic BP algorithm are done deeply. In the end, nonlinear homotopic algorithm including nonlinear normalization method to original samples is used to model the relation between height and current, current and time, and the curve of the oil cumulation. On the other hand, in order to optimize the outwork time of the electromotor, the general GA is researched thoroughly. At last the simulations to the all algorithms are studied, and the validity of the algorithms is proved. All referred above are significant and valuable to the theoretical guidance and application.
Finally, the configuration, interior structure, and installation of the control system are designed, which offer an agressive potent to the actualization of the system. After that the visual interface of software developing platform named VEGA between control system and the computer is designed for the analysis and demonstration of the system, containing the corresponding control function. In the dissertation, developing platform may be used to analyze the experiment data and demonstrate the high efficiency, accuracy and practicality of the
hardware, software algorithms scheme. The above methods are all realized depending on the hardware, software and VEGA platform. Ultimately, the oil extraction control system is successfully built, and the requirement of technology of the system is satisfied.
引文
李少远，席裕庚，陈增强.智能控制的新进展(Ⅰ) .控制与决策.2000，15(1)：1-5
T. L. Liao. Adaptive Robust Neural Tracking Control of A Class of Unknown Nonlinear Systems. Int. J. Systems Science. 1998,29(7): 731-743
廖俊，林建亚，任德祥.模糊神经网络快速学习算法的研究.控制与决策.1997，12(5)：606-609
W. A. Farag, V. H. Quintana, G.Lambert-Torres. A Genetic-Based Neural-Fuzzy Approach for Modeling and Control of Dynamical Systems. IEEE Trans. on NN. 1998,9(5): 756-767
Z. Michalewicz. Genetic Algorithms + Data Structures = Evolution Programs. Berlin: Springer-VERLAG, 1992:22-76
杨智民，王旭，庄显义.遗传算法在自动控制领域中的应用综述.信息与控制.2000，29(4)：329-339
S. H. Tan, Y. Yu. Adaptive Fuzzy Modeling of Nonlinear Dynamical Systems. Automatica, 1996,32(4): 637-643
C. H. Lee, S. D. Wang. A Self-Organizing Adaptive Fuzzy Controller. Fuzzy Sets and Systems. 1996, 80:295-313
S. B. Chen, L. Wu, Q. L. Wang. Self-Learning Fuzzy Neural Networks for Control of Uncertain Systems with Time Delays. IEEE Trans. on SMC.1997, 27(1): 142-148
Rumelhart D. E. et al. Learning Representation by BP Errors [J]. Nature (London), 1986, (7): 64-70
Fletcher R., et al. Function Minimization by Conjugate Gradients [J]. Computer Journal, 1964, (7): 265-271
Patrick P., et al. Minimization Method for Training Feed forward Neural
Network [J]. Neural Network, 1994, (7): 145-163
高小榕，杨福生.采用同伦BP算法进行多层前向神经网络的训练[J].计算机学报，1996，（9）：9-14
何光彩，洪炳熔.前馈神经网络的一种非线性同伦综合算法.计算机应用研究，1999（9）：9-10
彭松，方祖祥.BP神经网络学习算法的联合优化[J].电路与系统学报，2000，（3）：26-30
赵斌，吴中如.BP模型在大坝安全监测预报中的应用[J].大坝观测与土木测试，1999，（6）：1-4
S.Rudolph. On topology, size and generalization of non-linear feed-forward neural networks. Neurocomputing 1999, (16):1-22
王京慧，李宏光.递归复合型模糊神经网络结构研究。信息与控制，2003，（2）：181-184
Ljung L., et al. Adaptation and Tracking in System Identification-A Survey. Automatica. 1990, 26(1): 7-21
Buck. A. D. and A. C. Tsoi. An Adaptive Lattice Architecture for Dynamic Multi-layer Perceptions. Neural Computation. 1992, 4: 922-931
Frasconi P., M. Gori and G. Soda. Local Feedback Multi-layer Networks. Neural Computation. 1992, 4: 120-130
J. SK. Suykens, et al. Robust Local Stability of Multi-layer Recurrent NN. IEEE Trans. on NN. 2000, 11(1): 222-229
Willams R. J., et al. Experimental Analysis of the Real Time Recurrent Learning algorithm. Connection Science. 1989, 1(1): 87-111
Pineda F. J. Generalization of Back-propagation to Recurrent Neural Networks. Physical Review Letters. 1987, 59(19): 2229-2232
Antsaklis, P. J. Special Issue on Neural Networks. Control Systems Magazine. 1990, 10(3): 180-192
J. Zhang, A. J. Morres. Recurrent Neural-Fuzzy Networks for Nonlinear Process Modeling. 1999,10(2): 313-326
A. Hugget, et al. Global Optimization of a Dryer by Using Neural Network
and Genetic Algorithms. A. I. Ch. E. Journal. 1999, 45(6):1227-1238
Qin. S., et al. Comparison of four Neural Net Learning Methods for dynamic System Identification. IEEE Trans. on Neural networks. 1992, 3(1): 122-130
A. S. Poznyak. Nonlinear Adaptive Trajectory Tracking Using Dynamic Neural Network. IEEE Trans. on Neural Networks. 1999, 10(6): 1402-1411
M.Latron, et al. Modeling of Nonlinear Nonstationary Dynamic Systems with a Novel Class of Artificial Neural Networks. IEEE Trans. on NN. 1999, 10(2): 327-339
Kenneth F., et al. Neural Networks: Next Step for Simulation and Control. Power Engineering. 1991, 95(12): 41-45
O.stan and E. Kanen. A Local Linearized Least Squares Algorithm for Traning Feed-forward Neural Networks. IEEE Trans. on Neural Networks. 2000, 11(2): 487-495
Kirialkidis, kiriakos. Non-linear Control System Design via Fuzzy Modeling and Lmis. Int. j. of Control. 1999, 72(7): 676-685
I. Rivals, et al. Nonlinear Internal Model Control Using Neural Networks: Application to Process with Delay and Design Issues. IEEE Trans. on Neural Networks. 2000, 11(1): 80-90
N. S. Rubanov. A Large-wise Method and the Back Propagation Hybrid Approach to Learning a Feed-forward Neural Network. IEEE Trans. on Neural Network. 2000, 11(2): 495-305
Kramer, M. A., et al. Embedding Theoretical Models in Neural Networks. Proc. Of Amer. Contr. Conf., Chargo, 1992: 457-479
Psichogios, D. C., et al. A Hybrid Neural Networks – First Principles approach to Process Modeling. AICHE Natl. Meeting, Los Angeles, 1991: 393-399
王科俊等.神经网络建模、预测与控制.哈尔滨工程大学出版社.1996：20-40
Phansalker, V. V., et al. Analysis of the Back-Propagation Algorithm with Momentum. IEEE Trans. on NN. 1994, 5(3): 505-506
郭海丁,路志峰. 基于BP神经网络和遗传算法的结构优化设计.航空动力学报,2003,(2):216-220
S.Ghoshray, K. K. Yen. More Efficient Genetic Algorithm For Solving Optimization Problems. Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, v 5, 1995, p 4515-4520
John A.Miller, Walter D. Potter et al. An Evaluation of Local Improvement Operators for Genetic Algorithms. IEEE Transaction on system, man, and cybernetics, 1993(5): 1340-1351
Yi Shang, Guo Jie Li. New Crossover Operators In Genetic Algorithms. IEEE Int. Conf. On Tools for AI,1991:150-153
Alessio Gaspar, Philippe Collard. Time Dependent Optimization with a Folding Genetic Algorithm. IEEE Proceedings of the International Conference on Tools with Artificial Intelligence ，1997, p 125-132
J. A.Vasconcelos, J. A.Ramirez, et al. Improvements in Genetic Algorithms. IEEE Transaction on Magnetics,2001,p 3414-3417
周国芹,周春光,梁艳春,王艳.回溯遗传算法.吉林大学自然科学学报,1996(4):37-42
G. Rudolph, Convergence analysis of Canonical Genetic Algorithms. IEEE Trans. Neural Networks, 5(1994), 96-101
J. Suzuki, A Markov chain analysis on simple genetic algorithms. IEEE Trans. System, Man and Cybernetics, 25(1995), 655-659
J. Suzuki, A further result on the Markov chain model of genetic algorithms and its application to A simulated annealing – like strategy, IEEE Trans. System, Man and Cybernetics, 28(1998), 95-102
G. Rudolph, Finite Markov Chain results in evolution computation: a tour d’horizon. Fundamenta informaicae (in press), (1998), 1-22
G. Rudolph, Convergence Properties of Evolutionary Algorithms. Kokac, Hambrug, 1997
T. E. Davis, towards an extrapolation of the simulated anneals convergence theory onto the simple genetic algorithm. PhD thesis, University Florida, Gainesville, 1991
T. E. Davis and J. Principe, A Markov chain framework for the simple genetic algorithm. Evolutionary computation, 1(1993), 3, 269-288
S. W. Mahfoud and D. E. Goldberg, Parallel recombinative simulated annealing: A genetic algorithm. Parallel Computing, 21(1995), 1, 1-28
R. Cerf, The dynamics of mutation-selection algorithms with large population sizes. Ann. Inst. Henri Poincare, 32(1996), 455-508
R. Cerf, Asymptotic convergence of genetic algorithms. Adv. Appl. Prob., 30(1998), 521-550
J. A. Lozana, P. Larranaga, M. Grana and F. X. Albizuri, Genetic algorithms: Bridging the convergence gap. Theoretical Computer Science, (1999), 214
Jun He, Lishan Kang, on the convergence rate of genetic algorithms. Theoretical computer science, 229, (1999), 23-39
Il-Kwon, Jeong, Ju-Jang, Lee. Adaptive Simulated Annealing Genetic Algorithm for System Identification. Engineering Applications of Artificial Intelligence Volume: 9, Issue: 5, October, 1996, pp. 523-532
Wang, Q. J. Using genetic algorithms to optimise model parameters. Environmental Modelling and Software with Environment Data News Volume: 12, Issue: 1, 1997, pp. 27-34
A.H. Wright, Genetic Algorithms for Real Parameter Optimization, proceedings of First workshop on the Foundations of Genetic Algorithms and Classifier Systems, Indiana, 1990, pp. 205-218
R. Storn, K. Price, Differential evolution. A simple and efficient heuristic scheme for global optimization over continuous spaces, Journal of Global optimization 11 (1997) 341-359
J.A. Cabrera *, A. Simon, M. Prado. Optimal synthesis of mechanisms with genetic algorithms. Mechanism and Machine Theory 37 (2002)
1165-1177
A. C. NEARCHOU and N. A. ASPRAGATHOS. Application of genetic algorithms to point-to-point motion of redundant manipulators. Mech. Mach. Theory Vol.31, NO. 3, pp.261-270, 1996
Sánchez, M. Sagrario, Sarabia, Luis A. GINN (Genetic Inside Neural Network): towards a non-parametric training. Analytica Chimica Acta Volume: 348, Issue: 1-3, August 20, 1997, pp. 533-542
陈国良，王煦法，庄镇泉，王东升.遗传算法应用.北京：人民邮电出版社.1996
Pham, D.T., Karaboga, D. Self-tuning fuzzy controller design using genetic optimization and neural network modeling. Artificial Intelligence in Engineering Volume: 13, Issue: 2, April, 1999, pp. 119-130
T. L. Liao. Adaptive Robust Neural Tracking Control of A Class of Unknown Nonlinear Systems. Int. J. Systems Science. 1998,29(7): 731-743
廖俊，林建亚，任德祥.模糊神经网络快速学习算法的研究.控制与决策.1997，12(5)：606-609
W. A. Farag, V. H. Quintana, G.Lambert-Torres. A Genetic-Based Neural-Fuzzy Approach for Modeling and Control of Dynamical Systems. IEEE Trans. on NN. 1998,9(5): 756-767
Z. Michalewicz. Genetic Algorithms + Data Structures = Evolution Programs. Berlin: Springer-VERLAG, 1992:22-76
杨智民，王旭，庄显义.遗传算法在自动控制领域中的应用综述.信息与控制.2000，29(4)：329-339
S. H. Tan, Y. Yu. Adaptive Fuzzy Modeling of Nonlinear Dynamical Systems. Automatica, 1996,32(4): 637-643
C. H. Lee, S. D. Wang. A Self-Organizing Adaptive Fuzzy Controller. Fuzzy Sets and Systems. 1996, 80:295-313
S. B. Chen, L. Wu, Q. L. Wang. Self-Learning Fuzzy Neural Networks for Control of Uncertain Systems with Time Delays. IEEE Trans. on SMC.1997, 27(1): 142-148
Rumelhart D. E. et al. Learning Representation by BP Errors [J]. Nature (London), 1986, (7): 64-70
Fletcher R., et al. Function Minimization by Conjugate Gradients [J]. Computer Journal, 1964, (7): 265-271
Patrick P., et al. Minimization Method for Training Feed forward Neural
Network [J]. Neural Network, 1994, (7): 145-163
高小榕，杨福生.采用同伦BP算法进行多层前向神经网络的训练[J].计算机学报，1996，（9）：9-14
何光彩，洪炳熔.前馈神经网络的一种非线性同伦综合算法.计算机应用研究，1999（9）：9-10
彭松，方祖祥.BP神经网络学习算法的联合优化[J].电路与系统学报，2000，（3）：26-30
赵斌，吴中如.BP模型在大坝安全监测预报中的应用[J].大坝观测与土木测试，1999，（6）：1-4
S.Rudolph. On topology, size and generalization of non-linear feed-forward neural networks. Neurocomputing 1999, (16):1-22
王京慧，李宏光.递归复合型模糊神经网络结构研究。信息与控制，2003，（2）：181-184
Ljung L., et al. Adaptation and Tracking in System Identification-A Survey. Automatica. 1990, 26(1): 7-21
Buck. A. D. and A. C. Tsoi. An Adaptive Lattice Architecture for Dynamic Multi-layer Perceptions. Neural Computation. 1992, 4: 922-931
Frasconi P., M. Gori and G. Soda. Local Feedback Multi-layer Networks. Neural Computation. 1992, 4: 120-130
J. SK. Suykens, et al. Robust Local Stability of Multi-layer Recurrent NN. IEEE Trans. on NN. 2000, 11(1): 222-229
Willams R. J., et al. Experimental Analysis of the Real Time Recurrent Learning algorithm. Connection Science. 1989, 1(1): 87-111
Pineda F. J. Generalization of Back-propagation to Recurrent Neural Networks. Physical Review Letters. 1987, 59(19): 2229-2232
Antsaklis, P. J. Special Issue on Neural Networks. Control Systems Magazine. 1990, 10(3): 180-192
J. Zhang, A. J. Morres. Recurrent Neural-Fuzzy Networks for Nonlinear Process Modeling. 1999,10(2): 313-326
A. Hugget, et al. Global Optimization of a Dryer by Using Neural Network
and Genetic Algorithms. A. I. Ch. E. Journal. 1999, 45(6):1227-1238
Qin. S., et al. Comparison of four Neural Net Learning Methods for dynamic System Identification. IEEE Trans. on Neural networks. 1992, 3(1): 122-130
A. S. Poznyak. Nonlinear Adaptive Trajectory Tracking Using Dynamic Neural Network. IEEE Trans. on Neural Networks. 1999, 10(6): 1402-1411
M.Latron, et al. Modeling of Nonlinear Nonstationary Dynamic Systems with a Novel Class of Artificial Neural Networks. IEEE Trans. on NN. 1999, 10(2): 327-339
Kenneth F., et al. Neural Networks: Next Step for Simulation and Control. Power Engineering. 1991, 95(12): 41-45
O.stan and E. Kanen. A Local Linearized Least Squares Algorithm for Traning Feed-forward Neural Networks. IEEE Trans. on Neural Networks. 2000, 11(2): 487-495
Kirialkidis, kiriakos. Non-linear Control System Design via Fuzzy Modeling and Lmis. Int. j. of Control. 1999, 72(7): 676-685
I. Rivals, et al. Nonlinear Internal Model Control Using Neural Networks: Application to Process with Delay and Design Issues. IEEE Trans. on Neural Networks. 2000, 11(1): 80-90
N. S. Rubanov. A Large-wise Method and the Back Propagation Hybrid Approach to Learning a Feed-forward Neural Network. IEEE Trans. on Neural Network. 2000, 11(2): 495-305
Kramer, M. A., et al. Embedding Theoretical Models in Neural Networks. Proc. Of Amer. Contr. Conf., Chargo, 1992: 457-479
Psichogios, D. C., et al. A Hybrid Neural Networks – First Principles approach to Process Modeling. AICHE Natl. Meeting, Los Angeles, 1991: 393-399
王科俊等.神经网络建模、预测与控制.哈尔滨工程大学出版社.1996：20-40
Phansalker, V. V., et al. Analysis of the Back-Propagation Algorithm with Momentum. IEEE Trans. on NN. 1994, 5(3): 505-506
郭海丁,路志峰. 基于BP神经网络和遗传算法的结构优化设计.航空动力学报,2003,(2):216-220
S.Ghoshray, K. K. Yen. More Efficient Genetic Algorithm For Solving Optimization Problems. Proceedings of the IEEE International Conference on Systems, Man and Cybernetics, v 5, 1995, p 4515-4520
John A.Miller, Walter D. Potter et al. An Evaluation of Local Improvement Operators for Genetic Algorithms. IEEE Transaction on system, man, and cybernetics, 1993(5): 1340-1351
Yi Shang, Guo Jie Li. New Crossover Operators In Genetic Algorithms. IEEE Int. Conf. On Tools for AI,1991:150-153
Alessio Gaspar, Philippe Collard. Time Dependent Optimization with a Folding Genetic Algorithm. IEEE Proceedings of the International Conference on Tools with Artificial Intelligence ，1997, p 125-132
J. A.Vasconcelos, J. A.Ramirez, et al. Improvements in Genetic Algorithms. IEEE Transaction on Magnetics,2001,p 3414-3417
周国芹,周春光,梁艳春,王艳.回溯遗传算法.吉林大学自然科学学报,1996(4):37-42
G. Rudolph, Convergence analysis of Canonical Genetic Algorithms. IEEE Trans. Neural Networks, 5(1994), 96-101
J. Suzuki, A Markov chain analysis on simple genetic algorithms. IEEE Trans. System, Man and Cybernetics, 25(1995), 655-659
J. Suzuki, A further result on the Markov chain model of genetic algorithms and its application to A simulated annealing – like strategy, IEEE Trans. System, Man and Cybernetics, 28(1998), 95-102
G. Rudolph, Finite Markov Chain results in evolution computation: a tour d’horizon. Fundamenta informaicae (in press), (1998), 1-22
G. Rudolph, Convergence Properties of Evolutionary Algorithms. Kokac, Hambrug, 1997
T. E. Davis, towards an extrapolation of the simulated anneals convergence theory onto the simple genetic algorithm. PhD thesis, University Florida, Gainesville, 1991
T. E. Davis and J. Principe, A Markov chain framework for the simple genetic algorithm. Evolutionary computation, 1(1993), 3, 269-288
S. W. Mahfoud and D. E. Goldberg, Parallel recombinative simulated annealing: A genetic algorithm. Parallel Computing, 21(1995), 1, 1-28
R. Cerf, The dynamics of mutation-selection algorithms with large population sizes. Ann. Inst. Henri Poincare, 32(1996), 455-508
R. Cerf, Asymptotic convergence of genetic algorithms. Adv. Appl. Prob., 30(1998), 521-550
J. A. Lozana, P. Larranaga, M. Grana and F. X. Albizuri, Genetic algorithms: Bridging the convergence gap. Theoretical Computer Science, (1999), 214
Jun He, Lishan Kang, on the convergence rate of genetic algorithms. Theoretical computer science, 229, (1999), 23-39
Il-Kwon, Jeong, Ju-Jang, Lee. Adaptive Simulated Annealing Genetic Algorithm for System Identification. Engineering Applications of Artificial Intelligence Volume: 9, Issue: 5, October, 1996, pp. 523-532
Wang, Q. J. Using genetic algorithms to optimise model parameters. Environmental Modelling and Software with Environment Data News Volume: 12, Issue: 1, 1997, pp. 27-34
A.H. Wright, Genetic Algorithms for Real Parameter Optimization, proceedings of First workshop on the Foundations of Genetic Algorithms and Classifier Systems, Indiana, 1990, pp. 205-218
R. Storn, K. Price, Differential evolution. A simple and efficient heuristic scheme for global optimization over continuous spaces, Journal of Global optimization 11 (1997) 341-359
J.A. Cabrera *, A. Simon, M. Prado. Optimal synthesis of mechanisms with genetic algorithms. Mechanism and Machine Theory 37 (2002)
1165-1177
A. C. NEARCHOU and N. A. ASPRAGATHOS. Application of genetic algorithms to point-to-point motion of redundant manipulators. Mech. Mach. Theory Vol.31, NO. 3, pp.261-270, 1996
Sánchez, M. Sagrario, Sarabia, Luis A. GINN (Genetic Inside Neural Network): towards a non-parametric training. Analytica Chimica Acta Volume: 348, Issue: 1-3, August 20, 1997, pp. 533-542
陈国良，王煦法，庄镇泉，王东升.遗传算法应用.北京：人民邮电出版社.1996
Pham, D.T., Karaboga, D. Self-tuning fuzzy controller design using genetic optimization and neural network modeling. Artificial Intelligence in Engineering Volume: 13, Issue: 2, April, 1999, pp. 119-130