基于BP神经网络的水稻干燥智能控制研究
|
摘要
我国是粮食大国,粮食安全贮存问题关系国家前途和命运,而收获的粮食必须经过干燥达到安全水分后才能长期贮存,但目前我国粮食机械干燥还未实现真正的自动化控制。本研究目的在于提高稻谷干燥机连续式干燥的自动化控制水平,提高干燥后稻谷水分的一致性,增强粮食产后干燥设备在国际市场的竞争力。
本文以连续式水稻干燥机实物为研究对象,在实地采集数据的基础上,建立水稻连续干燥的物理模型,进而建立了数学模型,可计算连续干燥过程中水稻含水率变化情况,同时建立神经网络模型对干燥过程的控制进行仿真模拟。
在实测数据的基础上,参阅了大量的文献资料。以薄层干燥理论为指导,将数学计算结果和实测值进行对比验证,调整了计算稻谷连续干燥的公式参数,对水稻连续式干燥过程进行了数学建模。在数学模型的基础上,对整个连续干燥过程进行模拟计算,讨论和分析了环境温度、环境相对湿度、高温风温度、低温风温度、风量、谷物初始含水率以及排粮速度因素变化与水稻排粮终含水率变化的关系。同时基于神经网络理论,建立了干燥过程控制的BP神经网络模型,模拟干燥过程的智能控制,也就是通过控制干燥通风量来抵消其它因素变化所带来的影响,从而使水稻排出时的终含水率趋于一致。
本文以计算机和MATLAB为辅助工具,进行了连续干燥过程数学模型的模拟计算和干燥过程控制神经网络模型的仿真模拟。数学模型的计算结果和实测值基本吻合,可以用于连续式干燥过程的模拟计算,具有实际指导意义。BP神经网络控制模型通过学习后,预测值能够满足稻谷连续干燥过程控制的要求。最后设计了稻谷BP神经网络控制计算程序。
China has a large need for food grain, the security storage for grain is related to our nation's future and destiny. Grain must be dried to safe water content for long-term storage, but automation control is not really applied to machinery for grain drying. Purpose of this study is to improve the control level of automation for continuous flow grain dryers and moisture consistency of dried grain, as well as increasing the competitiveness of grain dryers in the international market.
Taking the real continuous flow paddy dryer as a research object, many measured data are collected. Based on a physical model, a mathematical model is made to compute the paddy moisture during the drying process and then a BP neural network model is built for the control simulation of the drying process.
On the basis of measured data and references and directed by the thin layer drying theory, through comparing between computed results and the measured results, some parameters in mathematical formulas are adjusted, then a mathematical model is made for continuous computation of the paddy drying process. According to simulation computation of this model, influences of many parameters including environment temperature, environment relative humidity, high and low temperature of drying medium, wind capacity, initial paddy moisture and paddy discharging speed (drying time) are discussed. A BP neural network model is built for computation of wind capacity in the control simulation of the drying process. That is, through the adjustment of amount of drying air to control the drying process as well as the paddy moisture and decrease the gap of the final moisture content of dried paddy.
With the assistance of computer and MATLAB, a simulation computation of the mathematical model and a control simulation of the BP neural network model are made. Conclusions are drawn that the mathematical simulation result for the final moisture content of paddy is well in line with the measured data and the predicted values of BP neural network model meets requirements of the drying process control. Finally, a program is designed to calculate the control process of BP neural network in paddy dyring.
本文以连续式水稻干燥机实物为研究对象,在实地采集数据的基础上,建立水稻连续干燥的物理模型,进而建立了数学模型,可计算连续干燥过程中水稻含水率变化情况,同时建立神经网络模型对干燥过程的控制进行仿真模拟。
在实测数据的基础上,参阅了大量的文献资料。以薄层干燥理论为指导,将数学计算结果和实测值进行对比验证,调整了计算稻谷连续干燥的公式参数,对水稻连续式干燥过程进行了数学建模。在数学模型的基础上,对整个连续干燥过程进行模拟计算,讨论和分析了环境温度、环境相对湿度、高温风温度、低温风温度、风量、谷物初始含水率以及排粮速度因素变化与水稻排粮终含水率变化的关系。同时基于神经网络理论,建立了干燥过程控制的BP神经网络模型,模拟干燥过程的智能控制,也就是通过控制干燥通风量来抵消其它因素变化所带来的影响,从而使水稻排出时的终含水率趋于一致。
本文以计算机和MATLAB为辅助工具,进行了连续干燥过程数学模型的模拟计算和干燥过程控制神经网络模型的仿真模拟。数学模型的计算结果和实测值基本吻合,可以用于连续式干燥过程的模拟计算,具有实际指导意义。BP神经网络控制模型通过学习后,预测值能够满足稻谷连续干燥过程控制的要求。最后设计了稻谷BP神经网络控制计算程序。
China has a large need for food grain, the security storage for grain is related to our nation's future and destiny. Grain must be dried to safe water content for long-term storage, but automation control is not really applied to machinery for grain drying. Purpose of this study is to improve the control level of automation for continuous flow grain dryers and moisture consistency of dried grain, as well as increasing the competitiveness of grain dryers in the international market.
Taking the real continuous flow paddy dryer as a research object, many measured data are collected. Based on a physical model, a mathematical model is made to compute the paddy moisture during the drying process and then a BP neural network model is built for the control simulation of the drying process.
On the basis of measured data and references and directed by the thin layer drying theory, through comparing between computed results and the measured results, some parameters in mathematical formulas are adjusted, then a mathematical model is made for continuous computation of the paddy drying process. According to simulation computation of this model, influences of many parameters including environment temperature, environment relative humidity, high and low temperature of drying medium, wind capacity, initial paddy moisture and paddy discharging speed (drying time) are discussed. A BP neural network model is built for computation of wind capacity in the control simulation of the drying process. That is, through the adjustment of amount of drying air to control the drying process as well as the paddy moisture and decrease the gap of the final moisture content of dried paddy.
With the assistance of computer and MATLAB, a simulation computation of the mathematical model and a control simulation of the BP neural network model are made. Conclusions are drawn that the mathematical simulation result for the final moisture content of paddy is well in line with the measured data and the predicted values of BP neural network model meets requirements of the drying process control. Finally, a program is designed to calculate the control process of BP neural network in paddy dyring.
引文
[1]李军富.我国谷物干燥机械的发展现状及对策.农机化研究.2006(9)：44-46页
[2]国家粮食局.实施农村粮食产后减损安全保障工程指导意见.国粮办展(2009)16号,2009-1-19
[3]顾冰,赵学工等.稻谷干燥智能控制系统的研究.电气控制.2007(3)：31-35页
[4]Li Changyou, Ban Hua. On-line moisture metering technology and device for paddy drying process. The Proceedings of the 5th Asia-Paciifc Drying Conference. World Scientiifc Publishing Co.Pte.Ltd,2007:490-495P
[5]李长友.谷物含水率在线检测装置[P].中国专利：03222718.3,2003-12-24
[6]李长友,班华.一种检测谷物含水率的方法及其装置[P].中国专利：200610123461.3,2007-05-16
[7]曹崇文,汪喜波.谷物干燥的自动控制.现代化农业.2002(2)：40-44页
[8]郭光辉.仿人智能控制算法的现状及趋势.安徽职业技术学院学报.2007(9).13-15页
[9]李国昉,齐玉斌等.谷物干燥仿人智能控制系统.农业机械学报.2007,38(1)：83-86页
[10]李国昉,李栋等.连续流动式粮食干燥机的仿人智能控制器.农业工程学报.2006(7)：105-107页
[11]齐玉斌,李国防等.基于虚拟仪器的谷物干燥智能控制器.粮食流通及仓储.2008(8).12-13页,16页
[12]周尚波,柳玉炯.基于EKF的模糊神经网络快速自组织学习算法.计算机应用.2010,27(7)：2462-2464,2471页
[13]HUANG Huan, WU Cong-xin. Approximation capabilities of multilayer fuzzy neural networks on the set of fuzzy-valued functions. Information Sciences, 2009,179(16):2762-2773P
[14]DENG Xing-sheng,WANG xin-zhou.Incremental learning of dynamic fuzzy neural networks for accurate system modeling.Fuzzy Sets and Systems,2009,160(7):972-987P
[15]RIGATOS G, ZHANG Q. Fuzzy model validation using the local statistical approach. Fuzzy Sets and Systems,2009,160(7):882-904P
[16]周岩,王盛等.基于改进粒子群算法的模糊神经网络及其在短时天气预报中的应用.计算机应用与软件.2010,27(5)：234-237页
[17]陈伯芳,尹平林等.基于模糊神经网络的温度控制系统研究.计算机与数字工 程.2010,38(7)：54-57页
[18]Fabry-asztalos L, Andonie R, et al. A genetic algorithm optimized fuzzy neural network analysis of the affinity of inhibitors for HIV-1 protease. Bioorganic & Medicinal Chemistry,2008,16(6):2903-2911P
[19]李业德,李业刚.顺流式谷物烘干机的模糊控制系统.农业工程学报.2002,18(1)：162-164页
[20]李长友,曹艳明.谷物循环干燥机控制系统硬件设计,农业机械学报.2002,33(6)：86-88页
[21]曹艳明,李长友.谷物循环干燥机控制系统软件设计.农业机械学报.2003,34(1)：83-85页
[22]李长友,班华.基于深层干燥解析理论的粮食干燥自适应控制系统设计.农业工程学报.2008(4)：142-146页
[23]王树东,唐伟强.应用自适应模糊控制在干燥机中的应用.电气传动.2007(10)：54-55页
[24]陈杰,薛彬.模糊控制的研究现状与展望.自动化与仪器仪表.2006(6)：1-3页,11页
[25]石云,陆金桂等.基于神经网络和遗传算法的锅炉燃烧建模与优化.计算机应用与软件.2010,27(6)：226-229页
[26]Gao Fang, Cui Gang, Liu Hong-wei. Integration of genetic algorithm and cultural algorithms for constrained optimization. Lecture Notes in Computer Science, 2006,42(34):817-825P
［27]龚红菊,陈坤杰.稻谷容重与水分的关系.粮食与饲料工业.2003(9)：19页
[28]杨明,张宣.BP神经网络算法预测及其在飞参数据分析中的应用.计算机应用与软件.2008(3)：169-170页
[29]孔范增,季仲梅等.基于BP神经网络的视频H.223协议标志符动态提取算法.计算机应用研究.2010,27(6)：2112-2115页
[30]梅华威,张铭泉.基于BP神经网络的手机病毒检测方法.计算机应用与软件.2010,27(7)：283-284,300页
[31]李鹏英,冯雅丽等.基于GA-BP神经网络的深海集矿机避障系统.中南大学学报(自然科学版).2010,41(4)：1374-1378页
[32]陈政,杨天奇.基于RBF神经网络的股票市场预测.计算机应用与软件.2010,27(6)：108-110页
[33]陶凤玲,袁俊英等.基于人工神经网络的龙羊峡水库入库径流预报.青海大学学报(自然科学版).2010,28(4)：40-43,97页
[34]蔡娜娜,陈月辉等.基于神经网络的蛋白质三级结构预测.计算机工程.2010,36(9)：176-177页
[35]贺尚红,杨昀梓.基于神经网络的混凝土泵车发动机万有特性建模与工况优化.中南大学学报(自然科学版).2010,41(4)：1398-1404页
[36]黄海燕,顾幸生.基于文化算法的神经网络及其在建模中的应用.控制与决策.2008,23(4)：477-480页
[37]Trung Thanh Nguyen, Xin Yao. Hybridizing cultural algorithms and local search. Lecture Notes in Computer Science,2006,42(24):586-594P
[38]Reynolds R G, Peng B. Knowledge learning and social swarms in culture allgorithms. J of Mathematical Sociology,2005,29(2):115-132P
[39]Yuan X H, Yuan Y B. Application of culture algorithm to generation scheduling of hydrothermal systems. Energy Conversion and Management,2006 (47):2192-2201P
[40]Andre V Abs da Crus, Marco Aurelio C Pacheco, et al. Cultural operators for a quantum-inspired evolutionary algorithm applied to numerical optimization problems. Lecture Notes in Computer Science,2005,35(62):1-10P
[41]Sasakawa T, Hu J L, Hirasawa K. A brainlike learing system with supervised, unsupervised, and reinforcement learning. Electrical Engineering in Japan, 2008,162(1):165-172P
[42]王燕妮,樊养余.改进BP神经网络的自适应预测算法.计算机工程与应用.2010,46(17)：23-26页
[43]李孝忠,张有伟.改进自适应遗传算法在BP神经网络学习中的应用.天津科技大学学报.2010,25(4)：64-67页
[44]CHEN Sheng, HONG Xia, LUK B L, et al. Orthogonal-least-squares regression:a unified approach for data modeling. Neurocomputing,2009,72(10-12):2670-2681P
[45]董国君,哈力木拉提·买买提.基于随机退火的神经网络学习算法及其应用.计算机工程与应用.2010,46(19)：39-42,47页
[46]钱晓东,肖强等.基于改进的RBF神经网络的人民币汇率预测研究.计算机工程与应用.2010,46(10)：229-231页
[47]吴良海.基于粒子群优化RBF神经网络原油含水率预测.计算机仿 真.2010(5)：261-263,300页
[48]A Shahsavand, A Ahmadpour. Application of optimal RBF neural networks for optimization and characterization of porous materials. Computers & Chemical Engineering,2005,29(10):2134-2143P
[49]Gh. A Montazer, Reza Sabzevari, Fatemeh Ghorbani. Three phase strategy for the OSD learning method in RBF neural networks. Neurocomputing,2009,72(7-9):1797-1802P
[50]李剑,江成顺等.基于优化RBF神经网络的集成算法及其在调制识别中的应用.信息工程大学学报.2010,11(4)：448-451页
[51]崔立堃,王伟等Hopfield神经网络在有限元求解中的应用.计算机工程与应用.2010,46(16)：46-47,59页
[52]刘琦,张引等.基于离散Hopfield网络求解极大独立集茎区选择算法以及在RNA级结构预测中的应用.计算机学报.2008,31(1)：51-58页
[53]Islam M M, Sattar M A, et al. A new adaptive merging and growing algorithm for designing artificial neural networks. IEEE Trans on Systems,Man,and Cybernetics-Part B:Cybernetics,2009,39(3):705-718P
[54]Kohonen T. Self-organizing neural projections. Neural Networks,2006,19(6):723-755P
[55]武俊丽,李建辉等.基于BP神经网络对时滞系统的模型参数辨识仿真.佳木斯大学学报.2010,28(4)：496-498,501页
[56]李萍,曾令可等.基于MATALAB的BP神经网络预测系统的设计.计算机应用与软件.2008,25(4)：149-150,184页
[57]高鹏毅,陈传波等.一种新的基于Agent的神经网络隐层节点数的优化算法.计算机工程与科学.2010,32(5)：30-33页
[58]Teoh E J, Tan K C, Xiang C. Estimating the number of hidden neurons in a feedforward network using the singular value decomposition. IEEE Trans on Neural Network,2006,17(6):1623-1629P
[59]Goh C-K, Teoh E-J, Tan K C. Hybrid multiobjective evolutionary designing for artificial neural networks. IEEE Trans on Neural Network,2008,19(9):1531-1547P
[60]邓万宇,郑庆华.神经网络极速学习方法研究.计算机学报.2010,33(2)：279-287页
[61]Tsai J-T, Chou J-H, Liu T-K. Tuning the structure and parameters of a neural network by using hybrid taguchi-genetic algorithm. IEEE Trans on Neural Networks, 2006,17(1):69-80P
[62]Ludermir T B, Yamazaki A, Zanchettin C. An optimization methodology for neural network weights and architectures. IEEE Trans on Neural Networks, 2006,17(6):1452-1457P
[63]王建军,徐宗本.多元多项式函数的三层前向神经网络逼近方法.计算机学报.2009(12)：2482-2488页
[64]吴跃波,杨景曙.基于函数逼近的多层前馈神经网络灵敏度分析.计算机工程与应用.2010,46(5)：36-39页
[2]国家粮食局.实施农村粮食产后减损安全保障工程指导意见.国粮办展(2009)16号,2009-1-19
[3]顾冰,赵学工等.稻谷干燥智能控制系统的研究.电气控制.2007(3)：31-35页
[4]Li Changyou, Ban Hua. On-line moisture metering technology and device for paddy drying process. The Proceedings of the 5th Asia-Paciifc Drying Conference. World Scientiifc Publishing Co.Pte.Ltd,2007:490-495P
[5]李长友.谷物含水率在线检测装置[P].中国专利：03222718.3,2003-12-24
[6]李长友,班华.一种检测谷物含水率的方法及其装置[P].中国专利：200610123461.3,2007-05-16
[7]曹崇文,汪喜波.谷物干燥的自动控制.现代化农业.2002(2)：40-44页
[8]郭光辉.仿人智能控制算法的现状及趋势.安徽职业技术学院学报.2007(9).13-15页
[9]李国昉,齐玉斌等.谷物干燥仿人智能控制系统.农业机械学报.2007,38(1)：83-86页
[10]李国昉,李栋等.连续流动式粮食干燥机的仿人智能控制器.农业工程学报.2006(7)：105-107页
[11]齐玉斌,李国防等.基于虚拟仪器的谷物干燥智能控制器.粮食流通及仓储.2008(8).12-13页,16页
[12]周尚波,柳玉炯.基于EKF的模糊神经网络快速自组织学习算法.计算机应用.2010,27(7)：2462-2464,2471页
[13]HUANG Huan, WU Cong-xin. Approximation capabilities of multilayer fuzzy neural networks on the set of fuzzy-valued functions. Information Sciences, 2009,179(16):2762-2773P
[14]DENG Xing-sheng,WANG xin-zhou.Incremental learning of dynamic fuzzy neural networks for accurate system modeling.Fuzzy Sets and Systems,2009,160(7):972-987P
[15]RIGATOS G, ZHANG Q. Fuzzy model validation using the local statistical approach. Fuzzy Sets and Systems,2009,160(7):882-904P
[16]周岩,王盛等.基于改进粒子群算法的模糊神经网络及其在短时天气预报中的应用.计算机应用与软件.2010,27(5)：234-237页
[17]陈伯芳,尹平林等.基于模糊神经网络的温度控制系统研究.计算机与数字工 程.2010,38(7)：54-57页
[18]Fabry-asztalos L, Andonie R, et al. A genetic algorithm optimized fuzzy neural network analysis of the affinity of inhibitors for HIV-1 protease. Bioorganic & Medicinal Chemistry,2008,16(6):2903-2911P
[19]李业德,李业刚.顺流式谷物烘干机的模糊控制系统.农业工程学报.2002,18(1)：162-164页
[20]李长友,曹艳明.谷物循环干燥机控制系统硬件设计,农业机械学报.2002,33(6)：86-88页
[21]曹艳明,李长友.谷物循环干燥机控制系统软件设计.农业机械学报.2003,34(1)：83-85页
[22]李长友,班华.基于深层干燥解析理论的粮食干燥自适应控制系统设计.农业工程学报.2008(4)：142-146页
[23]王树东,唐伟强.应用自适应模糊控制在干燥机中的应用.电气传动.2007(10)：54-55页
[24]陈杰,薛彬.模糊控制的研究现状与展望.自动化与仪器仪表.2006(6)：1-3页,11页
[25]石云,陆金桂等.基于神经网络和遗传算法的锅炉燃烧建模与优化.计算机应用与软件.2010,27(6)：226-229页
[26]Gao Fang, Cui Gang, Liu Hong-wei. Integration of genetic algorithm and cultural algorithms for constrained optimization. Lecture Notes in Computer Science, 2006,42(34):817-825P
［27]龚红菊,陈坤杰.稻谷容重与水分的关系.粮食与饲料工业.2003(9)：19页
[28]杨明,张宣.BP神经网络算法预测及其在飞参数据分析中的应用.计算机应用与软件.2008(3)：169-170页
[29]孔范增,季仲梅等.基于BP神经网络的视频H.223协议标志符动态提取算法.计算机应用研究.2010,27(6)：2112-2115页
[30]梅华威,张铭泉.基于BP神经网络的手机病毒检测方法.计算机应用与软件.2010,27(7)：283-284,300页
[31]李鹏英,冯雅丽等.基于GA-BP神经网络的深海集矿机避障系统.中南大学学报(自然科学版).2010,41(4)：1374-1378页
[32]陈政,杨天奇.基于RBF神经网络的股票市场预测.计算机应用与软件.2010,27(6)：108-110页
[33]陶凤玲,袁俊英等.基于人工神经网络的龙羊峡水库入库径流预报.青海大学学报(自然科学版).2010,28(4)：40-43,97页
[34]蔡娜娜,陈月辉等.基于神经网络的蛋白质三级结构预测.计算机工程.2010,36(9)：176-177页
[35]贺尚红,杨昀梓.基于神经网络的混凝土泵车发动机万有特性建模与工况优化.中南大学学报(自然科学版).2010,41(4)：1398-1404页
[36]黄海燕,顾幸生.基于文化算法的神经网络及其在建模中的应用.控制与决策.2008,23(4)：477-480页
[37]Trung Thanh Nguyen, Xin Yao. Hybridizing cultural algorithms and local search. Lecture Notes in Computer Science,2006,42(24):586-594P
[38]Reynolds R G, Peng B. Knowledge learning and social swarms in culture allgorithms. J of Mathematical Sociology,2005,29(2):115-132P
[39]Yuan X H, Yuan Y B. Application of culture algorithm to generation scheduling of hydrothermal systems. Energy Conversion and Management,2006 (47):2192-2201P
[40]Andre V Abs da Crus, Marco Aurelio C Pacheco, et al. Cultural operators for a quantum-inspired evolutionary algorithm applied to numerical optimization problems. Lecture Notes in Computer Science,2005,35(62):1-10P
[41]Sasakawa T, Hu J L, Hirasawa K. A brainlike learing system with supervised, unsupervised, and reinforcement learning. Electrical Engineering in Japan, 2008,162(1):165-172P
[42]王燕妮,樊养余.改进BP神经网络的自适应预测算法.计算机工程与应用.2010,46(17)：23-26页
[43]李孝忠,张有伟.改进自适应遗传算法在BP神经网络学习中的应用.天津科技大学学报.2010,25(4)：64-67页
[44]CHEN Sheng, HONG Xia, LUK B L, et al. Orthogonal-least-squares regression:a unified approach for data modeling. Neurocomputing,2009,72(10-12):2670-2681P
[45]董国君,哈力木拉提·买买提.基于随机退火的神经网络学习算法及其应用.计算机工程与应用.2010,46(19)：39-42,47页
[46]钱晓东,肖强等.基于改进的RBF神经网络的人民币汇率预测研究.计算机工程与应用.2010,46(10)：229-231页
[47]吴良海.基于粒子群优化RBF神经网络原油含水率预测.计算机仿 真.2010(5)：261-263,300页
[48]A Shahsavand, A Ahmadpour. Application of optimal RBF neural networks for optimization and characterization of porous materials. Computers & Chemical Engineering,2005,29(10):2134-2143P
[49]Gh. A Montazer, Reza Sabzevari, Fatemeh Ghorbani. Three phase strategy for the OSD learning method in RBF neural networks. Neurocomputing,2009,72(7-9):1797-1802P
[50]李剑,江成顺等.基于优化RBF神经网络的集成算法及其在调制识别中的应用.信息工程大学学报.2010,11(4)：448-451页
[51]崔立堃,王伟等Hopfield神经网络在有限元求解中的应用.计算机工程与应用.2010,46(16)：46-47,59页
[52]刘琦,张引等.基于离散Hopfield网络求解极大独立集茎区选择算法以及在RNA级结构预测中的应用.计算机学报.2008,31(1)：51-58页
[53]Islam M M, Sattar M A, et al. A new adaptive merging and growing algorithm for designing artificial neural networks. IEEE Trans on Systems,Man,and Cybernetics-Part B:Cybernetics,2009,39(3):705-718P
[54]Kohonen T. Self-organizing neural projections. Neural Networks,2006,19(6):723-755P
[55]武俊丽,李建辉等.基于BP神经网络对时滞系统的模型参数辨识仿真.佳木斯大学学报.2010,28(4)：496-498,501页
[56]李萍,曾令可等.基于MATALAB的BP神经网络预测系统的设计.计算机应用与软件.2008,25(4)：149-150,184页
[57]高鹏毅,陈传波等.一种新的基于Agent的神经网络隐层节点数的优化算法.计算机工程与科学.2010,32(5)：30-33页
[58]Teoh E J, Tan K C, Xiang C. Estimating the number of hidden neurons in a feedforward network using the singular value decomposition. IEEE Trans on Neural Network,2006,17(6):1623-1629P
[59]Goh C-K, Teoh E-J, Tan K C. Hybrid multiobjective evolutionary designing for artificial neural networks. IEEE Trans on Neural Network,2008,19(9):1531-1547P
[60]邓万宇,郑庆华.神经网络极速学习方法研究.计算机学报.2010,33(2)：279-287页
[61]Tsai J-T, Chou J-H, Liu T-K. Tuning the structure and parameters of a neural network by using hybrid taguchi-genetic algorithm. IEEE Trans on Neural Networks, 2006,17(1):69-80P
[62]Ludermir T B, Yamazaki A, Zanchettin C. An optimization methodology for neural network weights and architectures. IEEE Trans on Neural Networks, 2006,17(6):1452-1457P
[63]王建军,徐宗本.多元多项式函数的三层前向神经网络逼近方法.计算机学报.2009(12)：2482-2488页
[64]吴跃波,杨景曙.基于函数逼近的多层前馈神经网络灵敏度分析.计算机工程与应用.2010,46(5)：36-39页