大型非均质水驱油物理模拟系统研究
|
摘要
随着经济的飞速发展,我国经济对石油的依赖性越来越强,不断地研究高效的石油驱采技术至关重要。对于石油工作者来说,对石油的开采进行模拟,进而研究出石油开采工程中运移的一些规律,运用这些规律来指导实际的石油开采是一种行之有效的途径。
本课题基于电导率和饱和度的关系特点,针对实验室水驱油模拟的需求,选用SAMSUNG公司基于ARM7核的S3C44BOX芯片作为CPU,设计了一个数据采集系统,用来测量实验室大型模拟水驱油过程中的砂岩内部的含水饱和度。本文详细给出了硬件的设计和软件设计部分,文中都给出了较为详细的设计方法。
探针采用四线探针法的原理来设计,方法较为科学;为了能够使测量的电阻值更加准确,本文采用8个档位可调的恒流源给四线探针供电,给出了自动选择档位的软件流程和硬件的设计方法和原理;文中还给出了512路测量选通电路的设计方法,设计上的创新使得多路的测量准确可靠。
本文还给出了以ARM为核心器件的外围存储器和扩展接口的硬件设计方法,以及ARM的基于ADS编译环境下的软件设计的方法和流程。包括编译器的设置,boot loader的移植,程序的调试和Flash的下载。另外本文编写了PC机接受数据采集系统发送数据的VB程序。
最后本文基于上述数据采集系统所收集到的数据,利用径向基网络进行三维物理模型建模,分析了建模的结果并与传统的BP网络建模的预测结果进行了对比,通过对比结果表明,本模型采用RBF建模具有较强的实用性。最后采用对三维模型进行剖面分析的方法,绘制出二维的含水饱和度分布图,绘图较为直观地反映出水驱油过程中的油水分布情况。
As the economic developed rapidly, China's economic is becoming depended on oil more and more seriously, the efficiency of the oil-drive technology is becoming more and more essential. To the oil workers, simulate the water flooding process in the laboratory and find some laws about oil movement in the water flooding, then use these laws to guide the actual oil exploitation, which is a effective way to prompt the oil exploitation efficiency
As the relationship between the resistance and the water saturation, just it could work out the resistance rate by measuring the resistance value. The subject is based on the SAMSUNG company ARM7 S3C44B0x. it Designed a data collection system, used to measure the water saturation of water flooding simulation on large-scale sandstone in the laboratory. This paper described the system, including the hardware and software design part in detail.
Use the four-probe method to measure the resistance, which is more scientific. In order to be able to measure more accurately about the value of resistance, the paper used eight stalls adjustable constant current source, given the automatic selection process stalls in the software and hardware design methods and the principles; The paper also introduced the 512 measurement strobe circuit design, design innovation on the multi-channels makes the measurement accurate and reliable.
It also presented the design method of the external memory and expansion circuit interface, taking ARM as the core. And described the software design methods and processes in the ADS compiler environment, which including the compiler settings, boot loader transplantation, the process of debugging and Flash download. Also in this paper, a PC data acquisition system to accept data sent by the ARM is introduced which is program in the VB language.
Finally Based on the above data collection system to collect the data and model the module with the radial-based network, the modeling method is proposed. Analysis of the results of the modeling, and then made a Contrast of the forecast results between the traditional BP network model and RBF. By comparing the results, we can arrive at the conclusion that used RBF to model is Feasible. Finally, analyze the profile of three-dimensional model. Draw the two-dimensional map about the water saturation distribution, which reflected the water distribution situation of the certain profile in the water flooding process in certain time.
本课题基于电导率和饱和度的关系特点,针对实验室水驱油模拟的需求,选用SAMSUNG公司基于ARM7核的S3C44BOX芯片作为CPU,设计了一个数据采集系统,用来测量实验室大型模拟水驱油过程中的砂岩内部的含水饱和度。本文详细给出了硬件的设计和软件设计部分,文中都给出了较为详细的设计方法。
探针采用四线探针法的原理来设计,方法较为科学;为了能够使测量的电阻值更加准确,本文采用8个档位可调的恒流源给四线探针供电,给出了自动选择档位的软件流程和硬件的设计方法和原理;文中还给出了512路测量选通电路的设计方法,设计上的创新使得多路的测量准确可靠。
本文还给出了以ARM为核心器件的外围存储器和扩展接口的硬件设计方法,以及ARM的基于ADS编译环境下的软件设计的方法和流程。包括编译器的设置,boot loader的移植,程序的调试和Flash的下载。另外本文编写了PC机接受数据采集系统发送数据的VB程序。
最后本文基于上述数据采集系统所收集到的数据,利用径向基网络进行三维物理模型建模,分析了建模的结果并与传统的BP网络建模的预测结果进行了对比,通过对比结果表明,本模型采用RBF建模具有较强的实用性。最后采用对三维模型进行剖面分析的方法,绘制出二维的含水饱和度分布图,绘图较为直观地反映出水驱油过程中的油水分布情况。
As the economic developed rapidly, China's economic is becoming depended on oil more and more seriously, the efficiency of the oil-drive technology is becoming more and more essential. To the oil workers, simulate the water flooding process in the laboratory and find some laws about oil movement in the water flooding, then use these laws to guide the actual oil exploitation, which is a effective way to prompt the oil exploitation efficiency
As the relationship between the resistance and the water saturation, just it could work out the resistance rate by measuring the resistance value. The subject is based on the SAMSUNG company ARM7 S3C44B0x. it Designed a data collection system, used to measure the water saturation of water flooding simulation on large-scale sandstone in the laboratory. This paper described the system, including the hardware and software design part in detail.
Use the four-probe method to measure the resistance, which is more scientific. In order to be able to measure more accurately about the value of resistance, the paper used eight stalls adjustable constant current source, given the automatic selection process stalls in the software and hardware design methods and the principles; The paper also introduced the 512 measurement strobe circuit design, design innovation on the multi-channels makes the measurement accurate and reliable.
It also presented the design method of the external memory and expansion circuit interface, taking ARM as the core. And described the software design methods and processes in the ADS compiler environment, which including the compiler settings, boot loader transplantation, the process of debugging and Flash download. Also in this paper, a PC data acquisition system to accept data sent by the ARM is introduced which is program in the VB language.
Finally Based on the above data collection system to collect the data and model the module with the radial-based network, the modeling method is proposed. Analysis of the results of the modeling, and then made a Contrast of the forecast results between the traditional BP network model and RBF. By comparing the results, we can arrive at the conclusion that used RBF to model is Feasible. Finally, analyze the profile of three-dimensional model. Draw the two-dimensional map about the water saturation distribution, which reflected the water distribution situation of the certain profile in the water flooding process in certain time.
引文
[1]杨春梅,李洪奇.不同驱替方式下岩石电阻率与饱和度的关系[J].吉林大学学报(地球科学版).2005,9.35(5).
[2]钟蕴紫,孙耀庭,张俊杰等.低孔隙度低渗透率岩心水驱油岩电实验研究[J].测井技术.2005-05-016.
[3]E.H.萨弗诺夫,王成辉.阿尔拉油田提高采收率的物理化学方法[J].吐哈油气.006-02-031.
[4]李圣勇,李圣涛,陈馥.聚合物驱提高采收率发展现状与趋势[J].化工时刊.2005-08-013.
[5]熊伟,王业飞,焦翠等.聚合物驱后的高效洗油技术在孤岛油田的运用[J].断块油气田.2005-05-018.
[6]骆瑛.复杂断块高含水油藏加密井网提高采收率研究[J].小型油气藏.2007-03-017.
[7]常玲.薄层稠油油藏水平井开采技术[J].胜利油田职工大学学报.2008-01-028.
[8]呼惠娜,胡淑娟.注蒸汽加入烃类添加剂提高稠油采收率的改进分析模型和实验研究[J].国外油田工程.2008-02-004
[9]冈秦麟.化学驱油论文集一“八五”三次采油成果汇编[M].北京:石油工业出版社11998.
[10]Liu Yong-jian,Fan Hong-fu.The effect of hydrogen donor additive on the viscosity of heavy oil during steam stimulation[J].Energy& Fuel.2001,15(6):1475-1479.48.
[11]李宜强,孙连荣.三元复合驱物理模拟相似原理[J].大庆石油学院学报.2003,6.2(27)
[12]李宜强,张素梅.泡沫复合驱物理模拟相似原理[J].大庆石油学院学报.2003,6.2(27)
[13]孔祥言,陈峰磊.水驱油物理模拟理论和相似性准则[J].石油勘探开发.1997,12.6(24)
[14]Simon R,GraueD.Generalized correlations for predicting solubility,swelling and behavior of CO2 crude oil systems[J].J PT,1965,17(1):102-106.
[15]Chung T H,Jones R A.Measurements and correlations of the physical properties of CO2/heavy crude oil mixtures[R].SPE 15080,1988:822-828.
[16]董建华,刘鹏,王薇.地应力剖面在水力压裂施工中的应用[J].大庆石油学院学报.2005,29(2):40-42.
[17]陈天愚.套损问题影响因素分析[J].哈尔滨工程大学学报.2004,25(1):104-107.
[18]Arichie G.The electrical resistivity log as an aid in determining some reservoir characteristics[J].Transactions of American Institute of Mining Engineers.1942,146:54-62.
[19]郭秀军,张志阔,贾永刚等.黄河口饱和粉土的电性特征及其工程地质应用[J].岩土力学.Vol.28 N0.3 2007年3月.
[20]王家禄,沈平平,田玉玲等.应用微型探针测量油藏物理模拟饱和度变化[J].测井技术.28(2).2004-4
[21]沈平平,王家禄,田玉玲等.三维油藏物理模拟的饱和度测量技术研究[J].石油勘探与开发.004-S1-017
[22]晏敏,彭楚武等.智能四探针电阻率测试仪的研制[J].湖南大学学报(自然科学板).2005,10.32(5).
[23]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.2005.
[24]Haykin S.Neural Networks:A Comprehensive Foundmion.第2版[M].北京:清华大学出版社,2001.
[25]王镭,周国兴,吴启迪.人工神经网络理论在控制领域中的应用综述[J].同济大学学报.2001,29(3):357-361
[26]魏海坤.神经网络结构设计的理论与方法[M].北京:国防工业出版,2005.
[27]Hagan MT,Demuth HB,Beale MH 著,戴葵等译.神经网络设计[M].北京:机械工业出版社,2002
[28]McCulloch WS,Pitts W.A logical calculus of the ideas immanent in nervous activity[J].Bulletin of Mathematical Biophysics,1943,5(1):115-133.
[29]Wiener N.Cybernetics:Or Control and Communication in the Animal and the Machine[J].New York:Wiley,1948.
[30]Hebb DO.The Organization of Behavior:A Neuropsychological Theory[J].New York:Wiley,1949.
[31]Rosenblatt F.The perception:a probabilistic model for information storage and organization in the brain[J].Psychological Review,1958,65:386-408
[32]Widrow B,Hoff ME.Adaptive switching circuits[A].IRE-WESCON Convention Record,1960,4:96-104
[33]Minsky ML,Papert SA.Perceptrons:An Introduction to Computational Geometry.Cambridge[M],MA:MIT Press,1969
[34]Hopfiled JJ.Neural networks and physical systems with emergent collective computational abilities[J].Proceedings of the National Academy of Science,USA,1982,79(2):2554-2558.
[35]Hopfield JJ.Neurons with graded response have collective computational properties like those of two-state neurons[J].USA:Proceedings of the National Academy of Science,1984,81(1):3088-3092
[36]Kohonen T.Self-organized formation of topologically correct feature maps[J].Biological Cybernetics,1982,43:59-69
[37]Ackley DH,Hinton GE,Sejnowski YJ.A learning algorithm for Boltzmann machines[J].Cognitive Science,1985,9(1):147-169
[38]Rumelhart DE,Hinton GE,Williams RJ.Learning representations by back-propagating errors[J].Nature,1986,323:533-536
[39]Broomhead DS,Lowe D.Multi-variable functional interpolation and adaptive Networks [J].Complex System,1988,2(2):321-355
[40]]Poggio T,Girosi F.Networks for approximation and learning[J].Proceedings of the IEEE,1990,78:1481-1497
[41]Vapnik VN.Principles of risk minimization for learning theory.Advances in Neu nformation Processing Systems[M].San Mateo:Morgan Kaufmann,1992,4:831-838
[42]Vapnik VN.The Nature of Statistical Learning Theory[J].NewYork:Wiley,1995
[43]Moody JE,Darken CJ.Fast learning in networks of locally-tuned processing units[J].Neural Computation,1989,1(2):281-294
[44]Bishop CM.Improving the generalization properties of radial basis function neural networks[J].Neural Computation,1991,3(4):579-581
[45]Chert S,Cowan CFN.,and Grant PM.Orthogonal least squares learning algorithm for radial basis function networks[J].IEEE Transactions on Neural Networks,1991,2(2):302-309
[46]Chert S,Chng ES,Alkadhimi K.Regularized orthogonal least squares algorithm for radial basis function networks[J].International Journal of Control,1996,64(5):829-837
[47]Mao KZ.RBF neural network center selection based on fisher ratio class separability measure[J].IEEE transactions on Neural Networks,2002,13(5):1211-1217
[48]Schokopf R,Sung KK,Burges CJC,etc.Comparing support vector machines with Gaussian kernels to radial basis function classifiers[J].IEEE Transactions on Signal Processing,2000,45(11):2758-2765
[49]Shi D,Yeung DS,Gao J.Sensitivity analysis applied to the construction of radial basis function networks[J].Neural Networks,2005,18(7):951-957
[50]Xu L,Krzyzak A,Yuille A.On radial basis function nets and kernel regression:statistical consistency,convergence rates,and receptive field size[J].Neural Networks,1994,7(4):609-628
[51]Guang-bin H,Saratchandran P,Sundararajan N.A generalized growing and pruning RBF (GGAP-RBF)neural network for function approximation[J].IEEE Transactions on Neural Network,2005,16(1):57-67
[52]Mao KZ,Guang-Bin H.Neuron selection for RBF neural network,classifier based on data structure preserving criterion[J].IEEE Transaction on Neural Network,2005,16(6):1531-1540
[53]Sheta AF,Jong KD.Time-series forecasting using GA-tuned radial basis functions[J].Information Sciences,2001,133:221-228
[54]Mak MW,Cho KW.Genetic evolution of radial basis functions and genetic algorithm[A].International Conference on Neural Networks,IEEE,1997,4:2187-2192
[55]Whitehead BA,Choate DC.Cooperative-competitive genetic evolution of radial basis function centers and widths of time series prediction[J].IEEE Transactions on Neural Networks,1995,7(4):869-880
[56]Whitehead BA.Genetic evolution of radial basis function coverage using orthogonal niches[J].IEEE Transactions on Neural Networks,1996,7(6):1525-1528
[57]Billings SA,Zheng GL.Radial basis function network configuration using a genetic algorithm[J].Neural Networks,1995,8(6):877-890
[58]魏海坤,徐嗣鑫,宋文忠.RBF网学习的进化优选算法.控制理论与应用,2000,17(4):604-608
[59]王凌,郑大钟.径向基函数神经网络结构的混合优化策略.清华大学学报(自然科学版),1999,39(7):50-53
[60]Chen S,Wu Y,Luk BL.Combined genetic algorithm optimization and regularized orthogonal least squares learning for radial basis function networks.IEEE Transactions on Neural Networks,1999,.10(5):1239-124
[61]Barreto AMS,Barbosa HJC,Ebecken NFF.GOLS—Genetic orthogonal least squares algorithm for training RBF networks.Neurocomputing,2006,69:2041-2064
[62]Sanchez VDA.Robustization of a learning method for RBF networks.Neurocomputing,1995,7(9):85-94
[63]Chien-Cheng L,Pau-Choo C,Jea-Rong T,etc.Robust radial basis function neural networks.IEEE Transactions on Systems,Man,and Cybernetics,1999 29(6):674-685
[64]刘妹琴,廖晓昕.RBF神经网络的一种鲁棒学习算法.华中理工大学学报,2000,28(2):8-10
[2]钟蕴紫,孙耀庭,张俊杰等.低孔隙度低渗透率岩心水驱油岩电实验研究[J].测井技术.2005-05-016.
[3]E.H.萨弗诺夫,王成辉.阿尔拉油田提高采收率的物理化学方法[J].吐哈油气.006-02-031.
[4]李圣勇,李圣涛,陈馥.聚合物驱提高采收率发展现状与趋势[J].化工时刊.2005-08-013.
[5]熊伟,王业飞,焦翠等.聚合物驱后的高效洗油技术在孤岛油田的运用[J].断块油气田.2005-05-018.
[6]骆瑛.复杂断块高含水油藏加密井网提高采收率研究[J].小型油气藏.2007-03-017.
[7]常玲.薄层稠油油藏水平井开采技术[J].胜利油田职工大学学报.2008-01-028.
[8]呼惠娜,胡淑娟.注蒸汽加入烃类添加剂提高稠油采收率的改进分析模型和实验研究[J].国外油田工程.2008-02-004
[9]冈秦麟.化学驱油论文集一“八五”三次采油成果汇编[M].北京:石油工业出版社11998.
[10]Liu Yong-jian,Fan Hong-fu.The effect of hydrogen donor additive on the viscosity of heavy oil during steam stimulation[J].Energy& Fuel.2001,15(6):1475-1479.48.
[11]李宜强,孙连荣.三元复合驱物理模拟相似原理[J].大庆石油学院学报.2003,6.2(27)
[12]李宜强,张素梅.泡沫复合驱物理模拟相似原理[J].大庆石油学院学报.2003,6.2(27)
[13]孔祥言,陈峰磊.水驱油物理模拟理论和相似性准则[J].石油勘探开发.1997,12.6(24)
[14]Simon R,GraueD.Generalized correlations for predicting solubility,swelling and behavior of CO2 crude oil systems[J].J PT,1965,17(1):102-106.
[15]Chung T H,Jones R A.Measurements and correlations of the physical properties of CO2/heavy crude oil mixtures[R].SPE 15080,1988:822-828.
[16]董建华,刘鹏,王薇.地应力剖面在水力压裂施工中的应用[J].大庆石油学院学报.2005,29(2):40-42.
[17]陈天愚.套损问题影响因素分析[J].哈尔滨工程大学学报.2004,25(1):104-107.
[18]Arichie G.The electrical resistivity log as an aid in determining some reservoir characteristics[J].Transactions of American Institute of Mining Engineers.1942,146:54-62.
[19]郭秀军,张志阔,贾永刚等.黄河口饱和粉土的电性特征及其工程地质应用[J].岩土力学.Vol.28 N0.3 2007年3月.
[20]王家禄,沈平平,田玉玲等.应用微型探针测量油藏物理模拟饱和度变化[J].测井技术.28(2).2004-4
[21]沈平平,王家禄,田玉玲等.三维油藏物理模拟的饱和度测量技术研究[J].石油勘探与开发.004-S1-017
[22]晏敏,彭楚武等.智能四探针电阻率测试仪的研制[J].湖南大学学报(自然科学板).2005,10.32(5).
[23]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.2005.
[24]Haykin S.Neural Networks:A Comprehensive Foundmion.第2版[M].北京:清华大学出版社,2001.
[25]王镭,周国兴,吴启迪.人工神经网络理论在控制领域中的应用综述[J].同济大学学报.2001,29(3):357-361
[26]魏海坤.神经网络结构设计的理论与方法[M].北京:国防工业出版,2005.
[27]Hagan MT,Demuth HB,Beale MH 著,戴葵等译.神经网络设计[M].北京:机械工业出版社,2002
[28]McCulloch WS,Pitts W.A logical calculus of the ideas immanent in nervous activity[J].Bulletin of Mathematical Biophysics,1943,5(1):115-133.
[29]Wiener N.Cybernetics:Or Control and Communication in the Animal and the Machine[J].New York:Wiley,1948.
[30]Hebb DO.The Organization of Behavior:A Neuropsychological Theory[J].New York:Wiley,1949.
[31]Rosenblatt F.The perception:a probabilistic model for information storage and organization in the brain[J].Psychological Review,1958,65:386-408
[32]Widrow B,Hoff ME.Adaptive switching circuits[A].IRE-WESCON Convention Record,1960,4:96-104
[33]Minsky ML,Papert SA.Perceptrons:An Introduction to Computational Geometry.Cambridge[M],MA:MIT Press,1969
[34]Hopfiled JJ.Neural networks and physical systems with emergent collective computational abilities[J].Proceedings of the National Academy of Science,USA,1982,79(2):2554-2558.
[35]Hopfield JJ.Neurons with graded response have collective computational properties like those of two-state neurons[J].USA:Proceedings of the National Academy of Science,1984,81(1):3088-3092
[36]Kohonen T.Self-organized formation of topologically correct feature maps[J].Biological Cybernetics,1982,43:59-69
[37]Ackley DH,Hinton GE,Sejnowski YJ.A learning algorithm for Boltzmann machines[J].Cognitive Science,1985,9(1):147-169
[38]Rumelhart DE,Hinton GE,Williams RJ.Learning representations by back-propagating errors[J].Nature,1986,323:533-536
[39]Broomhead DS,Lowe D.Multi-variable functional interpolation and adaptive Networks [J].Complex System,1988,2(2):321-355
[40]]Poggio T,Girosi F.Networks for approximation and learning[J].Proceedings of the IEEE,1990,78:1481-1497
[41]Vapnik VN.Principles of risk minimization for learning theory.Advances in Neu nformation Processing Systems[M].San Mateo:Morgan Kaufmann,1992,4:831-838
[42]Vapnik VN.The Nature of Statistical Learning Theory[J].NewYork:Wiley,1995
[43]Moody JE,Darken CJ.Fast learning in networks of locally-tuned processing units[J].Neural Computation,1989,1(2):281-294
[44]Bishop CM.Improving the generalization properties of radial basis function neural networks[J].Neural Computation,1991,3(4):579-581
[45]Chert S,Cowan CFN.,and Grant PM.Orthogonal least squares learning algorithm for radial basis function networks[J].IEEE Transactions on Neural Networks,1991,2(2):302-309
[46]Chert S,Chng ES,Alkadhimi K.Regularized orthogonal least squares algorithm for radial basis function networks[J].International Journal of Control,1996,64(5):829-837
[47]Mao KZ.RBF neural network center selection based on fisher ratio class separability measure[J].IEEE transactions on Neural Networks,2002,13(5):1211-1217
[48]Schokopf R,Sung KK,Burges CJC,etc.Comparing support vector machines with Gaussian kernels to radial basis function classifiers[J].IEEE Transactions on Signal Processing,2000,45(11):2758-2765
[49]Shi D,Yeung DS,Gao J.Sensitivity analysis applied to the construction of radial basis function networks[J].Neural Networks,2005,18(7):951-957
[50]Xu L,Krzyzak A,Yuille A.On radial basis function nets and kernel regression:statistical consistency,convergence rates,and receptive field size[J].Neural Networks,1994,7(4):609-628
[51]Guang-bin H,Saratchandran P,Sundararajan N.A generalized growing and pruning RBF (GGAP-RBF)neural network for function approximation[J].IEEE Transactions on Neural Network,2005,16(1):57-67
[52]Mao KZ,Guang-Bin H.Neuron selection for RBF neural network,classifier based on data structure preserving criterion[J].IEEE Transaction on Neural Network,2005,16(6):1531-1540
[53]Sheta AF,Jong KD.Time-series forecasting using GA-tuned radial basis functions[J].Information Sciences,2001,133:221-228
[54]Mak MW,Cho KW.Genetic evolution of radial basis functions and genetic algorithm[A].International Conference on Neural Networks,IEEE,1997,4:2187-2192
[55]Whitehead BA,Choate DC.Cooperative-competitive genetic evolution of radial basis function centers and widths of time series prediction[J].IEEE Transactions on Neural Networks,1995,7(4):869-880
[56]Whitehead BA.Genetic evolution of radial basis function coverage using orthogonal niches[J].IEEE Transactions on Neural Networks,1996,7(6):1525-1528
[57]Billings SA,Zheng GL.Radial basis function network configuration using a genetic algorithm[J].Neural Networks,1995,8(6):877-890
[58]魏海坤,徐嗣鑫,宋文忠.RBF网学习的进化优选算法.控制理论与应用,2000,17(4):604-608
[59]王凌,郑大钟.径向基函数神经网络结构的混合优化策略.清华大学学报(自然科学版),1999,39(7):50-53
[60]Chen S,Wu Y,Luk BL.Combined genetic algorithm optimization and regularized orthogonal least squares learning for radial basis function networks.IEEE Transactions on Neural Networks,1999,.10(5):1239-124
[61]Barreto AMS,Barbosa HJC,Ebecken NFF.GOLS—Genetic orthogonal least squares algorithm for training RBF networks.Neurocomputing,2006,69:2041-2064
[62]Sanchez VDA.Robustization of a learning method for RBF networks.Neurocomputing,1995,7(9):85-94
[63]Chien-Cheng L,Pau-Choo C,Jea-Rong T,etc.Robust radial basis function neural networks.IEEE Transactions on Systems,Man,and Cybernetics,1999 29(6):674-685
[64]刘妹琴,廖晓昕.RBF神经网络的一种鲁棒学习算法.华中理工大学学报,2000,28(2):8-10