[1]乔俊飞,安茹,韩红桂.基于相对贡献指标的自组织RBF神经网络的设计[J].智能系统学报,2018,13(02):159-167.[doi:10.11992/tis.201608009]
 QIAO Junfei,AN Ru,HAN Honggui.Design of self-organizing RBF neural network based on relative contribution index[J].CAAI Transactions on Intelligent Systems,2018,13(02):159-167.[doi:10.11992/tis.201608009]
点击复制

基于相对贡献指标的自组织RBF神经网络的设计(/HTML)
分享到:

《智能系统学报》[ISSN:1673-4785/CN:23-1538/TP]

卷:
第13卷
期数:
2018年02期
页码:
159-167
栏目:
出版日期:
2018-04-15

文章信息/Info

Title:
Design of self-organizing RBF neural network based on relative contribution index
作者:
乔俊飞12 安茹12 韩红桂12
1. 北京工业大学 电子信息与控制工程学院, 北京 100124;
2. 计算智能与智能系统北京市重点实验室, 北京 100124
Author(s):
QIAO Junfei12 AN Ru12 HAN Honggui12
1. College of Electronic Information and Control Engineering, Beijing University of Technology, Beijing 100124, China;
2. Beijing Key Laboratory of Computation Intelligence and Intelligence System, Beijing 100124, China
关键词:
RBF神经网络相对贡献指标改进的LM算法结构设计出水氨氮收敛速度预测精度
Keywords:
RBF neural networkrelative contribution indeximproved LM algorithmstructure designammonia and nitrogen effluent parametersconvergence speedprediction accuracy
分类号:
TP183
DOI:
10.11992/tis.201608009
摘要:
针对RBF(radial basis function)神经网络的结构和参数设计问题,本文提出了一种基于相对贡献指标的自组织RBF神经网络的设计方法。首先,提出一种基于相对贡献指标(relative contribution,RC)的网络结构设计方法,利用隐含层输出对网络输出的相对贡献来判断是否增加或删减RBF网络相应的隐含层节点,并且对神经网络结构调整过程的收敛性进行证明。其次,采用改进的LM(Levenberg-Marquardt algorithm)算法对调整后的网络参数进行更新,使网络具有较少的训练时间和较快的收敛速度。最后,对提出的设计方法进行非线性函数仿真和污水处理出水参数氨氮建模,仿真结果表明,RBF神经网络能够根据研究对象自适应地动态调整RBF结构和参数,具有较好的逼近能力和更高的预测精度。
Abstract:
A design method for a self-organizing RBF Neural Network based on the Relative Contribution index is proposed with the aim of performing the structural design and parameter optimization of the Radial Basis Function (RBF) neural network. First, a self-organizing RBF network design method based on the Relative Contribution (RC) index is proposed. The relative contribution of the output of the hidden layer to the network output was used in order to assess whether a node of the hidden layer corresponding to the RBF network was inserted or pruned. Additionally, the convergence of the adjustment process of the neural structure was proven. Secondly, the adjusted network parameters were updated by the improved Levenberg-Marquardt (LM) algorithm in order to reduce the training time and increase the convergence speed of the network. Finally, the proposed algorithm was used in the simulation of the nonlinear function, and the modeling of the ammonia and nitrogen sewage effluent parameters. The simulation results revealed that the structure and parameters of the RBF neural network could be adjusted adaptively and dynamically according to the object under investigation, and that they had excellent approximation ability and higher prediction accuracy.

参考文献/References:

[1] QIAO J F, ZHANG Z Z, BO Y C. An online self-adaptive modular neural network for time-varying systems[J]. Neurocomputing, 2014, 125: 7-16.
[2] 韩红桂, 李淼, 乔俊飞. 基于模型输出敏感度分析的动态神经网络结构设计[J]. 电子学报, 2010, 38(3): 731-736.
HAN Honggui, LI Miao, Qiao Junfei. Design of dynamic neural network based on the sensitivity analysis of model output[J]. Acta electronica sinica, 2010, 38(3): 731-736.
[3] PLATT J. A resource-allocating network for function interpolation[J]. Neural computation, 1991, 3(2): 213-225.
[4] LU Y G, SUNDARARAJAN N, SARATCHANDRAN P. A sequential learning scheme for function approximation using minimal radial basis function neural networks[J]. Neural computation, 1997, 9(2): 461-478.
[5] PANCHAPAKESAN C, PALANISWAMI M, RALPH D, et al. Effects of moving the center’s in an RBF network[J]. IEEE transactions on neural networks, 2002, 13(6): 1299-1307.
[6] HUANG G B, SARATCHANDRAN P, SUNDARARAJAN N. A generalized growing and pruning RBF (GGAP-RBF) neural network for function approximation[J]. IEEE transactions on neural networks, 2005, 16(1): 57-67.
[7] GONZALEZ J, ROJAS I, ORTEGA J, et al. Multi-objective evolutionary optimization of the size, shape, and position parameters of radial basis function networks for function approximation[J]. IEEE transactions on neural networks, 2003, 14(6): 1478-1495.
[8] FENG H M. Self-generation RBFNs using evolutional PSO learning[J]. Neurocomputing, 2006, 70(1): 241-251.
[9] HAO C, YU G, XIA H. Online modeling with tunable RBF network[J]. IEEE transactions on cybernetics, 2013, 43(3): 935-947.
[10] LIAN J M, LEE Y G, SCOTT D, SUDHOFF, et al. Self-organizing radial basis function network for real-time approximation of continuous-time dynamical systems[J]. IEEE transactions on neural networks, 2008, 19(3): 460-474.
[11] YU H, REINER P D, XIE T T, et al. An incremental design of radial basis function networks[J]. IEEe transactions on neural networks and learning systems, 2014, 25(10): 1793-1803.
[12] CONSTANTINOPOULO C, LⅡKAS A. An incremental training method for the probabilistic RBF network[J]. IEEE transactions on neural networks, 2006, 17(4): 966-974.
[13] CHEN S, HANZO L, TAN S. Symmetric complex-valued RBF receiver for multiple-antenna-aided wireless systems[J]. IEEe transactions on neural networks, 2008, 19(9): 1659-1665.
[14] 王晓丽, 黄蕾, 杨鹏, 等. 动态RBF神经网络在浮选过程模型失配中的应用[J]. 化工学报, 2016, 3(67): 897-902.
WANG Xiaoli, HUANG Lei, YANG Peng, et al. Dynamic RBF neural networks for model mismatch problem and its application in flotation process[J]. CIESC journal, 2016: 3(67): 897-902.
[15] WILAMOWSKI B M, YU H. Improved computation for Levenberg-Marquardt training[J]. IEEE transactions on neural networks, 2010, 21(6): 930-937.
[16] MA C, JIANG L. Some research on Levenberg-Marquardt method for the nonlinear equations[J]. Applied mathematics and computation, 2007, 184(2): 1032-1040.
[17] XIE T T, YU H, HEWLETT J, et al. Fast and efficient second-order method for training radial basis function networks[J]. IEEE transactions on neural networks and learning systems, 2012, 23(4): 609-619.
[18] QIAO J F, HAN H G. Identification and modeling of nonlinear dynamical systems using a novel self-organizing RBF-based approach[J]. Automatica, 2012, 48(8): 1729-1734.
[19] CHEN C, WANG F Y. A self-organizing neuro-fuzzy network based on first order effect sensitivity analysis[J]. Neurocomputing, 2013, 118: 21-32.
[20] HAN H G, CHEN Q L, QIAO J F. An efficient self-organizing RBF neural network for water quality prediction[J]. Neural networks the official journal of the international neural network society, 2011, 24(7): 717-25.
[21] HUANG G B, SARATCHANDRAN P, SUNDARARAJAN N. An efficient sequential learning algorithm for growing and pruning RBF (GAP-RBF) networks[J]. IEEE transactions on systems, man, and cybernetics, Part B: (Cybernetics), 2004, 34(6): 2284-2292.
[22] CHO K B, WANG B Y. Radial basis function based adaptive fuzzy systems and their applications to system identification and prediction[J]. Fuzzy sets and systems, 1996, 83(3): 325-339.
[23] EBADZADEH M M, SALIMI-BADR A. CFNN: correlated fuzzy neural network[J]. Neurocomputing, 2015, 148: 430-444.
[24] An R, Li W J, Han H G, et al. An improved Levenberg-Marquardt algorithm with adaptive learning rate for RBF neural network[C]//IEEE Control Conference, [s.l.], 2016: 3630-3635.

相似文献/References:

[1]章 钱,李士勇.一种新型自适应RBF神经网络滑模制导律[J].智能系统学报,2009,4(04):339.
 ZHANG Qian,LI Shi-yong.A new adaptive RBFNN sliding mode guidance law for intercepting maneuvering targets[J].CAAI Transactions on Intelligent Systems,2009,4(02):339.
[2]张秀玲,陈丽杰,逄宗朋,等.RBF神经网络的板形预测控制[J].智能系统学报,2010,5(01):70.
 ZHANG Xiu-ling,CHEN Li-jie,PANG Zong-peng,et al.A predictive system for process control of flatness in rolling mills using a radial basis function network[J].CAAI Transactions on Intelligent Systems,2010,5(02):70.
[3]陈亮,何为,韩力群.RBF神经网络的行车路径代价函数建模[J].智能系统学报,2011,6(05):424.
 CHEN Liang,HE Wei,HAN Liqun.Radial basis function neural network modeling of the traffic path cost function[J].CAAI Transactions on Intelligent Systems,2011,6(02):424.
[4]陈亮,何为,韩力群.城市交通最优路径算法[J].智能系统学报,2012,7(02):167.
 CHEN Liang,HE Wei,HAN Liqun.Study on an urban transportation optimal path algorithm[J].CAAI Transactions on Intelligent Systems,2012,7(02):167.
[5]蒙西,乔俊飞,李文静.基于快速密度聚类的RBF神经网络设计[J].智能系统学报,2018,13(03):331.[doi:10.11992/tis.201702014]
 MENG Xi,QIAO Junfei,LI Wenjing.Construction of RBF neural networks via fast density clustering[J].CAAI Transactions on Intelligent Systems,2018,13(02):331.[doi:10.11992/tis.201702014]

备注/Memo

备注/Memo:
收稿日期:2016-08-29。
基金项目:国家自然科学基金重点项目(61533002,61225016);北京市教育委员会科研计划项目(km201410005002);高等学校博士学科点基金项目(20131103110016).
作者简介:乔俊飞,男,1968年生,教授,博士生导师,国家杰出青年基金获得者,教育部新世纪优秀人才,北京市精品课程负责人。主要研究方向为智能信息处理、智能优化控制。获教育部科技进步奖一等奖和北京市科学技术奖三等奖各1项,获得授权国家发明专利12项。发表学术论文近70篇,被SCI检索15篇;安茹,女,1990年生,硕士研究生,主要研究方向为智能控制理论及应用、非线性系统建模;韩红桂,男,1983年生,教授,博士生导师,先后入选香江学者计划、北京市科技新星计划等;主要研究方向为污水处理过程建模、优化与控制。获得授权国家发明专利13项,授权实用新型专利3项,授权软件著作权11项;研究成果获教育部科学技术进步奖一等奖等。近5年来发表学术论文35篇,撰写专著1部。
通讯作者:乔俊飞.E-mail:anru@emails.bjut.edu.cn.
更新日期/Last Update: 1900-01-01