智能系统学报2018，Vol.13Issue(2)：159-167,9.DOI:10.11992/tis.201608009
基于相对贡献指标的自组织RBF神经网络的设计
Design of self-organizing RBF neural network based on relative contribution index
摘要
Abstract
A design method for a self-organizing RBF Neural Network based on the Relative Contribution index is pro-posed 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 conver-gence of the adjustment process of the neural structure was proven. Secondly, the adjusted network parameters were up-dated by the improved Levenberg-Marquardt (LM) algorithm in order to reduce the training time and increase the con-vergence 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 struc-ture and parameters of the RBF neural network could be adjusted adaptively and dynamically according to the object un-der investigation, and that they had excellent approximation ability and higher prediction accuracy.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Key words
RBF neural network/relative contribution index/improved LM algorithm/structure design/ammonia and nitrogen effluent parameters/convergence speed/prediction accuracy分类
信息技术与安全科学引用本文复制引用
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(2):159-167,9.基金项目
国家自然科学基金重点项目(61533002,61225016) （61533002,61225016）
北京市教育委员会科研计划项目(km201410005002) （km201410005002）
高等学校博士学科点基金项目(20131103110016). （20131103110016）
