欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对基于T—S模型的模糊神经网络的局部逼近缺陷，提出了一种基于T—S模型的扩展型模糊神经网络，从训练样本特性和网络结构两个方面来提高网络模型的泛化能力．利用先验知识和模糊推理的方法对样本集进行分析和分类处理，使样本集更加规范；并采用模糊规则推理动态调整正则项系数的方法来减小网络结构，仿真结果表明，所提出的网络具有更快的收敛速度和良好的泛化能力．
英文摘要：
In order to overcome the drawback local approximation of the fuzzy neural network based on T-S model, an extended fuzzy neural network based on T-S model （EFNN-TS） is proposed in this paper. Two aspects of the characteristics of training samples and network structure are considered to enhance the generalization ability of the network. For normalizing the pattern set, the training patterns are classified and processed by using prior information of the patterns and fuzzy inference approach. Regularization is added whose coefficient can be adjusted dynamically by fuzzy rea- soning to simplify the structure of the network. The simulation results indicates that the proposed method has more rapid convergence and better generalization ability.
同期刊论文项目
同项目期刊论文
The hidden neurons selection of the wavelet networks using support vector machines and ridge regress
期刊信息
位置：