非线性系统建模的复合型模糊神经网络研究

针对非线性系统建模问题,提出了一类由函数逼近和规则推理网络构成的复合型模糊神经网络,其规则网络基于过程先验知识用于对操作区间的划分,而函数网络采用改进型模糊神经网络结构完成非线性函数逼近。采用一类非线性函数模型进行了仿真研究,结果表明,复合型模糊神经网络较之普通模糊神经网络在建模收敛速度和预测精度等方面都有较大的改善。

Compound fuzzy neural network proposed for modeling of nonlinear systems

A type of compound fuzzy neural network that has a rule network and a function network is proposed in this paper for building nonlinear system models. Based on process knowledge, the process operation is partitioned into a number of local operating regions by the rule network, and the local behavior of the process is approximated by the function network with improved fuzzy neural network structure. Simulation has been made with a nonlinear model , and the results of which show that the compound fuzzy neural network is better than the ordinary fuzzy neural network in convergence rate and predication precision.