针对一类能够有效引入过程先验知识的复合型模糊神经网络 ,研究了其动态结构 .通过对复合型模糊神经网络的函数网络的第二层引入动态递归环节 ,使其具有动态映射能力 ,实现了对动态系统的良好响应 .本文采用了动态非线性模型对其进行仿真研究 ,结果表明 ,对于处理动态非线性系统 ,此动态复合模糊神经网络较之静态网络在收敛速度、预测精度和网络规模等方面都有较大的改善 Aiming at a kind of compound fuzzy neural network which can effectively introduce the prior knowledge of the process, the dynamic structure of the complex fuzzy neural network is studied. By introducing dynamic recursion into the second layer of the functional network of the complex fuzzy neural network, The dynamic response of the dynamic system is studied.The dynamic nonlinear model is used to simulate the dynamic nonlinear system.The results show that compared with the static network, the dynamic complex fuzzy neural network has better performance in convergence speed, prediction accuracy and network size And so on have greatly improved