摘要
由于模糊算法过于依赖专家知识,参数决定过程中人为主观因素过大且无法进行自学习;而神经网络算法收敛速度较慢,还可能陷入局部最小.本文针对上述这两种算法在实际应用中的缺陷,引入并介绍了一种高效的粒子群算法,在采用Sugeno模糊推理计算模型建立模糊神经网络的基础上,利用粒子群算法收敛快、算法简单和全局寻优的优势,实现模糊神经网络的优化辨识,并进行了仿真实验.
Fuzzy algorithm excessively depends on the experience from experts and subjective factors and has not self - learning ability; nerve network has the problem of slow convergence and may be got in part minimum. The paper introduces particle swarm optimizer, which is global optimization, quick convergence and simple program, to solve the above - mentioned problems and achieve the identification of fuzzy nerve network based on Sugeno fuzzy control rule with particle swarm optimization. The paper has carried on the simulation experiment.
出处
《南昌大学学报（工科版）》
CAS
2006年第3期253-255,共3页
Journal of Nanchang University(Engineering & Technology)
关键词
粒子群优化
模糊神经网络
系统辨识
particle swarm optimization
fuzzy neural network
identification
