摘要
提出了一种基于人工免疫原理的最优RBF模糊神经网络控制器设计方案.首先给出了控制器结构,其次将免疫进化算法用于控制器参数的优化,设计了一种满足二次型性能指标的最优RBF模糊神经网络控制器.将该控制器用于控制实际倒立摆系统,并采用状态变量合成方法以大大减少模糊规则的数目,实验结果验证了该控制器的有效性.
A design scheme of optimal RBF fuzzy neural network controller is proposed based on artificial immune principle. The structure of the controller is given, then the immune evolutionary algorithm is used to optimize the parameters of the controller to design an optimal RBF fuzzy neural network controller that satisfies quadratic performance index. The designed controller is employed to control an actual inverted pendulum system, and the scheme of state variable synthesis is used to reduce the number of fuzzy rules greatly. Experiment results has verified the effectiveness of the designed controller.
出处
《信息与控制》
CSCD
北大核心
2004年第3期380-384,共5页
Information and Control
基金
国家自然科学基金资助项目(50138010)哈尔滨工业大学跨学科交叉性研究基金资助项目(HIT.MD2001.02)
关键词
免疫进化算法
模糊神经网络
模糊控制
人工免疫系统
immune evolutionary algorithm
fuzzy neural network
fuzzy control
artificial immune system
