自适应模糊神经网络控制系统的研究

自适应模糊神经网络控制器是由模糊控制和神经网络相结合构成,它不依赖被控对象的数学模型,并能自动产生模糊控制规则,又具有良好的自适应性,是目前受人们关注的课题。本文在对其分析的基础上又提出了卡尔曼滤波的学习算法,解决了原BP算法实时性差的问题,通过仿真实验说明了其优越性,并体现了模糊神经网络与最优控制相结合的思想。

Research of Adaptive Fuzzy Neural Network Control System

The fuzzy neural network controller is made up of fuzzy control and neural network control, which does not require accu- rate model of plant and produces fuzzy rules automatically .It has adaptive merits and draws a lot of concentrations. After analysis, the Kalman filter arithmetic is showed, which solves the problem of real time caused by BP arithmetic, we can show its superiority by simulation, and moreover the idea that fuzzy neural network and the optimal control is combined is embodied.