应用单层神经网络设计多变量自适应模糊控制器

提出一种应用单层神经网络设计多变量自适应模糊控制器的方法。应用单层神经网络可以学习多变量模糊控制规则中的未知参数，还可由它来实现多变量模糊推理过程。该方法能解决多变量模糊控制中普遍存在的规则获取困难和难于实现实时自适应等问题。仿真试验表明，所设计的多变量模糊控制器不仅实时性好，而且可得到满意的控制效果。

Design Multivariable Adaptive Fuzzy Controller Based on a Single-layer NeuralNetwork

A design method of multivariable adaptive fuzzy controller, based on a single-layer neural network, is proposed in this paper.The parameters of me fuzzy control rules of me controller can be learned by the learning slgorithm of the neural netowrk. and the inference process can be realized by the network. The presented design method can solve the difficulties in real-time adapting and in deducing the control rules Which are general problems existed in multivariable adaptive fuzzy ccontrol. The results of a simulation show that the method is effective.