基于粗糙集的模糊神经网络控制器的研究

针对现代工业过程的不断复杂化和其中非线性、不确定性因素的增加,给出了一种基于粗糙集的模糊神经网络控制器的设计方法,该方法将粗糙集理论与模糊神经网络结合起来,利用粗糙集从观测的输入输出数据中提取规则,并寻求最小规则集,解决了模糊神经网络"规则爆炸"问题。通过在MATLAB平台上进行仿真,结果表明该方法具有良好的控制能力,对于突加干扰具有良好自适应能力。

Research of fuzzy neural network controller based on rough set

Due to the complication of modern industrial process,and the increase in nonlinearity and uncertainty in it,a designing method of fuzzy neural network controller based on rough set is proposed in this paper.The method combines the Rough set with fuzzy neural network,which can derive control rules from input-output data effectively and find the minimal set of rules.And this method can solve the "rule explosion" problem.Results obtained from a simulation using MATLAB indicate that the proposed method has good-quality controlling ability and self-adaptive ability when encountering sudden disturbances.