基于神经模糊的矿井排水系统控制器的研究

A research on nerve-fussy based mine water discharge control system

结合流量的时变、滞后和非线性特性 ,提出了一种基于人工神经网络与模糊控制相结合的控制器。利用人工神经网络的自学习、自适应和并行处理的能力 ,将模糊控制规则转化为神经网络的学习样本 ,通过ANN的BP学习算法记忆这些规则样本。实验表明该控制器具有响应速度快。

On the basis of time-related change, delay and non-linear characters of flow, this paper proposes a kind of controller that is based on the combination of artificial nerve-network and fussy-control. The self-study, self-adapting and parallel processing ability of the artificial nerve network is used to transform the fussy control rules into study sample for the nerve network. Then, the rule samples are remembered by the BP study algorithm via the ANN. Experiments show that the controller is characterized of fast response and high accuracy, etc.