基于灰分反馈的FNN自适应配煤控制  

Control of FNN Self-Adapting Coal Blending Based on Ash Feedback

作　　者：王致杰[1] 李冬[2] 洪留荣[2] 孟江[2] 常彦伟[2] 

机构地区：[1]山东科技大学工程学院,山东泰安271021 [2]中国矿业大学信电学院,江苏徐州221008

出　　处：《选煤技术》2005年第6期39-41,共3页Coal Preparation Technology

摘　　要：结合灰分控制的时变、滞后和非线性特性,介绍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FNN的BP学习算法记忆这些规则样本。试验表明,该控制器具有响应速度快、精度高和鲁棒性的特点。Combined with ash control related time variation, residence and nonlinear property, the paper gives an introduction to a controller, which is based on the combination of artificial neural net and fuzzy control. By use of the self-learning ability of artificial neural net and self-adapting and parallel treatment ability, it allows fuzzy control law to be transformed to neural net learning specimen copy. Through BP learning calculation procedure of FNN, these regulation specimen copies are brought in memory. Test shows that the controller has many advantages such as high speed in response, higher precise and robustness etc.

关 键 词：温度 模糊控制 人工神经网络 RP算法 

分 类 号：TD941.6[矿业工程—选矿] TP18[自动化与计算机技术—控制理论与控制工程]