基于径基函数神经网络的精馏塔自适应控制  

Adaptive Control of Distillation Columns Based on RBF Neural Networks

作　　者：陈小红[1] 高峰[2] 钱积新[2] 孙优贤[2] 

机构地区：[1]同济大学CIMS研究中心,上海200092 [2]浙江大学工业控制研究所,杭州310027

出　　处：《控制理论与应用》1998年第2期226-231,共6页Control Theory & Applications

基　　金：国家自然科学基金!69334010

摘　　要：精馏塔是化工过程中最常用，最重要的操作单元，其本质的非线性及时变性使得对它的控制非常困难．本文提出了一种基于径基函数（RBF）网络的自适应控制方案．该方案简捷、可靠，具有很强的鲁棒性和抗干扰性能．将该方案应用在工业脱乙烷塔的控制中得到了令人满意的结果．Disillation columns are the most usual and important operating units in the process of chemicalengineering. They are non-linear and time-varying,and these characteristics make the design of the controlscheme very difficult. This paper proposed an adaptive control strategy based on radial basis function (RBF)neural network. The control scheme is simple,reliable and possesses strong robustness and disturbance rejection. Fairly good control results were obtained when the control scheme was applied to a distillation collumn.

关 键 词：精馏塔 RBF神经网络 化工过程 自适应控制 

分 类 号：TQ051.81[化学工程] TP273.2[自动化与计算机技术—检测技术与自动化装置]