人工神经元网络在碱回收苛化过程控制系统的研究  

Study of the Control Syetem for Artificial Neural Network in Alkali Recovery Caustic Process

作　　者：李明辉[1] 孟宪坤[1] 

机构地区：[1]陕西科技大学机电工程学院,陕西西安710021

出　　处：《郑州大学学报（工学版）》2011年第6期96-100,共5页Journal of Zhengzhou University（Engineering Science）

基　　金：陕西省科技计划自然基金项目资助(2011JM8001)

摘　　要：通过分析研究苛化工段的生产工艺及对控制系统的要求,结合影响苛化过程的主要因素与神经元网络的特性,提出了采用人工神经元网络建立苛化过程的数学网络模型,经MATLAB仿真验证了该模型符合实际的工艺情况,并对所确定的网络模型的输出和相应目标进行线性回归,进一步验证了该模型苛化度控制效果良好.在此模型基础上,通过对苛化过程优化问题的描述,以生产成本为目标函数,对苛化过程进行了操作优化的处理,经实际生产运行表明了该优化在满足苛化效果的同时,提高了系统的综合控制指标,节约了生产成本.Through analysis and study the production process and control system requirements of Caustic process,in combination with the main factors of impact on caustic process and the characteristics of neural networks,we proposed using artificial neural network to establish mathematical network model of caustic process,verified the model in line with the actual process conditions by MATLAB simulation,and made linear regression for the determined output of the network model and the corresponding target,further verified the caustic degree of the model has a good control effect.In this model,through described the problem of caustic process the cost of production as the objective function,we have optimized operation optimization for caustic process,and the production practice shows that the optimized meet the effects of caustic and improved the integrated control indicators of system,saving production costs.

关 键 词：苛化工段 神经元网络 过程控制 操作优化 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]