摘要
目的采用神经网络预测控制方法来解决铝电解过程中存在的时变和大时滞问题,提高其控制性能.方法提出了一种基于铝电解过程的神经网络预测控制算法,建立了神经网络预测模型,将神经网络和预测控制算法相结合.结果实现了铝电解过程的最优控制.神经网络预测模型的输出能够很好地跟踪铝电解生产过程,预测效果好.结论笔者提出的控制方案能够使铝电解过程很快进入稳态,超调量较小,提高了铝电解过程的动态和稳态性能.
Using neural network predictive control method can solve the problem of time variation and big delay which exit in the aluminum electrolysis process. That can improve the control charactristies, This paper presents an predictive control algorithm of neural network based on aluminum electrolysis process, establishes the predictive model of neural network, combines the neural network and predictive control algorithm, and realizes the optimize control of aluminum electrolysis process. The simulation result shows the predictive model can trace the aluminum electrolysis process. The predictive result is good. The method presented in this paper can make the aluminum electrolysis process stable, with low overshot and better dynamic and static performance,
出处
《沈阳建筑大学学报(自然科学版)》
EI
CAS
2006年第6期1023-1026,共4页
Journal of Shenyang Jianzhu University:Natural Science
基金
建设部基金项目(03-2-117)
辽宁省教育厅项目(202080720)
关键词
铝电解
神经网络
预测控制
模型辨识
aluminum electrolysis
neural network
predictive control
plant identification