重油分馏塔基于混沌神经网络的Laguerre函数模型自适应预测控制

A Laguerre Function Model Adaptive Predictive Control Strategy Based on a Chaotic Neural Network for Heavy Oil Distillation Column

  • 摘要: 基于Laguerre函数模型的自适应预测控制方法中的性能指标,在有约束的情况下往往难以达到全局极优,而混沌神经网络(CNN)可以有效地避免优化过程陷入局部极小.文章简介了Laguerre预测控制策略的基本方法和CNN的特点,着重提出了一种利用CNN对控制性能指标进行寻优的新颖策略.在重油分馏塔Shell模型上的仿真实验结果表明,这种混合智能控制策略比原有控制策略在控制品质上有显著提高.

     

    Abstract: Performance index of the adaptive predictive control strategy based on Laguerre model is hard to converge to global optima.And chaotic neural network(CNN)can effectively avoid local optima during the optimization process.In this paper,the adaptive predictive control strategy based on Laguerre model and the features of CNN are briefly introduced.And a novel strategy to apply CNN to optimize the performance index is emphatically brought forward.Simulation results on Shell model of heavy oil distillation column show that the control quality of this hybrid intelligent strategy is markedly improved compared with original control strategy.

     

/

返回文章
返回