论文部分内容阅读
基于动态 L evenberg- Marquardt(L M)算法 ,提出两步 L M方法建立非线性过程的循环神经网络模型。该模型能以足够的精度并行于过程运行 ,并能从过程的输入信息模拟过程未来的响应。研究了基于该模型的扩展 DMC预测控制策略 ,仿真结果表明该控制器的性能得到了很大提高。
Based on the dynamic L evenberg-Marquardt (L M) algorithm, a two-step L M method is proposed to build a neural network model of nonlinear process. The model runs with sufficient accuracy parallel to the process and simulates the process’s future response from the process’s input. The extended DMC predictive control strategy based on this model is studied. The simulation results show that the performance of this controller has been greatly improved.