循环神经网络建模在非线性预测控制中的应用

基于动态Ｌｅｖｅｎｂｅｒｇ－Ｍａｒｑｕａｒｄｔ（ＬＭ）算法，提出两步ＬＭ方法建立非线性过程的循环神经网络模型。该模型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。

Recurrent Neural Network Modeling and Its Application in Nonlinear Predictive Control

A two step Levenberg-Marquardt (LM) algorithm for designing a recurrent neural network process model was proposed based on dynamic LM algorithm. The obtained process model is sufficiently accurate to be used independently from the process, emulating the process response from only process input information. Implementation of a extend DMC predictive control strategy based on the identified recurrent neural network model was studied, and the simulation result shows the improvements in control performance.