基于GRU神经网络辨识系统的非线性MPC无静差跟踪
系统与控制
2022-01-26 v4 机器学习
系统与控制
摘要
由于黑盒建模能力,循环神经网络(RNNs)用于系统辨识近来受到越来越多关注。尽管RNN已成功应用于许多领域,但仅有少数工作致力于提供严谨的理论基础以证明其用于控制目的的合理性。本文旨在描述如何训练稳定的门控循环单元(GRUs,一种特定RNN架构)并将其用于非线性MPC框架中,以执行常值参考的无静差跟踪并保证闭环稳定性。所提方法在pH中和过程基准上测试,展现出卓越性能。
引用
@article{arxiv.2103.02383,
title = {Nonlinear MPC for Offset-Free Tracking of systems learned by GRU Neural Networks},
author = {Fabio Bonassi and C. F. Oliveira da Silva and Riccardo Scattolini},
journal= {arXiv preprint arXiv:2103.02383},
year = {2022}
}
备注
This work is the extended version of the article accepted at the Third IFAC Conference on Modelling, Identification and Control of Nonlinear Systems (MICNON 2021) for publication under a Creative Commons Licence CC-BY-NC-ND