A model-free approach for auto-tuning of model predictive control

A two-layer approach for the auto-tuning of model predictive control (MPC) is proposed. The bottom layer computes the weighting matrices of the cost function from a desired closed-loop bandwidth while the top layer aims at finding the optimal bandwidth. This optimum corresponds to the optimal balance between the robustness and nominal performance of the closed-loop system. To find the optimal bandwidth, the extremum seeking (ES) algorithm, a form of non-model-based adaptive optimisation, is proposed. The auto-tuning approach is tested on a binary distillation column model. It is shown that the auto-tuning approach enables the MPC system to track its optimal closed-loop bandwidth and therefore obtain the minimum output variance.

[1]  Yoshikazu Fukuyama,et al.  Automatic Tuning of Model Predictive Control Using Particle Swarm Optimization , 2007, 2007 IEEE Swarm Intelligence Symposium.

[2]  J. Maciejowski,et al.  Tuning MPC using H ∞ loop shaping , 2000 .

[3]  Sebastian Engell,et al.  Tuning MPC for desired closed-loop performance for MIMO systems , 2011, Proceedings of the 2011 American Control Conference.

[4]  Ying Tan,et al.  On the Choice of Dither in Extremum Seeking Systems: a Case Study , 2006, Proceedings of the 45th IEEE Conference on Decision and Control.

[5]  Milos S. Stankovic,et al.  Extremum seeking under stochastic noise and applications to mobile sensors , 2010, Autom..

[6]  B R Young,et al.  A tuning algorithm for model predictive controllers based on genetic algorithms and fuzzy decision making. , 2008, ISA transactions.

[7]  Jixin Qian,et al.  A Novel Robust Tuning Strategy for Model Predictive Control , 2006, 2006 6th World Congress on Intelligent Control and Automation.

[8]  John Bagterp Jørgensen,et al.  A tuning procedure for ARX-based MPC of multivariate processes , 2013, 2013 American Control Conference.

[9]  Corinne Alonso,et al.  Dynamic performance of maximum power point tracking circuits using sinusoidal extremum seeking control for photovoltaic generation , 2011 .

[10]  Chris Manzie,et al.  Newton-Like Extremum-Seeking for the Control of Thermoacoustic Instability , 2010, IEEE Transactions on Automatic Control.

[11]  Jay H. Lee,et al.  Tuning of model predictive controllers for robust performance , 1994 .

[12]  M. Soroush,et al.  Model Predictive Control Tuning Methods: A Review , 2010 .

[13]  J. Maciejowski,et al.  Robust finite horizon MPC without terminal constraints , 2000, Proceedings of the 39th IEEE Conference on Decision and Control (Cat. No.00CH37187).

[14]  Emad Ali,et al.  On-line tuning strategy for model predictive controllers , 2001 .

[15]  Quang N. Tran,et al.  MPC tuning based on impact of modeling uncertainty on closed-loop performance , 2012 .

[16]  M. Krstic,et al.  PID tuning using extremum seeking: online, model-free performance optimization , 2006, IEEE Control Systems.

[17]  I. Mareels,et al.  Extremum seeking from 1922 to 2010 , 2010, Proceedings of the 29th Chinese Control Conference.

[18]  Acpm Ton Backx,et al.  Generalized predictive control tuning by controller matching , 2015 .

[19]  Sten Bay Jørgensen,et al.  Tuning SISO offset-free Model Predictive Control based on ARX models , 2012 .

[20]  Ying Tan,et al.  On non-local stability properties of extremum seeking control , 2006, Autom..

[21]  Miroslav Krstic,et al.  Stability of extremum seeking feedback for general nonlinear dynamic systems , 2000, Autom..

[22]  Leyla Özkan,et al.  A frequency domain approach for MPC tuning , 2012 .