Publication

Dynamical tuning for MPC using population games: A water supply network application

Journal Article (2017)

Journal

ISA Transactions

Pages

175-186

Volume

69

Number

2017

Doc link

http://doi.org/10.1016/j.isatra.2017.03.027

File

Download the digital copy of the doc pdf document

Abstract

Model predictive control (MPC) is a suitable strategy for the control of large-scale systems that have multiple design requirements, e.g., multiple physical and operational constraints. Besides, an MPC controller is able to deal with multiple control objectives considering them within the cost function, which implies to determine a proper prioritization for each of the objectives. Furthermore, when the system has time-varying parameters and/or disturbances, the appropriate prioritization might vary along the time as well. This situation leads to the need of a dynamical tuning methodology. This paper addresses the dynamical tuning issue by using evolutionary game theory. The advantages of the proposed method are highlighted and tested over a large-scale water supply network with periodic time-varying disturbances. Finally, results are analyzed with respect to a multi-objective MPC controller that uses a static tuning.

Categories

control theory, predictive control.

Author keywords

Dynamical tuning

Scientific reference

J. Barreiro-Gomez, C. Ocampo-Martínez and N. Quijano. Dynamical tuning for MPC using population games: A water supply network application. ISA Transactions, 69(2017): 175-186, 2017.