Design of a Data-Driven Control System using a Multi-Objective Genetic Algorithm

In recent years, control design schemes for directly calculating control parameters from operational data have been realized and include the virtual reference feedback tuning (VRFT) method and the fictitious reference iterative tuning (FRIT) method. They were designed for objects that have a linear system. However, many objects in industry are nonlinear; hence, it is challenging to obtain good control performance by only applying fixed PID controllers. In this study, multiple linear systems as objects using multiple linear controllers are investigated. Specifically, it is necessary to solve two optimization problems of (i) the number of controllers (ii) the control parameters of each controller, and it is solving by using multi-objective genetic algorithm (MOGA) in this research. (C) 2019, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.