A Comparative Study of Flatness Controller Tuning Methods using Bio-Inspired Algorithms for a Coupled Tanks System

Gabriel da Silva Lima; Matheus Bawden Silverio de Castro; Vinicius Rafael de Freitas; Eduardo Perez Liberato; José Oniram de Aquino Limaverde Filho; Eugenio Libório Feitosa Fortaleza; Rafael Valladares de Almeida

doi:10.20906/SBAI-SBSE-2023/3954

Gabriel da Silva Lima Automation and Control Group (GRACO), Department of Mechanical Engineering, University of Brasília, Brasília, DF
Matheus Bawden Silverio de Castro Automation and Control Group (GRACO), Department of Mechanical Engineering, University of Brasília, Brasília, DF
Vinicius Rafael de Freitas Automation and Control Group (GRACO), Department of Mechanical Engineering, University of Brasília, Brasília, DF
Eduardo Perez Liberato Automation and Control Group (GRACO), Department of Mechanical Engineering, University of Brasília, Brasília, DF
José Oniram de Aquino Limaverde Filho Automation and Control Group (GRACO), Department of Mechanical Engineering, University of Brasília, Brasília, DF
Eugenio Libório Feitosa Fortaleza Automation and Control Group (GRACO), Department of Mechanical Engineering, University of Brasília, Brasília, DF
Rafael Valladares de Almeida Repsol Sinopec Brasil, Research and Development, Rio de Janeiro, RJ

DOI: https://doi.org/10.20906/SBAI-SBSE-2023/3954

Keywords: Liquid Level Control, Differential Flatness Theory, Particle Swarm Optimization, Gray Wolf Optimization, Nonlinear Coupled Tanks System

Abstract

Liquid level control has been a continuous challenge in a considerable number of industrial processes, especially when one has to simultaneously satisfy multiple control performance specifications. This paper presents a comparative study between Particle Swarm Optimization and Gray Wolf Optimization as tuning methodologies for a flatness-based controller, which has proved to be a useful tool to provide efficient solutions to tracking control tasks. Numerical simulations of a nonlinear coupled tanks system were performed to demonstrate the potential of using such bio-inspired algorithms as tuning methods for nonlinear controllers.