MPC-Tuning: Hybrid Optimization for Model Predictive Controllers

📖 Overview

This repository contains the source code for MPC-Tuning, developed by Sergio A. C. Giraldo (principal), Príamo A. Melo, and Argimiro R. Secchi at the Laboratory of Software Development for Process Control and Optimization (LADES), affiliated with the Programa de Engenharia Química (PEQ), Universidade Federal do Rio de Janeiro (UFRJ).

The purpose of this project is to provide a hybrid tuning methodology for Model Predictive Controllers (MPC), which optimizes both continuous and discrete tuning parameters to improve control performance while reducing computational effort.

This repository serves as a resource for researchers, control engineers, and practitioners working in process control, predictive modeling, and automation.

🚀 Features

• Hybrid tuning method for MPC optimization.
• Implementation with MATLAB's Model Predictive Control Toolbox.
• Support for both linear and nonlinear systems.
• Code adapted for practical applications and industrial simulations.

📦 Installation

To install and use this project, follow these steps:

Prerequisites

Ensure you have the following dependencies installed:

# Required MATLAB Toolbox:
MATLAB Model Predictive Control Toolbox

Usage

1. Clone the repository:

   git clone https://github.com/LADES-PEQ/MPC-Tuning.git

2. Navigate to the project directory:

   cd MPC-Tuning

3. Open MATLAB and add the project path:

   addpath(genpath('MPC-Tuning'))

4. Run an example:

   run('MPC-Tuning/Examples/Example1.m')

5. (Optional) Modify parameters in mpc_tuning.m to adapt the optimization to your system.

📂 Repository Structure

├── MPC-Tuning/           # Core tuning algorithms
└── DTC-GPC/              # Example implementation of DTC-GPC
└── Explicit-NMPC/        # Explicit Nonlinear MPC cases
└── Matlab-Toolbox/       # MPC examples using the MPC Toolbox - Matlab

✏️ Authors & Contributors

This project was developed by:

• Sergio A. C. Giraldo - Principal Researcher - sergio@peq.coppe.ufrj.br
• Príamo A. Melo - Research Collaborator
• Argimiro R. Secchi - Research Supervisor

We welcome contributions!.

🔬 References & Publications

If you use this work in your research, please cite the following publication:

• Sergio A. C. Giraldo, Príamo A. Melo, and Argimiro R. Secchi.
  "Tuning of Model Predictive Controllers Based on Hybrid Optimization",
  Processes, vol. 10, no. 2, 2022.
  DOI: 10.3390/pr10020351

BibTeX Citation

@article{Giraldo2022,
  author  = {Giraldo, Sergio A. C. and Melo, Príamo A. and Secchi, Argimiro R.},
  title   = {Tuning of Model Predictive Controllers Based on Hybrid Optimization},
  journal = {Processes},
  volume  = {10},
  year    = {2022},
  number  = {2},
  article-number = {351},
  url     = {https://www.mdpi.com/2227-9717/10/2/351},
  issn    = {2227-9717},
  doi     = {10.3390/pr10020351}
}

🛡 License

This work is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) License. You are free to:

• Use, modify, and distribute this code for any purpose.

• Cite the following reference when using this code:

  Sergio A. C. Giraldo, Príamo A. Melo, and Argimiro R. Secchi.
  "Tuning of Model Predictive Controllers Based on Hybrid Optimization",
  Processes, vol. 10, no. 2, 2022.
  DOI: 10.3390/pr10020351

See the full license details in the LICENSE file.

📞 Contact

For inquiries or collaborations, please contact Sergio A. C. Giraldo at sergio@peq.coppe.ufrj.br or open an issue in this repository.

⭐ Acknowledgments

• We acknowledge the support of the LADES Research Group, PEQ - UFRJ, and all contributors.
• This research was funded by the National Council for Scientific and Technological Development (CNPq) and the Brazilian Coordination for the Improvement of Higher Education Personnel (CAPES).