A Novel Tuning Approach for MPC Parameters Based on Artificial Neural Network: An application to FOPDT System

Houssam Moumouh , Nicolas Langlois (1) , Madjid Haddad

1 Pôle Automatique et Systèmes

A successful implementation of Model Predictive Control (MPC) requires appropriately tuned parameters. In this paper an Artificial-Neural-Network (ANN) based approach is presented and detailed in the case of a First Order Plus Dead Time (FOPDT) control-lable system. The original part of our approach lies in its capability to tune the MPC parameters using Particle-Swarm-Optimization (PSO) and Online-Sequential-Extreme-Learning-Machine(OS-ELM). This approach allows also to reach efficiently closed-loop stability. The effectiveness of our approach has been emphasized by comparing the obtained performances to other existing methods.

Keywords: Model Predictive Control; tuning parameters; Artificial Neural Network; FOPDT; PSO; OS-ELM

Domains: Engineering Sciences [physics]; Automatic

Houssam Moumouh, Nicolas Langlois, Madjid Haddad. A Novel Tuning Approach for MPC Parameters Based on Artificial Neural Network: An application to FOPDT System. 15th European Conference on Advanced Control and Diagnosis ACD'19, Nov 2019, Bologna, Italy. ⟨hal-02407974⟩