dc.contributor.author |
Guerazem, Said |
|
dc.contributor.author |
Benslimane, Tarik |
|
dc.contributor.other |
Boukhetala, Djamel, Directeur de thèse |
|
dc.contributor.other |
Achour, Hakim, Directeur de thèse |
|
dc.date.accessioned |
2024-10-07T13:21:39Z |
|
dc.date.available |
2024-10-07T13:21:39Z |
|
dc.date.issued |
2024 |
|
dc.identifier.other |
EP00730 |
|
dc.identifier.uri |
http://repository.enp.edu.dz/jspui/handle/123456789/11009 |
|
dc.description |
Mémoire de Projet de Fin d’Études : Automatique : Alger, École Nationale Polytechnique : 2024 |
fr_FR |
dc.description.abstract |
The main objective of this End of Studies project concerns the level control of the coupled tank systems. We begin with a description of the benchmark available in the laboratory of the Control Engineering Department, then we develop an analytical mathematical model of the system in question through which the description of the dynamic characteristics of the system has been carried out. A decentralised PI controller was adopted whilst the adjustment of its parameters has been carried out using a PSO algorithm. This control technique was taken as a reference for comparison with the techniques that we developed subsequently. Firstly, we develop two control techniques , linear and non-linear model predictive control (MPC) techniques. Additionally, we propose another approach based on recurrent neural networks (RNN) to predict the control inputs and reduce the computation time compared with conventional methods. Finally, we used Takagi-Sugeno type fuzzy systems to describe the non-linear model for multi-model control with stability analysis and trajectory tracking. Robustness tests have been carried out to evaluate the performance of each method |
fr_FR |
dc.language.iso |
en |
fr_FR |
dc.subject |
Quadruple |
fr_FR |
dc.subject |
Tank process |
fr_FR |
dc.subject |
Decentralized PI control Model Predictive Control (MPC) -Recurrent Neural Networks (RNNs) |
fr_FR |
dc.subject |
Model Predictive Control (MPC) |
fr_FR |
dc.subject |
Recurrent neural networks (RNNs) |
fr_FR |
dc.subject |
Takagi-Sugeno fuzzy models |
fr_FR |
dc.subject |
PDC control |
fr_FR |
dc.subject |
Linear matrix inequality (LMI) |
fr_FR |
dc.subject |
Particle Swarm Optimization (PSO) |
fr_FR |
dc.title |
Fuzzy predictive control of the coupled tanks process |
fr_FR |
dc.type |
Thesis |
fr_FR |