The main objective of this work is to design, adjust and validate a multivariable control structure for the control of a complex process. The process concerned is the temperature control of a PEM-type stack that performs the function of prime mover in a micro-CHP (micro Combined Heat and Power) system. It is a system that must supply electrical and thermal energy for domestic applications. Proper control of the thermal subsystem is essential for optimal operation. The work has to propose several alternatives for multivariable control. These structures must be optimally tuned to satisfy two objectives simultaneously: performance and control effort. The tuning problem is posed as a multiobjective optimization problem. In this particular case, there is uncertainty when working with a linear model when the process is non-linear. There is a part of the process dynamics that is not correctly modeled. A significant decrease in performance may occur when the designed controller is applied to the real process. Then an additional step in the controller validation consists of evaluating robustness for model uncertainties. The tasks to be performed in order to achieve the proposed objectives are: • To understand the system’s operation and to obtain linear models that adequately represent the system at least at one point of operation. • To understand the multiobjective optimization tools to be used for the approach and resolution of the problem of adjustment of the proposed control structures. • To define the multiobjective optimization problem for the tuning of multivariable controllers. • Multiobjective adjustment of the different multivariable control structures to be proposed. • Analysis of results and evaluation of the robustness for validating the proposed structure in the real process.
L’obiettivo principale di questo lavoro `e progettare, regolare e validare una struttura di controllo multivariabile per il controllo di un processo complesso. Il processo in questione `e il controllo della temperatura di uno stack di tipo PEM che svolge la funzione di motore primo in un sistema microCHP (micro Combined Heat and Power). E un sistema che deve fornire energia elettrica e termica per applicazioni domestiche. Il corretto controllo del sottosistema termico `e essenziale per un funzionamento ottimale. Il lavoro deve proporre diverse alternative per il controllo multivariabile. Queste strutture devono essere ottimizzate per soddisfare due obiettivi contemporaneamente: prestazioni e sforzo di controllo. Il problema del tuning si pone come un problema di ottimizzazione multiobiettivo. In questo caso particolare, c’`e incertezza quando si lavora con un modello lineare quando il processo non `e lineare. C’`e una parte della dinamica del processo che non `e modellata correttamente. Una diminuzione significativa delle prestazioni pu`o verificarsi quando il controller progettato viene applicato al processo reale. Quindi un ulteriore passaggio nella convalida del controller consiste nella valutazione della robustezza per le incertezze del modello. I compiti da svolgere per il raggiungimento degli obiettivi proposti sono: • Comprendere il funzionamento del sistema e ottenere modelli lineari che rappresentino adeguatamente il sistema almeno in un punto operativo. • Comprendere gli strumenti di ottimizzazione multiobiettivo da utilizzare per l’approccio e la risoluzione del problema dell’adeguamento delle strutture di controllo proposte. • Definire il problema di ottimizzazione multiobiettivo per la messa a punto di controllori multivariabili. • Adeguamento multiobiettivo delle diverse strutture di controllo multivariabili da proporre. • Analisi dei risultati e valutazione della robustezza per validare la struttura proposta nel processo reale.
Design of a multivariable control for the liquid cooling system of a 2KW PEM stack
EFTEKHARIPOUR, HAMED
2021/2022
Abstract
The main objective of this work is to design, adjust and validate a multivariable control structure for the control of a complex process. The process concerned is the temperature control of a PEM-type stack that performs the function of prime mover in a micro-CHP (micro Combined Heat and Power) system. It is a system that must supply electrical and thermal energy for domestic applications. Proper control of the thermal subsystem is essential for optimal operation. The work has to propose several alternatives for multivariable control. These structures must be optimally tuned to satisfy two objectives simultaneously: performance and control effort. The tuning problem is posed as a multiobjective optimization problem. In this particular case, there is uncertainty when working with a linear model when the process is non-linear. There is a part of the process dynamics that is not correctly modeled. A significant decrease in performance may occur when the designed controller is applied to the real process. Then an additional step in the controller validation consists of evaluating robustness for model uncertainties. The tasks to be performed in order to achieve the proposed objectives are: • To understand the system’s operation and to obtain linear models that adequately represent the system at least at one point of operation. • To understand the multiobjective optimization tools to be used for the approach and resolution of the problem of adjustment of the proposed control structures. • To define the multiobjective optimization problem for the tuning of multivariable controllers. • Multiobjective adjustment of the different multivariable control structures to be proposed. • Analysis of results and evaluation of the robustness for validating the proposed structure in the real process.File | Dimensione | Formato | |
---|---|---|---|
Thesis.pdf
Open Access dal 30/11/2023
Dimensione
3.88 MB
Formato
Adobe PDF
|
3.88 MB | Adobe PDF | Visualizza/Apri |
Executive_Summary.pdf
Open Access dal 30/11/2023
Dimensione
967.48 kB
Formato
Adobe PDF
|
967.48 kB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/198847