Ph.D. received on: 8/7/1997E-mail: wia@upiura.edu.pe
Tutor: Prof. Claudio Maffezzoni, Politecnico di Milano
___________________________________________________________________________________________________________
Modelling, simulation and control of a binary distillation column ___________________________________________________________________________________________________________Advisor:
Prof. Riccardo Scattolini, Università di Pavia
Referee:
Prof. Guido Guardabassi, Politecnico di Milano
Summary:
The Thesis deals with the modelling and control of a pilot binary distillation column located at the Dipartimento di Elettronica e Informazione, Politecnico di Milano. The research has implied a large experimental work. The plant model has been developed by resorting to first principle, physical laws (mass and energy balance equations). The model has been subsequently simplified for control purposes.
The control problem has first been solved by means of a multivariable decentralized PI structure, where the optimal couplings have been determined with the aid of some well known indices, such as RGA. In the final part of the work a multivariable predictive controller has been developped and implemented. The obtained results witness the superior performance provided by this technique.
MINORS
Software for real time control in a multitasking environment
Advisor: Prof. Lorenzo Mezzalira, Politecnico di Milano
The main goal of this research has been to study some guidelines for the development of software for real time control. In the proposed environment, the designer can easily include and modify any kind of functional block.
Implementation structures for PID control
Advisor: Prof. Gian Antonio Magnani
This study deals with the analysis of various schemes already proposed in the literature for the implementation of PID controllers. These schemes should provide an effective decoupling among the response to set-point changes and to load disturbances. Some formulas are also proposed which guarantee this decoupling effect for a specified class of plant models.
_______________________________________