Model Based Control of Refrigeration Systems

Thybo has supervised the work. I would especially like to express my gratitude to my good friend, dive mate, college and supervisor Claus Thybo, for his invaluable help, support and inspiration. His guidance and support has truly been ideal. Without his presence this project would not have been at all and it wouldn't have been so enjoyable. I would like to thank Jakob Stoustrup for sharing of his profound knowledge within control theory, that has improved the academic level of this thesis. Moreover I would like to thank Henrik Rasmussen, who has more than a lifetimes experience within application of control theory. His sincere and eager interest in my work has truly been a source of inspiration. I would like to thank Professor Manfred Morari for giving me the opportunity to stay at the Automatic Control Laboratory, ETH Zürich and for giving me valuable guidance while I was there. I was amazed by the unique scientific milieu and the high academic level there. I would especially like to thank Tobias Geyer, who contributed with many ideas and a lot of help, while I was at the ETH. I would furthermore like to thank him for taking of his valuable time at the ending of his Ph.D.-study to come and visit Danfoss and Aalborg University. I would like to thank the people at the department of ETH; Jo-han, Helfried, Miroslav and Frank for making it a really enjoyable and memorable stay. v Summary The subject for this Ph.D. thesis is model based control of refrigeration systems. Model based control covers a variety of different types of controls, that incorporates mathematical models. In this thesis the main subject therefore has been restricted to deal with system optimizing control. The optimizing control is divided into two layers, where the system oriented top layers deals with set-point optimizing control and the lower layer deals with dynamical optimizing control in the subsystems. The thesis has two main contributions, i.e. a novel approach for set-point optimization and a novel approach for desynchronization based on dynamical optimization. The focus in the development of the proposed set-point optimizing control has been on deriving a simple and general method, that with ease can be applied on various compositions of the same class of systems, such as refrigeration systems. The method is based on a set of parameter depended static equations describing the considered process. By adapting the parameters to …