Defesa de Dissertação de Mestrado – Tiago Araújo Elias – 05/09/2018
28/08/2018 14:47
| Defesa de Dissertação de Mestrado | |
| Aluno | Tiago Araújo Elias |
| Orientador | Prof. Julio Elias Normey-Rico, Dr. – DAS/UFSC |
| Data | 5/9/2018 (quarta-feira) – 14h30
Sala PPGEEL 2 (ao lado do Teixeirão) |
| Banca | Prof. Julio Elias Normey-Rico, Dr. – Presidente – DAS/UFSC;
Prof. Daniel Martins Lima, Dr. – UFSC/Blumenau; Prof. Eduardo Camponogara, Dr. – DAS/UFSC; Prof. Gustavo Artur de Andrade, Dr. – DAS/UFSC; Prof. Rodolfo Cesar Costa Flesch, Dr. – DAS/UFSC (suplente). |
| Título | Hybrid Model Predictive Control for Solar Fields |
| Abstract: This master thesis presents an advanced control algorithm for reducing heat losses caused by clouds in large-scale solar fields and an algorithm for defocusing the collectors in order to avoid oil decomposing. Large-scale solar fields can have partial cloud cover, and the covered part of the field works as an energy dissipator. If the volumetric flow is increased the output temperature rises, and the system loses energy. Limiting maximum fluid temperature is also a concern in solar systems, considering that the source of energy cannot be manipulated. There are two ways to prevent over-temperature: reduce solar energy input into the collector, or remove excess heat from the collector. The formulation of the algorithms is based on a Mixed Logical Dynamical (MLD) representation of the solar field plus the application of a Practical Nonlinear Model Predictive Controller (PNMPC) for calculating the optimal control action. The main purposes of the controllers are: (i) to deactivate fields with the inlet temperature greater than the outlet temperature and to manipulate the oil flow rate of the activated fields for tracking the reference of the field outlet temperature; (ii) to defocus the solar collectors when the output temperature reaches the maximum value and to manipulate the fluid rate to maintain the field output temperature in the desired reference. Simulation results using irradiation profiles with cloud variations are presented for illustrating the advantages of the proposed fields deactivation approach. The results of the defocusing approach are presented supposing extreme scenarios of overheating. | |
