Loading...

Model Predictive Control Based on Integer Programming

Ahmadi, Vahid | 2015

564 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 47979 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Farhadi, Alireza
  7. Abstract:
  8. Many industrial systems have on/off actuators. Traditionally, these systems are controlled using traditional control methods, like PID controller. The performance of this controller is not ideal for these systems, especially in term of transient response and power consumption. Due to the large number of constraints for these systems, to improve the performance of them, it is necessary to use the model predictive control technic. The decision variables of these systems are integer, hence it is necessary to develop the model predictive control method based on integer programming, which is the subject of this thesis. The main challenge in the development of this model predictive control method, is to find the best technic for solving the integer programming based optimization problem in each receding horizon, with low computational complexity. In this thesis, first some important and famous systems with on/off actuators are presented and their dynamic models are explained. Then a suitable model predictive control method based on integer programming is introduced. Finally to illustrate the performance of the developed model predictive control method, the temperature of an autoclave cylinder with linear dynamic model and the flow rate of a smart oil well system with non-linear dynamic model are controlled
  9. Keywords:
  10. Integer Programming ; Model Predictive Control ; Optimal Control ; On/Off Actuators

 Digital Object List

 Bookmark

No TOC