Loading...

Determination of Number and Location of Models to Improve Transient Response in a Multi-model Adaptive Control

Hosseinpour, Sajjad | 2017

563 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 49890 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Haeri, Mohammad
  7. Abstract:
  8. In recent decays, multiple model controllers have been attracted much attention due to their essential ability to control of non-linear systems with wide operating ranges. Number and location of models are two important factors. In this research an algorithm for determining number of models and their locations has been proposed which can be counted as an extension to existing algorithms. Utilizing the proposed algorithm, the minimum number of models which achieve closed-loop stability can be determined. If transient response of non-linear system with these models be not satisfactory, the number of models are increased by one using Binary Search method, and consequently the location of models would be determined by former algorithms. Moreover, if the transient response of non-linear system with these models does not satisfy the desirable conditions, once again the Binary Search is utilizing to increase the number of models. This algorithm will be continued till the desirable transient response for system output is achieved. After determining the number of models and their locations, MPC controllers are designed. Linear combination are used to combine designed MPC controllers. Number of models, their locations and also the weight of each model are specified using gap metric. The proposed algorithm is applied to four industrial practical models consisting 2 two-input two-output systems and 2 SISO systems. The significances of the proposed algorithm are demonstrated considering integral absolute transient error criteria
  9. Keywords:
  10. Adaptive Controller ; Transient Response ; Multiple Modeling ; Binary Algorithm ; Search Algorithm ; Gap Metric ; Weightny Models

 Digital Object List

 Bookmark

No TOC