Loading...

Improving Transient Response in Data-Driven Control Systems

Enayati Kafshgarkolaei, Amir | 2021

314 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 54210 (05)
  4. University: Sharif University of Technology
  5. Department: Electrical Engineering
  6. Advisor(s): Tavazoei, Mohammad Saleh
  7. Abstract:
  8. The term "data-driven" was first introduced in computer science and has entered the vocabulary of controlled engineering in recent years. So far, several classes of data-driven-control (DDC) methods have been introduced, which are identified by different names in the literature such as data-driven control, data-based control, modeless control, model-free adoptive control, iterative feedback tuning and virtual reference feedback tuning.Data-driven control includes all control theories and methods in which the controller directly and using the input-output data of the control system or knowing the processed data and without explicit knowledge of the mathematical model of the system (unlike Model-based control (MBC) methods are designed to ensure the stability of the closed-loop system and the achievement of the intended control objectives under reasonable assumptions. The main difference between MBC and DDC is that in MBC, assuming that a reasonable model for the process is available, the control system design approach is based on the use of this model; But in DDC, assuming that there is no reliable mathematical model for the process, the design approach is data-driven.Due to this difference in nature, DDC has many intrinsic features: (1) The controller in DDC methods does not explicitly use part or all of the process model. For this reason, the design of the control system has no dependence on the process model. (2) Stability and convergence in DDC methods do not depend on the accuracy of the model. (3) In the DDC approach, the problem of unmodulated dynamics and consistency that exists in traditional MBC theory will no longer exist. Despite the above inherent characteristics, data-driven controllers can produce undesirable transient responses compared to model-driven controllers.In this regard, in this project, the problem is studied in more detail and solutions based on the optimal adjustment of control parameters or modification of the controller structure to improve transient response (with indicators such as uplift or response speed) in data-driven control systems. presented
  9. Keywords:
  10. Transient Response ; Adaptive Control ; Data Driven Method ; Model Based Control ; Data Driven Control Systems

 Digital Object List

 Bookmark

No TOC