A new objective function design for optimization of secondary controllers in load frequency control

Yilmaz Z. Y., BAL G., Celik E., ÖZTÜRK N., GÜVENÇ U., Arya Y.

JOURNAL OF THE FACULTY OF ENGINEERING AND ARCHITECTURE OF GAZI UNIVERSITY, vol.36, no.4, pp.2053-2067, 2021 (SCI-Expanded) identifier identifier identifier

  • Publication Type: Article / Article
  • Volume: 36 Issue: 4
  • Publication Date: 2021
  • Doi Number: 10.17341/gazimmfd.841751
  • Journal Name: JOURNAL OF THE FACULTY OF ENGINEERING AND ARCHITECTURE OF GAZI UNIVERSITY
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Art Source, Compendex, TR DİZİN (ULAKBİM)
  • Page Numbers: pp.2053-2067
  • Keywords: Load frequency control, PID controller, objective function, two-area thermal power system, multi-source power system, AUTOMATIC-GENERATION CONTROL, DIFFERENTIAL EVOLUTION ALGORITHM, LEARNING BASED OPTIMIZATION, SYMBIOTIC ORGANISMS SEARCH, PI-PD CONTROLLER, POWER-SYSTEMS, PSO, PLUG, AGC
  • Gazi University Affiliated: Yes

Abstract

In this study, load frequency control (LFC) of two-area non-reheat thermal power system and multi-source power systems is addressed. A simple PID-structured controller is used as a secondary controller in these systems. To raise the performance of PID controller, a new multi-objective function is designed and PID controller parameters are acquired by minimizing the value of this function with symbiotic organisms search (SOS) algorithm. All electrical power systems simulated are modeled in MATLAB/Simulink environment and the optimizer is coded in MATLAB/M-file platform. In order to affirm the contribution of the work, results collected from each power system are compared with popular results published in prestigious journals. As per the comparative results, despite its simplicity, SOS:PID controller tuned via the proposed objective function is observed to result in better performance than other approaches in terms of oscillations, settling time, maximum overshoot and maximum undershoot time domain indicators of the frequency and tie-line power change curves.