GARM: A Stochastic Evolution based Genetic Algorithm with Rewarding Mechanism for Wind Farm Layout Optimization

Mohamed Mohandes, Salman A. Khan*, Shafiqur Rehman, Saudi Arabia, Ali Al-Shaikhi, Bo Liu, Kashif Iqbal

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review


Wind energy has emerged as a potential alternative to traditional energy sources for economical and clean power generation. One important aspect of wind energy generation is the layout design of the wind farm so as to harness maximum energy. Due to its inherent computational complexity, the wind farm layout design problem has traditionally been solved using nature-inspired algorithms. An important issue in nature-inspired algorithms is the termination condition, which governs the execution time of the algorithm. To optimize the execution time, appropriate termination conditions should be employed. This study proposes the concept of a rewarding mechanism to achieve optimization in termination conditions while maintaining the solution quality. The proposed rewarding mecha-nism,adopted from the stochastic evolution algorithm, is incorporated into a genetic algorithm. The proposed genetic algorithm with the rewarding mechanism (GARM) is empirically tested using real data from a potential wind farm site with different rewarding iterations.

Original languageEnglish
Pages (from-to)575-584
Number of pages10
JournalFME Transactions
Issue number4
StatePublished - 2023

Bibliographical note

Publisher Copyright:
© 2023 Faculty of Mechanical Engineering, Belgrade. All rights reserved.


  • Artificial Intelligence
  • Genetic Algorithms
  • Nature-inspired algorithms
  • Optimization
  • Stochastic Evolution
  • Wind energy
  • Wind farm layout design
  • Wind farm micrositing

ASJC Scopus subject areas

  • Mechanics of Materials
  • Mechanical Engineering


Dive into the research topics of 'GARM: A Stochastic Evolution based Genetic Algorithm with Rewarding Mechanism for Wind Farm Layout Optimization'. Together they form a unique fingerprint.

Cite this