Journal of the Korea Institute of Military Science and Technology (한국군사과학기술학회지)

The Korea Institute of Military Science and Technology (한국군사과학기술학회)
권호 표지
DOI QR코드

DOI QR Code

Comparison of Multi-Objective Optimization Results Using NSGA-Ⅱ Based on Military Wheeled Vehicle Endurance

군용 차륜 차량 내구시험 데이터 기반 NSGA-Ⅱ를 이용한 다목적 최적화 결과 비교

  • Gyeonghun Lee (School of Mechanical Engineering, Pusan National University) ;
  • Yoojeong Noh (School of Mechanical Engineering, Pusan National University) ;
  • Youngjin Kang (Research Institute of Mechanical Technology, Pusan National University) ;
  • Inho Baek (School of Mechanical Engineering, Pusan National University) ;
  • Jeonghwan Lee (6th R&D Institute - 4th Directorate, Agency for Defense Development) ;
  • Chiyoung Ryu (6th R&D Institute - 4th Directorate, Agency for Defense Development)
  • 이경훈 (부산대학교 기계공학부) ;
  • 노유정 (부산대학교 기계공학부) ;
  • 강영진 (부산대학교 기계기술연구원) ;
  • 백인호 (부산대학교 기계공학부) ;
  • 이정환 (국방과학연구소 제6기술연구원 4부) ;
  • 류치영 (국방과학연구소 제6기술연구원 4부)
  • Received : 2024.09.21
  • Accepted : 2025.01.13
  • Published : 2025.04.05
Download PDF
⟨ Previous Next ⟩

Abstract

This study aims to optimize the endurance test mode for military vehicles to more closely simulate actual operational conditions. To achieve this, a modified bi-objective optimization approach was developed, combining two system-level objectives for the relative damage of the chassis and powertrain with six corresponding component-level constraints. The modified bi-objective optimization employs the NSGA II(Non-dominated Sorting Genetic Algorithm II) to generate a diverse set of optimal solutions. The results, evaluated using the CRM(Coefficient of Residual Mass) and GMRAE(Geometric Mean Relative Absolute Error) metrics, compare this approach with bi-objective optimization focused solely on system-level objectives and six-objective optimization focused solely on component-level objectives. The findings show that the modified bi-objective approach provides a more accurate and reliable endurance test mode compared to both the bi-objective and six-objective methods.

Keywords

Acknowledgement

이 논문은 2023년 정부(방위사업청)의 재원으로 국방과학연구소 지원을 받아 수행된 연구임(UI230014UD).