기계적 메타구조 강화를 통한 접착제 기계적 특성 분석 = Analysis of the Mechanical Properties of Adhesives Reinforced with Mechanical Meta-Structures

다국어 초록 (Multilingual Abstract)

Abstract
In this study, an AI-based surrogate modeling framework combined with finite
element analysis was proposed to efficiently predict the mechanical properties of
meta-structure reinforced adhesives and to identify design configurations to identify
optimal design configurations that satisfy predefined target properties. The
geometric design variables of the meta-structure (Lx, Ly, D, S, and t) were used
as input features, while stiffness and Poisson’s ratio obtained from finite element
analysis were used as output responses. Various regression-based AI models were
trained, and the optimal predictive model for each mechanical property was
selected through based on cross-validation performance to construct the final
surrogate models.
Using the validated surrogate models, design candidates that simultaneously satisfy
the target stiffness and Poisson’s ratio were explored. The selected design
solutions were further verified through additional finite element analyses, showing
prediction errors within approximately 2–3% for both stiffness and Poisson’s ratio,
which confirms the effectiveness of the AI-based design exploration approach.
Furthermore,to provide physical insight into the design exploration result, analysis
of variance was performed. The proposed framework demonstrates the potential of
AI-based inverse design for efficient exploration of the design space of
meta-structure reinforced adhesives.

목차 (Table of Contents)

• 제 1 장 서 론 1
• 1.1 연구배경 1
• 1.2 문헌조사 2
• 1.3 연구목적 및 내용 4
• 제 2 장 메타구조 강화 접착제 유한요소해석 6
• 2.1 접착제 6
• 2.1.1 인장 물성 시험 방법 6
• 2.1.2 물성 시험 결과 8
• 2.2 메타구조 특성 10
• 2.2.1 메타구조란 10
• 2.2.2 메타구조 선정 10
• 2.3 메타구조 강화 접착제 시편 제작 방법 12
• 2.3.1 메타구조 제작 방법 12
• 2.3.2 메타구조 강화 접착제 제작 방법 13
• 2.4 해석 모델 정합성 검증 시험 14
• 2.4.1 인장 물성 (Tensile property) 시험 방법 14
• 2.4.2 변위 및 변형률 측정 방법 14
• 2.4.3 인장 특성 시험 결과 16
• 2.4.4 유한요소해석 20
• 2.4.5 경계조건 및 하중조건 21
• 2.4.6 해석 결과 22
• 2.5 메타구조 강화 접착제의 기계적 거동 분석 24
• 제 3 장 AI 학습 31
• 3.1 AI 모델 구성 및 학습 방법 31
• 3.2 AI 학습 결과 및 예측 성능 평가 34
• 3.3 AI 기반 역설계 절차 및 결과 36
• 제 4 장 결론 38
• 참고문헌 41
• Abstract 43
• 감사의글 48