인문학
사회과학
자연과학
공학
의약학
농수해양학
예술체육학
복합학
DBpia AI
AI 에이전트
AI 아이디어
AI 리더
AI 뷰어
Citeasy
크롬 서지관리 다운로드
지원사업
학술연구/단체지원/교육 등 연구자 활동을 지속하도록 DBpia가 지원하고 있어요.
커뮤니티
연구자들이 자신의 연구와 전문성을 널리 알리고, 새로운 협력의 기회를 만들 수 있는 네트워킹 공간이에요.
논문 기본 정보
- 자료유형
- 학술저널
- 저자정보
- 발행연도
- 2026.4
- 수록면
- 97 - 103 (7page)
이용수
초록· 키워드
This study developed an artificial neural network model to predict the fracture load of dissimilar flow-drill screw (FDS) connections. Experimental tests (Lap shear, Cross tension test) were conducted for 11 selected material combinations under up to 6 loading angle conditions, and a total of 45 fracture load data points were used for model training. The K-fold cross validation method (K-fold CV) was applied to address the limitations of the small dataset, and this strategy served as the basis for parametric optimization with respect to activation functions, hidden layer depth, and the number of nodes per layer. The results showed that the ReLU and Leaky ReLU activation functions achieved lower average RMSE values compared to the Sigmoid function, and the optimal network configuration was identified to ensure stable and accurate predictions. It was confirmed that the predictions obtained from the model with the optimized network configuration exhibited good agreement with the experimental results across the investigated material combinations and loading angle conditions. The optimized artificial neural network model can reliably estimate the fracture load of FDS joints for preliminary design, reducing reliance on repetitive, time-consuming, and material-intensive experimental tests. Consequently, the proposed approach provides a practical and efficient alternative to conventional experimental evaluation.
상세정보 수정요청해당 페이지 내 제목·저자·목차·페이지정보가 잘못된 경우 알려주세요!
목차
- Abstract
- 1. 서론
- 2. ANN 학습을 위한 실험 설계
- 3. 인공신경망 모델 설계 및 성능 평가
- 4. 결론
- REFERENCES
참고문헌
참고문헌 신청최근 본 자료
