논문 상세보기

필로티 건축물의 인공지능 기반 내진성능 평가를 위한 데이터 기반 부재의 단면 형상비 연구 KCI 등재

Effectiveness of Data-Driven Section Shape Ratios for Seismic Performance- Based Artificial Intelligence of Piloti-Type Buildings

  • 언어KOR
  • URLhttps://db.koreascholar.com/Article/Detail/438371
구독 기관 인증 시 무료 이용이 가능합니다. 4,000원
한국지진공학회 (Earthquake Engineering Society of Korea)
초록

Structures compromised by a seismic event may be susceptible to aftershocks or subsequent occurrences within a particular duration. Considering that the shape ratios of sections, such as column shape ratio (CSR) and wall shape ratio (WSR), significantly influence the behavior of reinforced concrete (RC) piloti structures, it is essential to determine the best appropriate methodology for these structures. The seismic evaluation of piloti structures was conducted to measure seismic performance based on section shape ratios and inter-story drift ratio (IDR) standards. The diverse machine-learning models were trained and evaluated using the dataset, and the optimal model was chosen based on the performance of each model. The optimal model was employed to predict seismic performance by adjusting section shape ratios and output parameters, and a recommended approach for section shape ratios was presented. The optimal section shape ratios for the CSR range from 1.0 to 1.5, while the WSR spans from 1.5 to 3.33, regardless of the inter-story drift ratios.

목차
A B S T R A C T
1. 서 론
2. 피해 예측을 위한 내진성능 평가
    2.1 문헌 검토
    2.2 반응 및 손상 요구 예측을 위한 내진 성능 평가
3. 내진성능 및 손상 요구예측을 위한 인공지능 기법개발
    3.1 입출력 데이터세트 구성
    3.2 인공지능 방법론
    3.3 서포트 벡터 머신 (SVM) 모델
    3.4 적응형 신경 퍼지 추론 시스템(ANFIS) 모델
4. IDR을 기반으로 한 단면 형상비의 내진성능
5. 결 론
감사의 글
REFERENCES
저자
  • 이가윤(세종대학교 건축공학과 딥러닝 건축연구소, 박사후연구원) | Lee Gayoon (Post-Doctoral Student, Deep Learning Architecture Research Center, Department of Architectural Engineering, Sejong University)
  • 토바오웍(세종대학교 건축공학과 딥러닝 건축연구소, 박사후연구원) | Quoc Bao To (Post-Doctoral Student, Deep Learning Architecture Research Center, Department of Architectural Engineering, Sejong University)
  • 조혜림(세종대학교 건축공학과 딥러닝 건축연구소, 석사과정) | Jo Hye-rim (Student, Deep Learning Architecture Research Center, Department of Architectural Engineering, Sejong University)
  • 신지욱(경상국립대학교 건축공학과 부교수) | Shin Jiuk (Professor, Department of Architecture, Gyeongsang National University)
  • 이기학(세종대학교 건축 공학과 딥러닝 건축연구소, 건축공학과 교수) | Kihak Lee (Professor, Deep Learning Architecture Research Center, Department of Architectural Engineering, Sejong University) Corresponding author