뉴랄-네트워크를 이용한 초고층 건물의 부등축소량 예측을 위한 알고리즘 개발
양원직, 김욱종, 이도범 et al. · 대한건축학회 학술발표대회 논문집 - 구조계 · 2006
The objective of this study was to develop and evaluate a Neural Network algorithm to predict the inelastic shortening of reinforced concrete members using the column shortening measuring data of a high-rise building. New training algorithms for the prediction of column shortening focuses on data processing and training methods and their validity is examined through training and predictions based on column shortening measuring data of high-rise buildings. In a Neural Network algorithm, the data to be trained uses a polynomial fit line instead of measuring data. The results show that the new training algorithms proposed in this study successfully predicted column shortening of high-rise buildings.
Cited by 0 works.