Title |
Prediction of Transfer Lengths in Pretensioned Concrete Members Using Neuro-Fuzzy System
|
Authors |
김민수(Minsu Kim) ; 한선진(Sun-Jin Han) ; 조해창(Hae-Chang Cho) ; 오재열(Jae-Yuel Oh) ; 김강수(Kang Su Kim) |
DOI |
https://doi.org/10.4334/JKCI.2016.28.6.723 |
Keywords |
ANFIS ; 퍼지 ; 전달길이 ; 뉴로-퍼지 ; 프리텐션 콘크리트부재 ANFIS ; fuzzy ; transfer length ; neuro-fuzzy ; pretensioned concrete member |
Abstract |
In pretensioned concrete members, a certain bond length from the end of the member is required to secure the effective prestress in the strands, which is defined as the transfer length. However, due to the complex bond mechanism between strands and concrete, most transfer length models based on the deterministic approach have uncertainties and do not provide accurate estimations. Therefore, in this study, Adaptive Neuro-Fuzzy Inference System (ANFIS), a Neuro-Fuzzy System, is introduced to reduce the uncertainties and to estimate the transfer length more accurately in pretensioned concrete member. A total of 253 transfer length test results have been collected from literatures to train ANFIS, and the trained ANFIS algorithm estimated the transfer length very accurately. In addition, a design equation was proposed to calculate the transfer length based on parametric studies and dimensional analyses. Consequently, the proposed equation provided accurate results on the transfer length which are comparable to the ANFIS analysis results.
|