Machine Learning-Aided Prediction of Post-Fire Shear Resistance Reduction of Q690 HSS Plate Girders
نویسندگان
چکیده
Fire has significant effects on the residual resistance of steel structures. It is necessary to accurately clarify its Q690 HSS plate girders, which have been widely used. In this paper, ultimate and effective service girders after a fire are obtained using material tests finite element (FE) analysis including parametric studies, where data 210 models were collected. The four key parameters (hw/tw ratio, a/hw exposure temperature cooling method) post-fire shear reduction roughly investigated by individual conditional expectation (ICE), showing most important factor. popular algorithms machine learning (ML), namely artificial neural network (ANN) support vector regression (SVR) algorithms, utilized in model training predict factors both resistance. Finally, results indicate that prediction ML shows much better performance than with traditional ordinary least squares (OLS) regression, SVR genetic algorithm (GA) provides highest accuracy. paper show superiority for solving problems structures, compared conventional methods such as linear regression.
منابع مشابه
A Proposed Method to Evaluate Ultimate Resistance of Plate Girders Subjected to Shear and Patch Loading (RESEARCH NOTE)
Experimental investigation of the ultimate resistance of slender, steel-plate girder web panels to combined shear-and-patch loading indicates significant interaction between shear loading and patch loading. However, an existing interaction formula is based on experimental results. Herein, an improved design procedure for slender plate girders subjected to combined shear and patch loading is pro...
متن کاملmodeling the patch load resistance of plate girders using a support vector machine
the support vector machine (svm) is a relatively new machine learning method which is increasingly being applied to engineering problems and have yielded encouraging results. because of complex behavior of elastoplastic of web panels of plate girders under patch loading, almost none of the proposed methods provides consistent and accurate predictions of patch load capacity. consequently, altern...
متن کاملSafety Analysis of the Patch Load Resistance of Plate Girders: Influence of Model Error and Variability
This study aims to undertake a statistical study to evaluate the accuracy of nine models that have been previously proposed for estimating the ultimate resistance of plate girders subjected to patch loading. For each model, mean errors and standard errors, as well as the probability of underestimating or overestimating patch load resistance, are estimated and the resultant values are compared o...
متن کاملModel Studies on Plate Girders
Tests on eighteen small scale models which simulate the elastic and post-buckling behaviour of plate girders when subjected to shear loading are reported and discussed. Themodels were fabricated of steel and Araldite; the major aim was to assess whether small scale models can be employed to study shear buckling problems. A secondary object was to examine whether araldite could be used for predi...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Buildings
سال: 2022
ISSN: ['2075-5309']
DOI: https://doi.org/10.3390/buildings12091481