A MACHINE LEARNING BASED APPROACH TO ESTIMATENONLINEAR DYNAMIC RESPONSE OF REINFORCEDCONCRETE MOMENT-RESISTING FRAMES
Publish Year: 1403
نوع سند: مقاله کنفرانسی
زبان: English
View: 54
This Paper With 8 Page And PDF Format Ready To Download
- Certificate
- من نویسنده این مقاله هستم
این Paper در بخشهای موضوعی زیر دسته بندی شده است:
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
SEE09_130
تاریخ نمایه سازی: 10 آبان 1403
Abstract:
This study advances structural engineering by demonstrating the efficiency of machine learning inreducing computational costs for time history analysis while ensuring reliable predictions. The goal isto provide engineers with a dependable tool for designing structures under dynamic loads. Machinelearning models, trained on a vast dataset of ۲۱۵,۰۰۰ reinforced concrete frames and earthquakerecords, excel in predicting base shear ratios with a ۹۰% coefficient of determination. Despite a slightreduction in accuracy when predicting maximum roof drift ratios, the models still exhibitcommendable predictive capability with coefficients of determination exceeding ۸۰%. The applicationon a seven-story building's collapse reliability index shows the trained machine with a ۷% error rate,performing ۱۳۰,۰۰۰ times faster than traditional time history analysis. This highlights the machine'sremarkable efficiency and potential for widespread application in structural engineering
Keywords:
Authors
Mehrdad Ghafarian Tamizi
M.Sc. Graduate, Civil and Environmental Engineering Department, Amirkabir University ofTechnology, Tehran, Iran,
Mehdi Banazadeh
Associate Professor, Civil and Environmental Engineering Department, Amirkabir University of Technology,Tehran, Iran,