مقاله انگلیسی رایگان در مورد پیش بینی مقاومت فشاری بتن بازیافتی – الزویر ۲۰۱۸

مقاله انگلیسی رایگان در مورد پیش بینی مقاومت فشاری بتن بازیافتی – الزویر ۲۰۱۸

 

مشخصات مقاله
انتشار مقاله سال ۲۰۱۸
تعداد صفحات مقاله انگلیسی ۸ صفحه
هزینه دانلود مقاله انگلیسی رایگان میباشد.
منتشر شده در نشریه الزویر
نوع نگارش مقاله مقاله پژوهشی (Research article)
نوع مقاله ISI
عنوان انگلیسی مقاله Compressive strength prediction of recycled concrete based on deep learning
ترجمه عنوان مقاله پیش بینی مقاومت فشاری بتن بازیافتی بر اساس یادگیری عمیق
فرمت مقاله انگلیسی  PDF
رشته های مرتبط مهندسی عمران، مهندسی کامپیوتر، فناوری اطلاعات
گرایش های مرتبط سازه، مدیریت ساخت، هوش مصنوعی، شبکه های کامپیوتری
مجله ساخت و ساز و مصالح ساختمانی – Construction and Building Materials
دانشگاه East China Jiaotong University – Nanchang – China
کلمات کلیدی بتن بازیافت شده، قدرت فشاری، مدل پیش بینی، یادگیری عمیق، شبکه عصبی کانولوشن
کلمات کلیدی انگلیسی Recycled concrete, Compressive strength, Prediction model, Deep learning, Convolution neural network
شناسه دیجیتال – doi
https://doi.org/10.1016/j.conbuildmat.2018.04.169
کد محصول E8662
وضعیت ترجمه مقاله  ترجمه آماده این مقاله موجود نمیباشد. میتوانید از طریق دکمه پایین سفارش دهید.
دانلود رایگان مقاله دانلود رایگان مقاله انگلیسی
سفارش ترجمه این مقاله سفارش ترجمه این مقاله

 

بخشی از متن مقاله:
۱٫ Introduction

Along with the rapid urban development and economic activities, the generation of construction and demolition (C&D) waste has increased substantially in many parts of the world. At the same time, large quantities of natural aggregates are extracted for construction every year [1]. The utilization of recycled aggregates (RAs) in concrete production can potentially conserve the nonrenewable natural resource of natural aggregates, eliminate unnecessary consumption of limited landfill areas and reduce energy consumption. Due to its benefits on preventing the shortage of natural aggregate and the deterioration of ecological environment caused by concrete waste, Recycled Aggregate Concrete (RAC) technology is considered as one of the main candidates for ecological concrete development [2]. However, the variability in the characteristics of RA and RAC prevents the use of RA further. For example, the use of RCA can lead to reduction of up to 40% in compressive strength [1,3]. Low density and high water absorption and porosity, mainly caused by the heterogeneous nature of RA, can influence the properties of fresh concrete and then reduce its workability [4–۶]. Over the last two decades, many investigators have made use of various methods to predict the properties of concrete with different components. The compressive strength of recycled concrete is closely related to these factors such as sand rate, watercement ratio, aggregate grade, aggregate type and substitution rate, mineral fine admixture variety and dosage [7,8]. However, the relationship between those factors and compressive strength shows a complex non-linear relationship, and there is still no definite theoretical formula which can accurately reflect their relationships [9,10]. In practice, substantial experiments have to be carried out to ensure the compressive strength of recycled concrete to meet the requirements. Nowadays various artificial intelligence algorithms, such as neural networks (NN) and support vector machines (SVM), are widely used in concrete strength prediction. In [11–۱۳], the artificial neural network serves as to predict the relationship between the different influencing factors and compressive strength of recycled concrete. The nonlinear mapping ability of BP (Back Propagation) neural network is adopted to establish the non-linear model between input variables and output variables. Thus accurate intensity prediction could be realized through a certain training and iteration. A 7-20-3 BP neural network model is employed in [14] to predict the recycled concrete slump. In [15], the neural network model and ultrasonic pulse velocity test are proposed to predict the concrete compressive strength. Although BP neural network shows good abilities on solving non-linear problems, it also exhibits some disadvantages including slow convergence, over learning and local optimization which will affect the accuracy and efficiency of prediction. In [16] the neural networks and the adaptive neuro-fuzzy inference system are combined to improve the capability of prediction model. In [17], artificial neural networks and regression techniques are used to analyses the relations between concrete components and concrete properties.

برچسب ها

ثبت دیدگاه