Prediction of Compressive Strength of Concrete by Data-Driven Models

Khademi, Faezehossadat; Architectural, Graduate Student, Civil,; Akbari, Mahmoud; Jamal, Sayed Mohammadmehdi

doi:10.26634/jce.5.2.3350

Cited by 27 publications

(7 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Al-Abdaly et al 50 reported that MLR algorithm (with R 2 = 0.64, RMSE = 8.68, MAE = 5.66) performed poorly in predicting the CS behavior of SFRC. Khademi et al 51 used MLR to predict the CS of NC and found that it cannot be considered an accurate model (with R 2 = 0.518). Moreover, according to the results reported by Kang et al 18 , it was shown that using MLR led to a significant difference between actual and predicted values for prediction of SFRC’s CS (RMSE = 12.4273, MAE = 11.3765).…”

Section: Resultsmentioning

confidence: 99%

Comparison of various machine learning algorithms used for compressive strength prediction of steel fiber-reinforced concrete

Pakzad

Roshan

Ghalehnovi

2023

Sci Rep

View full text Add to dashboard Cite

Adding hooked industrial steel fibers (ISF) to concrete boosts its tensile and flexural strength. However, the understanding of ISF’s influence on the compressive strength (CS) behavior of concrete is still questioned by the scientific society. The presented paper aims to use machine learning (ML) and deep learning (DL) algorithms to predict the CS of steel fiber reinforced concrete (SFRC) incorporating hooked ISF based on the data collected from the open literature. Accordingly, 176 sets of data are collected from different journals and conference papers. Based upon the initial sensitivity analysis, the most influential parameters like water-to-cement (W/C) ratio and content of fine aggregates (FA) tend to decrease the CS of SFRC. Meanwhile, the CS of SFRC could be enhanced by increasing the amount of superplasticizer (SP), fly ash, and cement (C). The least contributing factors include the maximum size of aggregates (Dmax) and the length-to-diameter ratio of hooked ISFs (L/DISF). Several statistical parameters are also used as metrics to evaluate the performance of implemented models, such as coefficient of determination (R2), mean absolute error (MAE), and mean of squared error (MSE). Among different ML algorithms, convolutional neural network (CNN) with R2 = 0.928, RMSE = 5.043, and MAE = 3.833 shows higher accuracy. On the other hand, K-nearest neighbor (KNN) algorithm with R2 = 0.881, RMSE = 6.477, and MAE = 4.648 results in the weakest performance.

show abstract

Section: Resultsmentioning

confidence: 99%

Comparison of various machine learning algorithms used for compressive strength prediction of steel fiber-reinforced concrete

Pakzad

Roshan

Ghalehnovi

2023

Sci Rep

View full text Add to dashboard Cite

show abstract

“…In order to model the concrete samples, the finite element approximation is used [12]. We consider a quadrilateral element (Fig.…”

Section: Annexmentioning

confidence: 99%

“…Deshpande et al [11] presented the findings of a study which carried out for modeled compressive strength of concrete using the techniques of Artificial Neural Network (ANN), Model Tree (MT) and Non-linear Regression. The prediction of compressive strength of concrete by data-driven models was used by Khademi et al [12]. Nitka et Tejchman [13] studied the modelling of the behaviour of plain concrete during uniaxial compression and tension using the discrete element method DEM.…”

Section: Introductionmentioning

confidence: 99%

Experimental study and modeling of the mechanical behavior of recycled aggregates-based high-strength concrete

Nehar

Benamara

2021

Frattura Ed Integrità Strutturale

View full text Add to dashboard Cite

Concrete, a material of prime importance, is widely used in various works. Among the raw materials composing concrete, aggregates come first. It is widely acknowledged that the consumption of natural aggregates increases with the growth in the amount of concrete needed. It has recently emerged that concrete waste can be recycled and reused in civil engineering works after a series of treatments. Moreover, in order to protect the environment and based on the principles of sustainable development, it was considered urgent to produce a High Strength Concrete incorporating recycled materials in addition to silica fume and a high-efficiency superplasticizer. This would certainly help to establish a harmonious sustainable development that guaranties the ecological balance and environmental protection, and prevents the depletion of natural resources. This study is part of a larger research program that that seeks to recover, recycle and valorize construction and demolition wastes. The main objective sought in this article is firstly to use aggregates from demolition concrete in the manufacture of a new concrete with high mechanical and rheological performance, and secondly, to model the behavior of this type of concrete using the Finite Element Method. This modeling aims to evaluate the maximum compressive strengths and compare them with those obtained experimentally.

show abstract

“…These models can give further information for a better understanding of the material properties [18]. Among the prediction models, ANN provides more accurate predictions for concrete mechanical properties [19]. In recent years, mechanical properties and the complex behavior of the concrete have been analyzed with the aid and abilities of the artificial intelligence-based methods [20,21].…”

Section: Introductionmentioning

confidence: 99%

Prediction of Bending Strength of Self-Leveling Glass Fiber Reinforced Concrete

Yıldızel¹,

Uzun²,

Çöğürcü³

2019

ijisae

View full text Add to dashboard Cite

Many studies have been conducted on the prediction of fiber reinforced concrete strength; however, there are very rare data concerning the prediction of bending strength values of self-leveling glass fiber reinforced concrete. And there is no study for prediction of bending strength of self-leveling glass fiber reinforced concrete from mixture ingredients and slump values. In the present study, relationships between the bending strength and the mixture proportions are explored. An artificial neural network model (ANN) is designed with an extensive experimental data including 395 four-point bending tests, and input parameters as white cement amount, maximum aggregate size, glass fiber content, water cement ratio, superplasticizer and metakaolin content and slump test results. Effect of each parameter on the bending strength is investigated with the developed model. An empirical and user-friendly formula was obtained with the generalization capabilities of the ANN. Results showed that the prediction results are in good agreement with the field data. And these numerical results with high efficiency can make it possible to use the neural design for real-life self-leveling glass fiber reinforced concrete applications.

show abstract

Prediction of Compressive Strength of Concrete by Data-Driven Models

Cited by 27 publications

References 22 publications

Comparison of various machine learning algorithms used for compressive strength prediction of steel fiber-reinforced concrete

Comparison of various machine learning algorithms used for compressive strength prediction of steel fiber-reinforced concrete

Experimental study and modeling of the mechanical behavior of recycled aggregates-based high-strength concrete

Prediction of Bending Strength of Self-Leveling Glass Fiber Reinforced Concrete

Contact Info

Product

Resources

About