Prediction of the Compressive Strength of Vibrocentrifuged Concrete Using Machine Learning Methods
The determination of mechanical properties for different building materials is a highly relevant and practical field of application for machine learning (ML) techniques within the construction sector. When working with vibrocentrifuged concrete products and structures, it is crucial to consider fact...
Main Authors: | , , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2024-02-01
|
Series: | Buildings |
Subjects: | |
Online Access: | https://www.mdpi.com/2075-5309/14/2/377 |
_version_ | 1797298695420510208 |
---|---|
author | Alexey N. Beskopylny Sergey A. Stel’makh Evgenii M. Shcherban’ Levon R. Mailyan Besarion Meskhi Irina Razveeva Alexey Kozhakin Anton Pembek Diana Elshaeva Andrei Chernil’nik Nikita Beskopylny |
author_facet | Alexey N. Beskopylny Sergey A. Stel’makh Evgenii M. Shcherban’ Levon R. Mailyan Besarion Meskhi Irina Razveeva Alexey Kozhakin Anton Pembek Diana Elshaeva Andrei Chernil’nik Nikita Beskopylny |
author_sort | Alexey N. Beskopylny |
collection | DOAJ |
description | The determination of mechanical properties for different building materials is a highly relevant and practical field of application for machine learning (ML) techniques within the construction sector. When working with vibrocentrifuged concrete products and structures, it is crucial to consider factors related to the impact of aggressive environments. Artificial intelligence methods can enhance the prediction of vibrocentrifuged concrete properties through the use of specialized machine learning algorithms for materials’ strength determination. The aim of this article is to establish and evaluate machine learning algorithms, specifically Linear Regression (LR), Support Vector Regression (SVR), Random Forest (RF), CatBoost (CB), for the prediction of compressive strength in vibrocentrifuged concrete under diverse aggressive operational conditions. This is achieved by utilizing a comprehensive database of experimental values obtained in laboratory settings. The following metrics were used to analyze the accuracy of the constructed regression models: Mean Absolute Error (<i>MAE</i>), Mean Squared Error (<i>MSE</i>), Root-Mean-Square Error (<i>RMSE</i>), Mean Absolute Percentage Error (<i>MAPE</i>) and coefficient of determination (<i>R</i><sup>2</sup>). The average <i>MAPE</i> in the range from 2% (RF, CB) to 7% (LR, SVR) allowed us to draw conclusions about the possibility of using “smart” algorithms in the development of compositions and quality control of vibrocentrifuged concrete, which ultimately entails the improvement and acceleration of the construction and building materials manufacture. The best model, CatBoost, showed <i>MAE</i> = 0.89, <i>MSE</i> = 4.37, <i>RMSE</i> = 2.09, <i>MAPE</i> = 2% and <i>R</i><sup>2</sup> = 0.94. |
first_indexed | 2024-03-07T22:39:45Z |
format | Article |
id | doaj.art-6fa1b52c8c3448a29cdf48867496d3d6 |
institution | Directory Open Access Journal |
issn | 2075-5309 |
language | English |
last_indexed | 2024-03-07T22:39:45Z |
publishDate | 2024-02-01 |
publisher | MDPI AG |
record_format | Article |
series | Buildings |
spelling | doaj.art-6fa1b52c8c3448a29cdf48867496d3d62024-02-23T15:10:03ZengMDPI AGBuildings2075-53092024-02-0114237710.3390/buildings14020377Prediction of the Compressive Strength of Vibrocentrifuged Concrete Using Machine Learning MethodsAlexey N. Beskopylny0Sergey A. Stel’makh1Evgenii M. Shcherban’2Levon R. Mailyan3Besarion Meskhi4Irina Razveeva5Alexey Kozhakin6Anton Pembek7Diana Elshaeva8Andrei Chernil’nik9Nikita Beskopylny10Department of Transport Systems, Faculty of Roads and Transport Systems, Don State Technical University, 344003 Rostov-on-Don, RussiaDepartment of Unique Buildings and Constructions Engineering, Don State Technical University, 344003 Rostov-on-Don, RussiaDepartment of Engineering Geology, Bases, and Foundations, Don State Technical University, 344003 Rostov-on-Don, RussiaDepartment of Unique Buildings and Constructions Engineering, Don State Technical University, 344003 Rostov-on-Don, RussiaDepartment of Life Safety and Environmental Protection, Faculty of Life Safety and Environmental Engineering, Don State Technical University, 344003 Rostov-on-Don, RussiaDepartment of Unique Buildings and Constructions Engineering, Don State Technical University, 344003 Rostov-on-Don, RussiaDepartment of Unique Buildings and Constructions Engineering, Don State Technical University, 344003 Rostov-on-Don, RussiaChair of Quantum Statistics and Field Theory, Faculty of Physics, Lomonosov Moscow State University, Leninskiye Gory, 1, 119991 Moscow, RussiaDepartment of Unique Buildings and Constructions Engineering, Don State Technical University, 344003 Rostov-on-Don, RussiaDepartment of Unique Buildings and Constructions Engineering, Don State Technical University, 344003 Rostov-on-Don, RussiaDepartment Hardware and Software Engineering, Faculty of IT-Systems and Technology, Don State Technical University, 344003 Rostov-on-Don, RussiaThe determination of mechanical properties for different building materials is a highly relevant and practical field of application for machine learning (ML) techniques within the construction sector. When working with vibrocentrifuged concrete products and structures, it is crucial to consider factors related to the impact of aggressive environments. Artificial intelligence methods can enhance the prediction of vibrocentrifuged concrete properties through the use of specialized machine learning algorithms for materials’ strength determination. The aim of this article is to establish and evaluate machine learning algorithms, specifically Linear Regression (LR), Support Vector Regression (SVR), Random Forest (RF), CatBoost (CB), for the prediction of compressive strength in vibrocentrifuged concrete under diverse aggressive operational conditions. This is achieved by utilizing a comprehensive database of experimental values obtained in laboratory settings. The following metrics were used to analyze the accuracy of the constructed regression models: Mean Absolute Error (<i>MAE</i>), Mean Squared Error (<i>MSE</i>), Root-Mean-Square Error (<i>RMSE</i>), Mean Absolute Percentage Error (<i>MAPE</i>) and coefficient of determination (<i>R</i><sup>2</sup>). The average <i>MAPE</i> in the range from 2% (RF, CB) to 7% (LR, SVR) allowed us to draw conclusions about the possibility of using “smart” algorithms in the development of compositions and quality control of vibrocentrifuged concrete, which ultimately entails the improvement and acceleration of the construction and building materials manufacture. The best model, CatBoost, showed <i>MAE</i> = 0.89, <i>MSE</i> = 4.37, <i>RMSE</i> = 2.09, <i>MAPE</i> = 2% and <i>R</i><sup>2</sup> = 0.94.https://www.mdpi.com/2075-5309/14/2/377vibrocentrifuged concretecompressive strength predictionmachine learninglinear regressionsupport vector regressionrandom forest |
spellingShingle | Alexey N. Beskopylny Sergey A. Stel’makh Evgenii M. Shcherban’ Levon R. Mailyan Besarion Meskhi Irina Razveeva Alexey Kozhakin Anton Pembek Diana Elshaeva Andrei Chernil’nik Nikita Beskopylny Prediction of the Compressive Strength of Vibrocentrifuged Concrete Using Machine Learning Methods Buildings vibrocentrifuged concrete compressive strength prediction machine learning linear regression support vector regression random forest |
title | Prediction of the Compressive Strength of Vibrocentrifuged Concrete Using Machine Learning Methods |
title_full | Prediction of the Compressive Strength of Vibrocentrifuged Concrete Using Machine Learning Methods |
title_fullStr | Prediction of the Compressive Strength of Vibrocentrifuged Concrete Using Machine Learning Methods |
title_full_unstemmed | Prediction of the Compressive Strength of Vibrocentrifuged Concrete Using Machine Learning Methods |
title_short | Prediction of the Compressive Strength of Vibrocentrifuged Concrete Using Machine Learning Methods |
title_sort | prediction of the compressive strength of vibrocentrifuged concrete using machine learning methods |
topic | vibrocentrifuged concrete compressive strength prediction machine learning linear regression support vector regression random forest |
url | https://www.mdpi.com/2075-5309/14/2/377 |
work_keys_str_mv | AT alexeynbeskopylny predictionofthecompressivestrengthofvibrocentrifugedconcreteusingmachinelearningmethods AT sergeyastelmakh predictionofthecompressivestrengthofvibrocentrifugedconcreteusingmachinelearningmethods AT evgeniimshcherban predictionofthecompressivestrengthofvibrocentrifugedconcreteusingmachinelearningmethods AT levonrmailyan predictionofthecompressivestrengthofvibrocentrifugedconcreteusingmachinelearningmethods AT besarionmeskhi predictionofthecompressivestrengthofvibrocentrifugedconcreteusingmachinelearningmethods AT irinarazveeva predictionofthecompressivestrengthofvibrocentrifugedconcreteusingmachinelearningmethods AT alexeykozhakin predictionofthecompressivestrengthofvibrocentrifugedconcreteusingmachinelearningmethods AT antonpembek predictionofthecompressivestrengthofvibrocentrifugedconcreteusingmachinelearningmethods AT dianaelshaeva predictionofthecompressivestrengthofvibrocentrifugedconcreteusingmachinelearningmethods AT andreichernilnik predictionofthecompressivestrengthofvibrocentrifugedconcreteusingmachinelearningmethods AT nikitabeskopylny predictionofthecompressivestrengthofvibrocentrifugedconcreteusingmachinelearningmethods |