Prediction on the mechanical strength of coal ash concrete using artificial neural network

Muhammad Nor Syahrul, Zaimi and Nur Farhayu, Ariffin and Sharifah Maszura, Syed Mohsin and Abdul Muiz, Hasim and Nurul Natasha, Nasrudin (2022) Prediction on the mechanical strength of coal ash concrete using artificial neural network. In: IET Conference Proceedings. 2022 Engineering Technology International Conference, ETIC 2022 , 7 - 8 September 2022 , Kuantan, Virtual. pp. 517-524., 2022 (22). ISSN 2732-4494 ISBN 978-183953782-0 (Published)

	Pdf Prediction on the mechanical strength of coal ash concrete.pdf Restricted to Repository staff only Download (1MB) | Request a copy
Preview	Pdf Prediction on the mechanical strength of coal ash concrete using artificial neural network_ABS.pdf Download (61kB) | Preview

Abstract

Machine learning approaches are essential for assessing the mechanical strength of concrete in civil engineering. With little work and expenditure, machine learning algorithms provide remarkable accuracy. However, these methods need information on the proportions of various components used including water, cement, aggregate, etc. This study uses a dataset that contains information on the composition of 105 distinct types of materials. The collection contains both conventional and cutting-edge materials that include fly ash (FA) and coal bottom ash (CBA) in addition to other essential components of concrete mix. Since CBA and FA are waste and by-products materials, adding it to the concrete mix helps create concrete that is environmentally beneficial. The prediction of concrete compressive strength is therefore made more difficult by the addition of more elements to the concrete. To maintain the safety of building projects, it is crucial to anticipate the compressive strength (CS) of concrete containing coal ash. This study presents the most accurate methods of the exact prediction in term of compressive and flexural strengths of concrete containing CBA and FA. The prediction incorporates the decision tree, linear regression (LR), and random forest through soft voting. This type of analysis involves one or more independent variables that may most accurately predict the value of the dependent variable and calculates the coefficients of the linear equation. The difference between the output values anticipated and those obtained is minimised by linear regression by fitting a straight line or surface. Model performance was evaluated using several well-known metrics, including R2, mean square error (MSE), mean absolute error (MAE), and root mean square error (RMSE) (R-square). With scores of 4.46 and 2.51, respectively, the random forest model beat the most sophisticated models, according to the findings. Random forest outperforms LR and decision tree in terms of computation efficiency. The application of 50 percent CBA into the concrete give an increasing strength compare to the other replacement in concrete along with duration of time. The mechanical strength prediction based on machine learning is more exact, accurate, and dependable than the standard concrete strength estimate.

Item Type:	Conference or Workshop Item (Lecture)
Additional Information:	Indexed by Scopus
Uncontrolled Keywords:	Coal bottom ash; Jupyter notebook; Machine learning; Mechanical strength; Regression model
Subjects:	T Technology > T Technology (General) T Technology > TA Engineering (General). Civil engineering (General)
Faculty/Division:	Faculty of Civil Engineering Technology
Depositing User:	Mr Muhamad Firdaus Janih@Jaini
Date Deposited:	30 Sep 2024 04:37
Last Modified:	30 Sep 2024 04:37
URI:	http://umpir.ump.edu.my/id/eprint/42017
Download Statistic:	View Download Statistics

Actions (login required)

View Item