PERFORMANCE EVALUATION OF DIFFERENT MACHINE LEARNING MODELS FOR COMPRESSIVE STRENGTH PREDICTION OF LATERIZED PORTLAND CEMENT-ACTIVATED POZZOLANIC CONCRETE
Abstract
Activated pozzolans/ geopolymer concrete is no doubt one of the major attempts to make the world liveable and ensure sustainable construction through the reduction of carbon footprint. A correct prediction of the compressive strength is important for the acceptability and use of such concrete within a complex system. In this study, four machine learning models consisting of support vector regression (SVR), artificial neural network (ANN), extreme gradient boosting (XGBoost), and random forest (RF), were employed to predict the compressive strength of hybrid laterized ordinary Portland cement-activated rice husk ash (OPC-ARha) concrete using eight input variables. Data from 192 concrete specimens were used for training and testing in the ratio of 153:39. The R2 values for SVR, ANN, XGBoost, and RF were 0.7172, 0.9496, 0.9200, and 0.9239 respectively, making ANN the model with the highest absolute fraction of variance (R2). Also from the result of the root mean squared error (RMSE), ANN proved better than the other models. However, the least mean absolute error (MAE) was obtained using RF. Feature importance of the eight variables used in the strength prediction showed that curing age is the most important factor followed by Portland cement content and water/binder ratio.
Keywords: Strength prediction, Machine learning, Activated rice husk ash, Laterized concrete, Concrete strength
Full Text:
PDFReferences
Awoyera, P.O; Kirgiz, M.S; Viloria, A and Ovallos-Gazabong D. (2020).Estimating Strength Properties of Geopolymer Self-compacting Concrete Using Machine Learning Techniques. Journal of Material Research and Technology, 9(4):9016–9028.
ASTM C618-19. Standard Specification for Coal Fly Ash and Raw or Calcinated Natural Pozzolan for Use in Concrete. ASTM International, PA.
ASTM C136/C136M-19. (2019). Standard Test Method for Sieve Analysis of Coarse and Fine Aggregates.
ASTM International, PA. 2019.
BS EN 196-1. (2005). Method of Testing Cement. British Standards Institution, London. 2005
BS EN 197-1:2000. (2000). Cement, Composition, Specification and Conformity Criteria for Common Cement. British Standards Institution, London. 2000.
BS 882:1992.(1992).Specification for Aggregates from Natural Sources for Concrete. British Standards Institution, London. 1992. W4 4AL.
Burns, E. (2022). An In-depth Guide to Machine Learning in the Enterprise.Available online at: www.techtarget.com. Assessed on August 26, 2022.
Denil, M ; Matheson, D. and Freitas, N.(2014).Narrowing the Gap: Random Forests in Theory and in Practice.Proceedings of the 31 st International Conference on Machine Learning, Beijing, China. 2014.
de-Prado-Gil, J; Palencia, C; Jagadesh, P; Martínez-García, R.A.(2022).Comparison of Machine Learning Tools That Model the Splitting Tensile Strength of Self-compacting Recycled Aggregate Concrete. Materials, 15(12):1-20.
Elhishi, S; Elashry, A.M; El-Metwally, S. (2023). Unboxing Machine Learning Models for Concrete Strength Prediction Using XAI. Scientific Reports, 13:19892.
Friedman, J. H. (2001). Greedy Function Approximation: A Gradient Boosting Machine. The Annals of Statistics, 29(5): 1189-1232.
Ghosh, A. and Ransinchung, R. N. (2022). Application of Machine Learning Algorithm to Assess the Efficacy of Varying Industrial Wastes and Curing Methods on Strength Development of Geopolymer Concrete. Construction and Building Materials, 341:127828
Ghorbani, B; Arulrajah, A; Narsilio, G; Horpibulsuk, S and Bo, M.W. (2021).Dynamic Characterization of Recycled Glass-Recycled Concrete Blends Using Experimental Analysis and Artificial Neural Network Modeling. Soil Dynamics and Earthquake Engineering, 2021;142.
Golafshani, E. M. and Kashani, A. (2022).Modeling the Compressive Strength of Concrete Containing Waste Glass Using Multi-Objective Automatic Regression.Neural Computing and Applications, 34(19):17107-17127.
Hashempour, M; Heidari, A and Shahi jounaghani, M.(2020). The Efficiency of Hybrid BNN-DWT for Predicting the Construction and Demolition Waste Concrete Strength. International Journal of Engineering Transactions B: Applications, 33(8): 1544-1552.
Imran, H; Al-Abdaly, N. M; Shamsa, M. H; Shatnawi, A; Ibrahim, M and Ostrowski, K. A.(2022). Development of Prediction Model to Predict the Compressive Strength of Eco-Friendly Concrete Using Multivariate Polynomial Regression Combined With Stepwise Method. Materials,15: 317.
Mai, H.V.T; Nguyen, T. A; Ly, H. B and Tran, V. Q.(2021). Prediction Compressive Strength of Concrete Containing GGBFS Using Random Forest Model. Advances in Civil Engineering.
Muliauwan, H. N; Prayogo, D; Gaby, G and Harsono, K. (2020).Prediction of Concrete Compressive Strength Using Artificial Intelligence Methods. IOP Conference Series 1625;Journal of Physics.
Pan, X; Xiao, Y; Suhail, S.A; Ahmad, W; Murali, G; Salmi, A and Mohamed, A. (2022). Use of Artificial Intelligence Methods for Predicting the Strength of Recycled Aggregate Concrete and The Influence of Raw Ingredients.Materials, 15: 4194.
Prayogo, D; Santoso, D.I; Wijaya, D; Gunawan, T and Widjaja JA. (2020).Prediction of Concrete Properties Using Ensemble MachineLearning Methods. IOP Conference Series 1625;Journal of Physics.
Li, Y; Zou, C; Berecibar, M; Naninimaury, E; Chan, J.C; Den Bossche, P.V; Van Mierlo, J and Omar N. (2018). Random Forest Regression for Online Capacity Estimation of Lithium-ion Batteries. Applied Energy, 232: 197–210.
Sagar, B. and Sivakumar, M.V.N. (2020).An Experimental and Analytical Study on Alccofine Based High Strength Concrete. International Journal of Engineering, Transactions A: Basics, 33(4):530-538.
Salimbahrami, S.R. and Shakeri, R. (2021). Experimental Investigation and Comparative Machine-Learning Prediction of Compressive Strength of Recycled Aggregate Concrete. Soft Computing, 25: 919–932
Shang M, Li H, Ahmad A, AhmadW, Ostrowski KA, Aslam F, Joyklad P, Majka TM. (2022) . Predicting the Mechanical Properties Of Rca-Based Concrete Using Supervised Machine Learning Algorithms. Materials,15: 647.
Shaqadan, A. (2016). Prediction of Concrete Mix Strength Using Random Forest Model. International Journal of Applied Engineering Research, 11(22): 11024-11029.
Singh, A. (2023).Support Vector Regression Model: A Regression-Based Machine Learning Approach. Available online at:https://medium.com/analytics-vidhya. Accessed on August 02, 2023.
Thang, V.L.; Cung, L. and Nguyen, D.S. (2022). An application of artificial neural network to predict the compressive strength of concrete using fly ash and stone powder waste products in central vietnam. International Journal of Engineering, Transactions B: Applications, 35(05):967-976.
Xu, J; Zhou, L; He, G; Ji, X; Dai, Y and Dang, Y. (2021). Comprehensive Machine Learning-based Model for Predicting Compressive Strength of Ready-mix Concrete. Materials,14(5):1068-86
Refbacks
- There are currently no refbacks.
Copyright © 2022-2025. Department of Civil Engineering, Nnamdi Azikiwe University. All Rights Reserved..