Predictive modelling of foam glass performance properties using linear regression-based machine learning models

Sergey Fedosov; Maksim Bakanov; Irina Grushko

doi:10.52957/2782-1919-2024-5-2-20-38

UDC

691.3 004.852

PREDICTIVE MODELLING OF FOAM GLASS PERFORMANCE PROPERTIES USING LINEAR REGRESSION-BASED MACHINE LEARNING MODELS Predictive modelling of foam glass performance properties using linear regression-based machine learning models

АА Аа

Published in Smart composite in construction · Pages 20–38 · Rubric: Construction materials and products

DOI: https://doi.org/10.52957/2782-1919-2024-5-2-20-38

Received: 02.03.2024 Accepted: 10.06.2024 Published: 23.06.2024

Authors

Sergey Fedosov

Maksim Bakanov

Irina Grushko

Sergey Fedosov

Academic degree: doctor of technical sciences

Science rang: academician Russian Academy of Architecture and Construction Sciences ; professor

Moscow State University of Civil Engineering Professor · kafedra tehnologiy i organizacii stroitel'nogo proizvodstva

Maksim Bakanov

Academic degree: doctor of technical sciences

Science rang: docent

Moscow State University of Civil Engineering

Irina Grushko

Platov South-Russian State Polytechnic University

ABSTRACT

Building construction requires the use of efficient thermal insulation materials such as foam glass in view of energy conservation. The paper considers predictive modelling of the performance properties of foam glass using machine learning models. The paper presents a mathematical description of the additives impact in the charge on the properties of foam glass. Nine charge compositions for foam glass synthesis were developed and their main microstructure parameters were determined. The authors tested the regression models using the Jupyter Notebook software environment and the SciKit-Learn library in the Python programming language. The paper analyses the regression equation coefficients and estimates the modelling error. The obtained results confirm the effectiveness of predictive modelling of foam glass performance properties on the basis of linear regression.

KEYWORDS

foam glass, microstructure, functional properties, machine learning, regression analysis

1. Sejdinović, B. (2023) Modern Thermal Insulation and Sound Insulation Materials, Lecture Notes in Networks and Systems, (539), pp. 218-233. https://doi.org/10.1007/978-3-031-17697-5_19.

2. Davraz, M., Koru, M., Akdağ, A.E., Kılınçarslan, Ş. & Delikanlı, Y.E. et al. (2022) An investigation of foaming additives and usage rates in the production of ultra-light foam glass, J. Therm. Analys. Calorimetr., (147), pp. 3567-3576. https://doi.org/10.1007/s10973-021-10781-8.

3. Bashiri, A., Amirhosseini, A., Mirkazemi, S.M. & Ghanbari, H. (2021) Effect of Temperature and Water Glass Addition on the Microstructure and Physical Properties of Soda–Lime Foam Glass, Glass Phys. Chem., (47), pp. 83-90. https://doi.org/10.1134/S1087659621020024.

4. Semukhin, B.S., Votinov, A.V. & Kazmina, O.V. (2020) Properties of Foamglass with Fullerene-like Mesostructure, Rus. Phys J., 63 (4), pp. 710-712 (in Russian).

5. Latyntseva, E. A., Podoynikova, Ya. R., Bezrukova, T.A. & Murtazina, A.A. (2020) Influence of raw materials on the properties of foam glass and development prospects, Constr. Mat. Prod., 3 (1), pp. 44-48 (in Russian).

6. Sorokin, D.S., Beregovoy, V.A. & Kapustin, A.E. (2019) Porous granular materials based on natural silicites // Engineering Journal of the Don, 2(53), p. 44 (in Russian).

7. Zhimalov, A.A., Nikishonkova, O.A., Spiridonov, Yu.A., Kosobudskii, I.D. & Vikulova, M.A. (2019) Physical-Chemical Studies of Gaizes as Alternative Raw Materials for the Production of Foam Glass and Foam Materials, Glass and Ceramics, (75), pp. 387-390. https://doi.org/10.1007/s10717-019-00091-9.

8. Liu, H., Tang, M., Wang, Z., Liu, W., Ma, Y. et al. (2022) Optimized mechanical properties and thermal insulation capacity of foam glass through K2Ti6O13 whiskers addition, J. Austral. Ceramic Soc., (58), pp. 1241 1248. https://doi.org/10.1007/s41779-022-00761-y.

9. Vedyakov, I., Vaskalov, V., Maliavski, N., Nezhikov, A. & Vedyakov, M. (2023) Granular Foam-Glass-Ceramic Thermal Insulation Based on Natural Quartz Sand, Lect. Not. Civ. Eng., (282), pp. 395-405. https://doi.org/10.1007/978-3-031-10853-2_37.

10. Fedosov, S.V. & Bakanov, M.O. "Theoretical and applied principles of high-temperature heat treatment processes in the production of thermal insulating foam glass" // Ustojchivoe razvitie regiona: arhitektura, stroitel'stvo, transport [Sustainable development of the region: architecture, construction, transport]. Mat. VII Mezhd. nauch.-prakt. konf. Tambov [Mat. VII Int. Sci. Pract. Conf. Tambov], 2020, pp. 40-43 (in Russian).

11. GOST 7076-99 Construction materials and products. Method for determining thermal conductivity and thermal resistance under stationary thermal conditions. M.: State Unitary Enterprise TsPPb, 2000 (in Russian).

12. GOST EN 1602-2011 Thermal insulation products used in construction. Method for determining apparent density. M.: Standartinform, 2012 (in Russian).

13. GOST 33949-2016 Heat-insulating foam glass products for buildings and structures. M.: Standartinform, 2019 (in Russian).

14. GOST 17177-94 Heat-insulating construction materials and products. Test methods. M.: IPK Publ. House of Standards, 2002 (in Russian).

15. GOST EN 1607-2011 Thermal insulation products used in construction. Method for determining tensile strength perpendicular to face surfaces. M.: Standartinform, 2012 (in Russian).

16. GOST EN 12430-2011 Thermal insulation products used in construction. Method for determining strength under concentrated load. M.: Standartinform, 2012 (in Russian).

17. Fedosov, S.V. & Bakanov, M.O. (2017) Modelling of Temperature Field Distribution of the Foam Glass Batch in Terms of Thermal Treatment of Foam Glass, Int. J. Comput. Civ. Struct Eng., 13(3), pp. 112-118.

18. Fedosov, S.V., Bakanov, M.O. & Domnina, K.L. (2020) Mathematical modeling of technological processes for producing heat-insulating cellular composites, News of the Kyrgyz. State Techn. Un-t im. I. Razzakova, 3(55), pp. 207-213 (in Russian).

19. Gutierrez, D.D. Inside BIG DATA. Guide to Predictive Analytics: TIBCO Spotfire Business Intelligence Platform. 2017. Available at: http://www.spotfiretibco.ru/wpcontent/uploads/2017/09/InsideBIGDATA.pdf. (in Russian) (accessed 12.02.2024).

20. Omar, N.S., Hatem, W.A. & Najy, H.I. (2019) Predictive modeling for developing maintenance management in construction projects, Civ. Eng. J., 5(4), pp. 892-900.

21. Moein, M.M., Saradar, A., Rakhmati K., Musavinedzhad S.Kh.G. & Bristow J. et al. (2023) Predictive models for concrete properties using machine learning and deep learning approaches: rev., J. Build. Eng., (63), p. 105444.

22. Mater, Ya., Kamel M., Karam A. & Bakhum E. (2023) ANN-Python prediction model for the compressive strength of green concrete, Constr. Innovation, 23(2), pp. 340-359.

23. Amin, M.N., Ivtikhar, B., Khan K., Javed M.F. & AbuArab A.M. et al. (2023) Prediction model for rice husk ash concrete using AI approach: Boosting and bagging algorithms, Structures, (50), pp. 745-757. https://doi.org/10.1016/j.istruc.2023.02.080.

24. Nazar, S., Tszyn Ya., Amin, M.N., Khan K. & Ashraf M. et al. (2023) Machine learning interpretable-prediction models to evaluate the slump and strength of fly ash-based geopolymer, J. Mat. Res. Techn., (24), pp. 100-124. https://doi.org/10.1016/j.jmrt.2023.02.180.

25. Fransson, E., Eriksson, F. & Erhart, P. (2020) Efficient construction of linear models in materials modeling and applications to force constant expansions, Npj Comput. Mater., (6), p. 135. https://doi.org/10.1038/s41524-020-00404-5.

26. Chore, H.S. & Shelke, N.L. (2013) Prediction of compressive strength of concrete using multiple regression model, Struct. Eng. Mechan., 45(6), pp. 837-851. https://doi.org/10.12989/SEM.2013.45.6.837.

27. Obianyo, I.I., Anosike-Francis, E.N., Ihekweme, G.O., Geng, Ya. & Jin, R. et al. (2020) Multivariate regression models for predicting the compressive strength of bone ash stabilized lateritic soil for sustainable building, Constr. Build. Mat., (263), p. 120677. https://doi.org/10.1016/j.conbuildmat.2020.120677

28. Jin, R., Chen, Q. & Soboyejo, A.B.O. (2018) Non-linear and mixed regression models in predicting sustainable concrete strength, Constr. Build. Mat., (170), pp. 142-152. https://doi.org/10.1016/j.conbuildmat.2018.03.063.

29. Matveev, M.A., Matveev, G.M. & Frenkel, B.N. Calculations on chemistry and glass technology: Ref. man. M.: Constr. Liter. Publ. House, 1972 (in Russian).

30. Pedregosa, F., Varoquaux G., Gramfort A., Mishel V. & Bertrand T. et al. (2011) Scikit-learn: Machine Learning in Python, J. Mach. Learn. Res., (12), pp. 2825-2830.

GOST

Fedosov S. V., Bakanov M. O., Grushko I. S. Predictive modelling of foam glass performance properties using linear regression-based machine learning models // . 2024. pp. 20-38. DOI: https://doi.org/10.52957/2782-1919-2024-5-2-20-38. (Date of access 19.08.2026).

MLA

Sergey Fedosov, Bakanov M. O., Grushko I. S.. "Predictive modelling of foam glass performance properties using linear regression-based machine learning models." (2024): 20-38. DOI: https://doi.org/10.52957/2782-1919-2024-5-2-20-38.

APA

Fedosov S. V., Bakanov M. O., Grushko I. S. (2024). Predictive modelling of foam glass performance properties using linear regression-based machine learning models. , 20-38. https://doi.org/10.52957/2782-1919-2024-5-2-20-38

Chicago

Sergey Fedosov, Bakanov M. O., Grushko I. S.. "Predictive modelling of foam glass performance properties using linear regression-based machine learning models." , no. (2024): 20-38. https://doi.org/10.52957/2782-1919-2024-5-2-20-38.

WIKI

{{cite journal | author = Fedosov S. V., Bakanov M. O., Grushko I. S. | title = Predictive modelling of foam glass performance properties using linear regression-based machine learning models | journal = | year = 2024 | pages = 20-38 | doi = 10.52957/2782-1919-2024-5-2-20-38 }}

Copy

Export: BibTeX EndNote RefMan RusMARC

PREDICTIVE MODELLING OF FOAM GLASS PERFORMANCE PROPERTIES USING LINEAR REGRESSION-BASED MACHINE LEARNING MODELS Predictive modelling of foam glass performance properties using linear regression-based machine learning models

Confirmation