AI Approaches for Geopolymer Composite Strengths Calculation

Geopolymers have come to light as a potential alternative to conventional cement in order to address issues related to cement production and consumption. They do this by employing aluminosilicate-rich waste sources. Geopolymer is composites (GPCs) are gaining prominence in both research and applicat...

Full description

Saved in:
Bibliographic Details
Published in2024 IEEE International Conference on Computing, Power and Communication Technologies (IC2PCT) Vol. 5; pp. 983 - 988
Main Authors Upendra Sharma, U.S, Alam, Intekhab, Ganguly, Saurav
Format Conference Proceeding
LanguageEnglish
Published IEEE 09.02.2024
Subjects
Alkali activation
Analytical models
Biological system modeling
Cement industry
Cementitious materials
Compressive strength
Computational modeling
Environmental impact
Extreme gradient boosting
Fly ash
Geopolymer composites
Gradient boosting
Machine learning
Predictive modeling
Predictive models
Support vector machine
Support vector machines
Online AccessGet full text

Cover

More Information
Summary:Geopolymers have come to light as a potential alternative to conventional cement in order to address issues related to cement production and consumption. They do this by employing aluminosilicate-rich waste sources. Geopolymer is composites (GPCs) are gaining prominence in both research and application. The traditional processes for the casting procedure, curing, plus inspecting GPC specimen need a significant amount of time, money, and effort. To improve research efficiency, new approaches are needed. In this research, we study the prediction of the compression strength (CS) of GPCs using predictive machine learning (ML) methods. According to our study, ensemble ML strategies outperform a single ML strategy in predicting the computational stability of GPCs. The R2 values for the SVM, GB, and XGB models. The models also displayed lower error values, such as mean absolute and root mean square errors, demonstrating the improved accuracy of ensemble ML approaches. Gravel (10-20 mm) and fly ash had conflicting impacts on the CS of GPCs; higher GPC concentrations reduced CS while higher fly ash concentrations increased it. The effects of gravel size (4-10 mm) were more negative than favourable. By providing rapid and economical techniques to assess material qualities, the adoption of methods of machine learning has an opportunity to revolutionise the construction sector.
DOI:10.1109/IC2PCT60090.2024.10486332