Machine Learning Prediction Model Integrating Experimental Study for Compressive Strength of Carbon-Nanotubes Composites

• Published in: Maǧallaẗ al-abḥath al-handasiyyaẗ 2024-09
• Main Authors: Manan, Aneel, Zhang, Pu, Ahmad, Shoaib, Umar, Muhammad, Raza, Ali
• Format: Article
• Language: English
• Subjects: CNT-reinforced cementitious composite; gene-expression programming; machine learning; mechanical property; parametric analysis
• ISSN: 2307-1877
• DOI: 10.1016/j.jer.2024.08.007

In this study, carbon nanotubes (CNTs) are added to improve the mechanical properties of concrete in different percentages from 0 to 2.0%. The mechanical performance of concrete was evaluated through compressive strength at different days of curing. The findings indicate that the compressive capacity of concrete improved with CNT. However, their complex structure and variation in properties present challenges that restrict their application. Therefore, the machine learning approach was used to develop a prediction model for the compressive strength of complex CNT composites. A comprehensive database of 295 points was created from the literature. Numerous models were developed using different hyper-parameters to get an optimized prediction. The evaluation of all the models was done using statistical parameters, sensitivity analysis and parametric. The experimental results of this study were used for validation of predicted results. The results indicate that the proposed prediction model is highly reliable. Additionally, the accuracy of the proposed model was tested with the experimental investigation and the strength of the prediction equation was checked by the comparison with the previously proposed equation. Lastly, a simple, accurate and efficient prediction is proposed for estimating the compressive strength of CNT composites. [Display omitted]