Work

Recycled aggregate concrete based on an improved SVM method a study on the prediction of compressive strength

Year: 2026

Type: article

Abstract: The 28-day compressive strength of recycled aggregate concrete (RAC) is a critical indicator for evaluating its engineering applicability. Therefore, developing an economical, convenient, and accurate strength prediction method is of great significance for promoting the large-scale utilization of recycled aggregates. In this study, a support vector machine (SVM) prediction model integrating the linear kernel function and the nonlinear kernel function (radial basis function, RBF) is built, and nine key mix proportion parameters, including cement content, fly ash content, and water content, are selected as input features. A total of 103 sets of experimental data were systematically collected from the literature for use in model training and validation. The results demonstrate that the SVM prediction model significantly outperforms conventional prediction methods. Specifically, the root mean square error (RMSE) of the SVM prediction model is reduced by 50.6%, and the coefficient of determination for the test set (R²p) reaches as high as 0.92326. These findings indicate that the proposed model exhibits high prediction accuracy, strong generalization capability, and no evident overfitting. The model provides an effective tool for mix proportion optimization and rapid performance evaluation of recycled aggregate concrete.

Source: Scientific Reports

DOI: 10.1038/s41598-026-68942-4

Authors: Chi Zhang, Yinyin Tang, Hanrui Fu

Institution: Xihua University

Language: English

Topic: Recycled Aggregate Concrete Performance

Subfield: Building and Construction

Field: Engineering

Domain: Physical Sciences

SDGs: Responsible consumption and production, Industry, innovation and infrastructure

Open Access status: gold

Estimated APC paid: $2,690