AUTHOR=Zhang Xinyi , Dai Chengyuan , Li Weiyu , Chen Yang 

TITLE=Prediction of compressive strength of recycled aggregate concrete using machine learning and Bayesian optimization methods

JOURNAL=Frontiers in Earth Science

VOLUME=Volume 11 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/earth-science/articles/10.3389/feart.2023.1112105

DOI=10.3389/feart.2023.1112105

ISSN=2296-6463

ABSTRACT=Recycled aggregate (RA) have been widely used in concrete preparation to reduce the environmental impact of construction waste. Compressive strength is an essential measure of the performance of concrete. In order to improve the accuracy of compressive strength prediction of recycled aggregate concrete (RAC), this paper uses machine learning combined with a hyperparameter optimization technique to predict the compressive strength of RAC. The experimental data of RAC mix ratio (the compressive strength of RAC, water-cement ratio (c/w), natural aggregate crushing rate (δ_N), RA crushing rate (δ_R), and RA replacement rate (ρ)) are collected, and the ratio method is treatment. The xgboost, random forest (RF), K-nearest neighbour (KNN), support vector machine regression (SVR), and gradient boosted decision tree (GBDT) RAC compressive strength prediction models were developed. Grid search (GS), Random search (RS), and Bayesian optimization (BO) algorithm to optimize the model hyperparameters. The results show that the ratio method data processing can improve the model's performance to some extent. The optimal combination of hyperparameters can be searched in the shortest time using the Bayesian optimization algorithm based on TPE (Tree-structured Parzen Estimator); the BO-TPE-XGBoost RAC compressive strength prediction model has higher prediction accuracy and generalization ability. This high-performance compressive strength prediction model provides a basis for RAC's research and practice and a new way to predict the performance of RAC.