Application of the bald search optimization-based regression analysis on properties of UHPC

To save time and do less experimental work, this study examined the effectiveness of constructing hybridized regression analysis on ultra-high-performance concrete (UHPC). Different physical substances and byproducts can be incorporated into this type of concrete. To achieve this, a dataset of 170 samples from various hybridized support vector regression (SVR) analyses was gathered from published articles. The best values of the SVR’s determinant components were then investigated using the meta-heuristic optimization methods like the bald eagle search algorithm (BES) and chimp optimization algorithm (ChOA). UHPC is a specialized type of concrete with unique properties, and accurately predicting these properties is crucial for ensuring the performance and reliability of structures made from UHPC. By utilizing optimization algorithms to fine-tune the regression model, the study aimed to achieve enhanced performance in terms of accuracy and reliability. The use of optimization algorithms in conjunction with regression analysis was also intended to save time and reduce the need for extensive experimental work. SVR_B achieved the greatest R² value in both the training and testing datasets, as well as the lowest values of error-based metrics in both datasets. The smallest value of performance index (PI) in both the training and testing dataset with a 0.0016 difference in the training dataset and 0.0078 difference in the testing dataset. The SVR_B has better performance compared to SVR_Ch. When compared to SVR_Ch and previously published studies, the hybridized SVR_B may achieve the greatest accuracy.