Using the meta-heuristic algorithms to optimize the radial basis function for predicting high-performance concrete compressive strength and slump flow

Concrete serves as a crucial foundational material in construction. While traditional concrete is commonly employed for standard building projects, it lacks the requisite strength for specialized structures like dams, silos, and skyscrapers. By introducing additives such as fly ash and blast furnace slag, HPC achieves a reduced water-cement ratio, enhancing compressive strength (CS), slump flow, and overall efficiency. To accurately determine HPC's CS and slump, using some machine learning (ML) approaches such as artificial neural networks (ANN) streamlines the process, offering substantial advantages over traditional laboratory methods, such as time and cost savings. One noteworthy neural network in this context is the radial basis function (RBF). By coupling RBF with the honey badger algorithm (HBA) and northern goshawk optimization (NGO), favorable and closely aligned results can be attained, mirroring laboratory experiment outcomes. The combination of the RBHB model yields key performance indicators signifying its effectiveness in evaluating HPC's slump and CS. In conclusion, this study underscores the potential for a more advantageous synergy between the HBA algorithm and the RBF model, ultimately producing highly favorable and reliable outcomes.