Modeling Strength of High-Performance Concrete Using an Improved Grammatical Evolution Combined with Macrogenetic Algorithm

Abstract

The main purpose of this paper is to propose an incorporating improved grammatical evolution (GE) into the genetic algorithm (GA), called GEGA, to estimate the compressive strength of high-performance concrete (HPC). The GE is a recently developed evolutionary programming type system. It is used to automatically discover complex relationships between significant factors and the strength of HPC. This method is transparent and can enhance our understanding of the mechanisms of HPC strength. A GA was used afterward with GE to optimize the appropriate function type and its associated coefficients. In addition, macroevolution algorithm was processed to improve search efficiency during the GA optimization procedure. The case study includes over 1,000 examples of HPC for which experimental data were available. This novel model, GEGA, can obtain a highly nonlinear mathematical equation for predicting the HPC’s compressive strength. The results show that GEGA has lower estimating errors, which outperforms another evolutionary strategy called genetic programming and two popular types of traditional multiple regression analysis.