A new heuristic algorithm for mix design of high-performance concrete

Abstract

Based on the latest developments in concrete technology, High-Performance Concrete (HPC) is characterized by superior properties such as strength, workability, and durability. The modeling and precise proportioning of the concrete mixture are important for expanding the application of HPC. There have been many efforts to develop an expert system for HPC mix proportioning, but the systems developed from these efforts are still somewhat complicated, time-consuming, and have uncertain tasks. In this paper, to provide a tool for the convenient design of HPC mixtures, mix proportioning programs using a new meta-heuristic algorithm called "Harmony Search", which was conceptualized based on a musical process of searching for a perfect harmony, is devised. The mix design programs developed by using harmony search algorithm were compared to other mix design approaches such as neural networks and genetic algorithm. The results showed that the harmony search algorithm can be a potentially strong tool for HPC mix proportioning.