Using second order non-linear regression model to optimize factors affecting the strength and spread ofhigh-performance concrete

Abstract

For conventional vibrated concrete (CVC), the concrete mix design process is relatively simple, usually required to test some of mixes with different cement ratios, the cost of one test is insignificant. Meanwhile, to design ultra-high performance concrete mix (HPC), it is necessary to determine the mix to meet technical requirements, of which the most important for HPC is the requirement for compressive strength and the spread. Currently, agencies, businesses, and researchers often have to spend a lot of effort and money to do experiments and practical tests. In this article, the author used the experimental planning method to find the optimal mix for initial ultra-high performance concrete for a given initial design concrete grades, meet the requirements of strength and spread with the least number of experiments. This result can be used for reference in the HPC mix design process to reduce experiment costs.