Strength monitoring and prediction of blended concrete systems from very early to delayed curing age using embedded piezo sensor data: An experimental and machine learning approach

Abstract

Concrete structures form the backbone of modern infrastructure, offering essential support for housing, water management and transportation systems. Effective monitoring of concrete strength development is essential to ensure safety and prevent damage during the construction phase. This study comprehensively evaluated the strength development of three blended concrete systems from very early (1–24 h), early age (1–5 days), later age (6–28 days) and delayed curing age (30–90 days) using an embedded piezo sensor (EPS), along with non-destructive, destructive tests, and a machine learning (ML) approach. The concrete mixtures incorporate fine and coarse aggregates along with Portland pozzolana cement (PPC), concrete enhancer (CE), and slag. Experimental results indicate that concrete prepared with PPC + CE + slag achieves the highest compressive strength, followed by concrete with PPC + CE, while PPC alone exhibits the lowest strength. EPS monitors phase transitions and strength development in blended concrete systems through observable shifts in conductance signatures, offering non-destructive insights into structural changes and strength development. Statistical and equivalent stiffness analysis further confirms the higher strength development in the PSC-C system followed by PC-C and PPC-C systems. Furthermore, various ML models are employed for strength prediction, with the random forest (RF) model demonstrating the highest accuracy of 0.97. Overall, EPS data provides a reliable non-destructive indicator of strength in blended concrete systems, while its integration with ML models enhances strength prediction capabilities. This approach advances the understanding of strength development during the curing process and provides valuable insights for contractors, engineers and researchers in the construction of sustainable concrete structures.