Multiscale structural health monitoring and strength forecasting of sustainable blended concretes using embedded piezoelectric sensors and AI-enhanced EMI analytics

Department of Civil Engineering, MNIT, Jaipur 302017, India

^* Corresponding author: pvramana.ce@mnit.ac.in

Published online: 7 January 2026

Abstract

The transition toward sustainable infrastructure demands innovative approaches to monitor and optimise the performance of low- carbon construction materials. This study presents a multiscale, sensor- integrated, and data-driven framework for structural health monitoring (SHM) and compressive strength prediction in blended concrete systems incorporating Ground Granulated Blast Furnace Slag (GGBS) and other supplementary cementitious materials (SCMs). Utilising embedded piezoelectric sensors (EPS) based on the Electro-Mechanical Impedance (EMI) technique, real-time impedance data were acquired across early-age (1–24 hours) and extended curing regimes (up to 90 days), capturing the evolution of microstructural stiffness and degradation states. A comprehensive experimental campaign was conducted across three concrete systems: PPC control, PPC with concrete enhancer, and GGBS-enhanced slag mix, subjected to progressive mechanical damage and aggressive chloride and sulphate exposures. EMI features such as RMSD, peak frequency shift, and impedance signature area were extracted and input into machine learning (ML) models, including Random Forest (RF), Support Vector Machines (SVM), and Artificial Neural Networks (ANN). The RF model yielded superior regression accuracy (R² = 0.95) for strength prediction and achieved 91% accuracy in classifying multistage damage states (healthy to fractured), confirming the viability of EMI-ML integration for in-situ diagnostics. Durability monitoring under coupled chemical- mechanical loading revealed accelerated degradation in control mixes, while GGBS systems exhibited superior resistance, validating the durability benefits of slag inclusion. The study establishes EMI-ML as a scalable methodology for continuous, non-destructive performance monitoring and predictive maintenance of sustainable concrete structures. The framework aligns with circular economy principles and digitised asset management, advancing next-generation intelligent infrastructure in the Industry 4.0 paradigm.