# Real-time concrete strength monitoring using piezoelectric sensors and deep learning

## TL;DR

This paper introduces a new method for real-time concrete strength monitoring using piezoelectric sensors and deep learning, offering a more accurate and scalable alternative to traditional testing.

## Contribution

The novel integration of piezoelectric sensors with deep learning for real-time, non-destructive concrete strength monitoring.

## Key findings

- The system achieved prediction errors within approximately 15% compared to standard compression tests.
- The technology has been incorporated into a new AASHTO standard (AASHTO T412).
- Validation across four highway projects confirmed the method's accuracy and scalability.

## Abstract

This study presents a transformative advancement in civil engineering by integrating artificial intelligence with infrastructure sensing to redefine concrete structures testing and monitoring. Traditional methods for evaluating concrete performance, largely unchanged for over a century, rely on labor-intensive, proxy-based techniques that are both time-consuming and limited in reliability. Our approach combines using piezoelectric sensors with AI-driven data analysis to enable real-time, in situ monitoring of structural conditions with enhanced accuracy and automation. By employing deep learning models to interpret electromechanical impedance signals, the system eliminates the need for destructive testing or human intervention, offering a scalable solution suitable for real-world deployment. Successfully validated across four large-scale highway construction projects, the system demonstrates prediction errors within approximately 15% when benchmarked against standard compression tests conforming to ASTM C39. Aspects of this technology, such as the underlying sensing principle have been incorporated into a new standard by the American Association of State Highway and Transportation Officials (AASHTO T412), representing a significant step toward the national standardization of this non-destructive testing method. Our findings propose a scalable method to integrate intelligent sensing into civil infrastructure system. This will enable the development of resilient and sustainable infrastructure, moving beyond traditional infrastructure monitoring.

Traditional concrete testing can be labour-intensive and limited in accuracy, consistency, and real-time applicability. The study uses piezoelectric sensors and deep learning for real-time monitoring of concrete strength, interpreting signals to achieve accurate predictions. This is further validated in highway projects.