Innovative approaches to concrete health monitoring: wavelet transform and artificial intelligence models

Abstract

The health monitoring of concrete structures is of principal concern to avoid major accidents. Presently, many large-scale structures have been constructed throughout the world and in India. Therefore, there is an urgent need for sensor-aided research to keep all these large infrastructural facilities for the long life in an uninterrupted manner. As per the available literature, the Acoustic Emission (AE) sensor data and its deployment for the development of an artificial intelligence (AI) model is most suitable for health monitoring of these types of structures. Researchers have used the signal processing method. However, the AI models have significantly reduced the effort as well as errors in the computation process. In this study, an experimental investigation is done using the AE system for data generation. A good number of concrete slabs of different grades were cast and used for generating data deploying the Pencil Lead Break (PLB) approach. The generated data was utilized for finding the damage location using the WT method and AI models. The developed AI model is more effective in the health monitoring of concrete structures as the error in calculation is less as compared to the WT method. The model is also validated by identifying the damage source (simulated) in the concrete slab. This approach can be utilized for real-time health monitoring of large-scale concrete structures comprised of slab-like components without any interruption. Results show promising trends for further research for making the health monitoring process in wider application of civil engineering structures.