Analytical Model and Data-driven Approach for Concrete Moisture Prediction

The advent of smart sensing technologies has opened up new avenues for addressing the billion dollar problem in the wastewater industry of H2S corrosion in concrete sewer pipes, where there is a growing interest in monitoring the environmental properties that govern the rate of corrosion. In this context, this paper proposes a methodology to predict the moisture content of concretes through data-driven approach by using Gaussian Process Regression modeling. The experimental program in this study practices measurements during wetting and drying phases of concrete. The obtained moisture data is used to train the prediction model against interpreted electrical resistivity data. The data of analytical model formulated from Archie’s Law is then analyzed with experimental and Gaussian Process prediction data.