Strength Prediction of Graphene Oxide Concrete Composite Using Machine Learning with EMI Data

In recent days, the development in integrating nanomaterials specifically graphene oxide with concrete has gained significant attention to improve the performance of concrete at a nano-scale level. This study presents compressive strength monitoring and prediction of graphene oxide concrete composites (GOCC) using the electro-mechanical impedance (EMI) technique via machine learning (ML) models. Graphene oxide (GO) is used as a nano additive in a concrete composite. Embedded piezoelectric sensors (EPS) have been used in this experiment to monitor and collect the EMI signatures of a GO concrete composites (GOCC) specimen during the 28 days of curing. To obtain the compressive strength of concrete, destructive testing was conducted during the strength growth. An equivalent stiffness parameter (ESP) is extracted from the raw signatures of EMI to monitor the gain in stiffness parameters. Additionally, using data on destructive compressive strength and ESP, ML models were created to predict the compressive strength. It has been investigated that incorporating GO with only 0.03% of cement into concrete shows a compressive strength increase of around 31% at 3 days and 25% at 28 days when compared with conventional concrete (CC). In predicting the compressive strength for GOCC, the developed ML models show promising results. Hence, for assessing and monitoring the strength growth development of concrete composite at a nano level the proposed method can be effectively utilized.