In the service life, RC beams may be subjected to some aggressive environment such as deicing salt and/or coastal/marine conditions, thus the steel reinforcement will be corroded due to the penetration of chloride ions and/or the carbonation of concrete cover. Corrosion may cause the failure modes of the beam to change from a flexural one to a shear one. Consequently, it is critical to accurately predict the shear strength of corroded RC (CRC) beams in their whole life-cycle. To this end, this work develops a time-dependent predictive model for the shear strength of CRC beams using one of the most representative ensemble machine learning algorithms, i.e., the gradient boosting regression tree (GBRT), in which a mechanical model is used to calculate the corrosion extent of reinforcements. The proposed time-dependent prediction model is capable to provide the shear strength of CRC beams with any given service time.
Reference:
Fu, Bo, and Feng, De-Cheng*. "A machine learning-based time-dependent shear strength model for corroded reinforced concrete beams". Journal of Building Engineering, Volume 36, April 2021, 102118. (https://doi.org/10.1016/j.jobe.2020.102118)