The growing demand for a more efficient maintenance of concrete bridges requires a model that tracks the deterioration of each bridge based on inspection data. Although it has been expected that machine learning could be applied to this problem, inspection data sparsely distributed over time are not suitable for machine learning in contrast to the continuous big data usually targeted. This study applies machine learning to a regression model of crack formation and propagation using inspection data to confirm the applicability. It includes the selection of the optimal algorithm, development of the model based on a novel methodology, and factor analysis using the model. Accordingly, the model was constructed by Gaussian process regression and it could appropriately extract the differences in the progress of crack damage due to multiple influential factors. The results demonstrate the excellent applicability of machine learning even to sparse data.

No abstract

No abstract

An enhanced multi-chemo-physical model for the time-dependent deformation of concrete is proposed based on thermodynamic state of moisture in micro-pores. The moisture migration mechanism is divided into 1) moisture transport through CSH gel grains and 2) water in motion within the inter-particle spaces of hydrate micro-products. The new kinematic model makes it possible to simulate both long-and short-term concrete creep. An enhanced mechanistic law of stress path dependency is introduced to cope with a wide variety of stress and ambient histories as well. Time-dependency at elevated temperature is also investigated with current high-accuracy thermo-hygro dynamics. The instantaneous plasticity in direct connection with evaporating moisture from CSH crystal inter-layers is incorporated into the predictive system. Although some mechanisms remain unverified, drying shrinkage and creep at high temperature are fairly simulated.

No abstract