A probability based model for the erosive wear of concrete by sediment bearing water

The purposes of this work were: (i) to understand how erosive wear is affected by the random meso structure of concrete, and (ii) to develop a model for erosive wear of concrete structures due to the action of sediment bearing water loads. An aggregate generation algorithm with certain novel features is used to generate a database comprising 400,000 aggregates with varying levels of angularity and flakiness. Using these aggregates, a probabilistic model for mass loss due to aggregate erosion was formulated. A Monte Carlo simulation is used to obtain the time to failure for single aggregates; these were used to obtain an estimate of slurry erosive loss from the concrete structure, where successive layers of aggregates are assumed to satisfy the ergodicity assumption. The mass loss estimates were compared with experimental results for a sediment bearing water jet. A good qualitative match was observed, with both experimental results and probabilistic estimates exhibiting very similar dependence on meso geometry.