With the increase of aged water pipes, especially ductile iron and cast iron pipes that account for 60% of thewater supply pipes in Japan, leakage incidents caused by pipe corrosion are increasingly concerned. Although a conventional method to predict corrosion pit depth of iron pipes has been used for a long period, it has two downsides: 1) prediction accuracy is insufficient, and 2) leakage frequency cannot be estimated quantitatively. Therefore, in this study, we have developed a new prediction model of corrosion pit depth with high accuracy by considering both lag-time periods due to the anticorrosive effect of external coating of pipes and deviation of corrosion rates. Based on the corrosion pit depth prediction model, another model, i.e., the leakage probability estimation model, was developed to estimate the leakage incidents quantitatively. By replacing aged pipes to new pipes according to the models developed in this study, the number of leakages can be reduced by 50% at best compared to the conventional model.

View full abstract

[in Japanese]

View full abstract