Performance evaluation of corrosion protective coatings on marine concrete piles: an experimental study utilizing artificial neural network and statistical analysis

Abstract

In coastal areas, reinforcement corrosion is a vital problem affecting the long-term durability of marine structures, particularly reinforced concrete piles. Marine piles subjected to alternate wetting and drying due the dynamic action of wave experience severe corrosion in the splash zone. This research is focused on the experimental investigation on performance of corrosion protective coatings on marine concrete piles. Wave generators were installed at a distance of ¼th of tank length from the left side of the tank to generate surface waves to assess the degree of corrosion in the simulated marine conditions. Polyurethane-coated concrete piles and the effect of coatings on the three different zones, atmospheric zone (AZ), splash zone (SpZ), and submerged zone (SuZ), were examined. The corrosion potential values of casted model piles were recorded using a half-cell potentiometer every 3 months in the three different zones. The results obtained were analyzed with the help of Artificial Neural Network model and the statistical analysis. The Artificial Neural Network model captured the corrosion potential values data from RSM optimization as indicated by the high R threshold (R > 0.99417), (R > 0.99815), (R > 0.99419). Also, data visualized statistical graphs of contour, surface plots, main plots, and time series. The performance of polyurethane coatings, especially in marine concrete piles, is excellent in preventing rebar corrosion.