A REVIEW OF PREDICTIVE MAINTENANCE APPROACHES FOR CORROSION DETECTION AND MAINTENANCE OF MARINE STRUCTURES

Abstract

Corrosion is a natural phenomenon that deteriorates and damages the surface of metallic material. Over time, the surface of the material deteriorates due to electrochemical reactions with the surrounding environment. If corrosion is not identified early on, it can become a major financial burden for industries, costing billions of dollars. Despite swift technological developments, preventing and maintaining corrosion progression with reactive maintenance remains difficult. Due to that, predictive maintenance has been developed to predict the deterioration, degradation, and fault over the remaining useful life of the material by using real-time data, historical data, simulation, modelling, and failure probability. Predictive maintenance allows inspectors to monitor the health and predict the corrosion level of the material. However, it is hard to predict the unexpected degradation of the material from the developed prediction model without considering the harsh environment and other external factors. Hence, there is a need to investigate these problems and their effect on predictive maintenance for corrosion detection and maintenance. Therefore, this paper reviews and compares the state-of-the-art predictive maintenance solutions developed to solve corrosion issues in various applications, industries, and academic research. The challenges and opportunities for the predictive maintenance application of corrosion detection and maintenance are also presented. This review will provide new and additional knowledge that can be used to develop prediction models for corrosion detection and maintenance, which will help prevent unexpected failures.