| Title |
Prediction of Ultraviolet Corrosion Levels of High Density Polyethylene using Artificial Intelligence |
| Authors |
서정원(Jeong Won Seo) ; 고진환(Jinhwan Koh) |
| DOI |
https://doi.org/10.5573/ieie.2024.61.8.21 |
| Keywords |
High density polyethylene; Ultraviolet rays; Artificial intelligence; Convolutional neural network; Corroison |
| Abstract |
Fiber-reinforced plastic (FRP) materials, which are mainly used in ship manufacturing, provide excellent durability and impact resistance, but long-term exposure to ultraviolet rays causes serious deformation, which complicates the processing of closed wires and is pointed out as one of the main causes of marine environmental pollution. As a countermeasure against this, the use of high-density polyethylene (HDPE), which is attracting attention as an eco-friendly and recyclable material, is increasing, but HDPE is also showing vulnerability to ultraviolet rays due to the limitations of heat resistance and weather resistance. In this paper, we apply the Convolutional Neural Network (CNN), an artificial intelligence technique, to analyze and predict HDPE corroded by exposure to ultraviolet rays over time. CNN has useful advantages for image classification and data learning in the field of deep learning, and through this technique, it was confirmed that after exposing HDPE to ultraviolet rays for a specific time using UV LAMP, it can be predicted with more than 90% accuracy by acquiring it with corroded image data and conducting training. Based on the research results, it is expected that it will be possible to predict the aging of ship materials and find effective ways to prevent the disposal of abandoned ships and marine environmental pollution in advance. |