Coating Degradation Prediction Based on Machine Learning Models: Considering Shielding Effects and Environmental Factors

Due to time-varying environmental factors, epoxy coatings usually exhibit a complex and nonlinear degradation process. This study attempts to utilize machine learning models, specifically Particle Swarm Optimization Backpropagation Neural Network, Particle Swarm Optimization Support Vector Machine, and Particle Swarm Optimization Random Forest, to predict the degradation behavior of epoxy coatings, considering shielding effects and environmental factors. The models are pre-trained with different input combinations, including ultraviolet index, temperature, humidity, and cumulative degradation value. Results show that training with both environmental factors and cumulative degradation can improve the performance of model, achieving an R2 of 0.8770 and a reduced mean square error. The testing dataset confirms that the proposed method, with a mean absolute error of 0.0113, outperforms than other methods. This study provides a new method for evaluating and optimizing the life of coatings in complex environments, which can contribute to the development of degradation prediction models.