Artificial Neural Network-Based Modeling of Sustainable Plasma Process Parameters for Predicting Wettability of Aircraft Composite Surfaces

Modeling the wettability behavior of Carbon Fiber Reinforced Polymer materials with atmospheric pressure plasma treatment is crucial for optimizing and sustaining efficient manufacturing processes, particularly in aerospace applications. In this study, plasma process parameters were modeled and 3D mapped to predict the wettability of CFRP using Artificial Neural Networks. The ANN model was trained with data obtained from these plasma parameters, and its performance was evaluated using correlation coefficients and error analysis. Contact angle measurements, surface roughness, and surface energy measurements were performed to evaluate the wettability of plasma-treated composites, while XPS and AFM analyses verified the chemical modifications and preservation of surface topography. These characterizations validated the reliability of the experimental inputs used for ANN modeling. The predictive performance of the model was independently confirmed by the high R² values obtained from target-output correlation plots, with 0.9934 for WCA and 0.9911 for DCA. The strong agreement between predicted and experimental values demonstrated the robustness of the ANN approach in capturing the complex behavior of plasma-treated CFRP surfaces. Furthermore, the model indicated that a nozzle speed of 10 mm/s and a nozzle-to-surface distance of 11 mm yielded optimal surface characteristics, while a nozzle speed of 30 mm/s and a distance of 20 mm corresponded to an insufficient surface energy region. The study highlights the effectiveness of atmospheric pressure plasma treatment and the predictive capability of the ANN model for optimizing wettability in CFRP materials, offering a valuable tool for surface property modification in industrial applications while promoting sustainability and cost-efficiency.