Predicting Pressure Sensitivity to Luminophore Content and Paint Thickness of Pressure-Sensitive Paint Using...

Area of Science:

• Materials Science
• Chemical Engineering
• Computational Modeling

Background:

• Pressure-sensitive paints (PSPs) are crucial for surface pressure measurements.
• Limited experimental data can hinder the development and optimization of PSPs.
• Predictive modeling can overcome data limitations in materials characterization.

Purpose of the Study:

• To develop and validate an artificial neural network (ANN) for predicting PSP pressure sensitivity.
• To investigate the influence of luminophore content and paint thickness on pressure sensitivity.
• To assess the effectiveness of data augmentation in improving ANN performance with limited datasets.

Main Methods:

• Construction and training of an artificial neural network (ANN).
• Utilizing experimental data including luminophore content and paint thickness.
• Application of a data augmentation technique to expand the dataset.
• Evaluation of ANN prediction accuracy using mean absolute percentage error.

Main Results:

• The ANN successfully predicted pressure sensitivity based on luminophore content and paint thickness.
• Predictions were within confidence intervals reflecting experimental errors.
• Data augmentation effectively increased the number of data points for training.
• The model demonstrated good predictive accuracy for PSP characteristics.

Conclusions:

• ANNs combined with data augmentation offer a powerful approach for characterizing PSPs.
• This methodology can improve the performance of PSPs for global surface pressure measurements.
• The study highlights the potential of computational methods to accelerate materials development.