Abstract:
In modern era, an aircraft experiences numerous adjustments during designing phase.
The advent of new features has made the aircraft’s design process more complicated
to make decision regarding selection of appropriate and parsimonious design. Artificial
Intelligence (AI) is one such approach to tackle this issue and attain reliable and efficient
approximation. AI is a learning-based program which mimics human intelligence. In
AI, Machine Learning (ML) models can help us to select the best geometric parameters
of the performance parameters at the time of design process.
In this study, different ML models are used to predict the accuracy of performance
parameters in the context of geometric parameters. These models are validated using the
four steps process i.e. checking the value of Adjusted-R2, Root Mean Square Error, Mean
Absolute Percentage Error and Nash-Sutcliffe Efficiency Coefficient. It was observed
that a Neural Network Regression model is best to predict the aircraft’s performance
parameter(s). Thus, AI holds a strong potential among aircraft design community during
preliminary design phase
