Derivation of a Wing-Weight Estimation Equation for a Roadable Personal Air Vehicle Using Regression Analyses of Data Obtained from Wing Planform Reference Points

Abstract

Estimating the wing weights of personal air vehicles (PAVs) is limited by lack of comparable data, as there were no PAV data available to the public. In this study, the limits were confirmed by verifying estimation equations published in the literature. Research was then conducted to derive an equation to estimate PAV wing weight to overcome such limits. To address difficulties which arose during the data collection process, research was conducted on the derivation of a PAV wing-weight estimation equation that could be used in the initial design stage, based on single-engine turbo-propeller aircraft with specifications similar to those of PAVs. The estimation equation was derived using a regression analysis. Typically, data estimation models created with statistical methods frequently result in the overfitting phenomenon, in which the model cannot properly respond to actual data. To address this problem, the model creation and evaluation steps were performed using the K-fold cross-validation method, which was made into a program using Microsoft Excel version 2016 Visual Basics for Application (VBA). The final model was verified using actual data not used in the estimation model creation process. The variations in the final model were examined using a Monte-Carlo simulation. A final equation was derived that also considered the weight increase caused by the installation of a folding mechanism.