Assessment of Semi-Empirical Wing-Mass-Estimation Methods for Commercial Transport Aircraft

The present contribution evaluates the responses produced by several semi-empirical wing-mass-estimation methods based on a wide range of parameter variations on the wing of an Airbus A320ceo-class aircraft model. A common wing-parametrization method is established to ensure the appropriate comparison between the results of each method. Six semi-empirical methods are introduced and compared against reported trends and a published surrogate model of a physics-based wing-weight-estimation framework. The investigated methods produce a 7.0% to 34.5% error relative to reference aircraft data for the baseline case. Variations with increasing aspect ratio show that most methods exhibit non-linear trends of decreasing gradients, opposing those found in the literature and numerical models. Removing the inertia relief considerations associated with wing fuel storage yields mass penalties ranging from 3.8% to 11.4% across the interval of wing aspect ratios explored, broadly matching published results. Alterations to the wing reference area and the spanwise location of the centre of pressure yield nearly linear and proportional changes to the wing mass for all the methods analysed, whereas the sensitivity to the engine’s spanwise location is found to be nearly linear or non-linear, depending on the method’s level of detail. To address the identified limitations, one of the simplest of the methods considered is systematically modified by restructuring its geometric terms and introducing a fuel mass relief, and is then calibrated against one of the most detailed, recovering the gradients and dry-wing sensitivity of the latter. This work shows that the simple semi-empirical methods commonly employed during the conceptual and preliminary aircraft design phases cannot fully represent the behaviour of methods of higher fidelity. Thus, future aircraft design studies should adopt more detailed wing-mass-estimation methods or systematically adapt existing ones.