Regression analysis of meteorological, aircraft trajectory, and ambient effects on sonic boom metric variability

Abstract

NASA’s Low-Boom Flight Demonstration mission is a major step toward commercial, overland supersonic flight. Certifying low-boom aircraft will require accurate measurements, including understanding uncertainty due to variations in meteorology, aircraft trajectory, and measurement environment. This work builds on preliminary work done with Least Absolute Shrinkage and Selection Operator (LASSO) regression [Durrant et al., J. Acoust. Soc. Am. 150, A259 (2021)] to analyze the variability present in NASA’s Quiet Supersonic Flights 2018 (QSF18) dataset. With the variability quantified and several factors having been identified as potential contributors through LASSO regression, least-squares regression is used on sets of these potential contributors to determine coefficient weights for each factor, including meteorological, aircraft trajectory, and ambient noise. Sonic boom metrics are then predicted using the results and the accuracy is compared to PCBoom predictions. Because the ambient noise is significant, the same regression techniques are used to examine which variables from a geographic information systems (GIS) dataset are correlated with the ambient noise. These techniques are likely to be useful for noise monitor placement planning and data interpretation during community testing of low-boom aircraft. [Work supported by NASA Langley Research Center through National Institute of Aerospace.]