Preliminary lasso regression analysis of environmental effects on sonic boom metric variability

Abstract

NASA’s Low-boom Flight Demonstration mission is a step toward commercial, overland supersonic flight. Certifying low-boom aircraft will require accurate measurement and understanding of uncertainty due to variations in meteorology, aircraft trajectory, and measurement environment. This work describes an analysis of the variability present in the Quiet Supersonic Flights 2018 (QSF18) dataset collected in Galveston, Texas, and an initial effort to correlate this variability with source and environmental factors. Although these data are for an F-18 executing a low-boom dive, no other low-boom dataset over a widespread populated area exists, and this study has the potential to guide future X-59 testing and data interpretation. Although several aircraft and meteorological variables are investigated for possible correlation, no single variable consistently explains the large variation in the measured booms. This has led to the use of lasso regression (LR) to identify the most relevant factors. Thus far, the fact that LR correlates boom Perceived Level (PL) with ambient PL is a strong indication of dataset contamination by significant ambient noise. Additionally, although several meteorological factors can contribute to boom metric variability, low-altitude wind speed is particularly important. [Work supported by NASA Langley Research Center through National Institute of Aerospace.]