A statistical evaluation on flight operational characteristics affecting aircraft noise during take-off. (May 2018)

• Record Type: Journal Article
• Authors: Gagliardi, Paolo; Teti, Luca; Licitra, Gaetano
• Is Part Of: Applied acoustics. Volume 134(2018)
• Page Start: 8
• Page End: 15
• Publication Date: 2018-05
• Subjects: Aircraft-noise prediction -- Principal Component Analysis -- Aircraft noise -- Noise modelling -- Noise abatement measures -- ADS-B systems; Acoustical engineering -- Periodicals; Periodicals; 620.2
• DOI: 10.1016/j.apacoust.2017.12.024
• Languages: English
• ISSNs: 0003-682X

Abstract

Highlights: ADS-B data are used as input of a predictive aircraft noise model. Principal component analysis is applied to avoid multicollinearity issues. Multiple linear regression method is implemented for SEL estimation. Flight-operational characteristics determining the most significant noise impact have been selected. Abstract: Aircraft noise immission level during the initial phase of take-off is influenced by several parameters, which often produce a significant fluctuation on measured noise levels at the noise monitoring terminal. This fluctuation is not only due to the different aircraft involved in the process, but it is also strongly dependent on the operational settings and characteristics of each take-off, even when the same aircraft type is analysed. The goal of this study is to investigate the relationships between the operational characteristics of flight departure and aircraft noise by means of a statistical approach in order to identify the parameters on which pilots could take action for noise-reduction purposes. The operational settings considered in the present work include actual take-off weight and ground run distance, lift-off aircraft ground speed and ground speed during the initial climb phase. Other variables, for instance source-receiver distance and weather conditions, such as air temperature, air density and headwind were also included in the analysis. The above mentioned parameters of B737-800 flight-departures were collected respectively during 15 days of July 2015 and September 2016. The data collected during the first session is the training set, while the second sample is the test set. Each sample of both datasets was joined with the corresponding noise level provided by the noise monitoring network. Principal Component Analysis and Multiple Linear Regression were performed in order to derive a simplified predictive noise model at a specific point on the ground. This method produced a good Sound Exposure Level estimation. The findings may also be useful to point out the operational characteristics causing the noisiest aircraft flyovers. Consequently, scheduled flight departures could be re-organized by introducing departure-direction and/or departure performance restrictions in order to minimize noise impact on the urban areas.