A Component Based Empirical Model for Airframe Noise Prediction

Abstract

This paper presents an empirical model for airframe noise prediction. The model is derived from a large database of airframe noise tests, involving various airplane models at various operating conditions. The model correlates far field noise not only to gross airplane parameters such as the dimensions of the high lift system and flight Mach number, but also to flow quantities that are physically responsible for the noise generation. For example, flap side edge noise is correlated to the strength of the vortex in the roll-up cross flow at the side edge. Noise data used in the development of the model are acquired by using phased microphone arrays, which enables the decomposition of the total noise into components, relating the noise to the individual components of the wing/high lift system. This is done by beam-forming the phased microphone array data onto the airplane, leading to source distributions on the airplane, which are then integrated over individual components and propagated to the far field to derive the noise from the components. The wing/high lift system in this prediction model is divided into six components, namely, the leading edge slat, the outboard flap side edge, the inboard flap side edge, the high-speed aileron, the wing trailing edge and a residue term. The first five components are chosen to represent the most dominant airframe noise sources and the last to account for areas not covered by the first five components. In this paper, the methodology and results of this component-based model is presented, including source identification by source strength maps, component integration to derive far field spectra, validation/calibration of the integrated spectra by conventional free field microphone data, extrapolation of small-scale model test data to full-scale conditions with Reynolds number dependent scaling laws and the correlation between noise and flow quantities. Validations of the predictions with flight test data are also given to show the accuracy of the developed prediction tool.