Helicopter Noise Source Separation Using an Order Tracking Filter

Abstract

Due to the importance of understanding the aeroacoustics of rotorcraft with continually changing noise sources, this paper presents a new technique for source separation from ground-based acoustic measurements. The source separation process is based on combining a time-domain de-Dopplerization method with the Vold–Kalman order tracking filter approach. This process can extract rotor harmonic noise even when the sources are continuously changing with time, including impulsive events such as blade–vortex interaction noise. The advantage of this approach over traditional methods such as harmonic averaging is that the phase and amplitude relationship of acoustic signals is preserved throughout the extraction process. The approach is applied to the measured acoustic data from a Bell 430 helicopter. The measured data were separated into main rotor harmonic, tail rotor harmonic, and broadband residual components. For steady-state conditions, the extracted components could be depropagated to form acoustic hemispheres showing the directivity of the separated main and tail rotor components. The source separation process was also applied to a maneuvering flight condition. Each component has different pulse shapes and directivity trends, consistent with aeroacoustic theory.