Developing a measurement system to identify noise sources on in-flight aircraft.
Building an application based on NI LabVIEW software that uses NI PXI hardware to acquire data from a phased microphone array on a runway.
To develop quieter passenger aircraft, we must identify all noise sources to improve our understanding of the noise generation mechanisms. While designing an aircraft, we can predict noise levels through numerical analysis and model tests. However, the properties and characteristics of actual aircraft noise can only be obtained by actual flight tests. Noise source localization through acoustic beamforming is a powerful tool to accomplish this. Beamforming is a method of mapping noise sources using a phased array of microphones and displaying their amplitude. Although we at JAXA have developed and improved this technique through both wind tunnel tests and flight tests using small scale model airplanes, we haven’t applied this technique to actual aircraft in flight. In 2009, we acquired a small Mitsubishi MU-300 Diamond business jet. In 2010, we set up a phased microphone array on a runway and began conducting noise source localization measurements to verify our current technology and to determine areas of improvement.
The phased array consists of many microphones distributed over a large diameter. Using the slight time lag for the sound waves emitted from the noise sources to arrive at each microphone, we can estimate the location and strength of each noise source. In this test, we designed the phased array to distinguish between two separated 1 kHz tones in 4 m distance on an aircraft flying at 120 m altitude. We designed an array consisting of 99 microphones arranged over a circular area with a 30 m diameter.
In-flight noise source localization tests must include the state of the jet engine; acoustic measurements; and the aircraft’s position, altitude, and speed while it flies over the phased array. We also record meteorological data such as wind direction and speed, temperature, and humidity, since the noise generated from the aircraft is attenuated by the atmosphere before reaching the microphones on the ground.
A key feature of our system is the simultaneous measurement of flight parameters with acoustics. We accomplish this using two line-scanning cameras placed with the phased array on the ground. These two cameras are directed upright and are staggered both in flight direction and in lateral direction, as shown in Figure 1. The cameras capture synchronized images of the passing aircraft, providing us 3D information that allows us to analyze the flight speed and flight altitude of the aircraft. The noise localization process is simultaneously executed with the acquired acoustic data on another computer. This data is combined together shortly after the aircraft flies over the phased array and is visualized as noise source maps overlapping the aircraft image.
We chose an NI PXI system with a variety of modules to meet our requirements for a compact system. Our system contains an embedded controller and an NI 8260, a 4-drive, in-chassis, high-speed data storage module. LabVIEW, the NI Vision Development Module, and the NI Sound and Vibration Measurement Suite provided an effective tool for developing our solution. The system provided quick setup, data logging, real-time monitoring, and data review. We mounted it on a mobile platform for portability.
The NI PXI-4498 dynamic signal acquisition (DSA) modules provide power to the microphones via integrated electronic piezoelectric (IEPE) conditioning and simultaneously acquire high-resolution data from them. Since each module can handle up to 16 channels, the seven modules can simultaneously sample 112 channels with room for expansion.
The NI PXI-1428 image acquisition modules allow line-scan digital imaging from line-scan cameras. These acquired images are processed automatically by the application we developed using the NI Vision Development Module. In addition, the NI PXI-6682H timing and synchronization module provides synchronization via GPS to timestamp the data and coordinate with avionics onboard the aircraft and on the ground.
We conducted flight tests November 16–18, 2010, at Taiki Aerospace Research Field in Hokkaido Prefecture. While the MU-300 business jet executed a series of repeated approaches and climbs to simulate takeoff and landing at various flight altitudes and speeds, we took measurements simultaneously on the ground and in the aircraft. Figure 6 shows the typical results of noise localization maps for 1,000 Hz and 2,000 Hz on the aircraft in a landing configuration. At 1,000 Hz, we confirmed noise sources with higher amplitudes (shown in red) near the main landing gear, outside edge of the flaps, and the engine nozzle. At 2,000 Hz, we observed noise sources near the engine nozzle, main landing gear, and nose landing gear, but no longer at the outside edge of the flaps. Through this test, we demonstrated the technology to localize multiple noise sources on an in-flight aircraft.
We are continuing to improve our technology to locate noise sources more accurately and to enable more detailed evaluation of noise source properties such as noise spectra.
Dr. Kenichiro Nagai
Japan Aerospace Exploration Agency (JAXA)