Imaging Acoustic Signature Analyses for Plant Turbo-machinery

Plant machinery, as well as rotating parts within aircraft, vehicles, and ships needs to be monitored for faults for both safety and economic reasons. Performing maintenance based on hours of operation rather than actual condition can lead to high costs by early replacement of good components and/or problems when a component fails before its time. Used for years for machine health monitoring, acoustic spectral signal analysis is similar to accelerometer-based vibration analysis with the advantage that it is not limited to recording casing vibrations at select locations and the sensor does not have to be attached to the machinery being monitored. Current state-of-the-art, however, has two shortcomings: Conventional Microphone Arrays are expensive, bulky, and considerable time is consumed in setting up an array grid. Systems require an expert human operator to interpret the data from these devices, and therefore is not suitable for automated, online monitoring.Progeny proposed to address these shortcomings by combining a compact, self-steering MEMS Acoustic Sensor Array with acoustic spectral analysis and automatic fault detection algorithms. We will prove the concept and advantage of our approach in Phase I and develop a complete, end-to-end real-time prototype system in Phase II.