A Principal Component Analysis of Acoustic Emission Signals from a Landing Gear Component

Acoustic emission monitoring was completed on a painted aerospace grade steel landing gear component undergoing fatigue loading until rupture. A post-test linear location analysis of the collected signals revealed eleven groups where high activity (greater than 2000 hits) occurred within a defined location, three of which corresponded in location to the position of fracture and final rupture of the specimen. Feature data, such as amplitude, rise-time, energy etc. were used to describe the identified signals in each group. A dimension reduction through principal component analysis of the feature data of all groups was performed. Results showed that high amplitude signals associated with four groups of signals arising from noise could be separated from the fracture groups. However four groups not associated with noise or the known positions of the fracture groups were not separable from the signals attributed to fractures. The paint layer of the specimen was removed and a magnetic particle investigation was completed that showed these four groups coincided with regions of additional fracture in the component. Introduction Messier-Dowty are the world's leaders in the design and manufacturer of aircraft landing gear. As part of each new design, the landing gear fitting is required to complete an airworthiness certification test. These tests can last up to five years, with 25% of that time assigned to periodic non-destructive testing (NDT). A method utilising acoustic emission (AE) to monitor the structure during loading (in order to reduce the need for periodic NDT investigations) is currently being researched, with a major aspect of the study being source identification. Source identification is a complex problem; in a landing gear investigation there are numerous sources of AE such as hydraulic noise, bearing noise, friction between components and movement of sliding tubes. In addition there are numerous propagation paths that can dramatically alter the signal with respect to distance. This paper expands on previous work using Principal component analysis (PCA) of feature data [1,2]. PCA is a method used to simplify high order data sets to lower dimensions to allow a simple analysis. It is performed by taking a rotation of the data around two orthogonal vectors which describe the largest amount of variance in the data. These are found as the eigenvectors of the data's covariance matrix; the eigenvectors with the largest corresponding eigenvalues are chosen. For a simple explanation of PCA, see Nabney [3]. In this paper the first investigation of the PCA used to remove noise sources from fracture sources in aircraft landing gear component is presented. Key Engineering Materials Vol. 347 (2007) pp 139-144 online at http://www.scientific.net © (2007) Trans Tech Publications, Switzerland Online available since 2007/Sep/15 All rights reserved. No part of contents of this paper may be reproduced or transmitted in any form or by any means without the written permission of the publisher: Trans Tech Publications Ltd, Switzerland, www.ttp.net. (ID: 130.203.133.33-17/04/08,16:17:05) Experimental Procedure Ten Physical Acoustics Ltd. (PAL) resonant sensors were attached to a landing gear lever link component (Fig. 1). The lever link was manufactured from aerospace grade steel, which has very high strength but a low fracture toughness which can result in a very small critical crack length. Depending on the stress intensity factor, critical crack lengths in the region of less than 1 mm are possible. Sensors were held in position via magnetic clamps or aluminium 'U' shaped clamps which were glued into position, with foam inserts used to prevent sensor grounding (Fig. 1). The 'U' shaped clamps were required due to the small width of material around the left hand lugs (Fig. 1) and the limited space between the clevis and the specimen. Grease was used as a couplant and installed sensor sensitivity was verified using the Hsu-Nielsen source (pencil lead fractures) technique [4,5]. Feature data (amplitude (dB), counts, duration (μs), energy (atto-J) and rise-time (μs) (Fig. 2)) was recorded using a 16-channel PAL DiSP system with a threshold of 45 dB for the total life of the test investigation. The threshold of the system was chosen to eliminate background noise while still maintaining the ability to record fractures in the component which occur above approximately 50 dB. The threshold however would alter the parameters of rise-time, duration and counts so needs to be kept constant during acquisition. Fig. 1: Test instrumentation and clamping methods 140 Damage Assessment of Structures VII