BASIC PRINCIPLES OF ACOUSTIC EMISSION TOMOGRAPHY

The present paper describes the basic principles of acoustic emission tomography. This method uses acoustic emission events as point sources and combines the usual iterative localisation algorithm of acoustic emission testing with algorithms for travel time tomography like ART (algebraic reconstruction technique). The procedure is equivalent to the solution of the generalised inverse localisation problem in locally isotropic heterogeneous media and leads to a new imaging technique where in addition to the source positions the volume of the specimen is visualised in terms of a locally varying wave speed distribution. It is shown by numerically obtained data sets that the algorithm leads to a more accurate localisation of acoustic emission events and offers totally new perspectives for acoustic emission imaging and for acoustic tomography in general. 1. Introduction and outline Localisation algorithms in acoustic emission (AE) testing mostly use the assumption of a homogeneous background medium with constant wave speed in order to determine the location of acoustic emission events. However in practice, the structures under investigation are inhomogeneous in many cases, i.e. wave speeds are changing in space and time due to heterogeneities of the microstructure (e.g. grains and pores), the effect of structural components (e.g. tendon ducts in concrete), and material changes caused by the damage mechanism itself (e.g. crack growth). These heterogeneities limit the accuracy of source localisation algorithms. In order to overcome these drawbacks the usual localisation algorithm of acoustic emission testing can be combined with travel time tomography by using the AE events as acoustic point sources. In this context a re-localisation, i.e. an update of the current source positions, has to be performed after each tomographic inversion resulting in an iterative procedure with alternating steps of source localisation and tomography. This method is in principle known from geophysics where earthquakes are located and used for tomographic imaging of the earth’s interior. Chapter 2 first summarises the fundamentals of iterative AE localisation and describes how the underlying equations can simply be generalised to heterogeneous media. Chapter 3 briefly describes the different approaches of tomographic imaging with diffracting and nondiffracting sources paying particular attention to algebraic reconstruction techniques (ART). Chapter 4 shows how the two concepts of localisation and travel-time tomography can be combined resulting in an iterative algorithm for acoustic emission tomography called AETOMO. In chapter 5 numerical AE data obtained by the elastodynamic finite integration technique (EFIT) are used to demonstrate the physical soundness of the proposed method. It is further shown that the AE-TOMO approach offers totally new perspectives, not only for acoustic emission imaging but also for traditional acoustic tomography since AE events can also be produced artificially at the outer surfaces of the specimen under investigation (e.g. by pencil lead breaks or hammer impacts). Finally an outlook is given how the present algorithms could be improved by using further advanced tomographic imaging techniques and how acoustic emission tomography can be verified experimentally.