Skip to main content

Understanding material failure by machine learning analysis of pattern strains

Primary supervisor

David Dowe


  • Michael Preuss

We examine data of metals with a desire to examine directions of strains and deformation.

Data available to us includes electron microscope images of materials, places where deformation has occurred and the directions and amounts of such deformation.  Much of this data is in static planar images.  But there is also data for out-of-plane movement (where the deformation has not entirely been in the plane) and there is also some sequential chronological time series data, where we can see a (kind of) movie of deformation taking place.

Student cohort

Double Semester


As a first step, we wish to infer a distribution of the directions of deformation displacement and the length of such deformations (or displacements).

As later steps, we will then look at time series data (or movies of images) and also at out-of-plane 3-dimensional deformations.

This will enable us to better understand deformation - with a view to anticipation, remedy and possibly prevention.


  Chen, Z. & Daly, S. H. Active Slip System Identification in Polycrystalline Metals by Digital Image Correlation (DIC). Exp Mech 57, 115–127 (2017).

 Comley, Joshua W. and D.L. Dowe (2005). ``Minimum Message Length and Generalized Bayesian Nets with Asymmetric Languages'', Chapter 11 (pp265-294) in P. Gru:nwald, I. J. Myung and M. A. Pitt (eds.), Advances in Minimum Description Length: Theory and Applications, M.I.T. Press (MIT Press), April 2005, ISBN 0-262-07262-9. [Final camera ready copy was submitted in October 2003.]

  Dowe, D.L., J.J. Oliver and C.S. Wallace (1996). MML estimation of the parameters of the spherical Fisher Distribution. In S. Arikawa and A. K. Sharma (eds.), Proc. 7th International Workshop on Algorithmic Learning Theory (ALT'96), Lecture Notes in Artificial Intelligence (LNAI) 1160, pp213-227, Sydney, Australia, 23-25 October 1996. [pp213-219, pp220-227; p213, p214, p215, p216, p217, p218, p219, p220, p221, p222, p223, p224, p225, p226, p227]

  Gioacchino, F. D. & Fonseca, J. Q. da. An experimental study of the polycrystalline plasticity of austenitic stainless steel. Int J Plasticity 74, 92–109 (2015).

  Gioacchino, F. D. & Fonseca, J. Q. da. Plastic Strain Mapping with Sub-micron Resolution Using Digital Image Correlation. Exp Mech 53, 743–754 (2013).

  Lunt, D. et al. Comparison of sub-grain scale digital image correlation calculated using commercial and open-source software packages. Mater Charact 163, 110271 (2020).

  Wallace, C.S. (2005), ``Statistical and Inductive Inference by Minimum Message Length'', Springer  (Link to the preface [and p vi, also here])

  Wallace, C.S. and D.L. Dowe (1994b), Intrinsic classification by MML - the Snob program. Proc. 7th Australian Joint Conf. on Artificial Intelligence, UNE, Armidale, Australia, November 1994, pp37-44

  Wallace, C.S. and D.L. Dowe (1999a). Minimum Message Length and Kolmogorov Complexity, Computer Journal (special issue on Kolmogorov complexity), Vol. 42, No. 4, pp270-283

 Wallace, C.S. and D.L. Dowe (2000). MML clustering of multi-state, Poisson, von Mises circular and Gaussian distributions, Statistics and Computing, Vol. 10, No. 1, Jan. 2000, pp73-83

Required knowledge

At least first year undergraduate mathematics, preferably more.

An ability to program.

Statistics, machine learning and/or data science at least to the level of an undergraduate degree.  At least an interest in angular data (i.e., distributions of angles - such as, e.g., hospital arrival distributions around a 24-hour clock).

At least an interest in materials, strains and material deformation.