FINITE SIMILITUDE THEORY FOR FRACTURE AND FAILURE

Abstract

Fracture mechanics is concerned with the response of defect ridden components, which under scaling suffer a phenomenon known as size effect, where the response of a material is influenced by its size. It can be observed that a component can fail before the limiting loads indicated by a smaller version of the same material. The difference can be explained by the more significant defects in the larger model, leading to failure at a lower load. An affliction of current scaling theories founded on dimensional analysis is that they can only accommodate scale/size effects ad hoc. Dimensional analysis ignores them completely and is one of the principal reasons why scaling is not presently a fundamental approach for fracture investigations. To overcome this limitation, a new method of scaling called finite similitude, involving additional experiments at scale, is examined as a possible alternative approach. It is shown in the thesis how two scaled experiments, as opposed to one, can better represent the response of cracked components. The finite similitude theory provides the means to link information from the scaled experiments. The theory provides the means to couple fields like stress, strain, and displacement in fracture mechanics. Although initially tested in quasistatic fracture, its scope is expanded to different fields in dynamic fracture and impact processes. This research presented in the thesis presents evidence of the efficacy of the finite similitude theory in fracture mechanics. Several standard specimens, with crack propagation, under different loading scenarios are examined for different ductile metals. Dynamic fracture mechanics is also investigated by applying the theory to the Charpy impact test, which serves to select different materials in physical modelling for more practical studies. An example of this is presented with a dynamic impact study of a thin-walled pressure vessel further to confirm the approach's relevance to an industrial setting. Through different tests, loading conditions, materials, boundary conditions and investigations, it is demonstrated in the thesis how the two-experiment approach to fracture mechanics provides higher accuracy than a single-scaled experiment.

Date of Award	31 Dec 2022
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Keith Davey (Supervisor) & Roohoolamin Darvizeh (Supervisor)