University of Southampton
FEM techniques for high stress prediction in accelerated fatigue simulation
Abstract
dc:description.abstractThis research is motivated by the need to accelerate fatigue analyses of complex mechanical systems, characterized by large numerical models, long time integration intervals and dynamic response. Numerical modelling is an essential tool for fatigue life determination of such systems. Despite the increased affordability of ever more capable hardware and software resources, however, the computational requirements combined with competitive time-to-market constraints remain a continued challenge for the structural durability engineer. In such an environment, the early detection of fatigue critical areas can lead to an informed reduction of the problem size, and a subsequent decrease in solution time and costs.<br/>The investigation focuses on the applicability and merits of accelerated simulation procedures aimed at the fast identification of a subset of critical regions, also known as hotspots. The work presents the theory and a numerical validation study to support a novel method for the identification of fatigue hotspots, to be determined prior to entering the time domain problem. An original statistical assessment of risks and benefits in fatigue simulation acceleration provides the means for damage prediction quantification and comparison.<br/>The proposed acceleration method is particularly suitable during the initial design stages of heavy-duty durability computations of complex mechanical structures typical of the transport and general machinery industry. The technique is applied to real life industrial cases in a comparative assessment with established practices, outlining applications, benefits and boundaries.
Degree
thesis:*- Name dc:type.qualificationname
- Ph.D.
- Level dc:type.qualificationlevel
- doctoral
- Grantor dc:publisher.institution
- University of Southampton
- Year dc:date.issued
- 2017
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Veltri, Marco
- Advisor dc:contributor.advisor
-
- Ferguson, Neil
- ORCID person:identifier.orcid
Rights
- Language dc:language
- English, eng