Blade vibrational amplitude in aero-engines bladed discs can be significantly amplified due to the uneven distribution of blade mechanical properties from the uncertain sources in manufacturing process, assembly tolerances and operational wear and tear. The possibly ensuing higher vibrational level on the blades is regarded as one of leading causes for the shortened high cycle fatigue life of a whole bladed disc system. This mistuning problem therefore has attracted tremendous attentions from aero-engine industries and research communities since 1960s. One of the key aspects of mistuning is how to quantify efficiently the effect of those uncertainties on the maximum blade dynamic response in a bladed disc. However, in spite of a great deal of research efforts poured into this topic, the mistuning problem is still considered a challenge from the design perspective, because of the unacceptable computational costs associated with the related dynamic analysis, even when using state-of-the-art reduced order models. The computational costs would be further increased when physical aspects like the non-linearity from the large geometry deformation of blades, contact friction in the dovetails and aerodynamic couplings are taken into account. The complexity of the simulation problem associated to mistuning increases when reliability analysis is also required. The main achievements of the present study . consist in the development of deterministic and probabilistic based reduced ordvr techniques to enhance the computational efficiency for linear . mistuning analysis of bladed disc systems. A novel parametric reduced order model using a simplified structural layout is presented to represent the equivalent dynamic behavior of a typical aero-engine blade. A comparative case study is then carried out to investigate the application of using this novel beam frame assembly to increase the computational efficiency for mistuning analysis of a whole bladed disc system. The results obtained from the beam frame assembly is benchmarked by three state-of-the-art finite element based reduced order models. After that, this study looks into the application of novel stochastic techniques for the reliability analysis of mistuned bladed discs in order to reduce the number of samples required by classical Monte Carlo Simulations. The feasibility of using subset simulation techniques is assessed for probabilistic analysis of a mistuned bladed disc system. The work finally investigates the efficient matrix inversion techniques to increase the ~omputational efficiency of stochastic analysis of bladed discs. Two classical matrix inversion techniques, namely Neumann expansion techniques and Sh~rman Morrison formula, are assessed respectively. Based on the results of the assessment, a more robust inversion technique based on Neumann expansion method and matrix factorization techniques is proposed and validated through two case studies.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:702730 |
| Date | January 2015 |
| Creators | Yuan, Jie |
| Publisher | University of Bristol |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
Page generated in 0.0016 seconds