Adaptive Risk Refinement Methodology for Gas Turbine Engine Rotor Disks
Jonathan P. Moody, Harry R. Millwater, Michael P. Enright · 2008
Probabilistic fracture mechanics is a well-established method for predicting the probability-of-fracture (POF) of gas turbine engine rotor disks subject to low-cycle fatigue. An adaptive risk refinement methodology (ARRM) was developed to automate zone discretization and refinement that are typically associated with this approach. ARRM generates initial meshes using an adaptive nodal selection feature designed to optimize accuracy and efficiency. Adaptive mesh refinement (AMR) is performed until a converged risk solution is obtained. ARRM employs several techniques including skeletonization, constrained Delaunay triangulation (CDT), superparametric interpolation, and adaptive mesh refinement. A numerical example is provided to illustrate the effectiveness of the methodology. Nomenclature Δσ = stress range ds = shifting distance F = fracture of an initial anomaly located in a zone K = number of zones le = projection path length m = iteration number P[Ai] = probability of having an anomaly in a zone P[Bi|Ai] = conditional probability-of-fracture given an anomaly in a zone P[Fi] = probability-of-fracture of an initial anomaly located in a zone Pdisk = probability-of-fracture in a disk POF = probability-of-fracture I.