Thermal cyclic behavior and lifetime prediction of self-healing thermal barrier coatings

Uncertainty Quantification of the Lifetime of Self-Healing Thermal Barrier Coatings Based on Surrogate Modelling of Thermal Cyclic Fracture and Healing

“…where the coefficient c allows to quantify the TGO growth rate for a given thermal cycle (see [5,38]). Accordingly, the variation in TGO growth rate has been quantified with the assumption that c varies uniformly in the range c min ; c max ½ , with c min ¼ 2:5 lm and c max ¼ 3 lm.…”

“…Each sample point corresponds to a computationally expensive simulation. For example, running a simulation with the computational model used in [5] with 10 processors in a cluster takes an average execution time of 60 h (not including pre and postprocessing). As a result, for the present work, a PCE order n = 2 has been chosen such that, in combination with the chosen number of input random variables K ¼ 5 and the oversampling ratio n p ¼ 2, it requires N ¼ 42 simulations.…”

“…The self-healing TBC micromechanical model developed in [5] has been utilized in order to obtain the number of cycles to TBC failure as a consequence of exposure to thermal cyclic behavior. Most of the applications of TBCs involve a typical thermal loading cycle consisting of three distinct phases: heating, dwelling and cooling.…”

“…As a result, the TBC lifetime is measured by the number of thermal cycles that the TBC can sustain until spallation. In order to arrest the cracks and potentially improve the TBC lifetime, researchers have investigated the implementation of self-healing particles within the coatings [3][4][5][6]. One of the most widely studied methods of incorporating self-healing properties in a TBC is to include encapsulated healing agents within the top coat, which is the location where cracks typically appear in TBCs (see Fig.…”

“…For the present work, the particle-based self healing thermal barrier coating system presented in [5] is used as the main tem- plate to generate a surrogate model for its lifetime. This is due to the fact that the corresponding computational framework was developed to simulate the complete thermal cycling evolution of the material system until failure, thus allowing to predict its lifetime.…”