| Popis: |
The possibility to use real operational data as an input for lifetime assessment methods is a key requirement in terms of both service applications as well as within the design of components by underlying specific service relevant scenarios. To address this, so called “Constitutive Viscoplastic Material Models” have been developed which represent a more generalized alternative for assessing turbo machinery components which undergo an irregular creep-fatigue loading. Based on several experimental and theory related national research programs, performed within the German working group W10 in the last years, the current status of the model development and the performance potentials are summarized in this paper. Within the first part, the general and developed model structure of one candidate material model is introduced by discussing different aspects of the equation system together with the specific practical related aspects. Secondly, the validation of this constitutive material model is shown by comparing the model results with a set of conducted complex experiments, like ansiothermal service like experiments performed on smooth, notched and biaxially loaded cruciform test samples. As the third focus, the applicability and the potential of using such a model for assessing real components will be discussed e.g. by introducing extrapolation or cycle jump concepts which allows to majorly reduce the calculation time without decreasing the result accuracy significantly. Finally, future potentials will be introduced with the goal to use such sophisticated models to train meta-models and finally allow for a machine-learning based on-site and service related on-line component assessment. |
