Speakers
Description
The research aims to develop a procedure, which can be used in the design and operation of nuclear power plants, which can predict the change in state of dangerous system components under specific load conditions. Fatigue life calculation methods available with simulation technology were applied to establish the procedure, and their outputs were incorporated into the output variables of a digital twin model of a steam generator. System simulation enables to perform the investigations in a more complex way, taking into account the combined effect of the system components, and thus enabling the output of the whole system to be examined for given inputs. A key stage in building a system simulation is developing a reduced-order model, a digital twin of the 3D system that captures its behaviour. Firstly, a geometric model of the steam generator, which is part of the block, must be generated. This is followed by the pre-processing of the finite element model using various mathematical modelling techniques. Finally, the digital twin model is created by running the finite element model at predefined learning points. Two simulation models have been used to calculate the fatigue life. The lifetime was calculated using a shell model built on the pressure vessel's midsurface, while the crack propagation was investigated using a body model and the submodelling method. A static, nonlinear five- pole digital twin model was constructed, which describes the finite element model behaviour for thermal and mechanical load combinations. With this surrogate model, it is possible to monitor the continuous state variation of the pressure vessel and, through this, to plan preventive maintenance. The outcome of this research is a simulation-based digital twin model for industrial applications in power plants. Implementing this model in the reliability-centred maintenance strategy makes a cutting-edge tool available that can optimise maintenance tasks.