Preliminary Research on Digital Twin of Main Thermodynamic Systems in Nuclear Power Plant for Thermal Performance Monitoring

The performance degradation and latent fault of equipment may cause unplanned outage or power reduction in nuclear power plant, as a result, the operation and maintenance costs increase. By developing a nuclear power plant digital twin, some potential abnormalities can be detected and predictive maintenance can be performed to avoid equipment failure. So, a digital twin for main thermodynamic systems in an operating nuclear power plant is built by using EPOS code to monitor its thermal performance. Firstly, a thermodynamic system simulation model containing primary loop, secondary loop and cooling system is developed according to design documents, and the design parameters under full load are used to calculate all the equipment’s nominal values. Then the thermal performance of specific equipment is identified under partial loads. Some fitting process are needed to find the correlations between different variables and the fitting curves will be embedded into the equipment model if necessary. Moreover, a data reconciliation technique is used to assimilate the system model depending on measurements of operating nuclear power plant and the digital twin is gained at last. The result shows that the digital twin can represent the thermal performance of real plant very well although a few of measure inconsistencies are detected during the data reconciliation process.