A Hybrid Modelling Approach to Developing Digital Twins of an Ultra-supercritical Steam Turbine Control Stage

Abstract

Modelling for digital twins is a key technology in smart power plants. A control stage is a key component for pressure and flow control in steam turbines of power plants. Its modelling and simulation for digital twins are important measures for steam turbine operation diagnosis and optimization. However, sequential operation modes of valves in a control stage result in dramatic changes in thermal and mechanical behaviours of the control stage, and modelling the change in behaviours still remains a challenge due to lack of measurements. In this paper, a hybrid modelling method based on the first-principle mechanism and actual operation data is proposed. Inlet flow of a control stage are obtained by calculating flow coefficient at the full range of sequential operation of valves. Then, characteristic functions of a control stage are obtained using the first-principle mechanism. In order to verify the reliability of the proposed method, a control stage of an ultra-supercritical steam turbine with typical sequential valves operation mode is modelled and simulated. Results show that the average value of the simulation relative error is less than 1%, which proves that the method has high simulation accuracy and reliability and lays a foundation for application of digital twins in smart power plants.