Prediction of the internal flow state of the hydraulic units in large pumping stations based on digital twins

Key Points

• This research aims to develop an efficient method to predict the internal flow states of hydraulic units using digital twins.
• Developed a flow state prediction method based on a surrogate model in a Simulink simulation environment.
• Constructed a physical field reduced-order model for dynamic interactions.
• Conducted system validation on a physical model experimental platform.
• Average relative error in predicting static pressure, velocity, and total pressure is maintained below 5%.
• Turbulence intensity prediction shows an average relative error of 8.5%.
• Simulation time per operational point averages approximately 0.1 seconds, indicating significant efficiency improvement.