Multi-objective optimization of guide-vane closure scheme with the consideration of pressure pulsations during the pump power trip of pump-turbines

To overcome the challenges associated with ensuring a power supply–demand balance, pumped-storage technology has been widely applied to regulate grid load fluctuations and enhance power system stability. This study focused on mitigating the pressure pulsations and complex flow phenomena caused by guide-vane closure during the pump-trip process of pumped-storage units. A multi-objective optimization framework based on the second-generation non-dominated sorting genetic algorithm was proposed, integrating a one- and three-dimensional coupled numerical model and a modified Suter-transform method to refine four-quadrant characteristic curves. The two-stage guide-vane closure pattern was optimized with the following four objectives: minimizing the spiral-casing pressure peak, draft-tube pressure fluctuation, unit speed overshoot, and pressure fluctuation. A response surface surrogate model was introduced to accelerate computational convergence. The results demonstrated that the optimized closure pattern effectively reduced the maximum reverse rotation speed to below the rated unit speed. In addition, the duration of the pump-braking mode, dominant frequency components of the pressure pulsations, and axial hydraulic-thrust fluctuations were significantly suppressed. Short-time Fourier transform and flow-field vector diagrams revealed that suppressing the pressure vortices in the vaneless zone and backflow at the runner inlet effectively mitigated the dynamic hydraulic loads. This method provides a practical approach for enhancing the operational stability and prolonging the service life of pumped-storage units.