Background: Centrifugal pump is widely used in industrial production as a key fluid conveying equipment. The transient characteristics during its start-up process may lead to vibration, noise and efficiency loss. Therefore, it is very important to optimize the start-up process. Objective: Optimize the structural parameters of the centrifugal pump, with the initial pulse values of both the head and efficiency as the optimization objectives, to enhance its transient performance. Methods: The centrifugal pump was numerically simulated by computational fluid dynamics method. In the optimization process, the BP neural network is established, and the genetic algorithm is used to optimize the parameters, which improves the efficiency and accuracy of parameter tuning. Results: The optimized head pulse value is reduced by 4.39 m, and the efficiency pulse value is reduced by 5.65 %. Make the start-up process of centrifugal pump more stable. Conclusion: Among the parameters that affect the stability of the centrifugal pump start-up process, it can be obtained that the three factors of impeller outlet width, blade outlet angle and blade wrap angle have the greatest influence on the fluctuating head and efficiency of the centrifugal pump. other: none
Read full abstract