Abstract
The distribution efficiency of the check valve directly affects the performance of the reciprocating pump. The flow coefficient is an important evaluation criterion for the flow capacity of the valve port, and it is of great significance to the design of the valve structure and even the control of cavitation. The traditional design uses flow coefficient as a fixed value, however, the flow state and flow coefficient will change during valve movement. In this study, a three-dimensional transient computational fluid dynamic model for high-pressure and large-flow reciprocating pump valve is established. The dynamic grid simulation method of coupling for the valves and plunger is innovatively proposed, and experimental verification was carried out. The flow state and pressure characteristics for the suction valve under high outlet pressure are analyzed, and the change rule of the suction coefficient is found. The research results show that the initial pressure of the plunger cavity prolongs the negative pressure duration of the plunger cavity when the valve is opened and increases the risk of cavitation of the valve. During the process from valve opening to maximum lift, the suction coefficient first increases and then decreases, and finally remains between 0.5 and 0.6. When the valve lift is large, two-stage throttling occurs, and the flow state will change from cylindrical jet on the lower surface of the valve disk to annular jet, which is beneficial to improve the suction coefficient.
An Integrated Approach Based on Improved CEEMDAN and LSTM Deep Learning Neural Network for Fault Diagnosis of Reciprocating Pump
