Abstract

Accurate prediction of gas-liquid pressurization performance of multistage centrifugal pumps is of vital importance to evaluate oil/gas production. In this paper, interstage boosting pressure and pressurization prediction of a 25-stage centrifugal pump under air-water condition were studied experimentally. A method was developed to measure gas volume fraction under high-pressure condition by using double-ring probes. Moreover, a new method was proposed to predict boosting pressure of multistage pump by training artificial neural network of boosting pressure in 3-stage booster units, updating interstage gas volume fraction and pressure. The assumption of isothermal compression was proposed to calculate the interstage gas volume fraction, which was verified by measuring gas volume fraction at pump outlet using double-ring probes with errors lower than ±15%. The increasing stage number led to the decrease of interstage gas volume fraction and better pressurization performance in downstream stages. Besides, the head degradation in downstream stages was obviously delayed with the increase of rotational speed due to the decreasing interstage gas volume fraction. For inlet pressure 0.5–2 MPa, rotational speed 2500–3500 rpm, liquid flow rate 14–26 m³·h−1 and inlet gas volume fraction 0–44.0%, the relative errors of the predicted boosting pressure of pumps were less than 20%.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call