An Improved Pressure Drop Prediction Model Based on Okiszewski's Model for Low Gasliquid Ratio Two-Phase Upward Flow in Vertical Pipe

Compared with experimental data of pressure drop of gas-liquid two-phase upward flow with low gas-liquid ratio in vertical pipe, the mean relative error of pressure drop predicted by Orkiszewski's model is 63.62%, the maximum relative error of the model is 98.07%. This paper first introduces the process of acquiring experimental data, then, combined with the experimental data, it is pointed out that the Orkiszewski's model has a small error in predicting the annular-mist flow pattern, but a large error in predicting the slug flow pattern and annular-slug transition flow pattern. The author analyzes the structure of Orkiszewski's model and points out that the formula of liquid distribution coefficient in slug flow pattern is complex and very important. In this paper, a new threshold value of liquid distribution coefficient is proposed and an improved Orkiszewski's model is obtained by particle swarm optimization. The calculated results of experimental data show that the average relative error of the new model is reduced to 25.28%, and the average relative error of the new model can be reduced from 76.17% to 17.21% for the slug flow pattern with continuous oil phase and total flow velocity greater than or equal to 3.048m/s.