Abstract: Metering the gas flow-rate in a wet gas flow by using a Venturi meter requires a correction of the meter over-reading to account for the liquids effect. This paper lists eight correlations and analyzes their characteristics and limitations. A new correlation was proposed based on the separated flow theory and neural network, and was verified by experiment.The error of this correlation was within 5%. Finally the performance of the new correlation and traditional correlations was compared with new independent data from the Tianjin University Wet Gas Loop and NEL report. The results showed that new correlation could predict wet gas Venturi meter over reading accurately with the conditions of pressure from 0. 15 MPa to 6. 0 MPa, diameter ratio β from 0. 4 to 0. 75, gas densiometric Froude number from 0. 5 to 5. 5, the modified Lockhart-Maretinelli parameter varied from 0. 002 to 0. 3, gas to total mass flowrate ratio from 0. 5 to 0. 99.

摘要： 文丘里流量计用于湿气测量需要对因液相的存在而产生的读数偏高（虚高）进行修正。文中分析了8个经典的文丘里湿气测量虚高模型的特点与局限，用分相模型理论结合神经网络曲线拟合方法得到了一个新的文丘里湿气测量模型，并对模型进行了验证，结果证明模型对文丘里湿气测量虚高特性的预测较准确，误差在5%以内。模型的适用范围宽，经大量实验验证并与8个传统模型作了对比，结果表明模型在压力P〖HTSS〗为0. 15～6. 0 MPa，孔径比β〖HTSS〗为0. 4～0. 75，气相Froude数Fr g为0. 5～5. 5，LM参数X为0. 002～0. 3，质量含气率x为0. 5～0. 99范围内均得到满意的预测结果。