基于高斯过程回归的调节阀组合 湿气计量方法研究.

Saved in:
Bibliographic Details
Title: 基于高斯过程回归的调节阀组合 湿气计量方法研究.
Alternate Title: Research on wet gas metering for regulating valve combination based on Gaussian process regression.
Authors: 梁裕如1 190187691@qq.com, 罗腾跃1, 孟祥振1, 韩建红1, 何鹏1, 王江源1, 艾昕宇1
Source: Chemical Engineering of Oil & Gas / Shi You Yu Tian Ran Qi Hua Gong. Jun2026, Vol. 55 Issue 3, p125-136. 12p.
Subject Terms: *Two-phase flow, *Flow measurement, *Gas extraction, *Measurement errors, *Pressure regulators, *Gaussian processes, *Flow meters
Abstract (English): Objective This studyaimsto address the common issue of large measurement errors in gas-liquid two-phase flow caused by low single-well production and fluctuating liquid production in a gas field. Method A wet gas dual-parameter combined measurement device based on a wellhead regulating valve was designed. Through air-water two-phase flow experiments, the variations of the valve differential pressure and pressure loss under different regulating valve openings, operating pressures, gas superficial velocities, and volumetric liquid fractions were analyzed.The valve differential pressure ratio was introduced, and the relationships among valve opening, differential pressure ratio, gas Froude number, gas over-reading, and liquid-to-gas mass flow rate ratio were investigated based on experimental results.Gaussian process regression method was used to establish models for gas over-reading and liquid-to-gas mass flow rate ratio, which were then solved iteratively to calculate the gas and liquid flow rates. Result The proposed combined metering model covers a wide range of gas-liquid two-phase flow patterns, with controllable prediction errors for gas and liquid flow rates. The average relative errors for gas and liquid flow rates are 5.43% and 15.39%, respectively. Conclusion The proposed meteringmodel effectively reduces the measurement errorsin gas-liquid two-phase flow, offering a new methodology forwet gas metering, and providing technical support for the low-cost development and refined management of gas fields. [ABSTRACT FROM AUTHOR]
Abstract (Chinese): 目的 解决某气田由于单井产量低、产液波动而普遍存在气液两相计量误差大的问题。方法 设计基于井口调节 阀的湿气双参数组合测量装置, 通过空气−水两相流室内实验, 分析不同调节阀阀门开度、运行压力、气相表观流速和体积 含液率条件下调节阀差压和压损的变化规律, 引入阀门差压比, 依据室内测试结果对阀门开度、阀门差压比、气体弗鲁德 数、气相虚高以及液气质量流量比之间的关系进行研究, 采用高斯过程回归方法分别建立气相虚高模型和液气质量流量 比双参数模型, 经迭代求解计算气相流量和液相流量。结果 该调节阀组合计量模型的气液两相流型覆盖范围较宽, 气相 流量和液相流量预测误差可控, 气相流量平均相对误差为 5.43%, 液相流量平均相对误差为 15.39%。结论 该计量模型有效 降低气液两相流量测量误差, 为湿气测量提供了一种新思路, 可为气田低成本开发和精细化管理提供技术支撑。 [ABSTRACT FROM AUTHOR]
Database: Energy & Power Source
Description
Abstract:Objective This studyaimsto address the common issue of large measurement errors in gas-liquid two-phase flow caused by low single-well production and fluctuating liquid production in a gas field. Method A wet gas dual-parameter combined measurement device based on a wellhead regulating valve was designed. Through air-water two-phase flow experiments, the variations of the valve differential pressure and pressure loss under different regulating valve openings, operating pressures, gas superficial velocities, and volumetric liquid fractions were analyzed.The valve differential pressure ratio was introduced, and the relationships among valve opening, differential pressure ratio, gas Froude number, gas over-reading, and liquid-to-gas mass flow rate ratio were investigated based on experimental results.Gaussian process regression method was used to establish models for gas over-reading and liquid-to-gas mass flow rate ratio, which were then solved iteratively to calculate the gas and liquid flow rates. Result The proposed combined metering model covers a wide range of gas-liquid two-phase flow patterns, with controllable prediction errors for gas and liquid flow rates. The average relative errors for gas and liquid flow rates are 5.43% and 15.39%, respectively. Conclusion The proposed meteringmodel effectively reduces the measurement errorsin gas-liquid two-phase flow, offering a new methodology forwet gas metering, and providing technical support for the low-cost development and refined management of gas fields. [ABSTRACT FROM AUTHOR]
ISSN:10073426
DOI:10.3969/j.issn.1007-3426.2026.03.014