基于回归模型的掺氢天然气压缩因子计算方法

Compression factor calculation method of hydrogen⁃doped naturalgas based on regression model

压缩因子回归模型结构简单、计算快捷,在常规天然气压缩因子计算中更易程序化。文中研究掺氢天然气的应用并与其他模型对比。回归模型包括三次方、幂指数等多种形式,如HY(Hall-Yarborough)、RK(Redlich-Kwong)模型等,各模型适用的压力和温度范围各不相同。采用文献391组掺氢天然气压缩因子实验数据,对比分析7种回归模型在不同掺氢比例、压力和温度范围下的计算精度,并与GERG-2008状态方程进行对比。其中压力范围为0.1—10 MPa、温度为210—350 K、掺氢摩尔分数0—100%。结果表明:RK的平均相对误差为1.36%、HY为1.46%,均小于GERG-2008(1.47%);在T>298.15 K时,HY模型的误差最小;在T<298.15 K时,RK模型的计算误差最小;在不同掺氢比例下,HY和RK的最大平均相对误差分别为2.73%和1.58%,小于GERG-2008(3.54%)。相较于其他回归模型,建立的HY-RK组合回归模型更适用于掺氢天然气的压缩因子计算。

The compression factor regression model is simple in structure,fast in calculation,and easier to program than conventional natural gas compression factor calculation.Hydrogen⁃doped natural gas was studied compared with other models.Regression models include various forms in cubic and power exponents,such as HY(Hall⁃Yarborough),RK(Redlich⁃Kwong)model etc.Each model is applicable to different pressure and temperature ranges.Based on the experimental data of 391 hydrogen⁃doped natural gas compression factors in literature,the calculation accuracy of 7 regression models under different hydrogen⁃doped ratios,pressures and temperatures was compared and compared with GERG⁃2008 equation of state.The pressure range was 0.1⁃10 MPa,the temperature was 210⁃350 K,and the hydrogen mole fraction was 0⁃100%.The comparison results show that the average relative error of RK is 1.36%and HY is 1.46%,both of which are smaller than that of GERG⁃2008(1.47%).When T>298.15 K,the error of HY model is the smallest.When T<298.15 K,the calculation error of RK model is the smallest.The maximum mean relative errors of HY and RK are 2.73%and 1.58%respectively,which is smaller than that of GERG⁃2008(3.54%).Compared with other regression models,the HY⁃RK combined regression model is more suitable for the compression factor calculation of hydrogen⁃doped natural gas.