Investigation on Adaptability of Physical Property State Equation Model for Hydrogen-Blended Natural Gas
Jiakang Wang,Xin Ouyang,Cheng Lei,Shiyao Peng,Zhiheng Wang,Jinhua Wang
DOI: https://doi.org/10.1016/j.ijhydene.2024.07.335
IF: 7.2
2024-01-01
International Journal of Hydrogen Energy
Abstract:During the transportation of hydrogen-blended natural gas (HBNG) through the natural gas pipeline network, the accuracy of the physical property model for real gas greatly influences the reliability of aerodynamic performance of compressor systems and the hydrothermal characteristics of pipeline networks. In this study, the adaptability of various equations of state (EOS) in predicting physical property parameters including compressibility factor, enthalpy difference, entropy difference, specific heat capacity, specific heat ratio and viscosity of natural gas (NG) and HBNG is evaluated based on the experimental data and NIST REFPROP firstly. The properties calculated using NIST REFPROP are compared with those from the EOSs, such as cubic EOSs (Redlich-Kwong EOS (RK), Soave-Redlich-Kwong EOS (SRK), Peng-Robinson EOS (PR)) and Soave-Benedict-Webb-Rubin (BWRS) EOS. After confirming the reliability of NIST REFPROP as the calculation benchmark, the accuracy of the four EOSs in calculating the physical properties of HBNG is assessed within the temperature range of 250 similar to 350 K, pressure range of 1 similar to 25 MPa, and hydrogen doping ratio of 0%-30%. Besides, the mixing rules of different equations of state (EOSs) are compared based on the NIST REFPROP calculation results. The results indicate that, except for specific heat capacity calculation, the mixing rule has little influence on the prediction accuracy of cubic EOS models, and the maximum calculation error between different mixing rules of the same EOS model does not exceed 0.5%. Under specific conditions, the traditional mixing rule exhibits better prediction accuracy and can be prioritized due to its simplicity. Furthermore, the pressure has a significant impact on the accuracy of EOS models, with the prediction accuracy decreasing as the pressure increased. The influence of temperature on the prediction accuracy is less significant than that of pressure. Generally, lower temperatures exhibit lower prediction accuracy, while higher temperatures show higher prediction accuracy. Among the cubic EOS models, the RK equation performs better than the SRK and PR EOS models. While the cubic EOSs show minimal differences in predicting specific heat ratio with an average relative error below 1%, the RK EOS shows significant advantages in predicting other physical parameters. The simple structure of RK EOS meets the calculation requirements for HBNG engineering. The comprehensive comparisons reveal that, the BWRS EOS has higher accuracy in predicting various physical properties parameters, with the maximum average relative error not exceeding 5% for each physical property parameter, and its stability is also better compared to the cubic EOSs. The further analysis shows that, the calculation of viscosity is influenced not only by the EOS but also by the viscosity model. When calculating viscosity using the LU-Model, the PR EOS provides the most accurate results. In contrast, when utilizing the LGE-Model or LBC-Model, the BWRS EOS demonstrates superior accuracy compared to other EOSs.