1 Introduction At present, various instruments based on various measurement principles are used in urban gas metering. However, after analysis and comparison, it was found that these gas flow measurement instruments all have the problem of flow lower limit bottleneck, that is, the flow below the instrument flow lower limit cannot be accurately measured or cannot be measured at all, and the problem of leakage of small flow of urban gas is prominent. In order to solve this problem, we proposed and designed a new type of urban gas flow meter that can realize wide range measurement of urban gas. The flow meter uses the fluid oscillation measurement principle to measure the gas in the high flow range, that is, it uses the measurement principle of vortex flow meter to complete the high flow measurement [1]. Due to the obstruction effect of the vortex generator, the flow field in the gas pipeline becomes extremely complex and it is difficult to analytically obtain the flow field distribution. Therefore, people's understanding of the formation and shedding of vortex after the vortex generator mostly depends on experience and experiments. This is very unfavorable for optimizing the design of new wide range gas flow meters [2][3]. Starting from the classical fluid dynamics equations describing vortex motion, this paper uses the computational fluid dynamics software FLUENT as a platform to numerically simulate the flow field around a square column of a wide-range gas flow meter, and compares the simulation results with the measured data. The results show that the numerical simulation results are in good agreement with the actual flow conditions of gas in the pipeline. 2 Numerical simulation model 2.1 Numerical simulation method for fluid vortex shedding phenomenon The vortex shedding characteristics and force conditions of flow around a single blunt body can be measured by experimental methods or calculated and analyzed by numerical simulation methods. There are two main types of numerical simulation methods: empirical model method and direct flow field simulation method [4][5]. Empirical model method: This type of method does not consider the specific fluid structure and regards the fluid and the oscillating objects in it as a whole system. Then, it is described by a set of suitable model equations, including some coefficients with known results or determined by experience, so that the motion characteristics of the system can be reproduced well after the solution, and a more intuitive understanding of the physical nature of the phenomenon itself can be obtained from the overall solution. At present, the wake oscillator model and related models are widely used and can be used to solve various specific problems. However, these methods rely heavily on experimental results and therefore have certain limitations. Direct flow field simulation methods include various finite difference methods, finite analysis methods, finite element methods, finite volume methods, spectral methods, and various forward and inverse methods based on boundary layer equations, as well as viscous/invisible disturbance methods, discrete vortex methods, or direct solutions to the Navier-Stokes equations. The computational process includes selecting the flow model (listing the governing equations), discretizing the equations, and solving the discrete equations. These methods can provide a detailed description of the entire flow field, but the calculations are quite complex, especially when the Reynolds number is high. The vortex method, finite difference method, and finite volume method are most commonly used in flow field calculations. Since this paper uses the commercial software FLUENT for fluid dynamics numerical simulation, the finite volume method for unsteady fluid dynamics problems will be emphasized. The finite volume method (FVM) is a type of numerical method that directly discretizes integral conservation laws on a selected control volume in physical space. Discretization refers to both dividing the computational domain into a network (or cells) and discretizing the integral conservation law into a system of linear or nonlinear algebraic equations. The Finite Volume Method (FVM) divides the computational domain into several regular or irregularly shaped cells or control volumes. After calculating the flow and momentum fluxes along the normal input (output) through the boundary of each control volume, the FVM performs flow and momentum balance calculations for each control volume separately, yielding the average pressure and velocity of each control volume at the end of the computational period. Therefore, FVM is essentially a regression of the control volume path used to derive the original differential equations, making its physical meaning more direct and clear. If the calculation of cross-boundary fluxes uses only the initial values ​​of the time period, it is an explicit FVM; conversely, when the values ​​at the beginning and end of the time period are involved, it is an implicit FVM. Because the fluxes transported across the interface between control volumes are equal in magnitude and opposite in direction for adjacent control volumes, the fluxes along all internal boundaries cancel each other out for the entire computational domain. For any region consisting of one or more control volumes, and even the entire computational domain, the physical conservation laws are strictly satisfied, there is no conservation error, and discontinuities can be correctly calculated. 2.2 Basic Control Equations of Fluid Mechanics The flow of fluid in a gas pipeline with a wide flowmeter is a time-varying eddy current. Since the fluid is in a turbulent state within the flow range of the numerical simulation, turbulence needs to be treated. There are many turbulence models in engineering calculations, and the selection of different turbulence models will directly affect the calculation results. After multiple trials, this paper selects the RNG k-method of turbulent motion in FLUENT to treat the turbulence in the numerical simulation. The basic idea of ​​RNG is to achieve a coarse-resolution description of the originally very complex system or process through a series of continuous transformations at arbitrary spatial scales. The RNG k-turbulence model is used to simulate the two-dimensional flow field around the square column. The RNG method is applied to the Reynolds-averaged NS equation and the turbulent energy k and its dissipation rate are introduced to obtain the following control equations [6][7]: Continuity equation: The mathematical models of flow around a square column and other blunt bodies are all based on the continuity equation on the law of conservation of mass, the motion equation on the law of conservation of momentum, and the constitutive equation on the first law of thermodynamics. The factor that determines the solution of the equation is the change of boundary conditions [8]. 2.3 Physical model, initial conditions and boundary conditions In this paper, the numerical simulation study of gas flow around a single blunt body in a square column used in pipeline gas metering will be carried out. Its physical model is shown in Figure 2-1, and the solution region is enlarged as shown in Figure 2-2. In this example, in order to compare the application of the DN200 wide-range gas flow meter, the width of the windward side of the square column is set to 0.05m, the width of the solution region is 0.20m, and the length is 0.50m. The vortex generator is placed in the center 0.10m away from the right inlet. To perform numerical calculations on the discretized fluid dynamics equations, boundary conditions, including initial conditions, need to be provided. ① Inlet Boundary: The inlet boundary is set with a velocity. Given the fluid velocity, seven velocity cases are used for comparison with measured values: 5.3 m/s, 10.5 m/s, 20.2 m/s, 29.8 m/s, 39.1 m/s, 50.6 m/s, and 65.7 m/s. ② Outlet Boundary: The outlet boundary condition is set as a pressure outlet, with a pressure of one atmosphere (zero gauge pressure). ③ Wall Conditions: Conditions are provided for both the fluid flow pipe wall and the rectangular prism wall. The same treatment method is used for both types of walls. ④ Mesh Generation: Due to the relatively simple geometric boundary of this problem, GAMBIT is used to divide the solution domain into a structured rectangular mesh. 3. Numerical Simulation Results The numerical simulation adopted a bluff body shape with a rectangular prism. The mesh generation is shown in Figures 2-1 and 2-2. A structured mesh was used, and the inlet boundary was set to a velocity inlet, with a given average fluid velocity. Figure 3 shows the velocity contour map of a typical vortex shedding cycle of the rectangular prism (natural gas) with an average inlet velocity of 39.1 m/s. Figure 3 illustrates that the turbulence model based on the RNG method can capture the instability and severe separation characteristics of the gas flow field around the rectangular prism, describing a complete vortex shedding process in the natural gas medium. Due to space limitations, only the velocity field contour map of a typical vortex shedding cycle of natural gas around the rectangular prism in the pipeline with an average inlet velocity of 39.1 m/s is given here. The dynamic pressure contour map, streamline map, and other velocity and gas medium correlation diagrams are similar. 4. Comparison of Numerical Simulation and Measured Results To compare with the measured results, the fluid medium used in the numerical simulation was also different, including natural gas and manufactured gas, with different physical property parameters. The temperature was set to the measured temperature (294.6 K), and the pressure was one standard atmosphere (101325 Pa). The measuring device used was a wide-range gas flow measurement standard device invested and established by Harbin Institute of Technology. This device was authorized by the Heilongjiang Provincial Bureau of Quality and Technical Supervision in August 2004. After more than two years of calibration and operation, and comparison with the calibration device of the National Crude Oil High-Flow Station (Daqing), it showed good stability and repeatability, with an accuracy of 0.5 grade. The comparison between numerical simulation and measured results is shown in Figure 4. To compare with the measured values, the average fluid velocity was taken as seven different values: 5.3 m/s, 10.5 m/s, 20.2 m/s, 29.8 m/s, 39.1 m/s, 50.6 m/s, and 65.7 m/s. 5. Conclusion This paper starts from the fundamental equations of classical fluid dynamics describing vortex motion and uses the computational fluid dynamics software FLUENT as a platform to numerically simulate the metering flow field of gas flowing around a single blunt body in a pipeline. The simulation results are compared with measured data. The results show that the numerical simulation results are in good agreement with the actual flow conditions of gas in the pipeline. The vortex escape frequency in the metering flow field does not depend on the physical properties of the medium. When the size of the vortex generator is constant, both the numerical simulation frequency and the measured frequency are proportional to the flow velocity of the measured medium. Numerical simulation methods can be applied to guide and optimize the structural design of new gas flow meters. References: [1] Li Chaohui, Dai Jingmin. A gas mass flowmeter with two measurement limits [J]. Measurement Techniques, 2005, 48 (5): 487-491 [2] Xia Taichun. Engineering Fluid Mechanics [M]. Shanghai: Shanghai Jiaotong University Press, 2006. [3] Tong Binggang, Zhang Bingxuan, Cui Erjie (eds.). Unsteady Flow and Vortex Motion. Beijing: National Defense Industry Press, 1993 [4] Sun Zhiqiang, Zhang Hongjian, Huang Yongmei. Numerical Simulation Study on Flow Field Characteristics of Vortex Flowmeter [J]. Automation Instrumentation, 2004, (25), 5:10-13 [5] Wang Yuancheng, Wu Wenquan. Numerical Simulation of Flow Around Blunt Body Based on RNG k-Turbulence Model [J]. Journal of Shanghai University of Science and Technology, 2004, (26), 6:519-523 [6] Yasutaka Nagano, Yoshihiro Itazu. Renormalization group Theory for turbulence: Assessment of the Yakhot Oiszag Smith theory [J]. Fluid Dynamics Research. 1997, 20 (6): 157-172 [7] Speziale G, Thangam S. Analysis of an RNG fased turbulence model for separated flows [J]. Int J Engrfg Sci. 1992, 30 (2): 1379-1384. [8] Peng Jiegang, Fu Xin, Chen Ying. Experimental study on fluid vibration characteristics of double blunt body vortex flowmeter [J]. Journal of Mechanical Engineering. 2002, (26), 8: 519-523 Editor: He Shiping