Research On Dynamic Simulation And Chocked Flow Characteristic Of High-pressure Natural Gas Network

Natural gas is a new clean energy resource, which has been attached great importance all over the world. Following the transition from the extensive economy and the adjustment of energy policy, the ratio of natural gas in the energy structure is increasing constantly with longer transmission pipeline and high-pressure gas pipe, and larger scaled gas storage facilities. The scientific calculation and design of the gas pipeline and the correct judgement and adjustment in the actual operation will directly affect the stable and safe operation of high-pressure gas pipeline and the normal supply of gas. Due to lacking the effective method and software in calculation of high-pressure gas, the hydraulic calculation is usually based on the constant flow in stead of variable flow. Thus there is a marginal difference with the actual pipe pressure and flow. Due to lacking the simulation tool, it is not possible to accurately assess the various working conditions of pipeline and control the pipeline operation. Therefore, it is necessary to conduct a deep study on the actual working conditions of high pressure pipeline, find out the operation laws of various parameters through simulation, and put forward the corresponding control strategy.This article focuses on the study of flow property of high-pressure gas pipeline, high-pressure gas storage, and chocking flow by adopting the methods of mathematical model, numerical analysis, dynamic simulation, and case studies. The major conclusions are following:1. The mathematical model is simplified as following:(1) constant temperature for the gas pipeline; (2) ignorance of altitude difference; (3) ignorance of linear velocity changes; (4) ignorance of mass flow changes; (5) linearization based onИ.А.Чарныйlinearization method and least square method. As an usual practice, the convective term, inertia, and neutral are ignored in the calculation of variable flow. In this article, the mathematical model is set up based on momemtum equation and energy equation without ignorance of any terms. Mothods of characteristic curves and implicit centered finite difference are used to solve the variable flow. Numerical analysis is used for the unsteady flow of ling-distance transmission pipeline, which is not necessary to simplify model equations with complicated mathematical conversions and can have better result. It will take longer time to solve the unknown parameters since the system of nonlinear equations is extremely large. In order to ensure numerical solution is absolutely stable, the knot timing and spacing should not be too small. Numerical solution takes the heat exchange between gas and outside environment into consideration and therefore can calculate the temperature distribution along the pipeline. Due to the fluctuations of gas supply and consumption, the pressure, flow, and temperature of the gas pipeline will be fluctuated from time to time. If there are nodes in the network, the changes at the nodes will be the most serious.2. In contrast with the implicit method, the main advantage of the characteristic method is to display boundary conditions with less required computer storage capacity. But the ratio between time step and pipe section is limited to a certain stable condition, the time step should be very small. If it is too long, it requires more time to calculate. When the characteristic method is used to solve the gas pipeline, difference equation is set up with reverse marching algorith differential grid. The system of ordinary differential equation is solved based on the primary condition and boundary condition. Due to the marginal changes of boundary node with timing, this method can increase the computational accuracy with easier solving process. This article also adopts the centered finite differential method to solve for slow transient flow model, and apply Newton-Raphson method to conduct iterative computation for high order non-linear equation group. The centered finite differential method is a kind of implicit type finite differential method, which has the advantages of high accuracy of second order and fast computation speed.3. When the gas network is simulated, it can adopts characteristic line method or implicit type finite difference method. This article combines the advantages of these two methods and concludes that the characteristic line method is better used on the boundary node of long-distance pipeline when the dynamic operation conditions of high-pressure gas network are simulated; under other conditions, the implicit type finite difference method is better used. The combination of these two methods can be complimentary with each other. When the dynamic simulation software developed with Visual C++ for branched and ring pipeline is adopted to calculate the example, it is found that the calculating result can satisfy with the actual engineering requirements.4. To replace usual comparision equation of state with SHBWR equation of state and to solve with Newton iterative method, it can greatly improve the computation speed and. accuracy of gas state paramenters. Due to the fact that the gas state of the outer ring high-pressure pipeline is always in a turbulance zone, relatively accurate Colebrook's Formula is chosen to compute the pipe friction factor with Newton iterative method.5. To conduct hydraulic analysis on gas transmission line, it can be started with the basic equation of pipeline and combine the actual equation of gas state with enthalpy equation to establish dynamic simulation mathematical model; to solve the friction factor with secant method, it is different from the ususal trial method and can increase the computation accuracy. To deform the formula of friction resistance coefficient and associate with pipe flow, it can be changed with the flow and decrease the error caused by the constant value input of friction resistance coefficient.6. To conduct thermodynamic analysis on the pipeline temperature, it works out the evolution of gas temperature. The computation result indicates that the pipe temperature solved separately can match with the steady flow model. It further concludes that it is feasible to solve the pipe temperature separately. This provides an effective method to solve the initial value of pipe temperature in multi-resource unsteady flow model.7. The equation of continuity, equation of momentum, and equation of energy applied to the multi-source and single-source mathematical model are the same. The major differences are treatment of boundary condition, solving method of initial value, and number of equations. This article puts forward the calculation method of initial value for multi-resource dynamic simulation, which is different from the single-resource simulation model. It indicates that the closure analysis of multi-resource dynamic simulation can be started with boundary condition and constraint condition. Based on the closure of euqation of multi-source gas network, it is necessary to add two equations to meet the solving conditions when adding one gas source:one is equation of flow or pressure, another is equation of temperature.8. The computation of multi-source pipe network can obtain the pressures and flows for the various gas sources, nodes, and pipe section at different profiles and moments. The pressures of various gas sources can be effectively adjusted and the flow can be rationally distributed in reponse to the pressure requirements of different locations. To adopt the data collected with SCADA to conduct simulation computation, and at the same time to compare the calculation result with the data collected, it can inspect the accuracy and reliability of data collection system.9. To conduct a study on the chocking phenomenon of dynamic simulation and gas flow in the multi-source high-pressure gas network, establish mathematical model of dynamic simulation and chocking flow in the high-pressure gas network, and to solve with Newton iteration method and four orders Rung-Kutta method, the reckoning is verified with actual examples. It indicates that the calculation result can be fully used for the design, construction, and operation control of gas pipe network.10. Based on the gasdynamics, a study is conducted on the chocking phenomenon at the variable sections and constant sections, and a reckoning is solved with four orders Rung-Kutta method for a constant section chocking model. Combining with an actual example, the evolution of the gas parameters in the chocking state is analyzed. The result indicates that under the conditions of gas-filling from compressor to CNG tanker and then to storage tank, the low-pressure end is inclined to be clogged up due to the rapid expansion because of the greater pressure difference at the two ends and smalller pipe diameter in the initial gas-filling stage. The formula for judgment of chocking is According to the computation, it may produce chocking phenomenon when the pipe terminal pressure ratio is between 0.32-0.41. Therefore, the designed flow shall be no more than the chocking flow, otherwise it may affect the normal gas supply. If the constant section occurred with chocking, the pressure and temperature are changed along the pipeline; if the pressure and temperatureare decreased along the pipeline, the Mach number and flow rate are increased, and the flow rate tends to sound velocity; sectional sound velocity is linked with density and temperature and decreased along the pipeline. The diameter of the pipe connected to the gas distribution station at the end of long-distance pipeline shall be not too small, otherwise it may cause chocking phenomenon and gas shortage at the peak time.11. A study is conducted on static gas storage and dynamic gas storage of high-pressure pipeline respectively. The computation formula is deduced for static gas storage based on the pipeline pressure evolution. The impacts of pipe length and pressure on static gas storage are also analyzed. Further analysis is conducted on the "emission" chocking phenomenon at the end of long-distance pipeline and its impact on static gas storage in accordance with the definition of dynamic gas storage and variable sections chocking model. The result indicates that the gas storage at the end of long-distance pipeline is usually computed in a static or dynamic state. The reckoning from the static state computation tends to be conservative, and the reckoning from the dynamic state computation is more accurate. The reckoning from the static state computation tends to have error. The static state computation is usually based on the maximum pressure at the starting point and minimum pressure at the end of pipeline, and the actual length of pipeline, which does not include maximum gas storage and optimum pipe length. If the pipeline is too long, it shall not only affect the storage capacity but also the gas storage. Therefore, it should be taken into consideration when the gas storage at the end of long-distance pipeline is designed.12. According to the dynamic model and computation method proposed in this article, a transient simulation and leakage analysis software is developed for the gas pipe network. The software has a computational simulation function on multi-source high-pressure gas network, realizes the historical data inquiry, new data computation and storage, and lays a foundation for the combination of load prediction and real-time simulation. Based on the computation of various working conditions of several high-pressure gas network, in comparison with the results calculated with foreign pipeline simulation softwares like PIPELINE, TGNET, and SPS, and the actual operation data of the city gas pipe, it indicates that relative deviation is within 6% compared to the foreign software, and within 3% compared to the actual operation data. The reckoning of pipe pressure indicates that the evolution trend of calculated pressures for every regulation stations is in line with the data collected and within 5% of error except for several nodes. It concludes that the software simulated result can reflect the evolution of pipe pressure. The software can be used for the dynamic simulation computation on the actual pipe network, meet the requirements of high-pressure network, and provide guidance to the intelligent operation.From the conclusions above, we can see that the model established and solution proposed can have higher-accurate computation result with rapid speed and reach the levels of the foreign similar studies. It reveals the evolution process of unsteady flow and gas storage and discharge in the city high-pressure gas network, and provides a theoretical basis for the design and control of high-pressure gas network. In combination with the SCADA data collection system, it can conduct a dynamic simulation on the high-pressure pipeline, and instructs on the design, operation condition simulation, and analysis of gas storage and discharge for the high-pressure gas network. This study has important values both in theory and pratice, which provides a new concept on the flowing of natural gas, theoretical basis for the design of high-pressure gas network, a new method for analysis of chocking phenomenon and high-pressure gas storage, and technical support to the national high-pressure gas transmission and storage.