(245f) Calculation on the Heat Transfer Correlations and Simulation Verification for Typical LNG Open Rack Vaporizer

Open rack vaporizer (ORV) is a type of commercial heat exchanger widely used in large gasification plants for base load Liquefied Natural Gas (LNG) receiving terminal. The heat source is seawater which is abundant and easily obtained. LNG in ORV is under supercritical pressure, which means there is no phase transition during the heat transfer progress, and the physical properties change dramatically, directly influencing the calculation. The aim of this study is to build a relatively accurate model for quantitative calculation for the heat transfer in ORV.

The whole block of the ORV is made of aluminum alloy. The heat transfer tubes of the ORV are placed in a line like a curtain, which are combined by means of the upper and lower header pipes in a single unit that referred to as a panel. There is a water spray device on the top of the vaporizer, and the water sprayed flows from top to bottom along the outer surface of the panel due to gravity. Supercritical LNG flows from bottom to top within the tube, during the progress the cold LNG absorbs heat and gets vaporized by the heat transferred from the seawater.

In order to achieve the localization of LNG vaporizers, the heat transfer correlations and calculation methods have been described and discussed for inside and outside of the tube in a typical ORV, and the calculation results are compared with the data of practical case. Meanwhile, numerical simulation is established by FLUENT to verify the heat transfer inside and outside of the tube.

The process where the heat transfers from seawater to LNG through the tube wall comprises three steps: (1) heat transfer from seawater to outer wall; (2) heat transfer from outer wall to inner wall; (3) heat transfer from inner wall to LNG. The heat transfer rate in the above three steps is equal, thus the heat transfer coefficient of the total process can be expressed by the formula containing convective heat transfer coefficient of outside and inside, and thermal conductivity of the wall.

Knowing the total heat transfer coefficient, the required heat transfer area of the vaporizer can be obtained through the temperature difference and heat transfer quantity, which can be obtained by subtraction of enthalpy of the appointed two states.

As the temperature difference of LNG between inlet and outlet is above 100K, and that the physical properties change dramatically near the pseudo-critical section, the area is divided into multiple regions and calculated separately in order to improve the accuracy. The final required area is the summation of the areas of all regions. The parameters of each region, such as temperature, velocity and convective heat transfer coefficient of LNG and water, can be obtained during the calculation. Thus, the distribution of flow field in the heat transfer tube can be presented.

LNG is a mixture of alkane, hydrogen sulfide, carbon dioxide, rare gases, etc. As the main composition, methane usually occupies 90% or more of LNG in volume fraction. Therefore, methane is used to replace LNG in the correlation and numerical simulation to simplify the calculation. A number of heat transfer correlations are adopted to optimize the calculation model: outside the tube, three correlations are compared to calculate the heat transfer of water; inside the tube, four correlations are selected to calculate the heat transfer of methane.

Physical model is established for the numerical simulation. The outermost layer of the tube is the falling film by water, and the thickness of the falling film is assumed 3mm which is obtained from observation of the actual vaporizer in receiving terminal; the middle layer is the solid structure of the tube made of aluminum alloy; the innermost layer is methane.

ANSYS ICEM is conducted to generate the mesh as the preprocessor of Fluent. Structured grids are established and a local grid refinement method is applied in the near wall region to improve the accuracy of the simulation. The boundary condition for the methane inlet and water inlet is the velocity inlet; the outlet boundary condition for the methane and water is the pressure outlet. The thermal condition of the outermost wall i.e. the interface of water and air is set as convection. SIMPLE algorithm is adopted for the pressure-velocity coupling in the simulation, and the second order upwind scheme is applied for the equations of momentum, turbulent kinetic energy and turbulent dissipation rate.

Through the verification of practical case and numerical simulation, the optimal correlations have been selected. “The flat plate in parallel flow correlation” is selected to calculate the heat transfer of water outside the tube with the minimum deviation. The combination of “D-B heat transfer correlation” for subcooled section and “heat transfer correlation for supercritical methane” for supercritical section has the highest accuracy to calculate the heat transfer of methane: the deviation for the methane temperature along the tube is 1.63%, and for heat transfer coefficient is 6.52%.

Effects of the inlet pressure of methane on heat transfer progress are investigated in this study. Deviations of temperature, convective heat transfer coefficient and velocity of methane between correlation calculation and numerical simulation in all working conditions are within 15%. In particular, the deviations of temperature are within 3% under different pressures. It can be concluded that the data of correlation calculation are fairly good agreement with the simulation data, and the results are of high reliability. The heat transfer coefficient decreases with the increasing pressure below the pseudo-critical temperature, while the change of the heat transfer coefficient is just the opposite above the pseudo-critical temperature. The curve of the heat transfer coefficient has a significant effect on the temperature. The maximum value of heat transfer coefficient occurs near the pseudo-critical temperature under different pressures, and the heat transfer coefficient decreases sharply near the pseudo-critical temperature, which results in the curve shape of the temperature: the rate of increase below the pseudo-critical temperature is much larger than that above the pseudo-critical temperature. Additionally, when the pressure decreases from 9MPa to 6.55MPa, the peak value of heat transfer coefficient has increased by about 22.2%, but the variation of temperature is not obvious. This is due to a significant increase in specific heat near the pseudo-critical region as pressure decreases. It also reveals that the outlet temperature of methane increases from 257.3K to 271.2K when the pressure increases from 6.55MPa to 9MPa. In addition, the higher the pressure is, the larger the methane density is. With the same mass flow, the velocity gradually decreases with increasing pressure.

Effects of the mass flow of methane are also investigated. The temperature along the tube rises more slowly with the increasing mass flow, and the outlet temperature of methane decreases from 271.8K to 235.5K when the mass flow increases from 0.0178kg/s to 0.0637kg/s. In addition, the heat transfer coefficient is significantly improved by the increasing mass flow due to the increment of turbulence effect, and the peak value almost occurs at the same temperature. Higher mass flow also means that the amount of heat absorbed by methane is increased. It is apparent that the increasing heat absorption capacity plays a major role in the process, which gives a reasonable explanation to the temperature curves: although the heat transfer coefficient is the largest under the highest mass flow, the required heat is also higher than the others, resulting in the gentlest temperature curve and lowest outlet temperature.

Effects of the inlet temperature of water are studied at last. The inlet water temperature has little effect on the distribution of methane temperature. In particular, the methane temperature has almost no difference in the first half of the tube at different inlet water temperature. The methane temperature increases with the increasing water temperature in the latter half. The temperature difference between methane and water is very large in the first half of the tube, which is slightly improved by the increase of water temperature, so it has very little effect on the heat transfer. As the temperature difference decreases, the effect of the elevated water temperature on heat transfer becomes increasingly apparent. For every 5 degrees’ increase in inlet water temperature, the methane outlet temperature increases by about 2 degrees.

Finally, the most accurate correlations and calculation method are selected, through which the heat transfer area of the ORV and the distribution of the temperature field and the convective heat transfer coefficient are obtained. In addition, the effects of different parameters on heat transfer of the tube are also investigated to meet the design requirement for effective LNG gasification.