142 / 2013-03-18 11:03:55

Study On Process Reliability Of Buried Lng Pipeline

LNG pipeline,process reliability

The volume proportion of the liquefied natural gas (LNG) and the natural gas under standard state can be reduced to about 1:600, thus LNG is transported in pipelines has remarkable advantages in some cases. The LNG is often transported after having been cooled to a certain temperature in cooling stations, and the flow assurance to “too cold state” has great significance in the design and management of LNG pipeline. Generally, the pipelining temperature against gasification is controlled according to the bubble point temperature of LNG. However, LNG is a multi-component mixture and has stochastic bubble point temperature when the components change randomly within a certain range on the one hand. On the other hand, the operation parameters of LNG pipeline have also stochastic fluctuations. Therefore, it is not only impossible but also unnecessary to keep the outlet temperature at an invariant value less than the bubble point temperature, which gives birth to assess the process reliability of buried LNG pipeline from a probabilistic point of view. A stochastic numerical simulation method is proposed for assessment and control of the outlet temperature, and a probability model of the outlet temperature more than “bubble point temperature - n oC” is developed. Calculations are carried out for different thicknesses of the insulation layers. Comparisons are made to the means and the standard deviations of the outlet temperatures obtained from the computation based on unsteady heat transfer and flow and that based on simplified steady computation. Necessity of the unsteady computation is demonstrated, and therefore an algorithm combining the numerical simulation and stochastic simulation is essential. By stochastic numerical simulation of various means of the LNG rate, the probabilities that the outlet temperature is more than “bubble point temperature - n oC” are computed, which gives a reference for industrial applications.
Acknowledgments
The research supported by the National Science Foundation of China (Grant No. 51005247) and supported by Beijing Nova Program (Grant No. 2010B068).