(595c) Operation Flexibility Study for Newly Designed Liquefied Natural Gas Receiving Terminals

Receiving terminal is one of the most important sectors along the whole liquefied natural gas (LNG) value chain.  And the major concern at receiving terminal is that a great deal of heating duty will be consumed during LNG re-gasification.  On the other hand, natural gas liquids (NGL) recovery process normally needs large quantity of cold energy for liquefaction and condensation.  NGL recovery section usually proceeds prior to liquefaction and products.  Moreover, a lot of shale gas plays distribute close to the receiving terminals in U.S. Currently, shale gas is usually burnt to generate electricity after simply treatment.  Apparently, it would be much more ideal if required utility can be provided conveniently for shale gas to conduct NGL recovery, since NGL has more economic values than that as part of the natural gas stream.  Hence, a novel design for receiving terminal has been proposed that integrates natural gas and shale gas NGL recovery process with LNG re-gasification process.  Simulation of this newly designed receiving terminal indicates that the cold energy from LNG re-gasification could be efficiently recovered by shale gas liquefaction and NGL fractionators’ condensers.

       The problem about this design is how to get the optimal operation condition under the uncertainty of LNG and shale gas flow rates and compositions.  However, the expected profit can be maximized under the stochastic optimization approach.  With the analysis of historic data and prior information, the probability distribution of LNG and shale gas flow rates and compositions can be estimated.  An MINLP model is developed with the integration of stochastic optimization approach.  Heat integration is the main concern in this MINLP model so that minimum utility will be consumed, in another word, maximum profit could be obtained.  Different operation conditions are calculated according to the changes on the flow rate and composition in order to attain the maximum average profit.  Numerous simulations have been conducted for building the model and the MINLP model is solved in GAMS.  The final results and analysis are also presented in this work.