Evaluation and Optimization of Processes in the Liquefied Natural Gas (LNG) Value Chain
MetadataVis full innførsel
As a cleaner fuel and an intermediate energy source on the way to a carbon-free society, the demand for natural is expected to grow considerably. Consequently, a larger production of liquefied natural gas (LNG) is also required due to its mobility with a high energy density per unit volume. To produce LNG and supply the liquid fuel to end-users, natural gas goes through a series of processes often referred to as the LNG value chain (gas production, pipeline transport of gas, natural gas liquefaction, ship transportation of LNG, regasification of LNG). Thus, a proper evaluation and improvement of systems in the chain is important to reduce the environmental footprints of LNG. Therefore, this thesis optimizes and evaluates various LNG systems to improve both their thermodynamic and economic effectiveness. For a fair comparison of LNG systems, the SQP algorithm (local solver) has been compared with two different global search algorithms (PSO and DIRECT) for the optimization of the processes. The local solver was found to be proper for any LNG system, while the near-global solvers give sub-optimal solutions for complex processes. To improve the liquefaction part of the LNG value chain, the different configurations of the dual hydrocarbon mixed refrigerant (DMR) technology have been optimized and compared as a promising base-load LNG process. The results proved that a large number of evaporation pressures for the refrigerants and phase separation of the cold refrigerant give a higher energy efficiency. The DMR process has also been compared with non-flammable liquefaction systems. The highest energy efficiency was still observed for the DMR process even though the non-flammable liquefaction systems have complex structures to improve their inherently low efficiency. However, the non-flammable processes can guarantee a higher safety for floating facilities than the hydrocarbon-based refrigeration systems. Different constraint formulations have also been assessed to make the DMR process more energy efficient. It was observed that the use of relaxed superheating constraints for the two MRs and the use of maximum heat exchanger conductance value constraints with relaxed minimum temperature difference constraints improve the efficiency of DMR processes. The LNG value chain can also be improved by the systems handling boil-off-gas (BOG) on LNG vessels since the gas is typically wasted after a part of it is being used as fuel for propulsion. This thesis has suggested BOG liquefaction systems based on self-liquefaction processes to prevent the loss of cargo. Economic optimization of the reliquefaction process proves that it can save around 10 % of total annual cost, compared to LNG carriers having only the fuel supply system. For complex LNG systems, exergy efficiency will be useful to reflect changes in the quality of products (heat, power, chemical materials) since energy efficiency only measures the quantity. Thus, an exergy efficiency (Exergy Transfer Effectiveness - ETE) developed in our research group has been thoroughly extended with general mathematical expressions to cover processes having changes in temperature, pressure, and chemical composition. The extended 𝐸𝑇𝐸 with a proper level of exergy decomposition was proven to be more consistent and accurate for a complex LNG process than other exergy efficiencies. By using exergy as a post design tool, the integration scheme of an air separation unit (ASU) and the LNG regasification step in the LNG chain has been suggested and evaluated to minimize the loss of LNG cold energy during conventional evaporation. The use of LNG cold energy in air separation was verified to be a proper solution for LNG regasification, increasing the extended 𝐸𝑇𝐸 of the ASU by 13 %. A sensitivity analysis with LNG supply pressure for the ASU system proved that the extended 𝐸𝑇𝐸 could properly reflect the quality changes in the products that occurred by the varying LNG pressure, while energy efficiency fails to do so. The extended 𝐸𝑇𝐸 has also been tested as objective function for the optimization of the DMR process integrated with NGL extraction. The optimization results indicate that the solutions from the exergy-based objective function give increased quality of products (LNG and NGL) with similar energy consumption as the results from an energy-based objective function. The results from the exergy-based objective function also confirm that upstream NGL extraction is thermodynamically better than the integrated systems unless the two processes are well heat integrated.