(7f) Optimizing the Debottleneck of Existing LNG Plants Using Dynamic Models

There is a significant emphasis on utilizing the existing natural gas assets to their maximum potential. The first step in achieving this is operating the existing LNG plants at their maximum capacity. Beyond that, there are economic incentives in debottlenecking the existing LNG plants to increase their capacity. There are various options to perform the debottlenecking of existing plants. The evaluation and selection of the best options to increase the plant capacity is important for optimum return on the investment. This paper discusses the methodology utilized to perform debottleneck of a LNG plants based on ConocoPhillips Optimized Cascade® process. Plantwide dynamic simulation model is used as a tool to perform the evaluation and optimization of various debottlenecks options.

It is important to identify the existing plant constraints before performing any debottleneck evaluations. These constraints vary with the operating conditions like ambient temperature, feed composition etc. The main constraints that limit the production in LNG plants are the refrigeration compressor drivers. Also, various other equipment limitations and hydraulic constraints may be present in the plant. Plantwide dynamic models provide a suitable tool to identify these existing plant constraints under various conditions. These models incorporate the equipment and hydraulics details and the various control schemes in the LNG plant. The models are built during the engineering and construction phase of the plant. The dynamic model is first baselined to reflect the actual plant behavior. The model parameters are tuned with the data that is reflective of the normal plant operation to obtain predictions within reasonable error margin.

Once the existing constraints are identified, various options to perform debottleneck of the LNG plant are studied. These include upgrades or modifications to existing equipments, replacing old equipment with new or changes in process configuration. For example, gas turbine drivers can be upgraded to those with higher power ratings. The dynamic model can be run with the actual field conditions to identify the plant performance with these debottleneck modifications. The incremental LNG production is identified for various times of the year to obtain an annual production increase. This is important to optimize the selection of LNG plant debottleneck modification that maximizes the return on capital investment.