(568o) Model Predictive Control of Dual Mixed Refrigerant Liquefaction Process for Stable and Economic Operation in Offshore FLNG

Dual mixed refrigerant (DMR) liquefaction cycle is one of the promising processes to produce LNG in offshore FLNG because of its high process efficiency and compactness. However, stable and economic operation of DMR liquefaction process is a key to harness its advantages in the offshore or marine operation environments.  For example, roll and pitch and heave movements are major disturbances in offshore FLNG process. Furthermore, frequent changes in LNG shipping schedule due to the uncertainty in offshore weather condition and the limited LNG storage volume can also lead to frequent LNG production load variations compared to the onshore LNG production. This study proposes a model predictive control (MPC) strategy for the DMR process in order to reject such process disturbances and minimize the energy consumption, production loss, subcooling or insufficient cooling of LNG under the load variation conditions.

 A rigorous process dynamic model was first constructed in Aspen HYSYS Dynamics^TMand MATLAB/Simulink environments that can simulate process disturbances including steep load variations. An energy optimized steady-state model was first developed to identify an appropriate nominal operating condition. The corresponding pressure and temperature conditions were used as a design basis for dynamic model. Using the step response data of from dynamic simulation software, multivariate transfer function matrix was identified for the purpose of control system design. In this study, LNG flowrate, cold mixed refrigerant liquid (CMRL) and cold mixed refrigerant vapor (CMRV) were manipulated to regulate LNG temperature, LNG flowrate, and warm side temperature difference of multi-stream heat exchanger and optimize accumulated energy consumption and production loss.

The dynamic behaviors and the features of process with MPC are compared with a conventional PI control scheme with loop pairing and show the applicability and advantage of implementing multivariate control for the offshore plant.