Optimal Control Structure Selection for Real-Time Optimization Systems

AGUSTÍN BOTTARI; ALEJANDRO G. MARCHETTI

Buenos Aires

Congreso; 26º Congreso Argentino de Control Automático (AADECA'18); 2018

Asociación Argentina de Control Automático

Control structure selection has an important influence on the economic performance of process systems in the presence of unmeasurable and measurable disturbances. Economic criteria has been incorporated in the design and control problem using different strategies. The backoff approach is based on the idea of minimizing the nominal economic loss that results from the need to backoff from the active constraints in order to avoid violating them in the presence of disturbances. The backoff approach determines the optimal control structure and fixed setpoint values that guarantee feasibility for all disturbance scenarios. On the other hand, under the assumption that a given set of disturbances and/or parameter values can be measured or estimated online, the presence of a real-time optimization layer aims at optimizing plant operation by computing optimal setpoints for the lower layer controllers. In this paper, we argue that the control structure implemented in the plant affectsthe economic performance that can be achieved by the real-time optimization system, and we formulate a control structure selection problem based on the backoff approach that takes into account the presence of the real-time optimization layer in the design of the control structure. The effectiveness of the approach is demonstrated in simulation for a linear example.