MODEL ADEQUACY ANALYSIS FOR REAL-TIME OPTIMIZATION VIA MODIFIER ADAPTATION

ALEJANDRO MARCHETTI; MAXIMILIANO GARCÍA; MARTA BASUALDO

Buenos Aires

Congreso; XXIIIº Congreso Argentino de Control Automático; 2012

Asociación Argentina de Control Automático (AADECA)

In order to optimize the plant operation in the presence of plant-model mismatch and unmeasured disturbances, real-time optimization schemes use some adaptation strategy based on measurements. Instead of updating the model parameters of the nonlinear first-principles model describing the steady-state behavior of the plant, the modifier-adaptation approach is to correct the constraints and gradients in the optimization problem by adapting the values of bias modifiers expressing the difference between the constraints and gradients of the plant and the model. The approach has the ability to converge to the plant optimum because it matches the first-order necessary conditions of optimality of the plant upon convergence. Based on second-order sufficient conditions of optimality, a model-adequacy criterion for modifier adaptation has been stated in the literature. However, this criterion has been incompletely stated, since it does not include conditions concerning an inadequate model.