This paper deals with the robust on-line diagnosis of abnormal situations in an industrial continuous styre11e polymerization reactor through a bank of unknown input observers (UIO) that supervise changes on the most relevant process parameters and external disturbances. A model predictive control (MPC) scheme is implemented aiming al to stabilize ihe system. This may become an additional difficulty because the detrimental effects of the feedback control on the detection of abnormal situations. In the design of the UIO's a lir.earized model of the process is utilized. The observers are tuned to supervise the change of a particular parameter of !he reactor model. The procedure takes into account possible uncenainties in these parameters such that a robust diagnosis strategy of the abnormal siiuation is obtained. Simulation results show a very promising perspective to ihe proposed strategy...

This paper deals with the robust on-line diagnosis of abnormal situations in an industrial continuous styre11e polymerization reactor through a bank of unknown input observers (UIO) that supervise changes on the most relevant process parameters and external disturbances. A model predictive control (MPC) scheme is implemented aiming al to stabilize ihe system. This may become an additional difficulty because the detrimental effects of the feedback control on the detection of abnormal situations. In the design of the UIO's a lir.earized model of the process is utilized. The observers are tuned to supervise the change of a particular parameter of !he reactor model. The procedure takes into account possible uncenainties in these parameters such that a robust diagnosis strategy of the abnormal siiuation is obtained. Simulation results show a very promising perspective to ihe proposed strategy...

This paper deals with the model re-identification in closed-loop systems with already existing MPC controllers. lt is assumed that the controller has a two-layer structure where the upper layer performs a simplified economic optimization in which the economic objective is represented as a linear combination of the process inputs. This is the case of several commercial MPC packages applied in industry. This work focuses on the case where the existing process model shows signs of deterioration and there is benefit in obtaining a new model. It is proposed a methodology where the test signal is introduced in the coefficients of the objective function of the economic layer. The approach allows the continuous operation of the system as the process constraints and product specification can be satisfied during the test. The application of the method is illustrated by simulation on a C3/C4 splitt...

This paper deals with the model re-identification in closed-loop systems with already existing MPC controllers. lt is assumed that the controller has a two-layer structure where the upper layer performs a simplified economic optimization in which the economic objective is represented as a linear combination of the process inputs. This is the case of several commercial MPC packages applied in industry. This work focuses on the case where the existing process model shows signs of deterioration and there is benefit in obtaining a new model. It is proposed a methodology where the test signal is introduced in the coefficients of the objective function of the economic layer. The approach allows the continuous operation of the system as the process constraints and product specification can be satisfied during the test. The application of the method is illustrated by simulation on a C3/C4 splitt...

Ver PDF

El objetivo de este trabajo es la re-identificación del modelo de proceso que se utiliza en controladores predictivos (MPC) ya existentes usando datos de operación en lazo cerrado. Se asume que el controlador tiene una estructura en dos capas, donde en la capa superior un simple algoritmo optimización económica determina un conjunto de valores óptimos en estado estacionario (“targets ”), los cuales son pasados al MPC para su implementación. Este es el caso de varios paquetes comerciales MPC aplicados en la industria. El presente trabajo enfoca el caso donde el modelo del proceso presenta signos de deterioro, por lo que obtener un nuevo modelo representa beneficios significativos en el procedimiento de re-comisionamiento del MPC. Se propone una nueva metodología para excitar el sistema en lazo cerrado, introduciendo señales de excitación persistente en la función objetivo de...

The objective of this work is the re-identification of the process model that is used in existing predictive controllers (MPC) using closed-loop operation data. The controller is assumed to have a two-layer structure, where in the upper layer a simple economic optimization algorithm determines a set of optimal steady-state values ​​("targets"), which are passed to the MPC for implementation. This is the case for several commercial MPC packages applied in the industry. This paper focuses on the case where the model represents significant benefits in the MPC re-commissioning procedure. A new methodology is proposed to excite the system in closed loop, introducing persistent excitation signals in the objective function of the upper layer of the MPC. This strategy allows the continuous operation of the system, respecting the constraints of the process and meeting the specifications of th...