Congress detail

Authors: Agustín Bottari; David Zumoffen; Alejandro Marchetti.

Title: Self-optimizing control structure selection based on active constraint regions.

Resumen: In process industries, the selection of the controlled variables (CV) and manipulated variables (MV) with their set point (SP) values, directly impacts on the operation costs of the plant. Defining the mentioned variables, together with the interaction between the CVs and MVs, determines a specific control structure (CS). Since the process is affected by disturbances, it may happen that under the effect of disturbances the implemented CS takes the controlled plant to an operating point that violates an operating constraint, or to an economical suboptimal point. In order to optimally select the CS and the set point values, while paying attention to the satisfaction of operating constraints, academics have proposed different approaches based on economics. Two strategies widely known are the back-off approach and the self-optimizing control. In a recent work (Bottari et al., 2019), it was argue that the steady-state back-off approach based on the minimization of the average cost can be seen as a global self-optimizing control strategy. This approach selects a fixed control structure with fixed set point values regardless of whether or not the set of active constraints of the optimization problem changes with the disturbance values. Based on this work, in the present paper we evaluate the economic improvement that can be achieved by changing the control structure and/or the set point values for different active constraint regions. The new formulations are applied to a forced-circulation evaporator, and the results show that an important economic benefit can be achieved.

Meeting type: Congreso.

Type of job: Artículo Completo.

Production: Self-optimizing control structure selection based on active constraint regions.


Meeting place: Santa Fe.

Organizing Institution: Asociación Argentina de Ingenieros Químicos.

It's published?: Yes

Publication place: Santa Fe

Meeting month: 8