(302d) Visualization of the Economic Impact of Process Uncertainty in Multivariable Control System Design

The notion of Backed-off Operating Point (BOP) selection, driven by the relationship between an Expected Dynamic Operating Region (EDOR) and the process constraint set, has been shown to be a powerful tool (both computationally and visually) for the economic analysis of dynamic processes (see for example [1, 2, 3]). In particular, [3] proposes a control system design method aimed at identifying variables of economic significance and minimizing the EDOR in those key directions, and thus providing maximum BOP profit. In the current paper, the procedure of [3] is extended to the cases of process uncertainty, specifically process and measurement biases as well as parametric model uncertainties. The proposed extension can also be used to visualize the impact and significance of possible uncertainties and thus will provide a vehicle to assess the importance of including candidate uncertainties in the model of the closed-loop system. The proposed scheme will be illustrated on a system of two non-isothermal tank reactors, from [4].

[1] Figueroa, J; P. Bahri; J. Bandoni; J. Romagnoli, ?Economic impact of disturbances and uncertain parameters in chemical processes ? a dynamic back-off analysis,? Comp. & Chem. Eng., vol 20(4), pp 453-461, 1996.

[2] Loeblein, C.; J. Perkins, ?Structural design for on-line process optimization: I. Dynamic economics of MPC,? AIChE J., vol 45(5), pp 1018-1029, 1999.

[3] Peng, J-K, A.Manthanwar & D.Chmielewski, ?On the Tuning of Predictive Controllers: The Minimum Back-off Operating Point Selection Problem,? Ind. Eng. Chem. Res., vol. 44, pp 7814-7822, 2005.

[4] Loeblein, C.; J. Perkins, ?Economic analysis of different structures of on-line process optimization systems,? Comp. & Chem. Eng., vol 20, pp S551-S556, 1996.