(434e) Interacting Dynamical System Control By Multi-Parametric Distributed Model Predictive Control

Traditionally, for the control of multi-variable processes, multiple single-loop proportional–integral–derivative (PID) controllers or model predictive controllers (MPC) are used in a decentralized control system (DCS) architecture to control the overall process. The main disadvantage of DCS is that interactions between several components (subsystems) of the process are not considered while calculating the control moves. The issues of DCS lead to the development of model-based centralized control architectures such as centralized MPC to control multiple variables simultaneously [1]. While centralized MPC accounts for interactions within a process and takes into consideration constraints, it has the shortcoming of scalability. When there is a substantial increase in the number of control variables, manipulated variables, or measurements, the computational time required to solve the optimization problem in real-time increases significantly, which could limit the ability of the control system to respond within the time limits set by the process dynamics and operating conditions. The shortcoming of the centralized MPC and the DCS lead to the development of model-based distributed control systems (DMPC) that use a set of controllers to control different subsystems of the process yet communicate with each other to account for the interactions in the process [2]. In this control architecture, the MPC controllers are not independent but the control output by each controller is passed on as an input to the rest of the controllers. This allows an MPC controller to optimize the control output of a subsystem while keeping the number of variables within acceptable limits and accounting for the actions of the other actuators in the overall process system [3, 4]. The limitation of the DMPC framework is that exchanging data among the MPC controllers and converging to an equilibrium solution leads to increased computational times.

In this work, we develop a DMPC system using multi-parametric model predictive controllers (mp-DMPC) to alleviate the online computational burden. In this approach, the optimal control problem of each distributed controller is solved explicitly (offline) by considering the associated state variables and the actions of the rest of the controllers as parameters, instead of solving the conventional MPC optimization problem repeatedly in real time [5, 6, 7, 8]. Therefore, the online computational costs of the DMPC control architecture are reduced. The developed mp-DMPC takes into account the interactions between different control loops, thereby improving the performance of the MPC control scheme, and thus the overall process. To illustrate the effectiveness of the developed approach, the mp-DMPC technique is implemented on a system of interacting continuous stirred tank reactors and separators.

References

[1] Mayne, D. Q., Rawlings, J. B., Rao, C. V., & Scokaert, P. O. (2000). Constrained model predictive control: Stability and optimality. Automatica, 36(6), 789-814.

[2] Camponogara, E., Jia, D., Krogh, B. H., & Talukdar, S. (2002). Distributed model predictive control. IEEE control systems magazine, 22(1), 44-52.

[3] Rawlings, J. B., & Stewart, B. T. (2008). Coordinating multiple optimization-based controllers: New opportunities and challenges. Journal of process control, 18(9), 839-845.

[4] Scattolini, R. (2009). Architectures for distributed and hierarchical model predictive control–a review. Journal of process control, 19(5), 723-731.

[5] Kouramas, K. I., Panos, C., Faísca, N. P., & Pistikopoulos, E. N. (2013). An algorithm for robust explicit/multi-parametric model predictive control. Automatica, 49(2), 381-389.

[6] Bemporad, A., Morari, M., Dua, V., & Pistikopoulos, E. N. (2002). The explicit linear quadratic regulator for constrained systems. Automatica, 38(1), 3-20.

[7] Diangelakis, N. A.; Avraamidou, S.; Pistikopoulos, E. N. (2016) Decentralized Multiparametric Model Predictive Control for Domestic Combined Heat and Power Systems. Industrial & Engineering Chemistry Research, 55 (12), 3313-3326.

[8] Pistikopoulos, E. N., Diangelakis, N. A., Oberdieck, R., Papathanasiou, M. M., Nascu, I., & Sun, M. (2015). PAROC—An integrated framework and software platform for the optimisation and advanced model-based control of process systems. Chemical Engineering Science, 136, 115-138.