Improving Economic Operation Performance of Self-Optimizing Control Structure by a Supervisory Control Layer

BOTTARI, AGUSTÍN; BRACCIA, LAUTARO; DAVID ZUMOFFEN

Buenos Aires

Congreso; WCCE11 - 11th WORLD CONGRESS OF CHEMICAL ENGINEERING; 2023

Asociación Argentina de Ingenieros Químicos

In the chemical and petrochemical industry, the involved processes are generally non-linear, multivariable, and subject to external disturbances. Finding the operating point which satisfies the production requirements (rate and quality) while the costs, energy consumption and/or environmental pollution are minimized, without violating any operational or safety restrictions, certainly requires a deep analysis and adopting a proper strategy to this end. The two-layer decision control system is the classical approach to address the optimal process operation. This system includes an optimization top layer and a regulatory control bottom layer. While the regulatory control layer rejects high-frequency disturbances and guarantees a safety and stable operation, the supervisory level, commonly, contains algorithms to pursue the changing optimal operating point of the process (because of external disturbance, plant-model mismatch or changes due planning and programming). Generally, the process control system is addressed by selecting the regulatory control layer based in on stability and controllability indices. However, the economic criteria or the existence of higher levels of control are ignored in these approaches. In this context, the economic-based regulatory control structure selection methods, like the back-off (BO) approach and the self-optimizing control (SOC), have been appeared in the last decades. In the BO and SOC methods the optimal control loops and set points values are selected by solving a mathematical programing problem with an economical operation cost objective. Moreover, self-optimizing control literature shows that the economic performance can be improved using virtual controllable variables that linearly combine output-input signals. However, a fixed control structure and set points values strategy may be far from an optimal operation for processes under changing conditions.Recently in [1] the authors formulate a mathematical programing problem which incorporates the supervisory level in the search for economic-optimal regulatory control loops using a classical input-output pairing. Based on their findings, it is shown that the use of classic control structures with a supervisory layer might still be far from an ideal optimal economic operation for some processes. In this sense, the main objective of the present work is the economical optimal hierarchical control structure system design, by considering virtual controllable variables.A novel mathematical programing problem for control structure selection is developed and tested for a (simulated) reactor separator recycle plant by using available commercial software. Results show that the use of virtual controllable variables, based in on lineal combination of measurables variables, together with a supervisory control layer could be the proper strategy to improve the economic performance of processes under changing operation conditions.References1. A. Bottari, D. A. R. Zumoffen, A. G. Marchetti, Economic control structure selection for two-layered real-time optimization systems, Industrial & Engineering Chemistry Research 59 (2020) 21413–21428.