Keynote II

Optimized adaptive control technology has experimentally proven to overcome the limitations of conventional and advanced control systems currently used in industry. It helps maximize process productivity, minimize energy consumption and emissions and, thus, represents a valuable tool for the control of climate change and achievement of sustainable economic growth. The concepts of modeling, estimation, prediction, predictive control and optimized control will progressively be introduced using practical examples of climatization processes. Next, the need for an adaptive system, to ensure precise predictive control for industrial processes, will be derived from a simple analysis of their cause-effect dynamic nature. Adding an adaptive system to predictive control will consequently lead to the methodology of adaptive predictive control and to the extended concept of Optimized Adaptive control.Practical implementation will be approached by penetrating the adaptive predictive (AP) controller structure, the blocks inside it, and the repetitive sequence of operations they perform periodically. A comparative analysis of AP controllers over conventional controllers generally used in industry will be undertaken in terms of control philosophy, stability and performance. A real time process simulation scenario will demonstrate the capacity of the AP controllers to identify and learn efficiently changes that may occur to the process dynamics and use this capacity to ensure the convergence of their performance towards optimized process control.Through the use of the available process knowledge, Adaptive Predictive Expert (ADEX) Control adds an expert component to AP control to facilitate the learning process and further improve its robustness, reliability and performance. The lecture will conclude illustrating the application of ADEX optimized adaptive control systems to several complex industrial processes (power plants, composites, nuclear and others) and analyzing their performance