Overall intelligent hybrid control system for a fossil-fuel power unit

In response to the multiple and tighter operation requirements already placed on power plants, and anticipating everyday variations on their quantity and relevance due to competition on deregulated energy markets, this dissertation contributes the Intelligent Coordinated Control System (ICCS) paradigm that establishes a reference framework for the design of overall control systems for fossil-fuel power units, and develops a minimum prototype (ICCS-MP) to show its feasibility.; The ICCS consists of a multiagent system organization structured as an open set of functionally grouped agent clusters in a two-level hierarchy. The upper level performs the supervisory functions needed to produce self-governing system behavior, while the lower level performs the fast reactive functions necessary for real-time control and protection.; The ICCS-MP greatly extends the concept of current coordinated control schemes and embraces a minimum set of ICCS functions for the power unit to participate in load-frequency control in deregulated power systems; provides the means to achieve optimal wide-range load-tracking in multiobjective operating scenarios. The ICCS-MP preserves the ICCS structure. Supervisory functions include optimization and command generation, learning and control tuning, and performance and state monitoring. Direct level control functions realize a nonlinear multivariable feedforward-feedback scheme. These functions are implemented in three modules: reference governor, feedforward control processor (FFCP), and feedback control processor (FBCP).; The reference governor provides set-point trajectories for the control loops by solving a multiobjective optimization problem that accommodates the operating scenario at hand. The FFCP facilitates achievement of wide-range operation; it is implemented as a fuzzy system that emulates the inverse static behavior of the power unit, and it is designed using neural networks. The FBCP provides disturbance and uncertainty compensation along the set-point trajectories; it includes fuzzy-PID controllers in a multiloop configuration and a multivariable feedforward interaction compensator, both scheduled in two-dimensions. Controllers are tuned using a genetic algorithm based procedure, and the compensator is designed using the relative gain array method.; Simulation results show that the proposed ICCS paradigm provides a suitable conceptual framework to manage the complexity and to integrate either algorithmic or heuristic techniques into a control system that provides high maneuverability and optimized unit operation.