Electrical Power Systems Management

The problem of ‘voltage collapse’ is a serious issue in isolated and complex systems like shipboard power systems, especially in military craft where the ability to sustain substantial damage to the power system without complete failure is vital. Intelligent reconfiguration of the power system including energy storage, shifting loads, or shedding loads that are non-critical to combat operations might enable a battle-impaired ship to continue performing essential operations. The concepts and methods of intelligent control have advanced significantly and we propose to use these new techniques to build an Electric Power System Management tool that will improve reliability and enable fight through capability of shipboard electric power systems. We will design, implement and evaluate a Shipboard Power System Management tool that will prevent loss of power in critical buses when damage conditions are encountered. Our approach is to optimally reconfigure a ship power network following the occurrence of disruptive events. We view the power system as a hybrid system involving highly nonlinear continuous dynamics described by differential-algebraic-equations and discrete dynamics characterized by a nondeterministic finite state machine. We will integrate the new hybrid systems finite horizon optimal controllers with our techniques for designing nonlinear and discrete event state estimators.