OPTIMAL POWER SYSTEM MANAGEMENT VIA MIXED INTEGER DYNAMIC PROGRAMMING

Power systems involve both continuous and discrete acting components and subsystems. In this work a logical specification is used to define the transition dynamics of the discrete subsystem. A computational tool that reduces the logical specification to a set of inequalities and the use of the transformed model in a dynamic programming approach to the design of optimal feedback controls arc described. An example of optimal load shedding for a power system with aggregate induction motor and constant admittance loads is given.