Ordinal-based genetic algorithm with cascaded encoding for reactive power optimization of electricity market

Abstract

Generators and capacitors are main reactive power sources for reactive load demands and system voltage stability. The cost of reactive power resource should be considered in dispatch optimization, based on which the reactive power market is built. A reactive power optimization model based on reactive power cost is proposed, which regards the on-off of capacitor as O-1 variable. The reactive power optimization becomes the non-linear programming of mixed integers, to which genetic algorithm with cascaded encoding is applied. The ordinal optimization theory is introduced to direct the individual selection and generation times of genetic algorithm. Its reliability is evaluated quantitatively. Simulation results of IEEE 30-bus system show its effectiveness. The operating conditions and relevant benefits are analyzed based on 5 cases.