Many-objective optimization of real and reactive power dispatch problems

Abstract

Power dispatch is one of the basic functions in power systems operation and planning activities. It involves security, performance, and economic aspects, while maintaining generation and demand balance. The Optimal Power Dispatch (OPD) and the Optimal Reactive Power Dispatch (ORPD) are formulations of the power dispatch that respectively address objectives related to efficiency, and voltage regulation and stability. Under most common approaches, both problems are independently solved, despite their strong connection. This paper addresses the problem of the simultaneous dispatch of real and reactive power considering four conflicting objectives: minimization of network power losses, maximization of system voltage stability, maximization of power factor on transmission network step-down transformers, and minimization of power dispatch economics. The mathematical model is developed based on the current regulatory framework of the Brazilian electricity sector. To solve the problem, an optimization methodology using the Non-dominated Sorting Genetic Algorithm III (NSGA-III) is proposed. This methodology is tested both on IEEE 30-bus and 57-bus systems and on a Brazilian 10351-bus system. For each system, a set of efficient solutions with good convergence and diversity is obtained. We highlight a 6.5% (30.1 MW) reduction of power losses in the Brazilian system, as well as a 2.8% reduction in the power dispatch cost and a increase of 0.6 in the power factor. The methodology showed that the simultaneous dispatch of real and reactive power is a successful approach to improving the performance of power systems in several criteria.