This paper proposes an improvement of an efficient multiobjective optimization algorithm, Non-dominated Sorting Genetic Algorithm II,NSGA-II, that has been here applied to solve the problem of optimal capacitors placement in distribution systems. The studied improvementinvolves the Crowded Comparison Operator and modifies it in order to handle several constraints. The problem of optimal location and sizing ofcapacitor banks for losses reduction and voltage profile flattening in medium voltage (MV) automated distribution systems is a difficultcombinatorial constrained optimization problem which is deeply studied in literature. In this paper, the efficiency of the proposed CrowdedComparison Operator, CCO1, is compared to the efficiency of another Crowded Comparison Operator, CCO2, whose definition derives from theconstraint-domination principle proposed by Deb et al. The two operators are tested on difficult test problems as well as on the optimal capacitorsplacement problem.
|Numero di pagine||11|
|Rivista||Advanced Engineering Informatics|
|Stato di pubblicazione||Published - 2006|
All Science Journal Classification (ASJC) codes
- Information Systems
- Artificial Intelligence