JOURNAL OF COMPUTERS (JCP)

ISSN : 1796-203X

Volume : 4 Issue : 9 Date : September 2009

**Distribution Network Reconfiguration Based on Particle Clonal Genetic Algorithm**

Yemei Qin and Ji Wang

Page(s): 813-820

Full Text: PDF (414 KB)

**Abstract**

Distribution network reconfiguration is an important aspect of automation and optimization of

distribution network system. To handle massive binary code infeasible solutions in distribution

network reconfiguration, a kind of sequence code is presented in which a loop is a gene and the

label of each switch in the loop is the gene value. To solve mutation probability and slow the

convergence of clonal genetic algorithm (CGA) in the later stage, in this paper particle clonal genetic

algorithm (PCGA) is proposed, in which we build particle swarm algorithm (PSO) mutation operator.

PCGA avoids the premature convergence of PSO and the blindness of CGA. It ensures evolution

direction and range based on historical records and swarm records. The global optimal solution

can be obtained with fewer generations and shorter searching time. Compared with CGA and clonal

genetic simulated annealing algorithm (CGSA), IEEE33 and IEEE69 examples show that PCGA can

cut the calculation time and promote the search efficiency obviously.

**Index Terms**

sequence code, distribution network reconfiguration, infeasible solution, PCGA

ISSN : 1796-203X

Volume : 4 Issue : 9 Date : September 2009

Page(s): 813-820

Full Text: PDF (414 KB)

distribution network system. To handle massive binary code infeasible solutions in distribution

network reconfiguration, a kind of sequence code is presented in which a loop is a gene and the

label of each switch in the loop is the gene value. To solve mutation probability and slow the

convergence of clonal genetic algorithm (CGA) in the later stage, in this paper particle clonal genetic

algorithm (PCGA) is proposed, in which we build particle swarm algorithm (PSO) mutation operator.

PCGA avoids the premature convergence of PSO and the blindness of CGA. It ensures evolution

direction and range based on historical records and swarm records. The global optimal solution

can be obtained with fewer generations and shorter searching time. Compared with CGA and clonal

genetic simulated annealing algorithm (CGSA), IEEE33 and IEEE69 examples show that PCGA can

cut the calculation time and promote the search efficiency obviously.