Faults locating of power distribution systems based on successive PSO-GA algorithm
Wenzhang Xu,Jiachun Li,Lv Yang,Quan Yu
DOI: https://doi.org/10.1038/s41598-024-61306-w
IF: 4.6
2024-05-18
Scientific Reports
Abstract:As the terminal of the power system, the distribution network is the main area where failures occur. In addition, with the integration of distributed generation, the traditional distribution network becomes more complex, rendering the conventional fault location algorithms based on a single power supply obsolete. Therefore, it is necessary to seek a new algorithm to locate the fault of the distributed power distribution network. In existing fault localization algorithms for distribution networks, since there are only two states of line faults, which can usually be represented by 0 and 1, most algorithms use discrete algorithms with this characteristic for iterative optimization. Therefore, this paper combines the advantages of the particle swarm algorithm and genetic algorithm and uses continuous real numbers for iteration to construct a successive particle swarm genetic algorithm (SPSO-GA) different from previous algorithms. The accuracy, speed, and fault tolerance of SPSO-GA, discrete particle swarm Genetic algorithm, and artificial fish swarm algorithm are compared in an IEEE33-node distribution network with the distributed power supply. The simulation results show that the SPSO-GA algorithm has high optimization accuracy and stability for single, double, or triple faults. Furthermore, SPSO-GA has a rapid convergence velocity, requires fewer particles, and can locate the fault segment accurately for the distribution network containing distorted information.
multidisciplinary sciences