Intelligent optimisation of self-healing strategy of distribution network based on differential evolution and advanced fractal

Traditional self-healing strategies for distribution networks are prone to getting stuck in local optima in complex power systems, resulting in the inability to find the globally optimal self-healing strategy. By introducing genetic algorithms and differential evolution optimisation, more effective self-healing strategies were sought to improve the robustness and anti-interference ability of the power system. The optimisation objective was to minimise the interruption time of power supply under the influence of faults, collect comprehensive data related to the power system, and apply genetic algorithm (GA), differential evolution (DE), and GA-DE to optimise the self-healing strategy of the distribution network. The experimental results indicate that the average convergence frequencies for GA, DE, and GA-DE were 716, 662, and 612, respectively. The average total power outage times were 195.3 seconds for GA, 178.8 seconds for DE, and 148.5 seconds for GA-DE. The GA-DE algorithm resulted in the fewest power outages over six years. By combining the strengths of GA and DE, the GA-DE algorithm enhances self-healing strategies in distribution networks, improving power system stability and reliability.