Path planning for electricity inspection UAV based on elite differential variance improved northern goshawk optimisation algorithm

UAV path planning is critical for efficient electricity inspection. This paper proposes an elite differential variance enhanced Northern Goshawk Optimization algorithm (EINGO) to address this challenge. first, population initialization via SPM chaotic mapping and refractive inverse learning to enhance diversity and convergence; then, exploration ratio factor and nonlinear factor is proposed for balanced global search; next elite differential variance and cosine based mechanisms to jump out of local optimal solution. validated on CEC-2014 and CEC-2017 benchmark functions, EINGO outperforms contrast algorithms accuracy and robustness.In UAV path planning tasks, EINGO reduces costs by 8.83%-26.14% compared to base methods, demonstrating significant practical value for power infrastructure inspection.