Research on adaptive collaborative dispatch optimization algorithms for drones in distribution networks
TL;DR
Simulation experiments comparing AACOES with particle swarm optimization, genetic algorithms, and traditional ant colony algorithms demonstrate its significant advantages in optimizing both single-unit performance and total flight distance, coupled with more stable convergence.
Abstract
This paper addresses the complex scheduling optimization problem in multi-UAV collaborative power line inspection by proposing an Adaptive Ant Colony Optimization Algorithm with Elite Strategy (AACOES). The study comprehensively considers multiple practical constraints, including UAV flight characteristics, battery endurance, and external wind conditions, to construct a scheduling optimization model closely aligned with real-world inspection operations. To overcome limitations in convergence speed and global search capability inherent in traditional ant colony algorithms, the proposed method incorporates an elite strategy and adaptive adjustment factors. It optimizes pheromone update rules, effectively enhancing colony diversity and accelerating convergence. This enables efficient identification of near-optimal solutions under multiple constraints. Simulation experiments comparing AACOES with particle swarm optimization, genetic algorithms, and traditional ant colony algorithms demonstrate its significant advantages in optimizing both single-unit performance and total flight distance, coupled with more stable convergence. This validates its effectiveness and practicality for multi-UAV collaborative inspection scheduling in complex environments, providing an efficient and reliable technical approach for real-world applications such as power line inspections.