Research on Multiple Air-To-Air Refueling Planning Based on Multi-Dimensional Improved NSGA-II Algorithm

Abstract

Reasonable air-to-air refueling planning (AARP) is essential for the successful completion of remote flight missions. A comprehensive task model for air refueling planning is proposed, and the key constraints are determined. The multi-objective optimization algorithm NSGA-II is improved from three distinct perspectives. The performance of the improved NSGA-II was evaluated by selecting test functions from the ZDT series for comparison against the original version. Simulation experiments demonstrate that the improved NSGA-II yields an increase in the average hypervolume index by approximately 10% to 18%, a decrease in the average spacing index by about 22% to 57%, and a reduction in the standard deviation of hypervolume by 27% to 76%. The obtained findings demonstrate that the improved NSGA-II variant exhibits superior convergence, uniformity, and universality. The airspace of the Americas was selected as the mission area to generate 50 AARP schemes for application. Five representative schemes with fuel consumption from 47,083 kg to 104,735 kg, corresponding to time consumption coefficient from 1.27 to 1.07, were chosen as alternatives. This research can enhance the efficiency and stability of air-to-air refueling planning, thereby serving as a valuable theoretical reference for selecting appropriate remote multi-point air refueling schemes.