粒子群优化的改进机场车辆调度模型研究

Particle swarm optimization to improve airport vehicle scheduling model research

随着航空事业的迅猛发展,机场车辆调度的安全性和时效性地位已日趋突显,传统的机场车辆调度采取First in first out策略,该策略算法简易,便于实施,缺陷是全部调度的分组被相同对待,无法为实时要求较高的业务提供时延保证,算法也不具有公正性。提出了一种基于粒子群优化的改进机场车辆调度模型,把粒子群已经搜索到的全局最优地点视为一个特殊的粒子,采用梯度降低策略寻优该粒子,全局寻优特性和梯度降低算法的邻域寻优特性相融合,以提升粒子群优化算法的全局寻优效率,减少机场车辆调度计算的时间。仿真实验表明:粒子群优化的改进机场车辆调度模型,能够减少传统调度方法的寻优轮换次数,进而缩短优化调度时间,有效缓解空中堵塞造成的资源浪费。

With the rapid development of aviation, the airport security and efficiency of the vehicle scheduling status has been increasingly highlighted, the traditional airport vehicle scheduling uses the strategy of First in First out, this strategy algorithm is simple, easy to implement, defect is all scheduling group being treated the same, cannot provide delay guarantee for real-time demanding business, algorithm with impartiality. This paper proposes an improved airport vehicle scheduling model based on optimization particle swarm, the particle swarm as a whole has to search the global optimal location as a special particle, using gradient descent strategy optimization of the particle, global optimization characteristics and optimization neighborhood of the gradient descent algorithm, feature fusion, to enhance the efficiency of particle swarm optimization algorithm of global optimization, reduce vehicle scheduling calculation time at the airport. Simulation experiments show that the particle swarm optimization to improve the airport vehicle scheduling model, can reduce the number of traditional scheduling method optimization iterations, and shorten the optimal operation time, effectively relieve air blockage caused by resources waste.