基于改进粒子群算法的行李条码阅读器布局优化

针对航空旅客托运行李时,检测行李条码的阅读器数量、位置、姿态存在很多不确定性问题,提出了动态种群-双适应值粒子群优化(DPDF-PSO)算法。首先,建立行李条码检测数学模型;然后,转化为约束优化问题;其次,通过标准粒子群优化(PSO)算法求解此优化问题;最后,依照模型特点对标准粒子群算法进行改进。仿真结果表明,与标准PSO算法相比,DPDF-PSO算法仿真时间降低了23.6%,目标函数值提高了3.7%。DPDF-PSO算法克服了标准粒子群优化算法中仿真时间慢、边界最优解难处理的缺点,阅读器布局方案能以较低的成本准确快速读取行李身份信息。

Optimization of baggage tag reader layout based on improved particle swarm optimization

When civil aviation passengers check in, various uncertainty problems exist in the baggage tag readers' number, position and angle. To solve the problems, the Dynamic Population-Double Fitness Particle Swarm Optimization (DPDF-PSO) algorithm was proposed. Firstly, the mathematical model of baggage tag detector was established, then it was transformed into an optimization problem; secondly, the optimization problem was solved by standard Particle Swarm Optimization (PSO) algorithm; finally, the standard PSO algorithm was improved in accordance with the model features. The simulation results show that compared with standard PSO algorithm, the simulation time of the DPDF-PSO algorithm reduced by 23.6%, the objective function value increased by 3.7%. DPDF-PSO algorithm overcomes the shortage of long simulation time and troublesome problem of optimal boundary solutions existed in standard PSO algorithm. Identity information can be read quickly and accurately by readers layout at a lower cost.