无线可充电传感器网络中的充电路径优化

Optimization of Charging Path of Wireless Rechargeable Sensor Network

在无线可充电传感器网络中,多充电小车路径规划是根据成本目标规划出最优充电路径的全局优化问题.针对该问题,仅仅考虑路径距离是不全面的.因此,本文提出将路径距离、充电时间和车辆成本组合为新的成本目标模型.同时,对于求解路径问题,采用改进的鸽群算法和遗传算法混合的方式求解.在鸽群算法中引入遗传算法的交叉变异操作,改善了最优解的质量,并且提出了基于自适应惯性权重的速度更新公式,以加快全局搜索的收敛速度.仿真实验表明,算法的混合克服了自身算法的不足.并且,改进后的混合算法实际充电成本更低.

In wireless rechargeable sensor network,multi-charge vehicle path planning is a global optimization problem to plan the optimal charging path according to the cost target.For this problem,only considering the path distance is not comprehensive.Therefore,this paper proposes to combine the path distance,charging time and vehicle cost as a newcost target model.At the same time,the improved pigeon colony algorithm and genetic algorithm are used to solve the path problem.The crossover mutation operation of genetic algorithm is introduced into the pigeon flock algorithm to improve the quality of the optimal solution,and a speed updating formula based on the adaptive inertia weight is proposed to speed up the convergence rate of global search.Simulation results showthat the hybrid algorithm overcomes the shortcomings of its own algorithm.Moreover,the actual charging cost of the improved hybrid algorithm is lower.