Most of the existing works on periodic charging consider that the mobile charger (MC) has enough energy to charge the sensors by travelling across the charging path. The existing works also assumed that the sensors always get full charge from MC and hence negligible dead period occurs for sensors in wireless sensor network (WSN). However, the above mentioned assumptions are not always practical. In this paper, we propose a novel periodic partial charging scheme using battery limited MC, which charges the sensors such that the total dead periods of the sensors is minimized. We use genetic algorithm (GA) to optimize the entire charging process. The simulation is performed to demonstrate the usefulness and competitiveness of the proposed scheme. We also verify the statistical significance of the simulation results using analysis of variance and post hoc least significant difference. The comparative analysis reveals that the proposed work improves the charging performance by reducing the dead period as well as number of dead sensors and maintain the perpetual network operation of WSN compared to the other existing works.

大多数现有的定期充电工作都认为移动充电器（MC）有足够的能量通过充电路径为传感器充电。现有的工作还假设传感器总是从MC获得完全充电，因此无线传感器网络 ( WSN )中的传感器发生的死区可以忽略不计。然而，上述假设并不总是可行的。在本文中，我们提出了一种使用电池受限MC的新型周期性部分充电方案，该方案对传感器进行充电，从而使传感器的总死区时间最小化。我们使用遗传算法（GA) 优化整个充电过程。进行仿真以证明所提出方案的有用性和竞争力。我们还使用方差分析和事后最小显着性差异来验证模拟结果的统计显着性。比较分析表明，与其他现有工作相比，所提出的工作通过减少死区时间和死传感器数量来提高充电性能，并保持WSN的永久网络运行。