Study on charging strategy of wireless rechargeable sensor networks based on dynamic inhomogeneous clustering

In wireless rechargeable sensor networks (WRSNs), high node mortality rate severely constrains the network performance. To address this problem, this paper proposes an innovative charging strategy based on dynamic inhomogeneous clustering (DICCS). The core of this strategy lies in dynamically adjusting the network clustering structure, which combines the dynamic changes of node energy, position and energy consumption rate to achieve the optimal division of clusters. Firstly, the improved k-means algorithm is used to perform dynamic inhomogeneous clustering of the network, determine the optimal number of clusters through iterative optimization, and introduce a weight function to synthesize the node's initial energy, residual energy, and the average intra-cluster distance to select the cluster head in order to balance the energy consumption. On this basis, DICCS plans efficient charging paths for mobile charging carts (MCs), designs charging dwell point selection mechanisms for single-node and multi-node clusters respectively, and dynamically adjusts the charging sequence based on the mixed priorities (distance, residual energy, and energy consumption rate). Simulation experiments show that DICCS significantly reduces the node mortality rate (only 4.3%) and charging waiting time, while optimizing the mobility cost of MCs, compared to strategies such as SAMER, VTMT, and FCFS. Its dynamic inhomogeneous clustering mechanism effectively mitigates the energy consumption imbalance problem, improves the network life cycle and stability, and provides an efficient solution for charging scheduling in heterogeneous dynamic WRSNs.