Research On Dynamic Charging Scheduling In Wireless Rechargeable Sensor Networks

Wireless rechargeable sensor networks,as a new product of combining of wireless charging technology with wireless sensor networks,not only break the bottleneck of energy demand of existing sensor nodes,but also provide researchers with a new way of charging.As a large number of charging schemes using wireless charging vehicles continue to emerge,the energy demand problem of nodes in the network is effectively alleviated.However,the generation of new charging demands during the charging process has been neglected in the current research.Therefore,based on the study and analysis of the above proposed problem,this paper constructs a many-objective dynamic charging scheduling model,thus converting the dynamic charging scheduling problem into a manyobjective dynamic optimization problem and optimally solving the charging scheme using an improved many-objective algorithm.The details of the work are as follows.(1)This paper proposes a single-vehicle-oriented dynamic charging scheduling model to address the problem of new charging demand generated in the charging process,which is ignored by the existing charging scheme.Firstly,the many-objective dynamic charging scheduling model is constructed by analyzing the problem of new charging demand generated during charging and many factors affecting the charging scheme.Second,due to the limitations of the existing many-objective algorithm to solve dynamic problems,an environmental response mechanism is introduced to dynamically respond to the generated new charging demand problems,and an improved evolutionary strategy is proposed because the traditional evolutionary approach will cause the phenomenon of individual coding redundancy.Finally,the single-vehicle-oriented dynamic charging scheduling model is transformed into a dynamic optimization problem,and an improved many-objective algorithm is used to achieve the optimal solution.The effectiveness of the proposed single-vehicle-oriented dynamic charging scheduling model is verified by simulation experimental results.(2)As the scale of the network continues to expand,the dynamic charging method using a single wireless charging vehicle will cause the death of a large number of sensor nodes while failing to meet the large-scale charging demand.Therefore,in order to develop a more efficient charging scheme,this paper proposes an adaptive multi-vehicle dynamic charging scheduling model.First,the multi-vehicle dynamic charging scheduling model is constructed by distributing the total charging demand to multiple vehicles to relieve the charging pressure,and considering the cost problem and energy utilization problem caused by increasing the number of vehicles,the total vehicle cost and energy utilization are introduced as the constraint objectives to minimize the number of vehicles.Secondly,under the above constraint,since the generation of new charging demand during charging will increase the vehicle charging load pressure,an adaptive mechanism is introduced to share the charging pressure by appropriately adding vehicles according to the vehicle load.Then,new environmental response mechanisms and probability-based evolutionary operator are designed due to the node assignment problem when new charging demands are generated and the inapplicability of the evolutionary strategy caused by the use of multi-vehicle charging mode.Finally,the multi-vehicle dynamic charging scheduling model is transformed into a dynamic optimization problem,and an improved manyobjective algorithm is used to achieve the optimal solution.The effectiveness of the proposed adaptive multi-vehicle dynamic charging scheduling model is verified by simulation experimental results.(3)Since the adaptive multi-vehicle charging approach used in the above proposed adaptive multi-vehicle charging scheduling model will dramatically increase the number of solutions in the decision space,resulting in a serious imbalance between convergence and diversity of the algorithm.Therefore,in order to improve the convergence of the algorithm and design more efficient charging schemes,a hybrid environment selection strategy based on pareto principle is designed in this paper.First,in order to select a more efficient charging scheme among a large number of solutions,an individual preference evaluation index consisting of node mortality and energy utilization is designed and individuals of the population are selected in the population together with one-byone selection strategy.Second,to ensure the balance of individual diversity and convergence in the selected population,the population size is divided according to pareto principle and selected by the above strategy.Simulation experiments show that the introduced hybrid environment selection strategy based on pareto principle can effectively improve the performance of the algorithm.