A charging method and system based on fine-grained power allocation in a wireless rechargeable sensor network

Abstract

The present invention relates to a charging method and system based on fine-grained power allocation in a wireless rechargeable sensor network, and belongs to the technical field of wireless sensor networks. The method includes: for the problem of minimizing the total charging cost based on dynamic power allocation in a wireless rechargeable sensor network, constructing an integer programming model based on charging completion time and full charging constraints; converting the integer programming model into a charging cost reduction maximization problem model, and converting the charging cost reduction maximization problem model into a submodule function problem model with multiple linear constraints; solving the submodule function problem model and outputting the optimal charging scheduling strategy. This method minimizes the charging service cost of the entire sensor network under the premise of completing full charging for all sensors in the sensor network, thereby improving charging utility and saving costs.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless sensor networks, and in particular relates to a charging method and system based on fine-grained power allocation in a wireless rechargeable sensor network. Background Art

[0002] At present, breakthroughs in wireless power transfer (WPT) technology have attracted widespread attention from academia and industry. Based on this technology, energy can be wirelessly transferred from power chargers to rechargeable devices, which provides a promising solution to the energy limitation problem of terminal devices. Since WPT technology has the advantages of high reliability and high efficiency of continuous power supply, using WPT technology to build rechargeable wireless sensor networks (WSNs) can easily extend the network life without spending a lot of manpower costs to replace batteries. Current research focuses on the scheduling of static chargers, including charger location placement, direction scheduling, power allocation and switching time, or mobile chargers access sensor devices for energy replenishment. They use the high mobility and flexibility of vehicles or drones equipped with chargers to optimize sensor sleep time, maximize on-demand charging requests, minimize charging delays, and maximize charging rewards.

[0003] Existing wireless sensor networks are widely used in monitoring environments such as oceans, volcanoes, ecosystems, bridges, etc. However, energy limitations have always hindered the efficient operation of these networks. This is because most sensor networks rely on batteries for power. Regular battery replacement not only causes frequent downtime of sensor devices, but also has high costs in harsh environments. Therefore, in recent years, with the rapid development of wireless charging technology, Wireless Rechargeable Sensor Networks (WRSNs) have gradually attracted the attention of the professional field.

[0004] A wireless rechargeable sensor network is a sensor network that uses wireless charging technology to recharge energy for sensor devices. Compared with traditional sensor networks, wireless rechargeable sensor networks have at least the following advantages: 1) Wireless charging eliminates complex cable connections, and devices of different brands and models can use the same charging device. 2) The use of wireless charging devices can avoid contact with the outside world, making the overall device better waterproof and dustproof. 3) It extends the working range of traditional wireless sensor networks, such as underwater work, pacemakers and other special applications. 4) Wireless charging does not produce waste batteries and is more environmentally friendly. 5) The charging scheduling of wireless charging is more timely and flexible, which can maximize the guarantee that the working time of the equipment is sufficient to meet the working needs.

[0005] Currently, the commercialization of wireless charging mostly stays at the sales of wireless charging equipment, with less consideration given to the scheduling of paid wireless charging services. Summary of the invention

[0006] The main purpose of the present invention is to overcome the shortcomings and deficiencies of the prior art and to provide a charging method and system based on fine-grained power allocation in a wireless rechargeable sensor network, which minimizes the charging service cost of the entire sensor network while ensuring that all sensors in the sensor network are fully charged, thereby improving charging efficiency and saving costs.

[0007] According to one aspect of the present invention, the present invention provides a charging method based on fine-grained power allocation in a wireless rechargeable sensor network, the method comprising the following steps:

[0008] S1: Aiming at the problem of minimizing the total charging cost based on dynamic power allocation in wireless rechargeable sensor networks, an integer programming model based on charging completion time and full charging constraints is constructed;

[0009] S2: Based on the characteristic that the charging cost of the sensor network has an upper limit, the integer programming model is transformed into a charging cost reduction maximization problem model, and the charging cost reduction maximization problem model is transformed into a submodule function problem model with multiple linear constraints;

[0010] S3: Solve the submodule function problem model and output the optimal charging scheduling strategy.

[0011] Preferably, the integer programming model is:

[0012]

[0013] Among them, B i 、E i Respectively represent the device v i Capacitance, current remaining capacity, Indicates charging stations j Power When providing charging service, rechargeable sensor devices v i The charging power that can be received, c k Indicates the charging station power P k The price of charging service for a time slot is provided below. Indicates charging stations j At power P k The charging service time provided below, T is the upper limit of charging time.

[0014] Preferably, the charging cost reduction maximization problem model is:

[0015]

[0016] in, For charging stations j In gear position P k The charging service time provided is relative to the charging station s in the maximum charging strategy j In gear position P k Charging service time provided The reduction in For all k≠d, there exists and

[0017] Preferably, the submodule function problem model is:

[0018]

[0019] in, is a real-valued function, w i (C) represents the rechargeable sensor device v due to the atomic charging scheduling strategy C i The reduction in the received charging energy, subset U j is a finite basis set based on atomic strategies.

[0020] Preferably, in the solution set obtained by solving the submodule function problem model, if there are constraints that are not suitable for the original problem model before transformation, a selection operation is performed to make the final solution meet the constraints.

[0021] According to another aspect of the present invention, the present invention also provides a charging system based on fine-grained power allocation in a wireless rechargeable sensor network, the system comprising:

[0022] A building block for constructing an integer programming model based on charging completion time and full charging constraints for the total charging cost minimization problem based on dynamic power allocation in wireless rechargeable sensor networks;

[0023] A conversion module, for converting the integer programming model into a charging cost reduction maximization problem model based on the characteristic that the charging cost of the sensor network has an upper limit, and converting the charging cost reduction maximization problem model into a submodule function problem model with multiple linear constraints;

[0024] The solution module is used to solve the submodule function problem model and output the optimal charging scheduling strategy.

[0025] Preferably, the integer programming model is:

[0026]

[0027] Among them, B i 、E i Respectively represent the device v i Capacitance, current remaining capacity, Indicates charging stations j Power When providing charging service, rechargeable sensor devices v i The charging power that can be received, c k Indicates the charging station power P k The price of charging service for a time slot is provided below. Indicates charging stations j At power P k The charging service time provided below, T is the upper limit of charging time.

[0028] Preferably, the charging cost reduction maximization problem model is:

[0029]

[0030] in, For charging stations j In gear position P k The charging service time provided is relative to the charging station s in the maximum charging strategy j In gear position P k Charging service time provided The reduction in For all k≠d, there exists and

[0031] Preferably, the submodule function problem model is:

[0032]

[0033] in, is a real-valued function, w i (C) represents the rechargeable sensor device v due to the atomic charging scheduling strategy C i The reduction in the received charging energy, subset U j is a finite basis set based on atomic strategies.

[0034] Preferably, in the solution set obtained by solving the submodule function problem model, if there are constraints that are not suitable for the original problem model before transformation, a selection operation is performed to make the final solution meet the constraints.

[0035] Beneficial effects: The charging method and system based on fine-grained power allocation in the wireless rechargeable sensor network proposed by the present invention improves the overall network charging service quality and efficiency, and can be widely used in the application scenario of providing on-demand pay charging services for terminal devices in the wireless rechargeable sensor network, which has certain practical significance and commercial value. Compared with other heuristic methods, the present invention has significant advantages in optimizing charging service costs.

[0036] The features and advantages of the present invention will become clear through reference to the following drawings and detailed description of specific embodiments of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a flow chart of a charging method based on fine-grained power allocation in a wireless rechargeable sensor network;

[0038] Figure 2 is a schematic diagram of a network model in this embodiment;

[0039] Figure 3 It is a schematic diagram of a charging system based on fine-grained power allocation in a wireless rechargeable sensor network. DETAILED DESCRIPTION

[0040] The following is a clear and complete description of the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0041] Example 1

[0042] Figure 1 This is a flow chart of a charging method based on fine-grained power allocation in a wireless rechargeable sensor network. Figure 1 As shown, this embodiment provides a charging method based on fine-grained power allocation in a wireless rechargeable sensor network, and the method includes the following steps:

[0043] S1: Aiming at the problem of minimizing the total charging cost based on dynamic power allocation in wireless rechargeable sensor networks, an integer programming model based on charging completion time and full charging constraints is constructed.

[0044] Specifically, for the problem of minimizing the total charging cost based on dynamic power allocation in a wireless rechargeable sensor network, an integer programming model based on charging completion time and full charging constraints is constructed; wherein, firstly, a network model of the wireless rechargeable sensor network is established, including charging stations, omnidirectional wireless chargers, rechargeable sensor devices and charging service providers; then, all sensor devices in the network are fully charged under the premise of limiting the completion of the task; finally, the integer programming model for minimizing the charging cost is obtained by combining the charging power switching time of the charging station and the upper limit of the charging time.

[0045] S2: Based on the characteristic that the charging cost of the sensor network has an upper limit, the integer programming model is transformed into a charging cost reduction maximization problem model, and the charging cost reduction maximization problem model is transformed into a submodule function problem model with multiple linear constraints.

[0046] Specifically, based on the characteristic that the charging cost of the sensor network in the constructed model has an upper limit, the problem is transformed into a problem of maximizing the reduction of charging costs, and it is proved that the problem is a submodule function problem with multiple linear constraints. Under the premise that the maximum charging time and the maximum charging power of the network are determined, the charging cost of the entire network has a maximum value. According to this property, the problem is transformed into a problem of maximizing the reduction of charging costs, and it is proved that the problem is a submodule function problem with multiple linear constraints by transforming various conditions in the problem.

[0047] S3: Solve the submodule function problem model and output the optimal charging scheduling strategy.

[0048] Specifically, based on the constructed submodular set function problem with multiple linear constraints, an approximation algorithm for this type of problem is given, and it is proved that the algorithm also has an approximation ratio for the original problem, thereby determining the final solution.

[0049] The submodular set function problem with multiple linear constraints is essentially a multidimensional knapsack problem with weights. To solve this problem, a maximum coverage algorithm with multiple packing constraints is used. This algorithm gives a polynomial time solution for the submodular problem with multiple knapsack constraints. The approximation algorithm of the approximation ratio is ε, which is a number between 0 and 1. The result obtained by the algorithm can be filtered to obtain the solution to the original problem, which is also proved to have Approximate ratio.

[0050] Under the premise of fully charging all sensors in the sensor network, the method minimizes the charging service cost of the entire sensor network, improves charging efficiency and saves costs.

[0051] Preferably, the integer programming model is:

[0052]

[0053] Among them, B i 、E i Respectively represent the device v i Capacitance, current remaining capacity, Indicates charging stations j Power When providing charging service, rechargeable sensor devices v i The charging power that can be received, c k Indicates the charging station power P k The price of charging service for a time slot is provided below. Indicates charging stations j At power P k The charging service time provided below, T is the upper limit of charging time.

[0054] Specifically, consider a network model consisting of a base station and n low-power rechargeable sensor devices V = {v1, v2, ..., v n}, which are deployed and operated by users to perform specific monitoring tasks. In addition, assume that there are m charging stations S = {s1, s2, ..., s m}, operated by a charging service provider (CSP), providing paid charging services. Each charging station is equipped with an omnidirectional wireless charger, and they all have d charging power levels P1, P2, ..., P d , and P1<P2<…<P d Adjustable; use c k (k=1, ..., d) represents the charging station with power P k The price of charging service for a time slot is provided below. In practice, it is usually followed that c1<c2<…<c d .

[0055] All charging stations are deployed and operated by a charging service provider CSP; the network will conduct charging scheduling at regular operating cycles.

[0056] The charging completion time is defined as the time from the first charging station starting to provide charging service to the last charging station ending charging service. All charging stations start charging service at the same time. The charging completion time is expressed as

[0057] The service quality of the charging task is limited. When the charging task is completed, all sensors in the sensor network need to be fully charged, that is:

[0058]

[0059] Among them B i 、E i Respectively represent the device v i Capacitance, current remaining capacity, Indicates charging stations j Power When providing charging service, rechargeable sensor devices v i The charging power that can be received.

[0060] Given a charging time upper limit T, a charging scheduling strategy for charging stations is established to maximize the network charging utility. The target problem: Fine-grained EconomicalCharging Scheduling based on Temporal-constraint (FECST) problem is modeled as an integer programming model to minimize the charging cost.

[0061] Preferably, the charging cost reduction maximization problem model is:

[0062]

[0063] in, For charging stations j In gear position P k The charging service time provided is relative to the charging station s in the maximum charging strategy j In gear position P k Charging service time provided The reduction in For all k≠d, there exists and

[0064] Specifically, the base station will periodically initiate charging commands to all sensor devices in the network. After receiving the charging command, each device will immediately report its charging needs to the base station. Based on the charging needs of all devices received and specific optimization goals, the base station will formulate an efficient charging scheduling strategy and provide it to the CSP to request charging services. After receiving the charging service request from the base station, the CSP will immediately forward the scheduling strategy to each charging station to allocate charging power and corresponding charging service time.

[0065] definition Indicates charging stations j At power P k The charging service time provided below is for any charging station s j , by executing the received charging scheduling strategy This strategy indicates that charging station s j The charging power will be adjusted to P1, P2, ..., P d , respectively provide corresponding charging service time That is j The total charging service time is

[0066] Under the premise that the maximum charging time and the maximum charging power of the network are determined, the charging cost of the entire network has a maximum value. Based on this property, the problem is transformed into a problem of maximizing the reduction of charging costs.

[0067] Since the charging unit price is c1<c2<…<c d , then the FECST problem must have a maximum cost solution, that is, each charging station works at the highest charging gear for T time slots: At this time, the entire sensor network reaches the maximum charging service cost: m·c d ·T.

[0068] For any set of feasible solutions to the FECST problem It can be found that it can be replaced by another set of relative charging scheduling strategies to describe, and they have the same meaning. Defined as the maximum charging cost solution Medium Charging Stations j In gear position P k The charging service time provided is relative to the charging station s in the maximum charging strategy j In gear position P k Charging service time provided The reduction of , and obviously has the following properties: 1) and 2) For all k≠d, there exists and

[0069] For rechargeable sensor devices in wireless sensor networks, the FECST problem has a constraint that all of them must be fully charged. Charging scheduling strategy corresponding to the converted problem This constraint can be expressed as follows:

[0070]

[0071] in Use the maximum cost feasible solution for the entire system When the i-th rechargeable sensor v i The received charging energy; therefore, the FECST problem can be transformed into an equivalent charging service fee reduction-amount Maximization (CSFRM) problem.

[0072] The goal is to find a relative charging scheduling strategy based on fine-grained power allocation. To maximize the final charging cost relative to the maximum charging service cost m·c d The reduction of T.

[0073] By transforming various conditions in the problem, it is proved that the problem is a submodular set function problem with multiple linear constraints.

[0074] For any given relative charging scheduling strategy For ease of description, use Denotes the charging scheduling strategy for any k∈D The charging scheduling time of the kth gear in By defining each k∈{0, 1, ..., d-1} The physical meaning of this construction is that no matter which time slot among the T time slots is adjusted for charging power, the final charging strategy is reflected in the same physical meaning, and thus a finite basis set based on the atomic strategy can be constructed. For charging stations j Any feasible relative charging scheduling strategy There must be a subset Such that for every k∈D there exists At this point, any feasible relative charging scheduling strategy can be found can be equivalently expressed as U1, U2, ..., U m A subset of the atomic charging scheduling strategy in . And for any Defining functions and as follows:

[0075]

[0076] in function Describes the implementation of the given relative charging scheduling strategy on the final charging strategy C j After the above, compared with the original best charging strategy Charging cost, charging strategy Cj How much the charging cost is reduced. Describes the implementation of the given relative charging scheduling strategy on the final charging strategy C j After that, for each rechargeable sensor device v i , how much the received charging energy is reduced respectively; and the constraint equation This is essentially equivalent to the following equation:

[0077]

[0078] Used to ensure that the given relative charging scheduling strategy is implemented in the final charging strategy C j After that, for each rechargeable sensor device v i , the charging energy it receives can fully charge the battery.

[0079] After further changing the formula, the original CSFRM problem can be transformed into the following expression:

[0080]

[0081] For each j∈M, there is a finite basis set based on atomic strategies And the goal is to start from U1, U2, ..., U m Find the optimal subset U′1, U′2, ..., U′ m ;

[0082] Preferably, the submodule function problem model is:

[0083]

[0084] in, is a real-valued function, w i (C) represents the rechargeable sensor device v due to the atomic charging scheduling strategy C i The reduction in the received charging energy, subset U j is a finite basis set based on atomic strategies.

[0085] Specifically, for any C∈U when j∈M j Using w i (C)(i∈N) represents the rechargeable sensor device v due to the atomic charging scheduling strategy C. i The reduction in the received charging energy is:

[0086]

[0087] where C∈Uj , and it must exist with w i (C)≥0, because In particular, At this point, we can further define the Multiple Knapsack constrained Submodular Function Maximization (MKSFM) problem.

[0088] For the MKSFM problem, there is a maximum coverage with multiple packing constraints (MCMP) algorithm that gives a polynomial time solution for the multiple knapsack constrained submodule problem. Approximation algorithm for approximation ratio, where ε is a number between 0 and 1.

[0089] Regulation is the atomic operation in the atomic strategy set that reduces the highest gear d to the kth gear. The total set of all atomic strategies is

[0090] These Q o (o∈{1,...,m·(d-1)}) is built on the basis set U j At this time, we can guess l pairs of matches (U j , Q o ) becomes a set B, define set B ε is the set of elements in B, set B s is the set of sets in B. And k = m·T is the number constraint that the final solution set needs to satisfy, that is, the number of atomic strategies in the final solution set for each charging station s j Cannot exceed T. And G r is the set of large sets that meet certain requirements in the conjecture, Q′ o For Q o Related but not in B ε The elements in x j ∈{0,1} is Q′ o Indicator of whether it is selected. j It is a 0 / 1 value that determines whether the element is selected with a probability. If y j =1, it means that the element is selected. If y j =0, the element is discarded.

[0091] If k ≤ ε -3+l Enumerate the subsets of the optimal solution and run MKP's PTAS to select the elements. (This guarantees an approximate ratio of (1-ε))); for the pairing (U j , Q o ) in each set B of maximum size l, and maximum size ε -4 Any Do the following:

[0092] For B, G r , is a set of randomly determined settings for the solution set. For any setting Q o (1≤o≤m·(d-1)), if Q o ∈B s Then add it to the solution. Otherwise, add Q o With probability (1-ε)x i Add to the solution set. If the resulting solution D contains more than k subsets, an empty solution is returned; otherwise, define make As a partial solution; initial design For any Q o ∈Q, if Q o ∈B s Q o Add to D; otherwise, add Q o (1-ε)x i probability is added to D; if the number of sets in D exceeds k, continue to the next iteration of the loop; when y j = 1, let A be the element U j ∈H; if A∪B ε The weight of the element in is greater than the weight of the current optimal solution, so choose A∪B ε And take D as the current optimal solution.

[0093] Combine A∪B ε And D can get the optimal solution

[0094] Preferably, in the solution set obtained by solving the submodule function problem model, if there are constraints that are not suitable for the original problem model before transformation, a selection operation is performed to make the final solution meet the constraints.

[0095] Specifically, the solution set given by the MCMP algorithm may not be suitable for the original problem, and a selection operation is given to make the final solution meet the constraints;

[0096] Here order And for any Define a set of real-valued functions as follows:

[0097]

[0098] where f top_T (U j ′) represents a filtering operation used to ensure that |U j ″|≤T to find a subset Make Maximize. In particular, if there exists |U j ′|≤T, then f top_T (U j ′)=U j ′; and if |U j ′|>T, then

[0099] Real-valued functions The meaning is: if the current relative charging scheduling strategy If the number of atomic dispatch strategies is less than or equal to T, then the scheduling strategy meets the constraints and can be used directly; but if the number of atomic dispatch strategies is greater than T, then there is obviously no such scheduling strategy, because the initial maximum charging strategy There are only T working time slots available for scheduling, so in this case, the T atomic scheduling strategies that can maximize the objective function are selected as the solution set.

[0100] This embodiment improves the overall network charging service quality and efficiency, and can be widely used in the application scenario of providing on-demand pay charging services for terminal devices in wireless rechargeable sensor networks, which has certain practical significance and commercial value. Compared with other heuristic methods, this embodiment has significant advantages in optimizing charging service costs.

[0101] In order to more fully understand the technical solution of the present invention, another example is given below.

[0102] See also Figure 2 , there are 2 wireless charging stations s1, s2 and 2 rechargeable sensor devices v1, v2 in the monitoring area. For simplicity and without loss of generality, it is assumed that each charging station has 2 levels (i.e. d = 2) of discrete charging power P1 and P2 (P1 < P2) that can be adjusted, and the corresponding unit time charging service price c1 = 1, c2 = 2 is set. Further, it is assumed that the relevant parameters are set as follows: B1 = B2 = 10, E1 = E2 = 5, And set the upper limit of charging completion time T=2.

[0103] The following planning model is obtained through modeling:

[0104]

[0105] Then we get the following problem model:

[0106]

[0107] Further transformation, we get the following problem model:

[0108]

[0109] Get the total set of all atomic strategies These Q o (o∈{1,2,3,4}) is based on the basis set U j At this time, it is pre-guessed that the matching (U j , Q o ) becomes a set B, define set B ε = {U1, U2, Q2, Q4} is the set of elements in B, and set B s ={(U1, Q2), (U2, Q4)} is the set of sets in B. This is because if (U1, Q1) or (U2, Q3) is used, the final result will not satisfy the full charging constraint. And k=m·T=4 is the number constraint that the final solution set needs to satisfy, that is, the number of atomic strategies in the final solution set for each charging station s j No more than 2. r is the set of large sets that meet certain requirements in the conjecture, Q′ o For Q o Related but not in B ε The elements in x j ∈{0,1} is Q′ o Indicator of whether it is selected. j It is a 0 / 1 value that determines whether the element is selected with a probability. If y j =1, it means that the element is selected. If y j =0, the element is discarded.

[0110] At this time, we first enumerate the strategies {(U1, Q1), (U1, Q2)} in U1, and first select (U1, Q1), which is not in B s In the solution set D, with probability not added, for (U1, Q2), it is in B s , add it to the solution set D. Then enumerate the strategies in U2 {(U2, Q3), (U2, Q4)}, first choose (U2, Q3), which is not in B s In the solution set D, with probability not added, for (U2, Q4), it is in B s, and add it to the solution set D. At this time, we get the solution set for the MKSFM problem

[0111] The current solution set meets the requirements of the original problem and directly obtains the optimal solution and

[0112] In order to evaluate the performance of this method in solving the charging cost minimization problem with fine-grained heterogeneity in prices in a wireless charging network, the performance of the method of this embodiment is compared with some common optimization algorithms after the number of rechargeable devices n, the number of wireless chargers m, the maximum battery capacity B of the rechargeable device, and the maximum number of charging time slots T in the network are changed to a certain extent.

[0113] In the simulation, the sensing area is a square area of ​​100m×100m, and the chargers and rechargeable devices are evenly distributed in the area. The default values ​​are set for the parameters in the network: the number of rechargeable devices n, the number of wireless chargers m, the maximum battery capacity B of the rechargeable device, and the maximum number of charging time slots T: n=200, m=15, the maximum charging power=1000J / time slot, the maximum number of charging time slots T=5, the number of charger gears d=5, and the charging power of each gear is evenly distributed according to the gear size, the maximum battery capacity of the rechargeable device B=50J, a=5, β=2, and the remaining power of the sensor at the beginning of charging is a random number between 0 and the maximum value. Since the charging unit price is heterogeneous between different gears, the lowest unit charging price is set to 1, and the unit prices of the remaining charging gears are 1.3, 1.6, 1.9, 2.2...

[0114] The following is a comparison between this method and three algorithms. They are: Greedy Algorithm (GA): Take turns to adjust the highest power atomic strategy of each charging station down by one charging gear. If a charging station cannot be adjusted down, it will be skipped until all charging stations cannot complete the adjustment operation; Randomized Algorithm (RDM): Take turns to select the highest power atomic strategy of the charging station, randomly select a gear with lower power than the atomic strategy to adjust down, if it cannot guarantee full charging after the adjustment, it will be skipped until all charging stations cannot complete the adjustment operation; Local Optimal Power Algorithm (FLA): Under the premise of ignoring the influence of other charging stations, select a charging gear for each charger to work all time slots. The selected solution is to work all time slots at this power to ensure that all sensors within the range of the charging station are fully charged.

[0115] The test results show that this method has great advantages over the baseline algorithm, and it has achieved the pre-selected goals well, demonstrating the feasibility and superiority of this method.

[0116] In the experiment, the number of rechargeable devices was increased from 150 to 300, and the charging costs of the four algorithms under different numbers of rechargeable devices were compared. The charging costs of this method in 7 different scenarios were 57955, 61100, 66960, 76370, 77600, 81260, and 80810, which were 9.97%, 15.12%, and 17.79% less than the other three baseline algorithms on average. The budget utilization efficiency of other baseline algorithms is shown in Table 1.

[0117] Table 1 The impact of the number of rechargeable devices on charging costs

[0118]

[0119] Then the number of chargers is increased from 10 to 25, and the charging costs of the four algorithms under different numbers of chargers are compared. The charging costs of this method in seven different scenarios are 61660, 67200, 66960, 64910, 65420, 62500, and 53020, which are 35.86%, 17.94%, and 23.09% less than the other three baseline algorithms on average. The budget utilization efficiency of other baseline algorithms is shown in Table II.

[0120] Table 2 The impact of the number of chargers on charging costs

[0121]

[0122] The maximum battery capacity of the rechargeable device is increased from 30J to 60J, and the charging costs of the four algorithms under different maximum battery capacities of the rechargeable device are compared. The charging costs of this method in 7 different scenarios are 34740, 41580, 50770, 60090, 66960, 73230, and 83810, which are 34.99%, 13.09%, and 22.93% less than the other three baseline algorithms on average. The budget utilization efficiency of other baseline algorithms is shown in Table III.

[0123] Table 3 Effect of maximum battery capacity of rechargeable devices on charging cost

[0124]

[0125] The number of charger gears is increased from 5 to 11, and the charging costs of the four algorithms under different charger gear numbers are compared. The charging costs of this method in 7 different scenarios are 66960, 64380, 57730, 53950, 52900, 47970, and 45680, which are 43.78%, 12.41%, and 33.74% less than the other three baseline algorithms on average. The budget utilization efficiency of other baseline algorithms is shown in Table IV.

[0126] Table 4 The impact of the number of charger gears on charging costs

[0127]

[0128] The maximum number of charging time slots is increased from 5 to 11, and the charging costs of the four algorithms under different maximum number of charging time slots are compared. The charging costs of this method in 7 different scenarios are 66960, 61150, 57730, 55290, 47270, 41290, and 35680, which are 32.19%, 16.22%, and 23.51% less than the other three baseline algorithms on average. The budget utilization efficiency of other baseline algorithms is shown in Table 5.

[0129] Table 5. The impact of the maximum number of charging time slots on charging costs

[0130]

[0131] Example 2

[0132] Figure 3 Figure 1 is a schematic diagram of a charging system based on fine-grained power allocation in a wireless rechargeable sensor network. Figure 3 As shown, this embodiment provides a charging system based on fine-grained power allocation in a wireless rechargeable sensor network, the system comprising:

[0133] A construction module 301 is used to construct an integer programming model based on charging completion time and full charging constraints for the total charging cost minimization problem based on dynamic power allocation in a wireless rechargeable sensor network;

[0134] A conversion module 302 is used to convert the integer programming model into a charging cost reduction maximization problem model based on the characteristic that the charging cost of the sensor network has an upper limit, and convert the charging cost reduction maximization problem model into a submodule function problem model with multiple linear constraints;

[0135] The solution module 303 is used to solve the submodule function problem model and output the optimal charging scheduling strategy.

[0136] Preferably, the integer programming model is:

[0137]

[0138] Among them, B i 、E i Respectively represent the device v i Capacitance, current remaining capacity, Indicates charging stations j Power When providing charging service, rechargeable sensor devices v i The charging power that can be received, c kIndicates the charging station power P k The price of charging service for a time slot is provided below. Indicates charging stations j At power P k The charging service time provided below, T is the upper limit of charging time.

[0139] Preferably, the charging cost reduction maximization problem model is:

[0140]

[0141] in, For charging stations j In gear position P k The charging service time provided is relative to the charging station s in the maximum charging strategy j In gear position P k Charging service time provided The reduction in For all k≠d, there exists and

[0142] Preferably, the submodule function problem model is:

[0143]

[0144] in, is a real-valued function, w i (C) represents the rechargeable sensor device v due to the atomic charging scheduling strategy C i The reduction in the received charging energy, subset U j is a finite basis set based on atomic strategies.

[0145] Preferably, in the solution set obtained by solving the submodule function problem model, if there are constraints that are not suitable for the original problem model before transformation, a selection operation is performed to make the final solution meet the constraints.

[0146] The specific implementation process of the functions realized by each module in this embodiment 2 is the same as the implementation process in embodiment 1, and will not be repeated here.

[0147] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. All equivalent structural changes made by using the contents of the present invention specification and drawings under the concept of the present invention, or directly / indirectly applied in other related technical fields are included in the patent protection scope of the present invention.

Claims

1. A charging method based on fine-grained power allocation in a wireless rechargeable sensor network, characterized in that: The method comprises the following steps: S1: Aiming at the problem of minimizing the total charging cost based on dynamic power allocation in wireless rechargeable sensor networks, an integer programming model based on charging completion time and full charging constraints is constructed; S2: Based on the characteristic that the charging cost of the sensor network has an upper limit, the integer programming model is transformed into a charging cost reduction maximization problem model, and the charging cost reduction maximization problem model is transformed into a submodule function problem model with multiple linear constraints; S3: solving the submodule function problem model and outputting an optimal charging scheduling strategy; The integer programming model is: ， Among them, , Respectively represent devices Capacitance, current remaining capacity, Indicates charging station Power When providing charging service, sensor devices can be charged The charging power that can be received, Indicates the power of the charging station The price of charging service for a time slot is provided below. Indicates charging station In power The charging service time provided below, is the upper limit of charging time; the model for maximizing the problem of reducing charging cost is: ， in, For charging stations In power The charging service time provided is relative to the charging station in the maximum charging strategy In power Charging service time provided The reduction in ; For all , all exist and ; The submodule function problem model is: ， in, is a real-valued function, Indicates that due to the atomic charging scheduling strategy Leading to rechargeable sensing devices The reduction in the received charging energy, subset , is a finite basis set based on atomic strategies.

2. The method according to claim 1, characterized in that In the solution set obtained by solving the submodule function problem model, if there is a situation that is not suitable for the constraint conditions of the original problem model before transformation, a selection operation is performed to make the final solution meet the constraint conditions.

3. A charging system based on fine-grained power allocation in a wireless rechargeable sensor network, characterized in that: The system comprises: A building block for constructing an integer programming model based on charging completion time and full charging constraints for the total charging cost minimization problem based on dynamic power allocation in wireless rechargeable sensor networks; A conversion module, for converting the integer programming model into a charging cost reduction maximization problem model based on the characteristic that the charging cost of the sensor network has an upper limit, and converting the charging cost reduction maximization problem model into a submodule function problem model with multiple linear constraints; A solution module is used to solve the submodule function problem model and output the optimal charging scheduling strategy; the integer programming model is: ， Among them, , Respectively represent devices Capacitance, current remaining capacity, Indicates charging station Power When providing charging service, sensor devices can be charged The charging power that can be received, Indicates the power of the charging station The price of charging service for a time slot is provided below. Indicates charging station In power The charging service time provided below, is the upper limit of charging time; the model for maximizing the problem of reducing charging cost is: ， in, For charging stations In power The charging service time provided is relative to the charging station in the maximum charging strategy In power Charging service time provided The reduction in ; For all , all exist and ; The submodule function problem model is: ， in, is a real-valued function, Indicates that due to the atomic charging scheduling strategy Leading to rechargeable sensing devices The reduction in the received charging energy, subset , is a finite basis set based on atomic strategies.

4. The system according to claim 3, characterized in that In the solution set obtained by solving the submodule function problem model, if there is a situation that is not suitable for the constraint conditions of the original problem model before transformation, a selection operation is performed to make the final solution meet the constraint conditions.

Images