A mobile charging scheduling method, device, network and application for optimizing monitoring utility

Abstract

The present invention belongs to the technical field of mobile charging scheduling, and specifically discloses a mobile charging scheduling method, device, network and application for optimizing monitoring utility. The method comprises: S1 defining a network model driven by monitoring utility; S2 defining an energy consumption model; S3 defining a monitoring utility model; S4 formalizing the problem of optimizing mobile charging scheduling by monitoring utility; S5 using regional discretization to transform the problem, constructing an auxiliary graph, and further transforming the problem. For the transformed problem, a monitoring utility optimization mobile charging scheduling algorithm is called to obtain a charging scheduling plan for a mobile charger. The mobile charging scheduling method, device, network and application provided by the present invention can determine the charging scheduling plan for a mobile charger without exceeding the energy capacity of the mobile charger, thereby maximizing the total monitoring utility of the sensor network.

Description

Technical Field

[0001] The present invention relates to the technical field of mobile charging scheduling, and specifically discloses a mobile charging scheduling method, device, network and application for optimizing monitoring utility. Background Art

[0002] Because of its advantages such as flexible deployment, wide coverage, high real-time performance and low cost, wireless sensor networks have been widely used in military monitoring, disaster prediction, biomedical health monitoring, and hazardous environment survey. At the same time, with the further development of wireless energy transmission technology, wireless rechargeable sensor networks have also been widely developed in real life and have been applied to drones, rechargeable robots, and radio frequency identification systems.

[0003] Mobile charging is a more appropriate choice in large-scale networks with a limited number of chargers. Previous studies on mobile charging only considered how to charge sensors without involving the impact of specific monitoring targets on monitoring utility, and only used energy to judge utility. In this case, each sensor is roughly considered equivalent. However, in reality, even if different sensors receive the same amount of energy, their utilities may be different. Therefore, we need to consider the differences between sensors. Summary of the invention

[0004] The technical problem to be solved by the present invention is to find a mobile loop for the mobile charger to charge the sensors on the path, and to maximize the total monitoring utility in the network without exceeding the energy capacity of the mobile charger.

[0005] In order to solve the above problems, the present invention discloses a mobile charging scheduling method, device, network and application for optimizing monitoring utility. In the present invention, we consider that there are some initial monitoring sub-areas that need to be monitored, and the unit area monitoring utilities of different initial monitoring sub-areas are different. By scheduling mobile chargers to charge sensors, the total monitoring utility is maximized without exceeding the energy capacity of the mobile chargers.

[0006] The technical solution of the present invention is as follows:

[0007] A mobile charging scheduling method for optimizing monitoring utility, comprising:

[0008] S1 defines the monitoring utility driven network model;

[0009] S2 defines the energy consumption model;

[0010] S3 defines the monitoring utility model;

[0011] S4 Formal monitoring utility optimization mobile charging scheduling problem;

[0012] S5 uses regional discretization to transform the problem, constructs an auxiliary graph, and performs further problem transformation. For the transformed problem, it calls the monitoring utility optimization mobile charging scheduling algorithm to obtain the charging scheduling plan for the mobile charger.

[0013] As a mobile charging scheduling method for optimizing monitoring utility described in the present invention, wherein: the network model driven by monitoring utility includes:

[0014] The set of sensors distributed on the two-dimensional plane R is O = {o1, o2, ..., o n}, where n is the number of sensors, for each sensor o j ∈O, the energy requirement is c(o j ), the initial battery power of all sensors is 0, and the monitoring radius is r; the initial monitoring sub-area set is Where m0 is the number of initial monitoring sub-areas, for each initial monitoring sub-area There is a monitoring utility per unit area w z , where z∈{1,2,…,m0}; a mobile charger starts from warehouse s0 and is allowed to stay at any position on the two-dimensional plane R to charge the sensors one-to-many. The maximum charging range of the mobile charger is D, and the energy capacity of the mobile charger is E; the scenario diagram is as follows Figure 1 shown.

[0015] As a mobile charging scheduling method for optimizing monitoring utility described in the present invention, wherein: the energy consumption model is defined, including:

[0016] Sensor j From any position s on the two-dimensional plane i The charging power obtained by ∈R is expressed as:

[0017]

[0018] Among them, α, β are constants determined by hardware and environment, d(s i ,o j ) is the charging position s i and sensor o j The distance between

[0019] When the mobile charger reaches the charging position s i When the energy demand of all sensors within the maximum charging range must be met, the set of sensors within the maximum charging range is defined as Λ(s i )={o j |d(s i ,o j )≤D}, the mobile charger is at the charging position s iThe charging time is:

[0020]

[0021] The transmitting power of the mobile charger is γ, at the charging position s i The charging energy consumption is γt(s i ); if an o j Belong to multiple Λ(s i ), when o j After being charged once, its energy requirement will become 0; in addition, s0 is not considered as a charging position, so the charging time of s0 position is zero;

[0022] Let L = {s0,s1,s2,…,s q ,s0} represents the mobile circuit of the mobile charger, where q is the number of charging positions in the circuit, and the unit mobile energy consumption of the mobile charger is μ. Then the total mobile energy consumption of L is Where d(s x ,s x:1 ) represents the charging position s x With charging position x:1 The distance between q , s0) represents the charging position s q The distance to warehouse s0;

[0023] The total energy consumption of the mobile loop L is:

[0024]

[0025] As a mobile charging scheduling method for optimizing monitoring utility described in the present invention, wherein: a monitoring utility model is defined, including:

[0026] like Figure 2 As shown in Figure 1, each sensor has a fixed monitoring range, which divides the initial monitoring sub-area into smaller sub-areas to form the final monitoring sub-area. Let the final monitoring sub-area set be Φ = {φ1, φ2, …, φ m}, where m is the number of final monitoring sub-areas, and each final monitoring sub-area φ l The area of ​​∈Φ is a l ; Use w l Represents the final monitoring sub-area φ l The monitoring utility per unit area is l The monitoring utility is a l w l ; With sensor o j The resulting monitoring sub-area whose monitoring ranges overlap is called the area belonging to sensor o j The final monitoring sub-area, let H(o j ) means it belongs to oj The final monitoring sub-area set of sensor o j The monitoring utility is:

[0027]

[0028] Let S represent the set of charging locations in L, then the set of all covered final monitoring sub-areas is

[0029] The total monitoring utility is:

[0030]

[0031] As a mobile charging scheduling method for optimizing monitoring utility described in the present invention, wherein: the formalized monitoring utility optimization mobile charging scheduling problem includes:

[0032] maxU(S) (6)

[0033]

[0034]

[0035] Among them, formula (6) aims to maximize the total monitoring utility, constraint (7) ensures that the total energy consumption of the mobile charger does not exceed the energy capacity, and constraint (8) ensures that the charging locations in S all belong to R.

[0036] As a mobile charging scheduling method for optimizing monitoring utility described in the present invention, wherein: the problem is transformed by using regional discretization, an auxiliary graph is constructed, and further problem transformation is performed. For the transformed problem, the monitoring utility optimization mobile charging scheduling algorithm is called to obtain a charging scheduling plan for the mobile charger, including:

[0037] The two-dimensional plane R is discretized into a uniform grid with a side length of δ, and then the discretized charging sub-areas are further divided by drawing a circle with a radius of D for each sensor; the distance from each charging position to the sensor in the same charging sub-area is also approximated accordingly; assuming that x charging sub-areas covering at least one sensor have been obtained, a position is randomly selected in each charging sub-area as a candidate charging position, and K = {k1, k2, ..., k χ} represents the set of candidate charging locations; then, the original problem is simplified to P1, where the number of charging locations is limited:

[0038] (P1):maxU(S) (9)

[0039]

[0040]

[0041] According to the original network, construct an auxiliary graph G and convert the charging position s i The charging energy consumption on the edge is distributed to the two edges connected to it, that is, for the edge (s i ,s i′ ), the energy consumption on the auxiliary graph G is Next, transform P1 into P2:

[0042] (P2):maxU(S) (12)

[0043]

[0044]

[0045] When a new charging location is added to the loop, the auxiliary map needs to be updated because the energy requirements of some sensors may become 0;

[0046] A1: Input parameters: sensor set O, warehouse s0, energy capacity E of mobile charger, transmission power γ of mobile charger, unit mobile energy consumption μ of mobile charger, final monitoring sub-area set Γ, candidate charging location set K, auxiliary graph G;

[0047] A2: Initialize the remaining candidate charging location set K * = K, initialize the charging position set in the mobile loop

[0048] A3: If Execute steps A4 to A7, otherwise execute step A9;

[0049] A4: Starting from s0, visit the nearest unvisited charging position in S in turn, and finally return to s0 to obtain loop L;

[0050] A5: For every k h ∈K * , define adding k h The temporary set S(k h ), let S(k h )=S∪{k h}, starting from s0, visit S(k h ) and finally returns to s0 to obtain the loop L(k h );

[0051] A6: Find the location with the maximum utility-to-energy ratio, i.e.

[0052] A7: If Then S = S(kh );

[0053] A8:K * =K * \{k h}, return to step A3;

[0054] A9: Starting from s0, visit the nearest unvisited charging position in S in turn, and finally return to s0 to obtain loop L;

[0055] A10: Output L.

[0056] Furthermore, the present invention discloses the application of the above method in charging scheduling of unmanned aerial vehicles.

[0057] Furthermore, the present invention discloses a charging control device, comprising:

[0058] 1) one or more processors;

[0059] 2) a storage device for storing one or more programs;

[0060] When the one or more programs are executed by the one or more processors, the one or more processors implement the above-mentioned mobile charging scheduling method.

[0061] Furthermore, the present invention also discloses a UAV parallel charging network, comprising the charging control device as described in claim 7.

[0062] Compared with the prior art, the present invention has the following beneficial effects:

[0063] 1. The total monitoring utility function U(S) is monotone, non-negative, and submodular.

[0064] According to the definition of the total monitoring utility function U(S), it is obvious that is monotone and non-negative. Next, we prove its submodular property, that is, prove that the following inequality for any and v∈V\B holds:

[0065] U(A∪{v})-U(A)≥U(B∪{v})-U(B) (15)

[0066] Consider the following three scenarios:

[0067] (1) If Obviously

[0068]

[0069] (2) If and have

[0070]

[0071] (3) If and because Let C = B\A, then

[0072]

[0073] In summary, beneficial effect 1 is established.

[0074] 2. When the total monitoring utility function U(S) is monotone, non-negative, submodular, and the algorithm for generating the moving loop has an approximation of θ, we have Where S * for The optimal solution, T(S) is the energy consumption of the shortest mobile loop when the input set is S, Y c = max{|S|:T(S)≤E},

[0075] Based on the existing theoretical results and beneficial effect 1, beneficial effect 2 is established.

[0076] 3. The time complexity of the monitoring utility optimization mobile charging scheduling algorithm is O(χ 3 ).

[0077] The running time of the algorithm is mainly determined by step A5, which is to find the moving loop. The time complexity of this step is O(χ 2 ), and because in the worst case, step A5 needs to be executed x times, the overall time complexity is O(x 3 ). BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 A schematic diagram of a scenario structure of a mobile charging scheduling method for optimizing monitoring utility of the present invention;

[0079] Figure 2 It is a schematic diagram of the final monitoring sub-area of ​​the present invention;

[0080] Figure 3 This is a flow chart of the monitoring utility optimization mobile charging scheduling algorithm in the present invention;

[0081] Figure 4 This is a total monitoring utility data diagram of changing the number of sensors in the third embodiment of the present invention;

[0082] Figure 5 A diagram of total monitoring utility data of changing energy demand in a third embodiment of the present invention;

[0083] Figure 6A total monitoring utility data diagram of changing the energy capacity of a mobile charger in a third embodiment of the present invention;

[0084] Figure 7 This is a total monitoring utility data diagram of changing the monitoring radius of the sensor in the third embodiment of the present invention;

[0085] Figure 8 This is a graph of total monitoring utility data for changing the side length of the discretized grid in the third embodiment of the present invention. DETAILED DESCRIPTION

[0086] It should be noted that the following embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

[0087] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0088] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0089] Example 1

[0090] The mobile charging scheduling method for optimizing monitoring utility described in the present invention specifically comprises the following steps:

[0091] S1: Define the monitoring utility driven network model;

[0092] The set of sensors distributed on the two-dimensional plane R is O = {o1, o2, ..., o n}, where n is the number of sensors, for each sensor o j ∈O, the energy requirement is c(o j ), the initial battery power of all sensors is 0, and the monitoring radius is r; the initial monitoring sub-area set is Where m0 is the number of initial monitoring sub-areas, for each initial monitoring sub-area There is a monitoring utility per unit area w z, where z∈{1,2,…,m0}; a mobile charger starts from warehouse s0 and is allowed to stay at any position on the two-dimensional plane R to charge the sensors one-to-many. The maximum charging range of the mobile charger is D, and the energy capacity of the mobile charger is E; the scenario diagram is as follows Figure 1 shown.

[0093] S2: define energy consumption model;

[0094] Sensor j From any position s on the two-dimensional plane i The charging power obtained by ∈R is expressed as:

[0095]

[0096] Among them, α, β are constants determined by hardware and environment, d(s i ,o j ) is the charging position s i and sensor o j The distance between

[0097] When the mobile charger reaches the charging position s i When , the energy requirements of all sensors within the maximum charging range must be met. The set of sensors within the maximum charging range is defined as

[0098] Λ(s i )={o j |d(s i ,o j )≤D}, the mobile charger is at the charging position s i The charging time is:

[0099]

[0100] The transmitting power of the mobile charger is γ, at the charging position s i The charging energy consumption is γt(s i ); if an o j Belong to multiple Λ(s i ), when o j After being charged once, its energy requirement will become 0; in addition, s0 is not considered as a charging position, so the charging time of s0 position is zero;

[0101] Let L = {s0,s1,s2,…,s q ,s0} represents the mobile circuit of the mobile charger, where q is the number of charging positions in the circuit, and the unit mobile energy consumption of the mobile charger is μ. Then the total mobile energy consumption of L is Where d(s x ,s x:1 ) represents the charging position sx With charging position x:1 The distance between q , s0) represents the charging position s q The distance to warehouse s0;

[0102] The total energy consumption of the mobile loop L is:

[0103]

[0104] S3: Define monitoring utility model;

[0105] like Figure 2 As shown in Figure 1, each sensor has a fixed monitoring range, which divides the initial monitoring sub-area into smaller sub-areas to form the final monitoring sub-area. Let the final monitoring sub-area set be Φ = {φ1, φ2, …, φ m}, where m is the number of final monitoring sub-areas, and each final monitoring sub-area φ l The area of ​​∈Φ is a l ;like Figure 3 The flow chart of the monitoring utility optimization mobile charging scheduling algorithm is shown in Figure 1. l Represents the final monitoring sub-area φ l The monitoring utility per unit area is l The monitoring utility is a l w l ; With sensor o j The resulting monitoring sub-area whose monitoring range overlaps is called the sub-area belonging to sensor o j The final monitoring sub-area, let H(o j ) means it belongs to o j The final monitoring sub-area set of sensor o j The monitoring utility is:

[0106]

[0107] Let S represent the set of charging locations in L, then the set of all covered final monitoring sub-areas is

[0108] The total monitoring utility is:

[0109]

[0110] S4: Formal monitoring utility optimization mobile charging scheduling problem;

[0111] maxU(S) (6)

[0112]

[0113]

[0114] Among them, formula (6) aims to maximize the total monitoring utility, constraint (7) ensures that the total energy consumption of the mobile charger does not exceed the energy capacity, and constraint (8) ensures that the charging locations in S all belong to R.

[0115] S5: Use regional discretization to transform the problem, build an auxiliary graph, and perform further problem transformation. For the transformed problem, call the monitoring utility optimization mobile charging scheduling algorithm to obtain the charging scheduling plan for the mobile charger;

[0116] The two-dimensional plane R is discretized into a uniform grid with a side length of δ, and then the discretized charging sub-areas are further divided by drawing a circle with a radius of D for each sensor; the distance from each charging position to the sensor in the same charging sub-area is also approximated accordingly; assuming that x charging sub-areas covering at least one sensor have been obtained, a position is randomly selected in each charging sub-area as a candidate charging position, and K = {k1, k2, ..., k χ} represents the set of candidate charging locations; then, the original problem is simplified to P1, where the number of charging locations is limited:

[0117] (P1):maxU(S)(9)

[0118]

[0119]

[0120] According to the original network, construct an auxiliary graph G and convert the charging position s i The charging energy consumption on the edge is distributed to the two edges connected to it, that is, for the edge (s i ,s i′ ), the energy consumption on the auxiliary graph G is Next, transform P1 into P2:

[0121] (P2):maxU(S)(12)

[0122]

[0123]

[0124] When a new charging location is added to the loop, the auxiliary map needs to be updated because the energy requirements of some sensors may become 0;

[0125] A1: Input parameters: sensor set O, warehouse s0, energy capacity E of mobile charger, transmission power γ of mobile charger, unit mobile energy consumption μ of mobile charger, final monitoring sub-area set Γ, candidate charging location set K, auxiliary graph G;

[0126] A2: Initialize the remaining candidate charging location set K * = K, initialize the charging position set in the mobile loop

[0127] A3: If Execute steps A4 to A7, otherwise execute step A9;

[0128] A4: Starting from s0, visit the nearest unvisited charging position in S in turn, and finally return to s0 to obtain loop L;

[0129] A5: For every k h ∈K * , define adding k h The temporary set S(k h ), let S(k h )=S∪{k h}, starting from s0, visit S(k h ) and finally returns to s0 to obtain the loop L(k h );

[0130] A6: Find the location with the maximum utility-to-energy ratio, i.e.

[0131]

[0132] A7: If Then S = S(k h );

[0133] A8:K * =K * \{k h}, return to step A3;

[0134] A9: Starting from s0, visit the nearest unvisited charging position in S in turn, and finally return to s0 to obtain loop L;

[0135] A10: Output L.

[0136] Example 2

[0137] A mobile charging scheduling method for optimizing monitoring utility described in the present invention comprises the following steps:

[0138] In the present embodiments S1, S2, and S3, there is a sensor set O={o1, o2, o3, o4} distributed on a 9m*9m two-dimensional plane R, and the initial monitoring sub-area set And a mobile charger warehouse s0. By combining the sensor monitoring range with the initial monitoring sub-area, the final monitoring sub-area set φ={φ1,φ2,φ3,φ4,φ5,φ6,φ7,φ8,φ9,φ 10 Table 1 shows the sensor related parameters, Table 2 shows the final monitoring sub-area related parameters, and Table 3 shows the mobile charger related parameters.

[0139] Table 1 Sensor related parameters

[0140]

[0141] Table 2 Related parameters of the final monitoring sub-area

[0142]

[0143] Table 3 Mobile charger related parameters

[0144]

[0145] Set α=90,β=10, then sensor o i From charging position i The charging power obtained at is expressed as:

[0146]

[0147] In S5 of this embodiment, δ=3m is taken, and then a set of candidate charging positions K={k1, k2, k3, k4, k5, k6, k7, k8, k9} can be obtained. Table 4 shows the coordinates of the candidate charging positions.

[0148] Table 4 Candidate charging locations

[0149]

[0150] Considering that there are 10 points in the auxiliary graph G, including warehouse s0, which will form 100 edges, the energy consumption of the edges will not be listed here. However, in the subsequent steps, an example of calculating the path cost will be given.

[0151] A1: Input parameters: sensor set O = {o1, o2, o3, o4}, warehouse s0, energy capacity E = 60KJ, and other parameters;

[0152] A2: Initialize the remaining candidate charging location set K * ={k1,k2,k3,k4,k5,k6,k7,k8,k9}, initialize the charging position set in the mobile loop

[0153] A3: If Execute steps A4 to A7, otherwise execute step A9; Go to step A4;

[0154] A4: Starting from s0, visit The nearest unvisited charging position in the loop is returned to s0, and the loop L = {s0} is obtained;

[0155] A5: For every k h ∈K * , define adding k h The temporary set S(k h ), let S(k h )=S∪{k h}, starting from s0, visit S(k h ) and finally returns to s0 to obtain the loop L(k h ); Taking k1 as an example, we can get L(k1)={s0,k1,s0};

[0156] A6: Find the location with the maximum utility-to-energy ratio, i.e. At this time, S(k2) = the utility energy consumption ratio corresponding to {k2}

[0157]

[0158] is the largest;

[0159] A7: If Then S = S(k h ); At this time, 246.25≤600,

[0160] S = {k2};

[0161] A8:K * ={k2}, return to step A3;

[0162] Repeat A4 to A8 until Enter A9;

[0163] A9: Starting from s0, visit the nearest unvisited charging positions in S = {k2, k5} in turn, and finally return to s0, obtaining the loop L = {s0, k5, k2, s0}. At this time, the total monitoring utility is 246.20;

[0164] A10: Return L = {s0, k5, k2, s0}.

[0165] Example 3

[0166] In order to verify and illustrate the technical effects used in this method, Example 3 involves two solutions for comparative testing with the method of the present invention:

[0167] Nearest neighbor algorithm: In each iteration, the nearest neighbor algorithm traverses all sensors that have not been visited. The mobile charger selects the nearest sensor as the charging location and goes to the sensor to charge. The iteration terminates when all sensors have been visited or the energy capacity is exceeded.

[0168] Greedy selection algorithm based on candidate charging locations: The greedy selection algorithm based on candidate charging locations traverses all unvisited candidate charging locations after discretization, and the mobile charger selects the charging location that maximizes the total monitoring utility. The iteration terminates when all candidate charging locations have been visited or the energy capacity is exceeded.

[0169] Parameter setting: The sensors are randomly distributed in a 100m*100m square two-dimensional plane. The default parameters are set as follows: the number of initial monitoring sub-areas is m0=4, the number of sensors is n=50, the energy capacity is E=2000KJ, and the energy demand is c(o j )∈[10,14]KJ, transmission power γ=15W, unit moving energy consumption μ=50J / m, discretized grid side length δ=2m, fixed constants α=90, β=10, maximum charging range D=2m, sensor monitoring radius r=3m. Next, we will change the values ​​of key parameters to explore their impact on the algorithm, and each measurement is the average of more than 100 random topologies.

[0170] like Figure 4 As shown in the figure, the total monitoring utility of the greedy selection algorithm based on candidate charging locations and the monitoring utility optimized mobile charging scheduling algorithm increases with the increase in the number of sensors. However, the total monitoring utility of the nearest neighbor algorithm decreases with the increase in the number of sensors. This is because the optimization goal of the nearest neighbor algorithm is not the total monitoring utility. When the number of sensors increases, there may be more sensors with lower monitoring utility in the mobile loop of the nearest neighbor algorithm. Specifically, the monitoring utility optimized mobile charging scheduling algorithm increases the total monitoring utility by an average of 180.28% and 10.62% respectively compared with the other two comparison algorithms.

[0171] like Figure 5 As shown in Figure 2, the total monitoring utility of all algorithms decreases as the energy demand of sensors increases. This is because the mobile charger can only charge fewer sensors due to the increasing energy demand for sensors. Specifically, the monitoring utility optimized mobile charging scheduling algorithm increases the total monitoring utility by 185.03% and 13.45% on average compared with the other two comparison algorithms, respectively.

[0172] like Figure 6As shown in the figure, as the energy capacity of the mobile charger increases, more sensors can be charged. Therefore, the total monitoring utility of all algorithms increases with the increase of the energy capacity of the mobile charger. Specifically, the monitoring utility optimized mobile charging scheduling algorithm increases the total monitoring utility by 189.11% and 7.76% on average compared with the other two comparison algorithms, respectively.

[0173] like Figure 7 As shown in Figure 2, the total monitoring utility of all algorithms increases with the increase of the monitoring radius of the sensor. This is because the sensor can achieve higher monitoring utility by covering a larger area. Specifically, the monitoring utility optimized mobile charging scheduling algorithm increases the total monitoring utility by 185.06% and 10.80% on average compared with the other two comparison algorithms, respectively.

[0174] like Figure 8 As shown in Figure 2, as the grid edge length increases, the total monitoring utility of the greedy selection algorithm based on candidate charging locations and the monitoring utility optimization mobile charging scheduling algorithm decreases accordingly. However, since the nearest neighbor algorithm is independent of discretization, its total monitoring utility remains unchanged. Specifically, the monitoring utility optimization mobile charging scheduling algorithm increases the total monitoring utility by an average of 155.32% and 8.08% respectively compared with the other two comparison algorithms.

[0175] From the above embodiments, it can be seen that the technical solution of the present invention achieves the following technical effects: finding a mobile loop for the mobile charger to charge the sensors on the path, and maximizing the total monitoring utility in the network without exceeding the energy capacity of the mobile charger.

[0176] The above are only a few preferred embodiments of the present invention, and the description is relatively specific and detailed, but it cannot be understood as limiting the scope of the present invention. It should be pointed out that for ordinary technicians in this field, several modifications and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention.

Claims

1. A mobile charging scheduling method for optimizing monitoring utility, characterized in that: include: S1 defines the monitoring utility driven network model; S2 defines the energy consumption model; S3 defines the monitoring utility model; S4 Formal monitoring utility optimization mobile charging scheduling problem; S5 uses regional discretization to transform the problem, constructs an auxiliary graph, and performs further problem transformation. For the transformed problem, it calls the monitoring utility optimization mobile charging scheduling algorithm to obtain the charging scheduling plan for the mobile charger; The monitoring utility driven network model in step S1 includes: The set of sensors distributed on the two-dimensional plane R is O = {o1, o2, ..., o n }, where n is the number of sensors, for each sensor o j ∈O, the energy requirement is c(o j ), the initial battery power of all sensors is 0, and the monitoring radius is r; the initial monitoring sub-area set is Where m0 is the number of initial monitoring sub-areas, for each initial monitoring sub-area There is a monitoring utility per unit area w z , where z∈{1,2,…,m0}; a mobile charger starts from warehouse s0 and is allowed to stay at any location on the two-dimensional plane R to charge sensors one-to-many. The maximum charging range of the mobile charger is D, and the energy capacity of the mobile charger is E; The energy consumption model described in step S2 includes: Sensor j From any position s on the two-dimensional plane i The charging power obtained by ∈R is expressed as: Among them, α, β are constants determined by hardware and environment, d(s i ,o j ) is the charging position s i and sensor o j The distance between When the mobile charger reaches the charging position s i When the energy demand of all sensors within the maximum charging range must be met, the set of sensors within the maximum charging range is defined as Λ(s i )={o j |d(s i ,o j )≤D}, the mobile charger is at the charging position s i The charging time is: The transmitting power of the mobile charger is γ, at the charging position s i The charging energy consumption is γt(s i ); if an o j Belong to multiple Λ(s i ), when o j After being charged once, its energy requirement will become 0; in addition, s0 is not considered as a charging position, so the charging time of s0 position is zero; Let L = {s0,s1,s2,…,s q ,s0} represents the mobile circuit of the mobile charger, where q is the number of charging positions in the circuit, and the unit mobile energy consumption of the mobile charger is μ. Then the total mobile energy consumption of L is Where d(s x ,s x+1 ) represents the charging position s x With charging position x+1 The distance between q , s0) represents the charging position s q The distance to warehouse s0; The total energy consumption of the mobile loop L is: The monitoring utility model described in step S3 includes: Each sensor has a fixed monitoring range, which divides the initial monitoring sub-area into smaller sub-areas to form the final monitoring sub-area. Let the final monitoring sub-area set be Φ = {φ1, φ2, …, φ m }, where m is the number of final monitoring sub-areas, and each final monitoring sub-area φ l The area of ​​∈Φ is a l ; Use w l Represents the final monitoring sub-area φ l The monitoring utility per unit area is l The monitoring utility is a l w l ; With sensor o j The resulting monitoring sub-area whose monitoring ranges overlap is called the area belonging to sensor o j The final monitoring sub-area, let H(o j ) means it belongs to o j The final monitoring sub-area set of sensor o j The monitoring utility is: Let S represent the set of charging locations in L, then the set of all covered final monitoring sub-areas is The total monitoring utility is: The formal monitoring utility optimization mobile charging scheduling problem described in step S4 includes: maxU(S)(6) Among them, formula (6) aims to maximize the total monitoring utility, constraint (7) ensures that the total energy consumption of the mobile charger does not exceed the energy capacity, and constraint (8) ensures that the charging locations in S all belong to R; The problem transformation is performed by using regional discretization in step S5, and an auxiliary graph is constructed to further transform the problem. For the transformed problem, the monitoring utility optimization mobile charging scheduling algorithm is called to obtain the charging scheduling plan of the mobile charger, including: The two-dimensional plane R is discretized into a uniform grid with a side length of δ, and then the discretized charging sub-areas are further divided by drawing a circle with a radius of D for each sensor; the distance from each charging position to the sensor in the same charging sub-area is also approximated accordingly; assuming that x charging sub-areas covering at least one sensor have been obtained, a position is randomly selected in each charging sub-area as a candidate charging position, and K = {k1, k2, ..., k χ } represents the set of candidate charging locations; then, the original problem is simplified to P1, where the number of charging locations is limited: (P1):maxU(S)(9) According to the original network, construct an auxiliary graph G and convert the charging position s i The charging energy consumption on the edge is distributed to the two edges connected to it, that is, for the edge (s i ,s i′ ), the energy consumption on the auxiliary graph G is Next, transform P1 into P2: (P2):maxU(S)(12) When a new charging location is added to the loop, the auxiliary map needs to be updated because the energy requirements of some sensors may become 0; A1: Input parameters: sensor set O, warehouse s0, energy capacity E of mobile charger, transmission power γ of mobile charger, unit mobile energy consumption μ of mobile charger, final monitoring sub-area set Γ, candidate charging location set K, auxiliary graph G; A2: Initialize the remaining candidate charging location set K * = K, initialize the charging position set in the mobile loop A3: If Execute steps A4 to A7, otherwise execute step A9; A4: Starting from s0, visit the nearest unvisited charging position in S in turn, and finally return to s0 to obtain loop L; A5: For every k h ∈K * , define adding k h The temporary set S(k h ), let S(k h )=S∪{k h }, starting from s0, visit S(k h ) and finally returns to s0 to obtain the loop L(k h ); A6: Find the location with the maximum utility-to-energy ratio, i.e. A7: If Then S = S(k h ); A8:K * =K * \{k h }, return to step A3; A9: Starting from s0, visit the nearest unvisited charging position in S in turn, and finally return to s0 to obtain loop L; A10: Output L.

2. Application of the method as claimed in claim 1 in UAV charging scheduling.

3. A charging control device, characterized in that: include: 1) one or more processors; 2) a storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the mobile charging scheduling method as claimed in claim 1.

4. A drone parallel charging network, characterized in that: Contains the charging control device as claimed in claim 3.

Images