Wireless sensor network charger deployment method based on improved pelican optimization algorithm

Abstract

The invention discloses a wireless sensor network charger deployment method based on an improved pelican optimization algorithm. The method comprises the steps of 1, performing mathematical modeling on a wireless rechargeable sensor network model; 2, modeling is carried out on a static charger deployment problem, an optimization target is determined to maximize the coverage rate of a static charger to a sensor node, and a fitness function is designed; and step 3, optimizing the node deployment position of the static charger on the basis of mathematical modeling in the step 1 by adopting a multi-strategy improved pelican optimization algorithm, and returning an optimal deployment result to a user. By adopting the deployment method provided by the invention, the number of static charging nodes can be reduced, and meanwhile, a better charging coverage rate is achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of wireless sensor networks, and in particular relates to a method for deploying chargers in a wireless sensor network based on an improved Pelican optimization algorithm. Background Art

[0002] In recent years, with the rapid development of Internet of Things technology, wireless sensor networks have been widely used in various fields as an important information transmission technology. It can form a multi-hop wireless self-organizing network through wireless communication, collect and process the collected local data through sensor nodes, and finally send the collected data to the end user. It has a wide range of applications in intelligent transportation, smart agriculture, military defense and scientific exploration. However, due to the small size of sensor nodes, the battery capacity is insufficient, which is the biggest problem that checks and balances the reliability of wireless sensor networks. When a sensor node loses energy, it is often impossible to replace its battery for cost considerations, so it can only be regarded as node death. When a large number of nodes in the network die and the network topology cannot be maintained, the life of the network will come to an end.

[0003] However, with the development of wireless sensor charging technology, wireless rechargeable sensor networks have become possible. Compared with wireless sensor networks, wireless rechargeable sensor networks can theoretically recharge nodes before they die, allowing the network to operate indefinitely. Wireless rechargeable sensor networks are a new technology developed on the basis of wireless sensor networks, which has brought important technological progress.

[0004] In order to effectively replenish the energy of sensor nodes, many researchers have proposed many solutions so far. Among them, the use of rechargeable wireless cars or the deployment of static chargers have attracted the most attention of researchers. These two solutions have different scopes of application. Using wireless cars to dynamically charge sensor nodes has higher efficiency in environments with small detection ranges and sparse deployment of sensor nodes. However, when the detection range is expanded and the nodes are densely deployed, even if the number of wireless cars is increased on a large scale for dynamic charging, the effect of extending the life of the network is still very limited. Especially when the sensor network is actually deployed, the environment is often very harsh, and it is difficult to ensure that the wireless car can operate stably in harsh environments. At this time, in order to extend the life of the network, it is preferred to deploy static chargers to charge the sensor nodes.

[0005] Static charger deployment algorithm is an important research direction in the field of wireless rechargeable sensor networks. Since static chargers are expensive, it is not realistic to deploy a large number of chargers in an area. Therefore, a static charger deployment method is urgently needed to ensure the network life by deploying as few chargers as possible. Summary of the invention

[0006] To solve the above technical problems, the object of the present invention is to provide a wireless sensor network charger deployment method based on an improved Pelican optimization algorithm.

[0007] The present invention provides a wireless sensor network charger deployment method based on an improved Pelican optimization algorithm, comprising:

[0008] Step 1: Mathematically model the wireless rechargeable sensor network model;

[0009] Step 2: Model the static charger deployment problem, establish the optimization goal to maximize the coverage of static chargers to sensor nodes, and design the fitness function;

[0010] Step 3: Using the multi-strategy improved Pelican optimization algorithm, based on the mathematical modeling in step 1, optimize the deployment location of static charger nodes and return the optimal deployment result to the user.

[0011] Furthermore, the step 1 is specifically as follows:

[0012] Step 1.1: Determine the parameters of the simulated wireless rechargeable sensor network: the monitoring area is a square area with a side length of L, the number of sensor nodes is n, the number of static charger nodes is N, and the charging radius of the static charger node is R;

[0013] Step 1.2: According to the given number of sensor nodes and the side length of the monitoring area, randomly generate a sensor node set U = {u1,u2,u3,···,u n}; The coordinates of the sensor node are u j =(x j ,y j ), all sensor nodes are generated in the monitoring area;

[0014] Step 1.3: Based on the generated sensor node coordinates, draw the distribution of sensor nodes in the monitoring area and return the drawing results to the user.

[0015] Furthermore, the step 2 is specifically as follows:

[0016] Step 2.1: Assume that the total number of deployed isomorphic static chargers is N in the monitoring area of ​​the wireless rechargeable sensor network with a side length of L. Assume that the static charger set is V = {v1,v2,v3,···,v N}, the position coordinates of the static charger are represented by v i =(x i ,y i );

[0017] Step 2.2: Discretize the monitoring area into n grid points to be covered. The grid points represent ordinary sensor nodes. The distance between the sensor node and the static charger node is defined as follows:

[0018]

[0019] Step 2.3: If the distance between a sensor node and a static charger node is less than or equal to the static charger radius R, the sensor node is considered to be covered by the static charger node. Using the Boolean model, the sensor node u j The static charger node v i The perceived probability is given by:

[0020]

[0021] Step 2.4: If a sensor node is sensed by multiple static charger nodes, the probability of the sensor node being jointly sensed is as follows:

[0022]

[0023] Step 2.5: The network coverage of the static charger nodes is expressed by dividing the sum of the perception probabilities of the sensor nodes that can be covered by all static charger nodes by the total number of all sensor nodes in the area; the higher the value of the network coverage, the more the sensor nodes are covered by static charger nodes. The calculation of the network coverage is shown in the following formula:

[0024]

[0025] Step 2.6: The static charger coverage optimization problem can be described as the location optimization of N static charger nodes in the monitoring area to achieve the maximum network coverage of all sensor nodes, that is, to maximize the sum of the coverage probabilities of all sensor nodes; therefore, this problem is transformed into a single-objective optimization problem, the goal is to find an optimal static charger deployment scheme to maximize the coverage and achieve the maximum value of the sum of the coverage probabilities of each sensor node in the network:

[0026]

[0027] When solving problems, intelligent optimization algorithms usually seek the minimum value in the optimization function. When seeking the maximum value, a conversion is required. Therefore, the final fitness function F(X) of the improved Pelican optimization algorithm is shown as follows:

[0028] F(X)=-R cov .

[0029] Furthermore, the step 3 is specifically as follows:

[0030] Step 3.1: Set the population size to M, the maximum number of iterations to T, and the algorithm upper bound u b =L, algorithm lower bound l b =0;

[0031] Step 3.2: Initialize the population. Each pelican individual is composed of the coordinate arrangement of all static charger nodes. Logistic mapping is added to the population initialization formula of the original pelican optimization algorithm to ensure that the initialized population has a better set of optimal points and prevent the problem of premature convergence caused by uneven distribution of the population in the solution space due to random initialization. The formula for population initialization is as follows:

[0032] X k = l b +Z k ·(u b -l b )

[0033] Z k =μ·Z k-1 ·(1-Z k-1 )

[0034] Among them, X k is the kth pelican individual in the population; Z k is a Logistic mapping, Z1 is randomly generated in the monitoring area; μ is the chaos coefficient;

[0035] Step 3.3: In the exploration phase, the weight factor ω is introduced on the basis of simulating the original pelican optimization algorithm to approach the prey to achieve the optimization goal. Since the update of the individual pelican position is closely related to the current pelican position, the nonlinear inertia weight factor ω is used to adjust the correlation between the pelican position update and the current pelican position information. The calculation method of the nonlinear inertia weight factor ω is as follows:

[0036]

[0037] Among them, T represents the highest number of iterations, t represents the current number of iterations, and e is a natural constant;

[0038] After introducing the weight factor ω, the update formula of the population in the exploration phase is:

[0039]

[0040] in, represents the value of the hth dimension of the kth pelican individual after the exploration phase update; x k,hrepresents the value of the hth dimension of the kth pelican individual before the exploration phase update; rand is a random number in the range [0,1]; I is a random integer of 1 or 2; p h is the coordinate of the prey in the hth dimension, and the prey coordinates are randomly generated in the search space; F p is the fitness value of the prey; F k is the fitness value of the kth pelican individual before the update;

[0041] Step 3.4: In the development stage, the pelican simulates the behavior of flying on the water to catch fish, and introduces the Cauchy mutation strategy. Each iteration compares the size relationship between the current individual fitness value of the pelican and the average fitness value of the population; if the current individual fitness value of the pelican is lower than the average fitness value of the population, it means that the current pelican individuals are in a clustered state. At this time, in order to increase the diversity of the pelicans, the Cauchy mutation strategy is adopted; if the current individual fitness value of the pelican is higher than the average fitness value of the population, the original pelican position update method is adopted. The improved formula is as follows:

[0042]

[0043] Among them, x best,h is the value of the hth dimension of the best individual in the current population, Cauchy(0,1) is a value variation using the Cauchy probability density function; F avg is the average fitness value of the population; is the fitness value of the kth pelican individual in the development stage; R is a constant;

[0044] In order to further improve the algorithm's optimization ability in the development stage, the update of the second half of the pelican individuals in the population is replaced by the mechanism of bringing food to a safe place in the second stage of the osprey optimization algorithm. The update formula for the second half of the pelican individuals in the development stage is as follows:

[0045]

[0046] Step 3.5: After completing one iteration, determine whether the maximum number of iterations has been reached. If not, continue to repeat steps 3.2 and 3.3. If the maximum number of iterations has been reached, save the optimal individual and optimal coverage, and display the final static charger deployment result to the user through software.

[0047] Furthermore, in the exploration phase of step 3.3, if the fitness value at the position is improved after the update, the updated coordinates are accepted, otherwise the updated coordinates are not accepted and the original values ​​are maintained. This process is expressed by the following formula:

[0048]

[0049] Among them, Xk is the kth pelican individual in the population; is the kth pelican individual after the exploration phase update; is the fitness value of the kth pelican individual after receiving the update in the exploration phase.

[0050] Furthermore, in the development phase of step 3.4, if the fitness value at the position is improved after the update, the updated coordinates are accepted, otherwise the updated coordinates are not accepted and the original values ​​are maintained. This process is expressed by the following formula:

[0051]

[0052] Among them, X k is the kth pelican individual in the population; is the kth pelican individual after the development phase update; is the fitness value of the kth pelican individual after being updated in the exploration and development phase.

[0053] The present invention discloses a method for deploying chargers in a wireless sensor network based on an improved Pelican optimization algorithm. Aiming at the problems of the original Pelican optimization algorithm in terms of solution accuracy and easy falling into local optimality, the proposed improved Pelican optimization algorithm introduces a fusion mechanism of a mutation strategy and an Osprey optimization algorithm, which effectively expands the search space and reduces the risk of falling into local optimality. In addition, the improved algorithm also has a significant improvement in computational efficiency, and a charger deployment scheme with high accuracy can be obtained with only 200 to 300 iterations. By optimizing the layout of the chargers, the method ensures that more than 90% of the sensor nodes can obtain power replenishment while reducing the number of static chargers, thereby significantly improving the life and stability of the network. This method not only reduces the deployment cost and improves the energy utilization efficiency, but also enhances the environmental adaptability and robustness of the network. It is suitable for a variety of environmental conditions and has good scalability. It can adapt to sensor networks of different sizes, and provides an efficient and intelligent power management solution for the fields of the Internet of Things, environmental monitoring, and smart cities. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 is a flow chart of a wireless sensor network charger deployment method based on an improved Pelican optimization algorithm of the present invention;

[0055] Figure 2 is a distribution diagram of sensor nodes in an example of the present invention;

[0056] Figure 3 is a flow chart of the improved Pelican optimization algorithm of the present invention;

[0057] Figure 4 This is a diagram of the final static charger deployment result in an example of the present invention. DETAILED DESCRIPTION

[0058] like Figure 1 As shown, the wireless sensor network charger deployment method based on the improved Pelican optimization algorithm of the present invention includes:

[0059] Step 1: mathematically modeling the wireless rechargeable sensor network model, wherein step 1 is specifically as follows:

[0060] Step 1.1: Determine the parameters of the simulated wireless rechargeable sensor network: the monitoring area is a square area with a side length of L, the number of sensor nodes is n, the number of static charger nodes is N, and the charging radius of the static charger node is R.

[0061] In specific implementation, L=100m, n=300, N=30, R=10m.

[0062] Step 1.2: According to the given number of sensor nodes and the side length of the monitoring area, randomly generate a sensor node set U = {u1,u2,u3,···,u n}; The coordinates of the sensor node are u j =(x j ,y j ). Ensure that the coordinates of the sensor nodes do not exceed the measurement boundary, that is, all sensor nodes are generated within the monitoring area. If the coordinates of a sensor node exceed the boundary, the program will automatically correct the error and regenerate it. The random generation of sensor nodes simulates the situation of deploying wireless sensor nodes in real production and life.

[0063] Step 1.3: Based on the generated sensor node coordinates, draw the distribution of sensor nodes in the monitoring area and return the drawing results to the user.

[0064] In the specific implementation, after generating the coordinates of all sensor nodes, a distribution diagram will be drawn and displayed to the user, such as Figure 2 At the same time, the sensor node coordinate set U will be used as an important parameter of the fitness function in the calculation process of the subsequent algorithm.

[0065] Step 2: Model the static charger deployment problem, establish the optimization goal to maximize the coverage of the static charger to the sensor nodes, and design the fitness function. The specific steps of step 2 are:

[0066] Step 2.1: Assume that the total number of deployed isomorphic static chargers is N in the monitoring area of ​​the wireless rechargeable sensor network with a side length of L. Assume that the static charger set is V = {v1,v2,v3,···,v N}, the position coordinates of the static charger are represented by v i =(x i,y i ).

[0067] In the specific implementation, in the subsequent algorithm calculation, in order to facilitate the calculation, the static charger set V is regarded as a row vector, according to {x1, y1, x2, y2, ···, x N ,y N} arrangement as the input vector X of the fitness function F(X).

[0068] Step 2.2: Discretize the monitoring area into n grid points to be covered. The grid points represent ordinary sensor nodes. The Euclidean distance between the sensor node and the static charger node is defined as follows:

[0069]

[0070] Step 2.3: If the distance between a sensor node and a static charger node is less than or equal to the static charger radius R, the sensor node is considered to be covered by the static charger node. Using the Boolean model, the sensor node u j The static charger node v i The perceived probability is given by:

[0071]

[0072] Step 2.4: If a sensor node is sensed by multiple static charger nodes, the probability of the sensor node being jointly sensed is as follows:

[0073]

[0074] Step 2.5: The network coverage of the static charger nodes is expressed by dividing the sum of the perception probabilities of the sensor nodes that can be covered by all static charger nodes by the total number of all sensor nodes in the area. The higher the value of the network coverage, the more static charger nodes cover the sensor nodes, which is more guaranteed for the extension of the network life. The calculation of the network coverage is shown in the following formula:

[0075]

[0076] Step 2.6: The static charger coverage optimization problem can be described as the position optimization of N static charger nodes in the monitoring area to achieve the maximum network coverage of all sensor nodes, that is, to maximize the sum of the coverage probabilities of all sensor nodes; therefore, this problem is transformed into a single-objective optimization problem, the goal is to find an optimal static charger deployment plan to maximize the coverage, that is, to achieve the maximum value of the sum of the coverage probabilities of each sensor node in the network through iterative solution of the optimization algorithm. The higher the coverage rate, the more sensor nodes can obtain power replenishment to extend the network life. Therefore, the optimization goal of the static charger deployment problem is to maximize the network coverage, that is:

[0077]

[0078] When solving problems, intelligent optimization algorithms usually seek the minimum value in the optimization function. When seeking the maximum value, a conversion is required. Therefore, the final fitness function F(X) of the improved Pelican optimization algorithm is shown as follows:

[0079] F(X)=-R cov .

[0080] Step 3: Using the multi-strategy improved Pelican optimization algorithm, based on the mathematical modeling in step 1, optimize the deployment location of static charger nodes and return the optimal deployment result to the user, such as Figure 3 As shown, the step 3 is specifically as follows:

[0081] Step 3.1: Set the population size to M = 30, the maximum number of iterations to T = 200, and the algorithm upper bound u b =L, algorithm lower bound l b =0.

[0082] Step 3.2: Initialize the population. Each pelican individual is composed of the coordinate arrangement of all static charger nodes. According to the initial definition of the number of static chargers N = 30, the dimension of each pelican individual is 60. Logistic mapping is added to the population initialization formula of the original pelican optimization algorithm to ensure that the initialized population has a better set of optimal points and prevent the problem of premature convergence caused by uneven distribution of the population in the solution space due to random initialization. The formula for population initialization is as follows:

[0083] X k = l b +Z k ·(u b -l b )

[0084] Z k =μ·Z k-1 ·(1-Z k-1)

[0085] Among them, X k is the kth pelican individual in the population; Z k is a Logistic map, Z1 is randomly generated in the monitoring area; μ is the chaos coefficient, and its value is 3 in this embodiment.

[0086] Finally, the initialized population set is returned as a matrix X, where each row vector of X represents a pelican individual X k , and then substitute X into the fitness function F(X) to get the initial target value F.

[0087]

[0088] Where X is the population matrix of pelicans; X k is the kth pelican individual, M is the population size, and m is the problem dimension. In this example, M = 30. m is twice the number of static chargers, that is, m = N × 2 = 60.

[0089]

[0090] Where F is the fitness value vector of the pelican population; F k is the fitness value of the kth pelican.

[0091] Step 3.3: In the exploration phase, the weight factor ω is introduced on the basis of simulating the original pelican optimization algorithm to approach the prey to achieve the optimization goal. Since the update of the individual pelican position is closely related to the current pelican position, the nonlinear inertia weight factor ω is used to adjust the correlation between the pelican position update and the current pelican position information. The calculation method of the nonlinear inertia weight factor ω is as follows:

[0092]

[0093] Among them, T represents the maximum number of iterations, t represents the current number of iterations, and e is a natural constant.

[0094] In the early stage of algorithm iteration, the value of ω is small, and the update of the optimal individual position is less affected by the current pelican position, which is conducive to the algorithm to search in a larger range and improve the global development ability of the algorithm; as the optimization process progresses, the value of ω gradually increases, and the update of the optimal individual position is more affected by the current pelican position, which narrows the optimization range of the algorithm and helps the algorithm search for the optimal solution, which not only improves the global exploration ability of the algorithm, but also improves the convergence speed of the algorithm.

[0095] After introducing the weight factor ω, the update formula of the population in the exploration phase is:

[0096]

[0097] in, represents the value of the hth dimension of the kth pelican individual after the exploration phase update; x k,h represents the value of the hth dimension of the kth pelican individual before the exploration phase update; rand is a random number in the range [0,1]; I is a random integer of 1 or 2, this parameter is randomly selected for each individual, when this parameter is 2, the individual can get more displacement to explore newer areas; p h is the coordinate of the prey in the hth dimension, and the prey coordinates are randomly generated in the search space; F p is the fitness value of the prey; F k is the fitness value of the kth pelican individual before the update. The above formula shows that if the objective function of the prey position has a better effect, that is, F p <F k , the population will move a greater distance toward the prey location, otherwise it will move a smaller distance toward the prey location.

[0098] In the exploration phase of step 3.3, if the fitness value at the position is improved after the update, the updated coordinates are accepted, otherwise the updated coordinates are not accepted and the original values ​​are maintained. This process is expressed by the following formula:

[0099]

[0100] Among them, X k is the kth pelican individual in the population; is the kth pelican individual after the exploration phase update; is the fitness value of the kth pelican individual after receiving the update in the exploration phase.

[0101] Step 3.4: In the development stage, the pelican simulates the behavior of flying on the water to catch fish, and introduces the Cauchy mutation strategy. Each iteration compares the size relationship between the current individual fitness value of the pelican and the average fitness value of the population; if the current individual fitness value of the pelican is lower than the average fitness value of the population, it means that the current pelican individuals are in a clustered state. At this time, in order to increase the diversity of the pelicans, the Cauchy mutation strategy is adopted; if the current individual fitness value of the pelican is higher than the average fitness value of the population, the original pelican position update method is adopted. The improved formula is as follows:

[0102]

[0103] Among them, x best,h is the value of the hth dimension of the best individual in the current population, Cauchy(0,1) is a value variation using the Cauchy probability density function; F avg is the average fitness value of the population; is the fitness value of the kth pelican individual in the development stage; R is a constant with a value of 0.2.

[0104] In order to further improve the algorithm's optimization ability in the development stage, the update of the second half of the pelican individuals in the population is replaced by the mechanism of bringing food to a safe place in the second stage of the osprey optimization algorithm. The update formula for the second half of the pelican individuals in the development stage is as follows:

[0105]

[0106] In the development phase of step 3.4, if the fitness value at the position is improved after the update, the updated coordinates are accepted, otherwise the updated coordinates are not accepted and the original values ​​are maintained. This process is expressed by the following formula:

[0107]

[0108] Among them, X k is the kth pelican individual in the population; is the kth pelican individual after the development phase update; is the fitness value of the kth pelican individual after being updated in the exploration and development phase.

[0109] Step 3.5: After completing one iteration, determine whether the maximum number of iterations has been reached. If not, continue to repeat steps 3.2 and 3.3; if the maximum number of iterations has been reached, save the optimal individual and the optimal coverage. After software simulation, the optimal individual will be converted into static charger deployment coordinates, and the result map will be drawn together with the sensor node deployment coordinates and returned to the user, such as Figure 4 shown.

[0110] The above description is only a preferred embodiment of the present invention and is not intended to limit the concept of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A wireless sensor network charger deployment method based on an improved Pelican optimization algorithm, characterized in that: include: Step 1: Mathematically model the wireless rechargeable sensor network model; Step 2: Model the static charger deployment problem, establish the optimization goal to maximize the coverage of static chargers to sensor nodes, and design the fitness function; Step 3: Using the multi-strategy improved Pelican optimization algorithm, based on the mathematical modeling in step 1, optimize the deployment location of static charger nodes and return the optimal deployment result to the user.

2. The wireless sensor network charger deployment method based on the improved Pelican optimization algorithm as claimed in claim 1, characterized in that: The step 1 is specifically as follows: Step 1.1: Determine the parameters of the simulated wireless rechargeable sensor network: the monitoring area is a square area with a side length of L, the number of sensor nodes is n, the number of static charger nodes is N, and the charging radius of the static charger node is R; Step 1.2: According to the given number of sensor nodes and the side length of the monitoring area, randomly generate a sensor node set U = {u1,u2,u3,···,u n }; The coordinates of the sensor node are u j =(x j ,y j ), all sensor nodes are generated in the monitoring area; Step 1.3: Based on the generated sensor node coordinates, draw the distribution of sensor nodes in the monitoring area and return the drawing results to the user.

3. The wireless sensor network charger deployment method based on the improved Pelican optimization algorithm as claimed in claim 2 is characterized in that: The step 2 is specifically as follows: Step 2.1: Assume that the total number of deployed isomorphic static chargers is N in the monitoring area of ​​the wireless rechargeable sensor network with a side length of L. Assume that the static charger set is V = {v1,v2,v3,···,v N }, the position coordinates of the static charger are represented by v i =(x i ,y i ); Step 2.2: Discretize the monitoring area into n grid points to be covered. The grid points represent ordinary sensor nodes. The distance between the sensor node and the static charger node is defined as follows: Step 2.3: If the distance between a sensor node and a static charger node is less than or equal to the static charger radius R, the sensor node is considered to be covered by the static charger node. Using the Boolean model, the sensor node u j The static charger node v i The perceived probability is given by: Step 2.4: If a sensor node is sensed by multiple static charger nodes, the probability of the sensor node being jointly sensed is as follows: Step 2.5: The network coverage of the static charger nodes is expressed by dividing the sum of the perception probabilities of the sensor nodes that can be covered by all static charger nodes by the total number of all sensor nodes in the area; the higher the value of the network coverage, the more the sensor nodes are covered by static charger nodes. The calculation of the network coverage is shown in the following formula: Step 2.6: The static charger coverage optimization problem can be described as the location optimization of N static charger nodes in the monitoring area to achieve the maximum network coverage of all sensor nodes, that is, to maximize the sum of the coverage probabilities of all sensor nodes; Therefore, this problem is transformed into a single-objective optimization problem, the goal is to find an optimal static charger deployment scheme to maximize the coverage rate and achieve the maximum value of the sum of the coverage probabilities of each sensor node in the network: When solving problems, intelligent optimization algorithms usually seek the minimum value in the optimization function. When seeking the maximum value, a conversion is required. Therefore, the final fitness function F(X) of the improved Pelican optimization algorithm is shown as follows: F(X)=-R cov 。 4. The wireless sensor network charger deployment method based on the improved Pelican optimization algorithm as claimed in claim 2 is characterized in that: The step 3 is specifically as follows: Step 3.1: Set the population size to M, the maximum number of iterations to T, and the algorithm upper bound u b =L, algorithm lower bound, b =0; Step 3.2: Initialize the population. Each pelican individual is composed of the coordinate arrangement of all static charger nodes. Logistic mapping is added to the population initialization formula of the original pelican optimization algorithm to ensure that the initialized population has a better set of optimal points and prevent the problem of premature convergence caused by uneven distribution of the population in the solution space due to random initialization. The formula for population initialization is as follows: X k =l b +Z k ·(u b -l b ) WITH k =μ·Z k-1 (1-Z k-1 ) Among them, X k is the kth pelican individual in the population; Z k is a Logistic mapping, Z1 is randomly generated in the monitoring area; μ is the chaos coefficient; Step 3.3: In the exploration phase, the weight factor ω is introduced on the basis of simulating the original pelican optimization algorithm to approach the prey to achieve the optimization goal. Since the update of the individual pelican position is closely related to the current pelican position, the nonlinear inertia weight factor ω is used to adjust the correlation between the pelican position update and the current pelican position information. The calculation method of the nonlinear inertia weight factor ω is as follows: Among them, T represents the highest number of iterations, t represents the current number of iterations, and e is a natural constant; After introducing the weight factor ω, the update formula of the population in the exploration phase is: in, represents the value of the hth dimension of the kth pelican individual after the exploration phase update; x k，h represents the value of the hth dimension of the kth pelican individual before the exploration phase update; rand is a random number in the range [0, 1]; I is a random integer of 1 or 2; p h is the coordinate of the prey in the hth dimension, and the prey coordinates are randomly generated in the search space; F p is the fitness value of the prey; F k is the fitness value of the kth pelican individual before the update; Step 3.4: In the development stage, the pelican simulates the behavior of flying on the water surface to fish, and introduces the Cauchy mutation strategy. Each iteration compares the size relationship between the current individual fitness value of the pelican and the average fitness value of the population; if the current individual fitness value of the pelican is lower than the average fitness value of the population, it means that the current pelican individuals are in a clustered state. At this time, in order to increase the diversity of the pelicans, the Cauchy mutation strategy is adopted; if the current individual fitness value of the pelican is higher than the average fitness value of the population, the original pelican position update method is adopted. The improved formula is as follows: Among them, x best,h is the value of the hth dimension of the best individual in the current population, Cauchy(0,1) is a value variation using the Cauchy probability density function; F avg is the average fitness value of the population; is the fitness value of the kth pelican individual in the development stage; R is a constant; In order to further improve the algorithm's optimization ability in the development stage, the update of the second half of the pelican individuals in the population is replaced by the mechanism of bringing food to a safe place in the second stage of the osprey optimization algorithm. The update formula for the second half of the pelican individuals in the development stage is as follows: Step 3.5: After completing one iteration, determine whether the maximum number of iterations has been reached. If not, continue to repeat steps 3.2 and 3.

3. If the maximum number of iterations has been reached, save the optimal individual and optimal coverage, and display the final static charger deployment result to the user through software.

5. The wireless sensor network charger deployment method based on the improved Pelican optimization algorithm as claimed in claim 4 is characterized in that: In the exploration phase of step 3.3, if the fitness value at the position is improved after the update, the updated coordinates are accepted, otherwise the updated coordinates are not accepted and the original values ​​are maintained. This process is expressed by the following formula: Among them, X k is the kth pelican individual in the population; is the kth pelican individual after the exploration phase update; is the fitness value of the kth pelican individual after receiving the update in the exploration phase.

6. The wireless sensor network charger deployment method based on the improved Pelican optimization algorithm as claimed in claim 4 is characterized in that: In the development phase of step 3.4, if the fitness value at the position is improved after the update, the updated coordinates are accepted, otherwise the updated coordinates are not accepted and the original values ​​are maintained. This process is expressed by the following formula: Among them, X k is the kth pelican individual in the population; is the kth pelican individual after the development phase update; is the fitness value of the kth pelican individual after being updated in the exploration and development phase.

Images