Method for deploying underwater detection nodes in designated area based on genetic algorithm

Abstract

The invention discloses an underwater detection node adaptive rapid deployment method based on a genetic algorithm, and belongs to the technical field of sensor network node deployment, and the method comprises the steps: collecting and preparing related data of a detection region and underwater nodes; preprocessing the water depth data of the detection area, and performing coordinate conversion and convolution operation; designing a fitness function and constraint conditions, and constructing a resource allocation model; and determining a node deployment method based on fitness function maximization by using an improved genetic algorithm. The underwater detection node self-adaptive rapid deployment method based on the genetic algorithm is high in decision making speed, short in optimization iteration time, high in search capability and capable of adapting to the conditions that the shape of a detection area is complex, many obstacles exist, and the shape is strange.

Description

Technical Field

[0001] The present invention relates to the technical field of sensor node deployment, and in particular to a method for deploying underwater detection nodes in a designated area based on a genetic algorithm. Background Art

[0002] Underwater detection node networks have broad application prospects in the fields of marine exploration, disaster warning and submarine detection. In the process of deploying underwater detection nodes, it is necessary to convert the underwater sensor network into a two-dimensional network topology to improve the effective coverage of underwater nodes at the bottom of the ocean, rivers, lakes or in shallow water and near the shore, so as to ensure good detection performance of network nodes. However, due to the complex shape of the detection area, many obstacles and strange shapes, these constraints make it difficult for traditional underwater deployment methods to obtain the optimal solution. In addition, due to the complex underwater terrain environment, the sudden change of water depth poses a severe challenge to the deployment of underwater nodes. Therefore, studying the deployment method of underwater detection nodes with the ability to adapt to complex terrain plays an important role in improving the detection and coverage effect of the detection network.

[0003] On the other hand, due to the large number of detection nodes that need to be deployed during regional detection, the traditional optimization deployment method based on geometric operations has the problems of complex calculations and long solution time. However, in the face of real-time and urgent scenarios such as submarine detection, it is often necessary to quickly decide on the deployment method of underwater detection nodes, which makes the traditional underwater detection node deployment method difficult to apply. At present, methods for rapid adaptive deployment of underwater detection nodes are still relatively rare. Therefore, studying an optimization deployment method with fast decision-making speed, short optimization iteration time, and strong search capability is of great research significance for solving the coverage deployment problem of underwater detection node networks. Summary of the invention

[0004] The purpose of the present invention is to provide a method for deploying underwater detection nodes in a designated area based on a genetic algorithm, which has the effect of improving the coverage effect of the node network in sea areas with complex terrain.

[0005] To achieve the above object, the present invention provides a method for deploying underwater detection nodes in a designated area based on a genetic algorithm, comprising the following steps:

[0006] S1. Collect and prepare relevant data of the detection area and underwater nodes, including the water depth threshold D for node layout p , the number of underwater detection nodes to be allocated n, the node detection radius r, and the obstacle boundary coordinate set C in the sea area o , the specified detection area boundary longitude and latitude coordinate set C;

[0007] S2, preprocessing the water depth data of the detection area, performing coordinate transformation and convolution operation;

[0008] S3, design fitness function and constraints, and build resource allocation model;

[0009] S4. Use the improved genetic algorithm to determine the node deployment method based on maximizing the fitness function.

[0010] Preferably, in step S1, the boundary coordinate set of obstacles in the sea area is Indicates that there are k obstacles in total.

[0011] Among them, obstacles Indicates that each obstacle is surrounded by different vertices;

[0012] The specified detection area boundary longitude and latitude coordinate set

[0013] in, represents the i-th boundary point of the detection area, represents the longitude coordinate, represents the latitude coordinate, and l is the total number of boundary points in the detection area.

[0014] Preferably, the specific steps of S2 are as follows:

[0015] S21. Find in set C λ min , max , forming a rectangular area Q, where Represents the minimum latitude of all coordinate points in set C, represents the maximum latitude of all coordinate points in set C, λ min represents the minimum value of the longitude of all coordinate points in set C, λ max Represents the maximum value of the longitude of all coordinate points in set C, for all coordinate points q in Q (qlon，qlat) The following formulas are established:

[0016]

[0017] S22. Determine the length w and width h of the rectangular area Q using the haversine formula;

[0018] S23, establishing a rectangular coordinate system and mapping boundary points;

[0019] S24, rasterize the rectangular area Q, divide the area Q into a grid of h rows and w columns at a step length of one kilometer, and map the rasterized area into a matrix A' with h rows and w columns, where the element A' in A' is (i,j) The value is the average depth of the area covered by the grid in the i-th row and j-th column;

[0020] S25, process the matrix A', invert all elements in the matrix A' to obtain the matrix A, which is A (i,j) =-A' (i,j) , 1≤i≤h,1≤j≤w, according to the node detection radius r, design the convolution kernel matrix K (2r+1)×(2r+1) , perform equal-size convolution operations on the matrix A to be convolved and the convolution kernel K to obtain the target matrix B of the same dimension as the matrix A.

[0021] Preferably, the formula for the length w of the rectangular area Q in step S22 is as follows:

[0022]

[0023] The formula for the width h of the rectangular area Q is as follows:

[0024]

[0025] Where R is the average radius of the Earth and [·] indicates rounding.

[0026] Preferably, the specific steps of S23 are as follows:

[0027] S231, establish a rectangular coordinate system, take the lower left corner of the rectangular area Q as the coordinate origin, and establish a rectangular coordinate system O with each kilometer as the step length xy , where the horizontal axis represents longitude and the vertical axis represents latitude. The horizontal axis value range is [0,w] and the vertical axis value range is [0,h];

[0028] S232, mapping boundary points, mapping the latitude and longitude coordinates of all boundary points to xy In the coordinate system, the centroid of the detection area is calculated, and the boundary points of the detection area are sorted according to the angle between each boundary point and the centroid. The scanning line method is used to extract all the coordinates in the detection area, and the coordinates of each obstacle are calculated. The overlapping parts are removed to form a set of arrangable coordinates E, which is defined as Where m is the total number of coordinates that can be arranged, is the latitude of the i-th arrangable coordinate, is the longitude of the i-th arrangeable coordinate.

[0029] Preferably, the convolution kernel matrix K (i,j) The value of is:

[0030]

[0031] Preferably, the specific steps of S3 are as follows:

[0032] S31. Define a set of coordinates to be arranged. The set of coordinates to be arranged for underwater nodes in the region is: in, Represents the coordinates of the i-th node layout;

[0033] S32. Design a fitness function. The designed fitness function is as follows:

[0034] f(n)=max(f1(n)-f2(n));

[0035] Among them, f1(n) is the sum of the water depths within the detection range of all nodes; f2(n) is the penalty function for overlapping detection ranges of nodes;

[0036] S33, setting a constraint condition, wherein the constraint condition is set as a water depth constraint, ensuring that all nodes are placed below the required water depth, and using a penalty function method for the constraint;

[0037] S34. Establish an underwater node resource allocation model based on the designed fitness function and constraints.

[0038] Preferably, the specific formula of f1(n) in S32 is as follows:

[0039]

[0040] The specific formula of f2(n) is as follows:

[0041]

[0042] Among them, r is the node detection radius, D i represents the i-th element in set D, set D = {d ij |d ij <2r}, 1≤i <n,i<j≤b,d ij is the Euclidean distance between any two nodes i and j, expressed as

[0043] Preferably, the specific steps of S4 are as follows:

[0044] S41, initialize the population, use integer coding, the chromosome length is n, where the i-th gene position represents the i-th node, and the value range of its gene value x is [1, m], which means E x The coordinates in , m is the number of elements in set E; initialize a population with N chromosomes according to the encoding rule;

[0045] S42, perform population iteration, set the termination condition of the iteration, when the termination condition is met, the algorithm ends and returns the optimal individual found;

[0046] S43, perform fitness evaluation, calculate the fitness value of each individual according to the designed fitness function, and perform statistical analysis on the current population, recording its optimal individual and average fitness;

[0047] S44, performing a crossover operation, using a crossover operator to randomly generate two random numbers that are less than the length of the chromosome, independently selecting two mothers from the population, and exchanging the genes between the gene positions corresponding to the random numbers;

[0048] S45, performing mutation operation, using mutation operator, randomly selecting an individual from a chromosome population, and duplicating it as Cr, performing high probability mutation on Cr, obtaining mutant individual Cr1, performing two-point crossover on Cr and Cr1, obtaining new individuals Cr2 and Cr3, and finally comparing Cr, Cr1, Cr2 and Cr3, and selecting the individual with the highest fitness to replace the original individual;

[0049] S46. Perform a selection operation and use a selection operator to merge the parent population with the population that has completed the crossover and mutation operations to obtain a population of size 2N, and select N chromosomes from it as a new population according to the tournament selection method; if the new generation population contains the global optimal chromosome, return to S42; otherwise, replace the chromosome with the worst fitness in the new population with the global optimal chromosome, and return to S42.

[0050] Preferably, the iteration condition set in S42 is specifically as follows:

[0051] The number of population iterations reaches the set value: G = N G , where G is the number of iterations, N G is the maximum number of iterations;

[0052] Evolution stagnates when the range of the global optimal individual of the population after h consecutive iterations starting from the i-th generation is less than the set threshold, which can be expressed as:

[0053] |f * (i+h)-f * (i)|≤V,i∈[1,N G -h],h∈[1,N G ];

[0054] Among them, f * (n) is the global optimal individual fitness function value of the nth iteration, V is the threshold for judging whether the evolution is stagnant, and h is the upper limit of the evolution stagnation counter.

[0055] Therefore, the present invention adopts the above-mentioned method for deploying underwater detection nodes in a designated area based on a genetic algorithm, which has the following beneficial effects:

[0056] (1) The present invention has a fast deployment decision speed and significantly improves the time for determining the node deployment location. According to simulation verification, the deployment time can be completed within 2 seconds.

[0057] (2) The genetic algorithm used in the present invention is highly robust and adaptable to the detection space, obstacles and various constraints, and can be implemented in parallel, so that excellent solutions can be searched out more quickly in a multi-processor or distributed environment;

[0058] (3) The present invention uses an optimized fitness function design, taking the average depth within the detection range as part of the fitness function, thereby eliminating the influence of poor detection effect caused by excessive changes in terrain depth, and adopts convolution operation for preprocessing, so that the preprocessed map data can be saved locally, reducing the program optimization iteration time;

[0059] (4) The present invention innovatively uses breeding mutation operators to replace traditional mutation operators, thereby enhancing the local search capability without changing the global search capability.

[0060] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 A flowchart of a method for deploying underwater detection nodes in a designated area based on a genetic algorithm according to the present invention;

[0062] Figure 2 A population iteration curve diagram of a method for deploying underwater detection nodes in a designated area based on a genetic algorithm according to the present invention;

[0063] Figure 3 The present invention discloses an underwater node arrangement coordinate diagram of a method for deploying underwater detection nodes in a designated area based on a genetic algorithm. DETAILED DESCRIPTION

[0064] The following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. 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.

[0065] Example

[0066] like Figure 1 As shown, the present invention provides a method for deploying underwater detection nodes in a designated area based on a genetic algorithm, comprising the following steps:

[0067] S1. Collect and prepare relevant data of the detection area and underwater nodes, including the water depth threshold D for node layout p , the number of underwater detection nodes to be allocated n, the node detection radius r, and the obstacle boundary coordinate set C in the sea area o , specifies the set of longitude and latitude coordinates C of the detection area boundary.

[0068] Coordinate set of obstacles in the sea area

[0069] The obstacles It means that each obstacle is surrounded by different vertices, and k represents the number of obstacles;

[0070] The specified detection area boundary longitude and latitude coordinate set

[0071] in represents the i-th boundary point of the detection area, represents the longitude coordinate, represents the latitude coordinate, and l is the total number of boundary points in the detection area.

[0072] In this embodiment, the depth threshold D of the node arrangement p =2000, the number of underwater detection nodes to be allocated n = 20, the node detection radius r = 15,

[0073] Obstacle boundary latitude and longitude coordinates set in

[0074] The detection area boundary coordinate set C = {(120.03°, 21.6°), (120.87°, 21.2°), (121.65°, 21.2°), (121.65°, 19.94°), (121.36°, 20.3°), (120.37°, 19.49°), (119.66°, 19.89°), (119.66°, 20.31°)}.

[0075] S2. Preprocess the water depth data of the detection area and perform coordinate transformation and convolution operations.

[0076] S21. Find in set C λ min , max , forming a rectangular area Q, where Represents the minimum latitude of all coordinate points in set C, represents the maximum latitude of all coordinate points in set C, λ min represents the minimum value of the longitude of all coordinate points in set C, λ max Represents the maximum value of the longitude of all coordinate points in set C, for all coordinate points q in Q (qlon，qlat) The following formulas are established:

[0077]

[0078] In this embodiment, the detection area is found in set C with a minimum longitude of 119.66°, a minimum latitude of 19.49°, a maximum longitude of 121.65°, and a maximum latitude of 21.6°, forming a rectangular area Q with longitude from 119.66° to 121.65° and latitude from 19.49° to 21.6°.

[0079] S22. Determine the length w and width h of the rectangular area Q using the haversine formula;

[0080] The formula for the length w of the rectangular area Q is as follows:

[0081]

[0082] The formula for the width h of the rectangular area Q is as follows:

[0083]

[0084] Wherein, R is the average radius of the earth, R = 6371.4 km, and [·] indicates rounding operation.

[0085] In this embodiment,

[0086]

[0087] S23. Establish a rectangular coordinate system and map the boundary points.

[0088] S231, establish a rectangular coordinate system, take the lower left corner of the rectangular area Q as the coordinate origin, and establish a rectangular coordinate system O with each kilometer as the step length xy , where the horizontal axis represents longitude and the vertical axis represents latitude. The horizontal axis value range is [0, w] and the vertical axis value range is [0, h]. In this embodiment, the longitude coordinate range is [0, 209] and the latitude coordinate range is [0, 235].

[0089] S232, mapping boundary points, mapping the longitude and latitude of all detection area boundary points to xy In this embodiment, their coordinates are {(39,235), (126,191), (207,191), (209,51), (178,91), (75,0), (0,45), (0,92)}. Map the vertex coordinates of all obstacles to O xy Below, its coordinates are {(41,38), (49,24), (65,24), (73,38), (65,53), (49,53)}.

[0090] At the same time in O xy In the coordinate system, the centroid of the detection area is calculated as Then, the boundary points of the detection area are sorted according to the angle between each boundary point and the centroid, and the scanning line algorithm is used to scan horizontally line by line to calculate the intersection of the scanning line and the polygon, and then all the coordinates in the detection area are extracted according to the parity of the intersection. Similarly, all the coordinates in each obstacle are calculated, and the parts of all the coordinates in the detection area that overlap with all the coordinates in each obstacle are removed, that is, the set of all the coordinates that can be arranged in the detection area is E = {(75,0), (76,0), ..., (41,234)}, which is a total of 33018 arrangable points.

[0091] S24, rasterize the rectangular area Q, divide the area Q into a grid of h rows and w columns at a step length of one kilometer, and map the rasterized area into a matrix A' with h rows and w columns, where the element A' in A' is (i,j) The value is the average depth of the area covered by the grid in the i-th row and j-th column. In this embodiment, the grid has 235 rows and 209 columns.

[0092] S25, process the matrix A', invert all elements in the matrix A' to obtain the matrix A, that is, convert the underwater depth data into positive numbers and the above-water values ​​into negative numbers, that is;

[0093] A (i,j) =-A' (i,j) , 1≤i≤235,1≤j≤209;

[0094] According to the node detection radius r, design the convolution kernel matrix K (2r+1)×(2r+1) In this embodiment, the node detection radius r is 15 kilometers, so the convolution kernel matrix K 31×31 , where K (i,j) The value of is:

[0095]

[0096] The convolution matrix A is convolved with the convolution kernel K to obtain the target matrix B with the same dimension as the matrix A.

[0097] S3. Design fitness functions and constraints and build resource allocation models.

[0098] S31. Define a set of coordinates to be arranged. The set of coordinates to be arranged for underwater nodes in the region is: in, Represents the coordinates of the i-th node arrangement.

[0099] S32. Design a fitness function. The designed fitness function is as follows:

[0100] f(n)=max(f1(n)-f2(n));

[0101] Among them, f1(n) is the sum of the water depths within the detection range of all nodes; f2(n) is the penalty function for overlapping node detection ranges.

[0102] The specific formula of f1(n) is as follows:

[0103]

[0104] The specific formula of f2(n) is as follows:

[0105] f2(n)=∑ i＝1 r 4.5 / (D i +1), D i ∈D;

[0106] Among them, r is the node detection radius, D i represents the i-th element in set D, set D = {d ij |d ij <2r}, 1≤i <n,i<j≤n,d ij is the Euclidean distance between any two nodes i and j, expressed as

[0107] S33, setting constraints, where the constraints are set as water depth constraints, ensuring that all nodes are placed below the required water depth, and using a penalty function method to enforce the constraints.

[0108] S34. Establish an underwater node resource allocation model based on the designed fitness function and constraints.

[0109] S4. Use the improved genetic algorithm to determine the node deployment method based on maximizing the fitness function.

[0110] S41, initialize the population, use integer coding, the chromosome length is n, where the i-th gene position represents the i-th node, and the value range of its gene value x is [1, m], which means E x , m is the number of elements in set E; a population with N chromosomes is initialized according to the encoding rule. In this embodiment, the value range of its gene value x is [1,33018], and the number of chromosomes is 50.

[0111] S42, perform population iteration, set the termination condition of the iteration, when the termination condition is met, the algorithm ends and returns the optimal individual found;

[0112] The specific iteration conditions are as follows:

[0113] (1) The number of population iterations reaches the set value: G = N G , where G is the number of iterations, N Gis the maximum number of iterations; in this embodiment, the maximum number of population iterations is set to 200 times, that is, G=200.

[0114] (2) Evolution stagnates, that is, starting from the i-th generation, the range of change of the global optimal individual of the population after h consecutive iterations is less than the set threshold, which can be specifically expressed as:

[0115] |f * (i+h)-f * (i)|≤V,i∈[1,N G -h],h∈[1,N G ];

[0116] Among them, f * (n) is the global optimal individual fitness function value of the nth iteration, V is the threshold for judging whether the evolution is stagnant, and h is the upper limit of the evolution stagnation counter.

[0117] In this embodiment, |f * (i+2)-f * (i)|≤10 -6 , i∈[1,198].

[0118] S43, perform fitness evaluation, calculate the fitness value of each individual according to the designed fitness function, and perform statistical analysis on the current population, recording its optimal individual and average fitness;

[0119] S44, perform a crossover operation, use a crossover operator, randomly generate two random numbers that are smaller than the chromosome length, the chromosome length of this embodiment is 20, independently select two mothers from the population, swap the genes between the gene positions corresponding to the random numbers, and set the probability of each individual being selected for the crossover operation to 0.7.

[0120] S45. Perform mutation operation. Use mutation operator to randomly select an individual from a chromosome population and copy it as Cr. Perform mutation with probability 0.1 on each gene position on Cr to obtain mutant individual Cr1. Perform two-point crossover on Cr and Cr1 to obtain new individuals Cr2 and Cr3. Finally, compare Cr, Cr1, Cr2 and Cr3, and select the individual with the highest fitness to replace the original individual.

[0121] S46, perform a selection operation, use the selection operator, merge the parent population with the population that completes the crossover and mutation operations to obtain a population of size 2N, and select N chromosomes from it as a new population according to the tournament selection method. If the new generation population contains the global optimal chromosome, return to S42; otherwise, replace the chromosome with the worst fitness in the new population with the global optimal chromosome, and return to S42.

[0122] In this embodiment, the population size is 100 and the number of chromosomes is 50.

[0123] The simulation environment is PyCharm 2024.2.1, Python 3.10, the processor is R5 5600H 3.30GHz, the memory is 16.0GB, and the running time is 1.81s.

[0124] The simulation layout results are shown in Table 1:

[0125] Table 1 Allocation results

[0126] (13,82) (15,52) (162,64) (68,84) (148,29) (70,18) (40,96) (44,65) (66,115) (96,114) (71,49) (158,171) (111,192) (103,11) (16,116) (116,52) (81,191) (134,79) (41,32) (104,83)

[0127] Table 2 Population iteration log

[0128]

[0129] The population iteration curve is as follows Figure 2 As shown, the horizontal axis is the number of iterations, the vertical axis is the fitness function value, the yellow line is the global optimal individual iteration curve, and the blue line is the population average fitness value iteration curve;

[0130] The simulation layout coordinates are obtained as Figure 3 As shown, the red part is the coordinate that meets the constraints, the white part is the coordinate that does not meet the constraints, and the white hexagonal area is the obstacle.

[0131] Therefore, the present invention adopts the above-mentioned method for deploying underwater detection nodes in a designated area based on a genetic algorithm, which can solve the problem of rapid deployment of multi-constrained underwater nodes under a two-dimensional topological structure.

[0132] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that they can still modify or replace the technical solution of the present invention with equivalents, and these modifications or equivalent replacements cannot cause the modified technical solution to deviate from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for deploying underwater detection nodes in a designated area based on a genetic algorithm, characterized in that: The following steps are involved: S1. Collect and prepare relevant data of the detection area and underwater nodes, including the water depth threshold D for node layout p , the number of underwater detection nodes to be allocated n, the node detection radius r, and the obstacle boundary coordinate set C in the sea area o , the specified detection area boundary longitude and latitude coordinate set C; S2, preprocessing the water depth data of the detection area, performing coordinate transformation and convolution operation; S3, design fitness function and constraints, and build resource allocation model; S4. Use the improved genetic algorithm to determine the node deployment method based on maximizing the fitness function.

2. According to claim 1, a method for deploying underwater detection nodes in a designated area based on a genetic algorithm is characterized in that: In step S1, the coordinate set of the obstacle boundary in the sea area Indicates that there are k obstacles in total. Among them, obstacles Indicates that each obstacle is surrounded by different vertices; The specified detection area boundary longitude and latitude coordinate set in, represents the i-th boundary point of the detection area, represents the longitude coordinate, represents the latitude coordinate, and l is the total number of boundary points in the detection area.

3. According to the method for deploying underwater detection nodes in a designated area based on a genetic algorithm according to claim 1, it is characterized in that: The specific steps of S2 are as follows: S21. Find in set C λ min , max , forming a rectangular area Q, where Represents the minimum latitude of all coordinate points in set C, represents the maximum latitude of all coordinate points in set C, λ min represents the minimum value of the longitude of all coordinate points in set C, λ max Represents the maximum value of the longitude of all coordinate points in set C, for all coordinate points q in Q (qlon，qlat) The following formulas are established: S22. Determine the length w and width h of the rectangular area Q using the haversine formula; S23, establishing a rectangular coordinate system and mapping boundary points; S24, rasterize the rectangular area Q, divide the area Q into a grid of h rows and w columns at a step length of one kilometer, and map the rasterized area into a matrix A' with h rows and w columns, where the element A' in A' is (i,j) The value is the average depth of the area covered by the grid in the i-th row and j-th column; S25, process the matrix A', invert all elements in the matrix A' to obtain the matrix A, which is A (i,j) =-A' (i,j) , 1≤i≤h,1≤j≤w, according to the node detection radius r, design the convolution kernel matrix K (2r+1)×(2r+1) , perform equal-size convolution operations on the matrix A to be convolved and the convolution kernel K to obtain the target matrix B of the same dimension as the matrix A.

4. The method for deploying underwater detection nodes in a designated area based on a genetic algorithm according to claim 3, characterized in that: The formula for the length w of the rectangular area Q in step S22 is as follows: The formula for the width h of the rectangular area Q is as follows: Where R is the average radius of the Earth and [·] indicates rounding.

5. The method for deploying underwater detection nodes in a designated area based on a genetic algorithm according to claim 3, characterized in that: The specific steps of S23 are as follows: S231, establish a rectangular coordinate system, take the lower left corner of the rectangular area Q as the coordinate origin, and establish a rectangular coordinate system O with each kilometer as the step length xy , where the horizontal axis represents longitude and the vertical axis represents latitude. The horizontal axis value range is [0,w] and the vertical axis value range is [0,h]; S232, mapping boundary points, mapping the latitude and longitude coordinates of all boundary points to xy In the coordinate system, the centroid of the detection area is calculated, and the boundary points of the detection area are sorted according to the angle between each boundary point and the centroid. The scanning line method is used to extract all the coordinates in the detection area, and the coordinates of each obstacle are calculated. The overlapping parts are removed to form a set of arrangable coordinates E, which is defined as Where m is the total number of coordinates that can be arranged, is the latitude of the i-th arrangable coordinate, is the longitude of the i-th arrangeable coordinate.

6. The method for deploying underwater detection nodes in a designated area based on a genetic algorithm according to claim 3, characterized in that: Convolution kernel matrix K (i,j) The value of is:

7. The method for deploying underwater detection nodes in a designated area based on a genetic algorithm according to claim 1, characterized in that: The specific steps of S3 are as follows: S31. Define a set of coordinates to be arranged. The set of coordinates to be arranged for underwater nodes in the region is: in, Represents the coordinates of the i-th node layout; S32. Design a fitness function. The designed fitness function is as follows: f(n)=max(f1(n)-f2(n)); Among them, f1(n) is the sum of the water depths within the detection range of all nodes; f2(n) is the penalty function for overlapping detection ranges of nodes; S33, setting a constraint condition, wherein the constraint condition is set as a water depth constraint, ensuring that all nodes are placed below the required water depth, and using a penalty function method for the constraint; S34. Establish an underwater node resource allocation model based on the designed fitness function and constraints.

8. The method for deploying underwater detection nodes in a designated area based on a genetic algorithm according to claim 7, characterized in that: The specific formula of f1(n) in S32 is as follows: The specific formula of f2(n) is as follows: f2(n)=∑ i＝1 r 4.5 / (D i +1),D i ∈D; Among them, r is the node detection radius, D i represents the i-th element in set D, set D = {d ij |d ij <2r}, 1≤i <n,i<j≤n,d ij is the Euclidean distance between any two nodes i and j, expressed as 9. The method for deploying underwater detection nodes in a designated area based on a genetic algorithm according to claim 1, characterized in that: The specific steps of S4 are as follows: S41, initialize the population, use integer coding, the chromosome length is n, where the i-th gene position represents the i-th node, and the value range of its gene value x is [1, m], which means E x The coordinates in , m is the number of elements in set E; initialize a population with N chromosomes according to the encoding rule; S42, perform population iteration, set the termination condition of the iteration, when the termination condition is met, the algorithm ends and returns the optimal individual found; S43, perform fitness evaluation, calculate the fitness value of each individual according to the designed fitness function, and perform statistical analysis on the current population, recording its optimal individual and average fitness; S44, performing a crossover operation, using a crossover operator to randomly generate two random numbers that are less than the length of the chromosome, independently selecting two mothers from the population, and exchanging the genes between the gene positions corresponding to the random numbers; S45, performing mutation operation, using mutation operator, randomly selecting an individual from a chromosome population, and duplicating it as Cr, performing high probability mutation on Cr, obtaining mutant individual Cr1, performing two-point crossover on Cr and Cr1, obtaining new individuals Cr2 and Cr3, and finally comparing Cr, Cr1, Cr2 and Cr3, and selecting the individual with the highest fitness to replace the original individual; S46. Perform a selection operation and use a selection operator to merge the parent population with the population that has completed the crossover and mutation operations to obtain a population of size 2N, and select N chromosomes from it as a new population according to the tournament selection method; if the new generation population contains the global optimal chromosome, return to S42; otherwise, replace the chromosome with the worst fitness in the new population with the global optimal chromosome, and return to S42.

10. The method for deploying underwater detection nodes in a designated area based on a genetic algorithm according to claim 9, characterized in that: The iteration conditions set by S42 are as follows: The number of population iterations reaches the set value: G = N G , where G is the number of iterations, N G is the maximum number of iterations; Evolution stagnates when the range of the global optimal individual of the population after h consecutive iterations starting from the i-th generation is less than the set threshold, which can be expressed as: |f * (i+h)-f * (i)|≤V,i∈[1,N G -h],h∈[1,N G ]; Among them, f * (n) is the global optimal individual fitness function value of the nth iteration, V is the threshold for judging whether the evolution is stagnant, and h is the upper limit of the evolution stagnation counter.

Images