Electromagnetic environment investigation and evaluation method and system based on vehicle-mounted electromagnetic detection equipment

Abstract

The application provides an electromagnetic environment investigation and evaluation method and system based on a vehicle-mounted electromagnetic detection device, and belongs to the technical field of electromagnetic environment monitoring. The method comprises the following steps: determining a target monitoring point set and corresponding geographical position coordinates of a target region based on spatial geographical information of the target region; determining a patrol path of the vehicle-mounted electromagnetic detection device by using a preset path planning algorithm based on the geographical position coordinates of the target monitoring point set; collecting electromagnetic environment monitoring data at each monitoring point in the target monitoring point set when a vehicle loaded with the electromagnetic environment detection device travels along the patrol path; processing the electromagnetic environment monitoring data of each monitoring point to obtain an electromagnetic environment data set; and analyzing the electromagnetic environment data set to obtain an electromagnetic environment evaluation result of the target region. The electromagnetic environment investigation and evaluation method and system based on the vehicle-mounted electromagnetic detection device can comprehensively, efficiently and accurately evaluate the electromagnetic environment of the target region.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electromagnetic environment monitoring, and in particular to an electromagnetic environment investigation and evaluation method and system based on a vehicle-mounted electromagnetic detection device. BACKGROUND

[0002] At present, vehicle-mounted electromagnetic detection has become a new direction of electromagnetic environment investigation due to its strong mobility and flexible deployment characteristics. By mounting high-sensitivity detection equipment on a vehicle, dynamic inspection of a target area can be performed to efficiently collect multi-dimensional electromagnetic data. However, existing vehicle-mounted electromagnetic environment evaluation methods generally have problems such as rough path planning, low data processing accuracy, and single evaluation dimension.

[0003] Therefore, there is an urgent need for an electromagnetic environment investigation and evaluation method and system based on a vehicle-mounted electromagnetic detection device to achieve comprehensive, efficient, and accurate evaluation of the electromagnetic environment of a target area. SUMMARY

[0004] To solve the above technical problems, the present application provides an electromagnetic environment investigation and evaluation method and system based on a vehicle-mounted electromagnetic detection device.

[0005] In a first aspect, the present application provides an electromagnetic environment investigation and evaluation method based on a vehicle-mounted electromagnetic detection device, comprising:

[0006] Based on the spatial geographic information of the target area, a target monitoring point set of the target area and its corresponding geographic location coordinates are determined;

[0007] Based on the geographic location coordinates of the target monitoring point set, a preset path planning algorithm is used to determine a patrol path of the vehicle-mounted electromagnetic detection device;

[0008] When a vehicle loaded with an electromagnetic environment detection device travels along the patrol path, electromagnetic environment monitoring data is collected at each monitoring point in the target monitoring point set;

[0009] The electromagnetic environment monitoring data of each monitoring point is processed to obtain an electromagnetic environment data set;

[0010] The electromagnetic environment data set is analyzed to obtain an electromagnetic environment evaluation result of the target area.

[0011] Optionally, the target monitoring point set of the target area is determined based on the spatial geographic information of the target area, comprising:

[0012] The target area is spatially gridded according to the area of the target area and a preset spatial resolution threshold to generate an initial monitoring point set;

[0013] Screening the initial monitoring point set based on the spatial geographic information of the target area, and removing the monitoring points in the impassable area to obtain a screened monitoring point set;

[0014] Simulating field intensity distribution of the preset frequency band signal in the target area according to an electromagnetic wave propagation attenuation model to obtain a blind area with a field intensity value less than a detection threshold;

[0015] Supplementing monitoring points in the blind area based on a spatial adaptive supplement method to obtain a supplemented monitoring point set;

[0016] Combining the screened monitoring point set and the supplemented monitoring point set to obtain a target monitoring point set.

[0017] The present application can effectively cover the target area by grid division, screening of monitoring points in the impassable area, determination of the blind area based on simulation of field intensity distribution, and supplement of monitoring points, thereby improving the accuracy and integrity of electromagnetic environment data collection.

[0018] Optionally, the spatial adaptive supplement method for supplementing monitoring points in the blind area comprises:

[0019] Dividing the blind area into sub-areas and calculating the simulated field intensity value of the center point of the sub-area;

[0020] When the simulated field intensity value of the sub-area is less than the detection threshold, the sub-area is further subdivided into smaller sub-areas;

[0021] At the center point of each subdivided sub-area, the difference between the simulated field intensity value and the detection threshold is calculated, wherein the difference is the detection threshold minus the simulated field intensity value;

[0022] Selecting the center point with a difference greater than a preset difference threshold as a supplemented monitoring point;

[0023] Repeating the iterative subdivision process and calculating the spatial density of the current supplemented monitoring point set, wherein the spatial density is calculated based on the number of monitoring points per unit area;

[0024] If the spatial density is less than a preset blind area monitoring density threshold, the sub-area with a simulated field intensity value less than the detection threshold is further subdivided, and the supplemented monitoring point set and the spatial density value are updated after the subdivision;

[0025] When the spatial density value reaches the blind area monitoring density threshold or the number of iterations reaches a preset maximum number of iterations, the subdivision is stopped.

[0026] The application can supplement monitoring points in the blind area by using a space adaptive supplement method, can adaptively adjust the distribution of monitoring points according to the field strength of different positions in the blind area, and further improves the accuracy and comprehensiveness of monitoring, which helps to more accurately grasp the electromagnetic environment status of the blind area.

[0027] Optionally, the method for setting the preset difference threshold value comprises:

[0028] Obtaining a spatial field strength gradient statistical value of the blind area, the spatial field strength gradient statistical value being obtained by calculating the change rate of the simulated field strength values of the center points of adjacent sub-regions in the blind area;

[0029] Adjusting the preset difference threshold value according to the spatial field strength gradient statistical value;

[0030] The method for adjusting the preset difference threshold value according to the spatial field strength gradient statistical value comprises:

[0031] When the field strength gradient statistical value is greater than a first statistical value threshold value, reducing the preset difference threshold value by a first step length;

[0032] When the field strength gradient statistical value is less than a second statistical value threshold value, increasing the preset difference threshold value by a second step length.

[0033] The application can adjust the preset difference threshold value according to the spatial field strength gradient statistical value of the blind area, which can adapt to regions with different electromagnetic environment change degrees. When the field strength gradient statistical value is large, i.e., the electromagnetic environment changes dramatically, the difference threshold value is reduced, so that more points can be selected as supplement monitoring points, thereby more finely monitoring the electromagnetic environment change; on the contrary, the difference threshold value is increased, which can reduce unnecessary monitoring points in the region with a gentle electromagnetic environment change, thereby improving the monitoring efficiency while ensuring the monitoring accuracy.

[0034] Optionally, the geographic position coordinates of the target monitoring point set are used to determine the inspection path of the vehicle-mounted electromagnetic detection equipment by using a preset path planning algorithm, which comprises:

[0035] Converting the geographic position coordinates of the target monitoring point set into latitude and longitude coordinates;

[0036] Generating feasible path nodes according to the spatial geographic information of the target region and the vehicle passing constraint condition;

[0037] Constructing a weighted topological network based on the target monitoring point set and the feasible path nodes;

[0038] input the weighted topological network into the ant colony algorithm to drive path generation to generate a patrol path meeting the vehicle passing constraint condition by a multi-objective optimization function; wherein the multi-objective optimization function comprises: minimization of total travel distance of the patrol path, minimization of vehicle energy consumption estimation based on total travel distance of the patrol path, and minimization of variation coefficient of distance between consecutive monitoring points on the path.

[0039] The application determines the patrol path of the vehicle-mounted electromagnetic detection device by using a preset path planning algorithm, and can generate a patrol path meeting the vehicle passing constraint condition and comprehensively considering factors such as travel distance, energy consumption and monitoring point distance uniformity by converting geographic position coordinates, generating feasible path nodes, constructing a weighted topological network and combining the ant colony algorithm and the multi-objective optimization function. This not only improves the detection efficiency of the vehicle-mounted electromagnetic detection device, but also reduces the detection cost.

[0040] Optionally, the geographic position coordinates based on the target monitoring point set determine the patrol path of the vehicle-mounted electromagnetic detection device by using a preset path planning algorithm, and further comprise:

[0041] When the ant colony algorithm generates multiple patrol paths, a path evaluation model is used to calculate the comprehensive score of each patrol path.

[0042] The patrol path with the highest comprehensive score is selected as the final patrol path.

[0043] If the difference between the comprehensive scores of the multiple patrol paths is less than a first difference threshold, the patrol path with the minimum total travel distance among the multiple patrol paths is selected as the final patrol path.

[0044] The application calculates the comprehensive score by using a path evaluation model when the ant colony algorithm generates multiple patrol paths, and selects the optimal path according to the score. This way can filter out the most suitable patrol path from multiple paths, further optimize the patrol scheme, and enable the vehicle-mounted electromagnetic detection device to perform electromagnetic environment detection in the best path, thereby improving the overall benefit of the detection work.

[0045] Optionally, the electromagnetic environment monitoring data of the monitoring points is processed to obtain an electromagnetic environment data set, comprising:

[0046] The electromagnetic environment monitoring data of the monitoring points is subjected to data cleaning to obtain a sub-data set.

[0047] The sub-data set is subjected to time and space correlation and feature extraction to generate an electromagnetic environment feature matrix.

[0048] The electromagnetic environment feature matrix and geographic information system data of the target region are subjected to fusion processing to obtain an electromagnetic environment data set.

[0049] The application processes electromagnetic environment monitoring data of each monitoring point, removes noise and invalid data through data cleaning to improve data quality, mines useful information in the data through time and space correlation and feature extraction, and then fuses with geographic information system data to combine electromagnetic environment data and geographic information, so that the electromagnetic environment status of the target region can be more intuitively and comprehensively reflected.

[0050] Optionally, the analysis on the electromagnetic environment data set to obtain the evaluation result comprises:

[0051] The electromagnetic environment data set is filtered based on a preset evaluation index system to obtain a first electromagnetic environment data set;

[0052] The parameters of the electromagnetic environment evaluation model are iteratively optimized through a genetic algorithm to obtain a target electromagnetic environment evaluation model;

[0053] The first electromagnetic environment data set is input into the target electromagnetic environment evaluation model to obtain an evaluation result of the target region; wherein the electromagnetic environment evaluation model is an extreme gradient boosting algorithm, and the evaluation result is a comprehensive quality evaluation result of the electromagnetic environment of the target region.

[0054] The application filters the electromagnetic environment data set based on a preset evaluation index system, can remove irrelevant or interference data, and highlights key information. The parameters of the electromagnetic environment evaluation model are optimized through a genetic algorithm, which can improve the accuracy and adaptability of the model. The filtered data is input into the optimized extreme gradient boosting algorithm model, and a more reliable comprehensive quality evaluation result of the electromagnetic environment of the target region can be obtained.

[0055] Optionally, the iteratively optimizing the parameters of the electromagnetic environment evaluation model through the genetic algorithm to obtain the target electromagnetic environment evaluation model comprises:

[0056] A set of to-be-optimized hyperparameters of the extreme gradient boosting algorithm is determined, and the set of to-be-optimized hyperparameters comprises a tree number, a tree depth, a learning rate and a minimum loss drop threshold for splitting;

[0057] The fitness function is defined as a comprehensive evaluation error on a historical electromagnetic reference data set as a fitness evaluation index, and the comprehensive evaluation error is a weighted sum of a normalized mean square error and a spectral feature importance deviation;

[0058] The population size is determined based on the number of frequency bands of the target region;

[0059] The mutation probability is determined based on the electromagnetic field strength gradient in the electromagnetic environment data set;

[0060] The crossover probability is determined based on the spatial autocorrelation coefficient of the electromagnetic field strength of the target region;

[0061] iteratively optimize parameters of the electromagnetic environment evaluation model based on the fitness function, the mutation probability and the crossover probability;

[0062] terminating the optimization when the continuous optimal fitness change rate is less than a first change rate threshold or a maximum iteration number is reached, to obtain a globally optimal hyperparameter; and constructing a target electromagnetic environment evaluation model according to the globally optimal hyperparameter.

[0063] The application clearly defines the specific process of iteratively optimizing the parameters of the electromagnetic environment evaluation model by the genetic algorithm, and for the key hyperparameters of the extreme gradient boosting algorithm, the globally optimal hyperparameter can be effectively searched by defining the fitness function and determining the population size, mutation probability and crossover probability according to the target region frequency band number, electromagnetic field strength gradient and spatial autocorrelation coefficient, so as to construct a more optimal target electromagnetic environment evaluation model, thereby improving the precision and reliability of the electromagnetic environment evaluation.

[0064] A second aspect of the embodiment of the application provides an electromagnetic environment investigation and evaluation system based on a vehicle-mounted electromagnetic detection device, including:

[0065] A data processing module is configured to determine a target monitoring point set of a target region and corresponding geographic location coordinates of the target monitoring point set based on spatial geographic information of the target region.

[0066] A path planning module is configured to determine a patrol path of the vehicle-mounted electromagnetic detection device by using a preset path planning algorithm based on the geographic location coordinates of the target monitoring point set.

[0067] A collection control module is configured to control a vehicle loaded with the electromagnetic environment detection device to collect electromagnetic environment monitoring data at each monitoring point in the target monitoring point set when the vehicle travels along the patrol path.

[0068] A collection data processing module is configured to process the electromagnetic environment monitoring data of each monitoring point to obtain an electromagnetic environment data set.

[0069] A data analysis module is configured to analyze the electromagnetic environment data set to obtain an electromagnetic environment evaluation result of the target region.

[0070] A third aspect of the embodiment of the application provides an electronic device including a memory, a processor and a computer program stored in the memory and running on the processor, and the processor implements the steps of the above-mentioned electromagnetic environment investigation and evaluation method based on a vehicle-mounted electromagnetic detection device when executing the computer program.

[0071] In a fourth aspect, the application provides a computer readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-mentioned electromagnetic environment investigation and evaluation method based on a vehicle-mounted electromagnetic detection device.

[0072] The electromagnetic environment investigation and evaluation method and system based on a vehicle-mounted electromagnetic detection device provided by the application has the advantages that the application accurately determines the monitoring point set and its geographic position coordinates according to the spatial geographic information of the target area, generates an optimal inspection path by combining a preset path planning algorithm, and greatly improves the inspection efficiency of the vehicle-mounted electromagnetic detection device. Secondly, the electromagnetic environment monitoring data is collected at each monitoring point and processed by the system, so that a comprehensive and accurate electromagnetic environment data set can be obtained, and data omission or deviation can be avoided. By deeply analyzing the electromagnetic environment data set, multi-dimensional evaluation results including an electromagnetic environment distribution map and interference source positioning information are output, which not only can intuitively present the current situation of the electromagnetic environment of the target area, but also can predict the evolution trend and assess the health risk level, so as to provide scientific and reliable decision-making basis for communication network optimization, interference management, policy making and the like, and realize comprehensive, efficient and accurate evaluation of the electromagnetic environment of the target area. BRIEF DESCRIPTION OF DRAWINGS

[0073] Figure 1 A flowchart of an electromagnetic environment investigation and evaluation method based on a vehicle-mounted electromagnetic detection device provided by an embodiment of the application is shown in the figure.

[0074] Figure 2 A structural block diagram of an electromagnetic environment investigation and evaluation system based on a vehicle-mounted electromagnetic detection device provided by an embodiment of the application is shown in the figure.

[0075] Figure 3 A schematic block diagram of an electronic device provided by an embodiment of the application is shown in the figure. DETAILED DESCRIPTION

[0076] In the following description, specific details are set forth such as particular system configurations, techniques, etc., in order to provide a thorough understanding of the embodiments of the application. However, it will be apparent to those skilled in the art that the application can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the application with unnecessary detail.

[0077] In order to make the objectives, technical solutions and advantages of the application clearer, the following will combine the accompanying drawings to make a detailed description. Figures 1-3 The application is described by specific embodiments.

[0078] Reference should be made to Figure 1 , Figure 1A flowchart of an electromagnetic environment investigation and evaluation method based on a vehicle-mounted electromagnetic detection device is provided for an embodiment of the present application. The method comprises:

[0079] S101: Based on the spatial geographic information of the target area, determine the target monitoring point set of the target area and its corresponding geographic location coordinates.

[0080] In this embodiment, the spatial geographic information of the target area is obtained through satellite remote sensing images, geographic information system (GIS) data, unmanned aerial vehicle surveying and mapping, and other multi-source data fusion methods. According to the characteristics of the terrain, geomorphology, electromagnetic sensitive facility distribution, population activity density, and other characteristics of the target area, the grid division method, key area point distribution method, or mixed point distribution method is used to scientifically set up monitoring points in the target area, obtain the target monitoring point set, and obtain the geographic location coordinates of each monitoring point. The monitoring point information database can also be established to record the number, location, and surrounding environment information of the monitoring points.

[0081] S102: Based on the geographic location coordinates of the target monitoring point set, use a preset path planning algorithm to determine the inspection path of the vehicle-mounted electromagnetic detection device.

[0082] In this embodiment, the path planning algorithm can be a genetic algorithm or an ant colony algorithm. According to the traffic conditions data, road traffic restriction information (such as height limit, weight limit, one-way line, etc.), and the endurance of the vehicle-mounted electromagnetic detection device, the shortest total driving distance, the optimal detection time, and the minimum energy consumption are used as the optimization target to generate the inspection path of the vehicle-mounted electromagnetic detection device. In addition, the generated inspection path can be manually adjusted and optimized according to the actual situation to generate the final inspection path.

[0083] S103: Control the vehicle loaded with the electromagnetic environment detection device to drive along the inspection path, and collect electromagnetic environment monitoring data at each monitoring point in the target monitoring point set.

[0084] In this embodiment, when the vehicle loaded with the electromagnetic environment detection device drives along the inspection path, electromagnetic environment monitoring data is collected at each monitoring point in the target monitoring point set. High-precision, wide-band electromagnetic environment detection equipment such as electromagnetic radiation measuring instruments, electric field strength measuring instruments, and magnetic field strength measuring instruments is configured, and the equipment is calibrated and tested to ensure the measurement accuracy and reliability of the equipment. When the vehicle drives to each monitoring point, the electric field strength, magnetic field strength, electromagnetic spectrum, and other electromagnetic environment monitoring data of multiple frequency bands are collected according to the preset sampling frequency, sampling time, and sampling method. During the collection process, the collection time, equipment number, and monitoring point number are recorded in real time, and the collected data is preliminarily checked for quality to eliminate abnormal data.

[0085] S104: Process the electromagnetic environment monitoring data of each monitoring point to obtain an electromagnetic environment data set;

[0086] In this embodiment, the collected original electromagnetic environment monitoring data is preprocessed, including data filtering, denoising, normalization and other operations, to eliminate noise and interference in the data and obtain an electromagnetic environment data set. The integrity, accuracy and consistency of the electromagnetic environment data set can also be evaluated, and data that does not meet the quality requirements can be supplemented or corrected.

[0087] S105: Analyze the electromagnetic environment data set to obtain an electromagnetic environment evaluation result of the target area.

[0088] In this embodiment, the electromagnetic environment data set is analyzed in multiple dimensions; according to the relevant electromagnetic environment standards, specifications of the country and industry, and the functional positioning and electromagnetic environment requirements of the target area, the electromagnetic environment quality level, electromagnetic interference level, and electromagnetic environment safety risk of the target area are evaluated; a detailed electromagnetic environment evaluation report is generated, and the report content includes evaluation data and evaluation results.

[0089] From the above, it can be concluded that the present application accurately determines the monitoring point set and its geographic coordinates according to the spatial geographic information of the target area, and generates an optimal inspection path combining a preset path planning algorithm, which greatly improves the inspection efficiency of the vehicle-mounted electromagnetic detection equipment. Secondly, the electromagnetic environment monitoring data is collected at each monitoring point and systematically processed, which can obtain comprehensive and accurate electromagnetic environment data sets, avoiding data omission or deviation. Through deep analysis of the electromagnetic environment data set, multi-dimensional evaluation results including electromagnetic environment distribution map and interference source positioning information are output, which not only can intuitively present the current situation of the electromagnetic environment of the target area, but also can predict the evolution trend and assess the health risk level, providing scientific and reliable decision-making basis for communication network optimization, interference management, policy making, etc., and realizing comprehensive, efficient and accurate evaluation of the electromagnetic environment of the target area.

[0090] In one embodiment of the present application, based on the spatial geographic information of the target area, the target monitoring point set of the target area is determined, including:

[0091] According to the area of the target area and the preset spatial resolution threshold, the target area is spatially gridded to generate an initial monitoring point set;

[0092] Based on the spatial geographic information of the target area, the initial monitoring point set is screened to remove monitoring points in non-passable areas to obtain a screened monitoring point set;

[0093] According to the electromagnetic wave propagation attenuation model, the field strength distribution of the preset frequency band signal in the target area is simulated to obtain a blind area with a field strength value less than a detection threshold;

[0094] The monitoring points in the blind area are supplemented based on the spatial adaptive supplement method to obtain a supplemented monitoring point set.

[0095] The screening monitoring point set and the supplemented monitoring point set are combined to obtain a target monitoring point set.

[0096] In the embodiment, a spatial resolution threshold is set according to the geographical range of the target area and the accuracy requirement of the electromagnetic environment investigation, wherein the spatial resolution threshold is determined according to the detection capability of the vehicle-mounted electromagnetic detection device, the investigation cost and the complexity of the electromagnetic environment of the area; if the target area is a city dense area and the electromagnetic environment is complex, a smaller spatial resolution threshold can be set, for example, 100 m x 100 m; if the target area is a vast rural area, a larger spatial resolution threshold can be set, for example, 500 m x 500 m.

[0097] In the embodiment, a regular grid division algorithm is used to divide the target area into a plurality of equal-area grids, and the center position of each grid is taken as an initial monitoring point to obtain an initial monitoring point set, and the grid number and geographical coordinates corresponding to each initial monitoring point are recorded. The initial monitoring point set is screened based on the spatial geographical information of the target area to remove the monitoring points in the impassable area to obtain a screening monitoring point set: a three-dimensional geographical information model of the target area is constructed by using the spatial geographical information; whether the position of each initial monitoring point is passable is determined according to the passing performance parameters of the vehicle-mounted electromagnetic detection device; wherein the passing performance parameters include: vehicle height, width, minimum turning radius, climbing ability, etc.; the initial monitoring points located in the impassable areas such as rivers, lakes, mountains, buildings, military restricted areas, and the monitoring points that cannot be reached by the vehicle-mounted electromagnetic detection device due to factors such as narrow roads and steep slopes are removed from the initial monitoring point set, and the remaining monitoring points constitute the screening monitoring point set.

[0098] In the embodiment, an electromagnetic wave propagation attenuation model is determined according to the target area, for example, the target area is a city environment, a rural environment, etc., and the model is calibrated according to parameters such as building material and height, vegetation coverage, etc.; a plurality of preset frequency bands can be set, and the propagation of signals in each frequency band in the target area is simulated by using an electromagnetic simulation software to calculate the field strength values at different positions; the field strength values are compared with a detection threshold to mark the areas with field strength values less than the detection threshold as blind areas, and a blind area distribution map is drawn, wherein the detection threshold is set according to the sensitivity of the electromagnetic environment detection device and the investigation requirement.

[0099] In the embodiment, the screening monitoring point set and the supplementary monitoring point set are combined into a target monitoring point set: the screening monitoring point set and the supplementary monitoring point set are integrated to remove duplicate monitoring points; the target monitoring point set is sorted according to geographical coordinates to generate an ordered target monitoring point list; the target monitoring point set is checked for reasonableness to check whether the monitoring points are uniformly distributed and cover key areas, and if there is anything unreasonable, manual adjustment and optimization are performed.

[0100] In an embodiment of the present application, the spatial adaptive supplementary method is used to supplement monitoring points in the blind area, including:

[0101] The blind area is divided into sub-regions, and the simulated field strength value of the center point of each sub-region is calculated;

[0102] When the simulated field strength value of a sub-region is less than the detection threshold, the sub-region is further subdivided into smaller sub-regions;

[0103] At the center point of each sub-region generated by each subdivision, the difference between the simulated field strength value and the detection threshold is calculated, where the difference is the detection threshold minus the simulated field strength value;

[0104] The center points with a difference greater than a preset difference threshold are selected as supplementary monitoring points;

[0105] The subdivision process is repeated, and the spatial density of the current supplementary monitoring point set is calculated, where the spatial density is calculated based on the number of monitoring points per unit area;

[0106] If the spatial density is less than a preset blind area monitoring density threshold, the sub-regions with a simulated field strength value less than the detection threshold are further subdivided, and the supplementary monitoring point set and the spatial density value are updated after the subdivision;

[0107] When the spatial density value reaches the blind area monitoring density threshold or the number of iterations reaches a preset maximum number of iterations, the subdivision is stopped.

[0108] In this embodiment, based on the geometry and area of the blind area, the blind area is divided into multiple sub-regions by using irregular polygon division or quadtree division algorithm. By using the pre-calibrated electromagnetic wave propagation attenuation model, combined with the parameters of terrain, ground objects and other parameters in the sub-region, the simulated field strength value of the center point of each sub-region is calculated, and the number, range, center point coordinates and simulated field strength value of each sub-region are recorded. When the simulated field strength value of the sub-region is less than the detection threshold, in order to more accurately capture the electromagnetic environment information, the sub-region is further subdivided into smaller sub-regions. The subdivision method can use equal proportion division or non-uniform division according to the complexity of the electromagnetic environment in the sub-region, and more detailed division can be performed near the electromagnetic interference source. In this embodiment, the difference between the simulated field strength value and the detection threshold is calculated at the center point of each sub-region generated by each subdivision, wherein the difference is the detection threshold minus the simulated field strength value, and the difference threshold is determined according to the electromagnetic environment detection accuracy requirement and the device measurement error. In this embodiment, the center points with a difference greater than a preset difference threshold are selected as supplementary monitoring points. The electromagnetic signals at the positions of these supplementary monitoring points are weak and need to be monitored. In order to avoid the concentration of monitoring points, the distance between the candidate points and the selected supplementary monitoring points is also considered when selecting, so as to ensure that the distance between the points is not less than twice the effective detection radius of the vehicle-mounted electromagnetic detection device.

[0109] In this embodiment, in each iteration of the iterative subdivision process, the spatial density of the current supplementary monitoring point set is calculated based on the number of monitoring points in the unit area. In this embodiment, the actual area of the sub-region and the number of monitoring points included in the sub-region are counted when calculating, and if the shape of the sub-region is irregular, the area can be estimated by using the grid approximation method. This embodiment can also establish a spatial density change curve to record the spatial density value after each iteration, so as to analyze the change trend of the monitoring point distribution.

[0110] In this embodiment, the termination step of the repeated iteration subdivision includes: if the spatial density of the current supplementary monitoring point set is less than a preset blind area monitoring density threshold, the sub-region with a simulated field strength value less than the detection threshold is further subdivided, and the supplementary monitoring point set and the spatial density value are updated after the subdivision. The blind area monitoring density threshold is determined according to the functional positioning of the target area and the electromagnetic environment management requirement, for example, a higher monitoring density threshold is set for a communication base station dense area. At the same time, in order to prevent excessive calculation caused by high electromagnetic environment complexity, a maximum iteration number is set. When the spatial density value reaches the blind area monitoring density threshold or the iteration number reaches the preset maximum iteration number, the subdivision is stopped, and the final determined supplementary monitoring point set is numbered and labeled with geographic coordinates.

[0111] In an embodiment of the present application, it further includes: after stopping the subdivision, if the iteration number reaches the preset maximum iteration number and the spatial density does not reach the blind area monitoring density threshold, the current supplementary monitoring point is output.

[0112] In an embodiment of the present application, the method for setting the preset difference threshold value comprises:

[0113] The spatial field strength gradient statistical value of the blind area is obtained by calculating the change rate of the simulated field strength values of the center points of the adjacent sub-regions in the blind area.

[0114] The preset difference threshold value is adjusted according to the spatial field strength gradient statistical value.

[0115] The preset difference threshold value is adjusted according to the spatial field strength gradient statistical value, which comprises:

[0116] When the field strength gradient statistical value is greater than the first statistical value threshold, the preset difference threshold value is reduced by a first step size.

[0117] When the field strength gradient statistical value is less than the second statistical value threshold, the preset difference threshold value is increased by a second step size; wherein the greater the field strength gradient statistical value, the more intense the electromagnetic environment change.

[0118] In the embodiment, the center point coordinate correlation table of the sub-regions in the blind area is established to clearly define the corresponding relationship of the center points of the adjacent sub-regions. The spatial field strength gradient statistical value is obtained by calculating the change rate of the simulated field strength values of the center points of the adjacent sub-regions in the blind area,

[0119] The specific calculation method is as follows: for the adjacent sub-regions A and B, the center points are PA and PB, the simulated field strength values are EA and EB, and the spatial field strength gradient G = distance(PA, PB) | EA-EB |, wherein distance(PA, PB) is the Euclidean distance between the two points.

[0120] The field strength gradients of the center points of all the adjacent sub-regions in the blind area are statistically analyzed to calculate statistical quantities such as the average value and the standard deviation, and the spatial field strength gradient statistical value is comprehensively obtained.

[0121] In the embodiment, the first statistical value threshold and the second statistical value threshold need to be set according to the electromagnetic environment characteristics of the target area, the performance parameters of the detection equipment, and the investigation accuracy requirements. For an area with a relatively stable electromagnetic environment and high equipment detection accuracy, the first statistical value threshold can be appropriately increased to reduce unnecessary monitoring point supplementation; the second statistical value threshold can be relatively reduced to avoid missing important monitoring points due to small changes in the field strength gradient. For example, in a suburban area with a gentle electromagnetic environment change, the first statistical value threshold can be set to 0.5 (unit: dB / m), and the second statistical value threshold can be set to 0.1 (unit: dB / m); while in a complex electromagnetic environment area such as the city center, the first statistical value threshold can be set to (unit: dB / m), and the second statistical value threshold can be set to 0.3 (unit: dB / m). In addition, the first statistical value threshold and the second statistical value threshold can be adjusted through historical electromagnetic environment monitoring data and actual detection experience.

[0122] In the embodiment, the determination of the first step length and the second step length needs to balance the efficiency and accuracy of the monitoring point supplement. Among them, the first step length should be set according to the change trend of the electromagnetic environment when the field strength gradient statistical value is large and the detection accuracy requirement. If the field strength gradient changes dramatically, the first step length can be set to a smaller value, for example, 0.05 (unit: dB), to more accurately capture the changes in the electromagnetic environment. The second step length is set to avoid excessive supplement of monitoring points when the field strength gradient changes slightly, and the second step length is greater than the first step length, for example, 0.8 (unit: dB). The embodiment can establish a step length adjustment mechanism to correct the first step length and the second step length according to the actual adjustment effect and the distribution of the monitoring points.

[0123] In the embodiment, the preset difference threshold value is adjusted according to the field strength gradient statistical value: when the field strength gradient statistical value is greater than the first statistical value threshold value, it indicates that the electromagnetic environment in the blind area changes dramatically. To ensure that the areas with weak electromagnetic signals can be monitored comprehensively, the preset difference threshold value is reduced by the first step length, so that more points that meet the conditions are selected as supplement monitoring points. When the field strength gradient statistical value is less than the second statistical value threshold value, it indicates that the electromagnetic environment is relatively stable. To avoid excessive density of monitoring points, the preset difference threshold value is increased by the second step length. After adjusting the preset difference threshold value each time, the sub-regions in the blind area are re-evaluated to determine whether the selection of the supplement monitoring points needs to be adjusted.

[0124] In a possible implementation, the embodiment verifies the effect after the adjustment of the preset difference threshold value is completed. The coverage and uniformity indexes of the supplement monitoring point set are calculated by simulating the distribution of the supplement monitoring points under the new preset difference threshold value. The coverage is measured by calculating the proportion of the blind area covered by the supplement monitoring points in the total blind area; the uniformity is calculated by using the distance-based uniformity evaluation method to calculate the coefficient of variation of the distance from each supplement monitoring point to its nearest neighbor. If the coverage and uniformity do not meet the preset standard, the first and second statistical value thresholds, the first step length and the second step length are re-evaluated, and the adjustment of the preset difference threshold value is performed again until the requirements are met. The preset standard is that the coverage is greater than 90% and the uniformity coefficient of variation is less than 0.3.

[0125] In an embodiment of the present application, based on the geographic location coordinates of the target monitoring point set, a preset path planning algorithm is used to determine the inspection path of the vehicle-mounted electromagnetic detection device, including:

[0126] Converting the geographic location coordinates of the target monitoring point set into latitude and longitude coordinates;

[0127] Generating feasible path nodes according to the spatial geographic information of the target area and the vehicle traffic constraint conditions;

[0128] Constructing a weighted topological network based on the target monitoring point set and the feasible path nodes;

[0129] The weighted topological network is input into the ant colony algorithm, and a path meeting the vehicle passing constraint condition is generated by driving a multi-objective optimization function; wherein the multi-objective optimization function includes: minimization of the total travel distance of the inspection path, minimization of the vehicle energy consumption estimation based on the total travel distance of the inspection path, and minimization of the variation coefficient of the distance between consecutive monitoring points on the path.

[0130] In the embodiment, the geographical position coordinates of the target monitoring point set are converted into latitude and longitude coordinates by using a Gauss-Krueger projection conversion method. Specifically, the conversion parameter equation between the plane rectangular coordinate system and the geographical coordinate system is established, the original coordinate data of the target monitoring point is projected and calculated, the coordinate conversion is realized, and the error correction of the converted latitude and longitude coordinates is performed.

[0131] In the embodiment, the spatial geographical information of the target region and the vehicle passing constraint condition are used to generate feasible path nodes, wherein the spatial geographical information includes road network data, terrain data, building distribution data, etc., and the vehicle passing constraint condition includes road height limit, width limit, forbidden period, maximum vehicle cruising range, etc.; the spatial analysis function of the geographic information system is used to perform rasterization processing on the road network of the target region, and it is judged whether it is passable in each grid unit according to the vehicle passing constraint condition. If it is passable, the center position of the grid unit is taken as a feasible path node. At the same time, obstacles are identified in the non-road region of the target region, such as rivers, lakes, mountains, etc., and the nodes in the impassable region are excluded.

[0132] In the embodiment, the weighted topological network is constructed based on the target monitoring point set and the feasible path nodes. In the construction process, the straight-line distance between nodes is calculated, and the actual length of the road, the road congestion situation, the road surface condition and other factors are combined to assign a weight value to each connection edge. The road congestion situation is dynamically updated by real-time traffic data, and the road surface condition is evaluated according to historical maintenance records and field survey data. For the road with serious damage, the weight value of the corresponding connection edge is increased. At the same time, special identification is set for the target monitoring point, so as to give priority to in path planning;

[0133] The embodiment inputs a weighted topological network into an ant colony algorithm, and drives path generation to meet vehicle passing constraint conditions by a multi-objective optimization function; the multi-objective optimization function includes: minimization of total travel distance of the inspection path, minimization of vehicle energy consumption estimation based on the total travel distance of the inspection path, and minimization of the coefficient of variation of the distance between consecutive monitoring points on the path; during the operation of the ant colony algorithm, pheromone evaporation coefficient and pheromone update rules are set to guide the ant colony to search for an optimal path in the weighted topological network; to improve the convergence speed of the algorithm, an elite strategy is used to enhance the pheromone of the historical optimal path; when calculating the vehicle energy consumption estimation, the load, engine power, travel speed and other parameters of the vehicle are considered to establish a vehicle energy consumption model; for minimization of the coefficient of variation of the distance between consecutive monitoring points on the path, the order of the path nodes is adjusted to make the monitoring points more evenly distributed, facilitating analysis and comparison of electromagnetic environment data; in addition, after the path is generated, the generated inspection path is verified for feasibility, and if the path violates the vehicle passing constraint conditions, the weight coefficients of the multi-objective optimization function are adjusted again, and the ant colony algorithm is run again until an inspection path that meets the conditions is generated.

[0134] In an embodiment of the present application, when the ant colony algorithm generates multiple inspection paths, the comprehensive score of each inspection path is calculated by a path evaluation model.

[0135] The inspection path with the highest comprehensive score is selected as the inspection path.

[0136] If the difference between the comprehensive scores of the multiple inspection paths is less than a first difference threshold, the inspection path with the minimum total travel distance among the multiple inspection paths is selected as the final inspection path.

[0137] In the embodiment, when the ant colony algorithm generates multiple inspection paths, the comprehensive score of each inspection path is calculated by a path evaluation model.

[0138] The path evaluation model analyzes multiple evaluation indexes, including: total travel distance of the path, estimated travel time, vehicle energy consumption, path reliability, and degree of satisfaction of the timeliness of the monitoring task. In the embodiment, each evaluation index is given a weight according to the actual situation of the target area and the monitoring task requirements. For example, in a monitoring task with a tight time requirement, the weight of the estimated travel time can be set to 0.3; in an energy-saving scenario, the weight of the vehicle energy consumption is set to 0.3; the weight of the total travel distance of the path is set to 0.2; the weight of the path reliability is set to 0.15; and the weight of the degree of satisfaction of the timeliness of the monitoring task is set to 0.05.

[0139] The embodiment ranks the comprehensive scores of each inspection path calculated in the selection of the inspection path with the highest comprehensive score as the final inspection path, selects the path with the highest score as the preliminary determined inspection path, and records the detailed information of the path, including the node sequence passed by the path, the driving distance of each road segment, the predicted driving time, energy consumption and other data.

[0140] In the embodiment, if the comprehensive scores of the multiple inspection paths differ by less than the first difference threshold, the inspection path with the minimum total driving distance among the multiple inspection paths is selected as the final inspection path. The first difference threshold is set according to the accuracy requirement of the path selection of the actual monitoring task. When the difference values of the comprehensive scores of the multiple candidate paths are all less than the first difference threshold, it indicates that the paths are relatively close in comprehensive performance, and at this time, the path with the minimum total driving distance is selected from the paths as the final inspection path. If there are multiple paths with the same total driving distance, the predicted driving time of the paths is further compared, and the path with the shortest predicted driving time is selected. If the predicted driving times are also the same, a path is randomly selected as the final inspection path.

[0141] In an embodiment of the present application, the electromagnetic environment monitoring data of each monitoring point is processed to obtain an electromagnetic environment data set, including:

[0142] The electromagnetic environment monitoring data of each monitoring point is subjected to data cleaning to obtain a sub-data set;

[0143] The sub-data set is subjected to time and space correlation and feature extraction to generate an electromagnetic environment feature matrix;

[0144] The electromagnetic environment feature matrix and geographic information system data of the target region are subjected to fusion processing to obtain the electromagnetic environment data set.

[0145] In the embodiment, a data cleaning rule library is established, wherein the cleaning rule library includes multiple data cleaning strategies, and corresponding cleaning rules are set for different types of electromagnetic environment monitoring data. Different types of electromagnetic environment monitoring data include electric field intensity data, magnetic field intensity data and spectrum data.

[0146] In the embodiment, for anomaly value detection, a statistical method is used to calculate the mean and standard deviation of the data, and data exceeding the range of mean ± 3 times standard deviation is determined as an anomaly value. For missing value processing, a nearest point interpolation method is used, and according to the time and space information of the monitoring points around the missing data point, multiple monitoring point data closest to the missing data point are selected, and interpolation is performed by weighted average.

[0147] In the embodiment, the sub-data set is subjected to time and space correlation and feature extraction to generate an electromagnetic environment feature matrix, including:

[0148] Based on the geographical position coordinates and time stamp of the monitoring points, the cleaned sub-data set is time and space aligned;

[0149] Based on the environmental temperature and humidity recorded by the vehicle-mounted electromagnetic detection equipment and / or the device state parameters of the vehicle-mounted electromagnetic detection equipment, the aligned sub-data set is compensated and calibrated;

[0150] Feature extraction is performed on the compensated and calibrated sub-data set to obtain electromagnetic characteristic parameters;

[0151] According to the electromagnetic characteristic parameters and the monitoring point set, an electromagnetic environment feature matrix is generated;

[0152] Specifically, the geographical position coordinates of each monitoring point are imported into a geographic information system, and the spatial positioning of the monitoring points is performed using the spatial analysis function of the GIS. At the same time, according to the time stamp information, the electromagnetic environment monitoring data in the sub-data set is sorted in chronological order. In view of the slight difference in data collection time of different monitoring points, a time interpolation method is used for unification. For example, if the collection time interval of two adjacent monitoring points is 5 minutes, and the target time interval is set to 1 minute, then through linear interpolation, new time series data is generated based on the original data. In the spatial dimension, the Euclidean distance between each monitoring point is calculated to construct a spatial distance matrix, and the spatial relationship between the monitoring points is clarified, thereby realizing the accurate alignment of the sub-data set in time and space.

[0153] The vehicle-mounted electromagnetic detection equipment of the present embodiment has a built-in temperature and humidity sensor that records environmental temperature and humidity data in real time, and the operating parameters of the equipment itself (such as sensor sensitivity, amplifier gain, and other device state parameters) are also recorded. The present embodiment can pre-establish a correlation model between environmental temperature and humidity, device state parameters, and electromagnetic signal measurement error, which is obtained by training a large amount of calibration experimental data under different environmental conditions and device states. The aligned sub-data set and its corresponding environmental temperature and humidity and device state parameters are input into the correlation model for processing to compensate and calibrate the electromagnetic signal data, eliminate the influence of environmental factors and changes in the state of the equipment itself on the measurement results, and improve the accuracy and reliability of the data.

[0154] The present embodiment uses digital signal processing technology to pre-process the compensated and calibrated electromagnetic signal data, including filtering, noise reduction, and other operations to remove interference signals. Fast Fourier transform is used to convert time-domain signals to frequency-domain signals to extract frequency components, power spectral density, and other frequency-domain features of the signals. Time-domain statistical features of the signals are calculated, including mean, variance, peak value, and kurtosis. For the waveform features of the electromagnetic signal, morphological analysis method is used to extract parameters such as rise time, fall time, and pulse width. The various time-domain and frequency-domain features extracted above are integrated to obtain a set of electromagnetic characteristic parameters that describe the characteristics of the electromagnetic signal.

[0155] The embodiment generates an electromagnetic environment feature matrix according to electromagnetic characteristic parameters and a set of monitoring points. Specifically, the monitoring points are indexed as rows, and various extracted electromagnetic characteristic parameters are indexed as columns to construct a two-dimensional matrix. The electromagnetic characteristic parameters corresponding to each monitoring point are sequentially filled into the corresponding positions in the matrix. If there are monitoring data at multiple time points, the matrix is expanded in time sequence to generate an electromagnetic environment feature matrix that can comprehensively reflect the electromagnetic environment features of the target region.

[0156] In the embodiment, the geographic information system data includes topographic data, land use data, infrastructure data, and the like of the target region. The embodiment adopts a feature-level fusion method to extract and preprocess key features in the geographic information system data, so as to match the dimensions and data formats of the electromagnetic environment feature matrix. The electromagnetic environment feature matrix and the geographic information system feature data are fused by using a data fusion algorithm. For example, according to the importance of different features to the electromagnetic environment evaluation, each feature is assigned a corresponding weight in the weighted fusion algorithm, and a fused feature vector is obtained by weighted summation. The fused feature vector is organized according to the geographical position and time sequence of the monitoring points to construct a complete electromagnetic environment data set.

[0157] In an embodiment of the present application, the electromagnetic environment data set is analyzed to obtain an evaluation result, including:

[0158] The electromagnetic environment data set is filtered based on a preset evaluation index system to obtain a first electromagnetic environment data set;

[0159] The parameters of the electromagnetic environment evaluation model are iteratively optimized by a genetic algorithm to obtain a target electromagnetic environment evaluation model;

[0160] The first electromagnetic environment data set is input into the target electromagnetic environment evaluation model to obtain an evaluation result of the target region. The electromagnetic environment evaluation model is an extreme gradient boosting algorithm, and the evaluation result is a comprehensive quality evaluation result of the electromagnetic environment of the target region.

[0161] In the embodiment, an evaluation index system is constructed according to national and industry-related electromagnetic environment standards, the functional positioning of the target area, and the actual needs of electromagnetic environment monitoring. For basic indexes such as electric field intensity and magnetic field intensity, threshold values are strictly set according to national standards. The embodiment can also add specific evaluation indexes in different functional areas according to customer needs, such as adding harmonic distortion rate, voltage fluctuation and flicker in industrial areas, and adding low-frequency electromagnetic field exposure level index in residential areas. At the same time, expert knowledge and historical monitoring data are introduced to evaluate the importance of each index, and the analytic hierarchy process is used to determine the weight of each index to construct the evaluation index system. According to the evaluation index system, the data in the electromagnetic environment data set is screened, and the data records that do not meet the standards or are invalid are removed, and the effective data related to the evaluation indexes are retained, so as to obtain a first electromagnetic environment data set, and the screened data is normalized to unify the dimension and value range of the data.

[0162] In the embodiment, the first electromagnetic environment data set is input into the electromagnetic environment evaluation model to obtain the prediction result of the model. The prediction result is analyzed and processed, the evaluation index system and the weight are combined, and the electromagnetic environment comprehensive quality evaluation score of the target area is calculated. According to the score, the electromagnetic environment quality of the target area is divided into different levels, and a detailed evaluation report is generated. The levels include excellent, good, medium, and poor, and the report content includes specific values of each evaluation index, comprehensive score, and quality level.

[0163] In an embodiment of the present application, the parameters of the electromagnetic environment evaluation model are iteratively optimized by a genetic algorithm to obtain a target electromagnetic environment evaluation model, including:

[0164] The set of to-be-optimized hyperparameters of the extreme gradient boosting algorithm is determined, and the set of to-be-optimized hyperparameters includes the number of trees, the depth of the tree, the learning rate, and the minimum loss drop threshold for splitting;

[0165] The fitness function is defined as the comprehensive evaluation error on the historical electromagnetic reference data set as the fitness evaluation index, and the comprehensive evaluation error is the weighted sum of the normalized mean square error and the spectral feature importance deviation;

[0166] The population size is determined based on the number of frequency bands of the target area;

[0167] The mutation probability is determined based on the electromagnetic field strength gradient in the electromagnetic environment data set;

[0168] The crossover probability is determined based on the spatial autocorrelation coefficient of the electromagnetic field strength of the target area;

[0169] The parameters of the electromagnetic environment evaluation model are iteratively optimized based on the fitness function, the mutation probability, and the crossover probability;

[0170] When the continuous optimal fitness change rate is less than the first change rate threshold or the maximum number of iterations is reached, the optimization is terminated, and a global optimal hyperparameter is obtained; and a target electromagnetic environment evaluation model is constructed according to the global optimal hyperparameter.

[0171] In this embodiment, the set of hyperparameters to be optimized of the extreme gradient boosting algorithm is determined, and the set of hyperparameters to be optimized includes the number of trees, the depth of the tree, the learning rate, and the minimum loss drop threshold for splitting. The value range of each hyperparameter is determined: the value range of the number of trees is set to [50, 500], which is based on historical experimental data and model complexity. A smaller number of trees will lead to underfitting, and too many trees will increase the computational cost and the risk of overfitting; the value range of the tree depth is [3, 10], a shallower tree depth helps to prevent overfitting, and a deeper tree depth can improve the fitting ability of the model to complex data; the value range of the learning rate is [0.01, 0.3], a too large learning rate will make the model training not converge, and a too small learning rate will lead to a slow training speed; the value range of the minimum loss drop threshold for splitting is [0, 1], which determines the minimum requirement for loss drop when splitting the tree node, and affects the complexity of the model.

[0172] In this embodiment, the fitness function is defined as the comprehensive evaluation error on the historical electromagnetic reference data set as the fitness evaluation index, and the comprehensive evaluation error is the weighted sum of the normalized mean square error and the spectral feature importance deviation. When calculating the normalized mean square error, for each sample in the historical electromagnetic reference data set, the difference between the model prediction value and the actual value is calculated, the difference is squared, and then the mean value is calculated, and then divided by the variance of the actual value. The spectral feature importance deviation is obtained by calculating the difference between the weight distribution of the model to the spectral feature and the ideal weight determined based on expert knowledge or historical data, and the Euclidean distance is used to measure the degree of deviation.

[0173] In this embodiment, when the number of frequency bands in the target area is small, the population size can be set to a small population size, because fewer frequency bands mean that the data features are relatively simple, and a smaller population size can reduce the amount of calculation and meet the optimization requirements; when the number of frequency bands is large, in order to cope with more complex data features, a larger population size is needed to explore a better combination of hyperparameters.

[0174] In the embodiment, when calculating the electromagnetic field intensity gradient, the finite difference method is used to calculate the gradient value of the electromagnetic field intensity data of adjacent monitoring points in the electromagnetic environment data set, and the average electromagnetic field intensity gradient of the entire data set is obtained. A function relationship between the gradient value and the mutation probability is established. When the average electromagnetic field intensity gradient value is a first gradient value, the first gradient value is small, and the mutation probability is set to a first range, for example, the first range is [0.01, 0.05]. Because a small gradient indicates that the data is relatively stable, a low mutation probability can prevent the algorithm from over-exploring and slow down the convergence. As the gradient value increases, the mutation probability increases accordingly. When the gradient value is greater than the first gradient value, the mutation probability is set to a second range, which can be [0.05, 0.1]. Because a large gradient means that the data changes dramatically, a high mutation probability helps the algorithm to jump out of the local optimum and explore better solutions.

[0175] In the embodiment, the spatial statistical method is used to calculate the spatial autocorrelation coefficient of the electromagnetic field intensity of the target area, which reflects the correlation of the electromagnetic field intensity in space. When the spatial autocorrelation coefficient is high, it indicates that the electromagnetic field intensity has strong similarity in spatial distribution, and a high crossover probability can be set at this time because it can speed up the search speed of the algorithm around the current optimal solution. When the spatial autocorrelation coefficient is low, a low crossover probability can be set at this time because a low crossover probability enables the algorithm to explore the solution space more extensively and avoids falling into a local optimum.

[0176] In the embodiment, the population is initialized, each individual is composed of a set of parameter values in the set of hyperparameters to be optimized, and the initial population is generated according to the determined population size. The selection operation is performed, the roulette selection method is used, the probability of being selected is calculated according to the fitness value of the individual, the probability of being selected by the individual with higher fitness is greater, and the selected individual enters the next generation population. The crossover operation exchanges genes of the selected individuals according to the set crossover probability to generate new individuals. The mutation operation randomly changes the genes of the individuals with the set mutation probability to introduce new genetic information. The selection, crossover and mutation operations are repeated, the fitness value of each individual in the new population is calculated every time an iteration is completed, and the optimal individual of the population is updated.

[0177] In an embodiment of the present application, the evaluation result further includes: an electromagnetic environment distribution map, interference source positioning information, a spectrum occupation report, and an electromagnetic environment evolution trend prediction.

[0178] In an embodiment of the present application, the electromagnetic environment data set is analyzed to generate an electromagnetic environment distribution map, and the method further includes:

[0179] The electromagnetic field intensity data in the electromagnetic environment data set is mapped to the raster layer of the geographic information system;

[0180] The electromagnetic field intensity data is processed by the inverse distance weighted algorithm for spatial interpolation to generate the electromagnetic environment distribution map.

[0181] In the embodiment, the electromagnetic field intensity data in the electromagnetic environment data set is mapped to the raster layer of the geographic information system; the resolution of the raster layer in the geographic information system is set according to the area of the target region and the monitoring accuracy requirement, the geographic position coordinates of each monitoring point in the electromagnetic environment data set are matched with the raster layer, the electromagnetic field intensity data of the corresponding monitoring point is assigned to the raster unit where the monitoring point is located, so as to realize the preliminary spatial mapping of the data, and the raster layer is initialized and rendered to distinguish different electromagnetic field intensity ranges by different colors. In the embodiment, the inverse distance weighted algorithm is used to perform spatial interpolation processing on the electromagnetic field intensity data to generate an electromagnetic environment distribution map. In the implementation process of the inverse distance weighted algorithm, the number of adjacent points participating in the interpolation calculation is determined, and 4-8 adjacent points are set according to the raster resolution and the monitoring point density; the distance weight index is set, and the value range is usually 2-4, which determines the influence degree of the distance on the interpolation result, and the greater the index, the higher the weight of the adjacent point. For each raster unit to be interpolated, the Euclidean distance between the raster unit and the surrounding monitoring points is calculated, the weight of each monitoring point is calculated according to the distance and the weight index, the electromagnetic field intensity estimation value of the raster unit is calculated by weighted average, the interpolation calculation is performed on all raster units, and finally a continuous electromagnetic environment distribution map is generated, which is visualized in the geographic information system, and the warmer the color of the electromagnetic environment distribution map, the higher the electromagnetic field intensity.

[0182] In an embodiment of the present application, the electromagnetic environment data set is analyzed to generate a spectrum occupation report, which further comprises:

[0183] Extracting the signal intensity and occupation time length of each frequency band in the electromagnetic environment data set;

[0184] Calculating the spectrum occupation rate and interference probability in the preset frequency band;

[0185] Generating a spectrum occupation space-time distribution report according to the spatial position.

[0186] In the embodiment, the electromagnetic environment data set is analyzed, and the signal intensity data of each frequency band at different time points is extracted according to the preset frequency band division rule; the frequency band division rule is divided according to the communication frequency band and the power frequency band and the like. In the embodiment, when the signal intensity in the frequency band is greater than the set signal intensity threshold, it is determined that the frequency band is in the occupied state, and at this time, the cumulative occupation time length of each frequency band in the monitoring time period is counted. The spectrum occupation rate calculation formula of the embodiment is: occupation rate = frequency band occupation time length / total monitoring time length x 100%. In the embodiment, the historical interference times of the frequency band, the number of times that the current signal intensity exceeds the normal range and the like are taken as the input probability statistical model to obtain the interference probability of the frequency band.

[0187] The embodiment fuses the spectrum occupancy rate and interference probability data of different spatial positions (i.e., each monitoring point) with the spatial information of the geographic information system, and makes a spectrum occupancy space-time distribution chart in the time dimension, to show the spectrum occupancy and interference risk in different regions and time periods.

[0188] In an embodiment of the present application, the analysis of the electromagnetic environment data set to generate the interference source positioning information further includes:

[0189] The electromagnetic environment data set is analyzed by using a time difference positioning algorithm to analyze multi-source signals, to identify and locate the position coordinates and signal strength of the electromagnetic interference source.

[0190] In the embodiment, a cross-correlation algorithm is used to calculate the time difference of the same interference signal received by different monitoring points, and multiple sampling average processing is performed on the time difference calculation result to improve the calculation accuracy. Based on the hyperbolic positioning principle, a hyperbolic equation set is constructed according to the time difference information of multiple monitoring points, and the position coordinate estimate value of the interference source is obtained by solving the equation set; to further optimize the positioning result, the least square method is introduced to fit and correct the position coordinates. At the same time, according to the interference signal strength received by the monitoring point and the signal propagation attenuation model, the initial signal strength of the interference source is back calculated, and the position coordinates and signal strength information of the interference source are finally determined, and the position of the interference source is marked in the geographic information system, to generate an interference source positioning report.

[0191] In an embodiment of the present application, the analysis of the electromagnetic environment data set to obtain the electromagnetic environment evolution trend prediction further includes:

[0192] The electromagnetic environment data set is input into a machine learning model to predict the electromagnetic environment change trend in a specified future time period; wherein the machine learning model includes a long short-term memory neural network model.

[0193] In the embodiment, the electromagnetic environment data set is preprocessed, the data is divided into a training set, a validation set and a test set, the training set data is input into the long short-term memory neural network model for training, in the training process, the mean square error is used as the loss function, and the Adam optimizer is used to update the model parameters; the model is evaluated by using the validation set, and the hyperparameters are adjusted to prevent model overfitting. When the model training converges, the test set data is input into the model for prediction, to obtain the electromagnetic environment parameter prediction value in a specified future time period; the prediction result is displayed in the form of a chart, combined with historical data and current trend analysis and interpretation, to generate an electromagnetic environment evolution trend prediction report.

[0194] In an embodiment of the present application, the electromagnetic environment investigation and evaluation method based on the vehicle-mounted electromagnetic detection device further comprises obtaining a comprehensive index based on the electromagnetic environment distribution map, the interference source positioning information, the spectrum occupation report, the electromagnetic environment evolution trend prediction and the preset comprehensive index system, wherein the lower the comprehensive index, the better the electromagnetic environment of the target region.

[0195] Specifically, the comprehensive index system is shown in Table 1,

[0196] Table 1, comprehensive index system

[0197]

[0198] In a specific embodiment, it is assumed that we need to perform electromagnetic environment investigation and evaluation on a new district (target region) of a city, wherein the new district has an area of 30 square kilometers and includes commercial areas, residential areas and science and technology parks and other regions.

[0199] I. Determine the target monitoring point set

[0200] According to the area of the new district and the preset spatial resolution threshold, the new district is spatially gridded to generate an initial monitoring point set. Then, based on the spatial geographic information of the new district, the monitoring points located in the non-passable regions such as rivers and parks are removed to obtain a screened monitoring point set.

[0201] Using an electromagnetic wave propagation attenuation model, the field strength distribution of the preset frequency band signal in the new district is simulated, and blind areas with field strength values less than the detection threshold are identified. These blind areas are divided into sub-regions, and the simulated field strength values of the center points of the sub-regions are calculated. When it is found that the simulated field strength values of some sub-regions are less than the detection threshold, these sub-regions are further subdivided. At the center points of the sub-regions generated in each subdivision, the difference between the simulated field strength value and the detection threshold is calculated, and the center points with a difference greater than a preset difference threshold adjusted by obtaining the spatial field strength gradient statistical value of the blind area are selected as the supplemented monitoring points.

[0202] In the iteration process, the preset difference threshold is adjusted according to the field strength gradient statistical value. When the field strength gradient statistical value is greater than a first statistical value threshold, the preset difference threshold is reduced by a first step size; when it is less than a second statistical value threshold, the preset difference threshold is increased by a second step size; wherein the first step size is greater than the second step size. The subdivision process is repeatedly iterated, and the spatial density of the current supplemented monitoring point set is calculated. If the spatial density is less than a preset blind area monitoring density threshold, the subdivision is continued. The supplemented monitoring points in the blind area are finally determined, and a supplemented monitoring point set is obtained. The screened monitoring point set and the supplemented monitoring point set are combined, and the duplicate monitoring points are removed to obtain a target monitoring point set.

[0203] II. Determine the patrol path

[0204] Coordinate conversion and network construction: converting the geographic position coordinates of the target monitoring point set into latitude and longitude coordinates, generating feasible path nodes according to the spatial geographic information of the new area and the vehicle passing constraint conditions, and constructing a weighted topological network based on the target monitoring point set and the feasible path nodes.

[0205] The weighted topological network is input into the ant colony algorithm, and a multi-objective optimization function containing minimization of the total driving distance of the inspection path, minimization of the vehicle energy consumption estimation based on the total driving distance of the inspection path, and minimization of the variation coefficient of the distance between consecutive monitoring points on the path is used to generate the inspection path. When the inspection path is multiple, the comprehensive score of each inspection path is calculated through the path evaluation model. Since the comprehensive scores of two inspection paths differ by less than a first difference threshold, the one with the smallest total driving distance is selected as the final inspection path.

[0206] III. Collecting and processing data

[0207] The vehicle loaded with electromagnetic environment detection equipment is controlled to drive along the finally determined inspection path, and the monitoring points in the target monitoring point set are respectively collected electromagnetic environment monitoring data.

[0208] The electromagnetic environment monitoring data of each monitoring point is subjected to data cleaning to remove outliers and noise, obtaining a sub-data set. The sub-data set is subjected to time and space correlation and feature extraction to generate an electromagnetic environment feature matrix. The electromagnetic environment feature matrix and the geographic information system data of the new area are subjected to fusion processing to obtain a complete electromagnetic environment data set.

[0209] IV. Analyzing data to obtain evaluation results

[0210] Based on the preset evaluation index system, the electromagnetic environment data set is filtered to obtain a first electromagnetic environment data set. The global optimal hyperparameters of the extreme gradient boosting algorithm are determined through the electromagnetic environment data set and the genetic algorithm, and the target electromagnetic environment evaluation model is constructed according to the global optimal hyperparameters.

[0211] The first electromagnetic environment data set is input into the target electromagnetic environment evaluation model, and finally the comprehensive quality evaluation result of the electromagnetic environment of the new area of the city is obtained. The evaluation result shows the level of the overall electromagnetic environment of the new area and the information of electromagnetic signal interference, etc.

[0212] Further, the evaluation result also includes: electromagnetic environment distribution map, interference source positioning information, spectrum occupation report and electromagnetic environment evolution trend prediction.

[0213] Further, based on the electromagnetic environment distribution map, the interference source positioning information, the spectrum occupation report and the electromagnetic environment evolution trend prediction, and the preset comprehensive index system, a comprehensive index is obtained, wherein the lower the comprehensive index, the better the electromagnetic environment of the target area.

[0214] The electromagnetic environment investigation and evaluation method based on the vehicle-mounted electromagnetic detection device corresponding to the above embodiment, Figure 2 A structural block diagram of an electromagnetic environment investigation and evaluation system based on a vehicle-mounted electromagnetic detection device is provided for an embodiment of the present application. For ease of illustration, only parts related to the embodiments of the present application are shown. Referring to Figure 2 An electromagnetic environment investigation and evaluation system 20 based on a vehicle-mounted electromagnetic detection device includes a data processing module 21, a path planning module 22, a collection control module 23, a collection data processing module 24, and a data analysis module 25.

[0215] The data processing module 21 is configured to determine a target monitoring point set of a target region and its corresponding geographical position coordinates based on the spatial geographical information of the target region.

[0216] The path planning module 22 is configured to determine a patrol path of the vehicle-mounted electromagnetic detection device based on the geographical position coordinates of the target monitoring point set using a preset path planning algorithm.

[0217] The collection control module 23 is configured to control the vehicle loaded with the electromagnetic environment detection device to collect electromagnetic environment monitoring data at each monitoring point in the target monitoring point set when the vehicle travels along the patrol path.

[0218] The collection data processing module 24 is configured to process the electromagnetic environment monitoring data of each monitoring point to obtain an electromagnetic environment data set.

[0219] The data analysis module 25 is configured to analyze the electromagnetic environment data set to obtain an electromagnetic environment evaluation result of the target region.

[0220] Referring to Figure 3 , Figure 3 A schematic block diagram of an electronic device is provided for an embodiment of the present application. As shown in the embodiment of the present application, Figure 3 The electronic device 300 can include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The above-mentioned processors 301, input devices 302, output devices 303, and memories 304 complete mutual communication through a communication bus 305. The memory 304 is used to store a computer program, and the computer program includes program instructions. The processor 301 is used to execute the program instructions stored in the memory 304. The processor 301 is configured to invoke the program instructions to execute the functions of each module in the above-mentioned device embodiments, for example Figure 2 The functions of the data processing module 21, the path planning module 22, the collection control module 23, the collection data processing module 24, and the data analysis module 25 shown in the above-mentioned device embodiments.

[0221] It should be appreciated that in the embodiments of the present application, the processor 301 can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic components, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0222] The input device 302 can include a touchpad, a fingerprint collection sensor (for collecting fingerprint information and direction information of a fingerprint of a user), a microphone, etc., and the output device 303 can include a display (LCD, etc.), a speaker, etc.

[0223] The memory 304 can include a read-only memory and a random access memory, and provide instructions and data for the processor 301. A portion of the memory 304 can also include a non-volatile random access memory. For example, the memory 304 can also store device type information.

[0224] In specific implementations, the processor 301, the input device 302, and the output device 303 described in the embodiments of the present application can execute any implementation described in the embodiments of the electromagnetic environment investigation and evaluation system based on the vehicle-mounted electromagnetic detection device provided by the embodiments of the present application, and can also execute the implementation of the electronic device described in the embodiments of the present application, which will not be described here.

[0225] In another embodiment of the present application, a computer readable storage medium is provided, which stores a computer program. The computer program includes program instructions, which, when executed by a processor, implement all or part of the processes of the above-mentioned embodiment methods. The computer program can also instruct related hardware to complete the implementation. The computer program can be stored in a computer readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various method embodiments can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, executable files or some intermediate form. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0226] The computer readable storage medium can be an internal storage unit of the electronic device of any of the preceding embodiments, such as a hard disk or a memory of the electronic device. The computer readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. Further, the computer readable storage medium can include both the internal storage unit and the external storage device of the electronic device. The computer readable storage medium is used to store the computer program and other programs and data required by the electronic device. The computer readable storage medium can also be used to temporarily store data that has been output or will be output.

[0227] Those skilled in the art can appreciate that the units and algorithm steps of the examples described in conjunction with the embodiments disclosed herein can be realized in electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been described in general terms in the above description. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0228] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the electronic device and the units described above can refer to the corresponding processes in the above-mentioned method embodiments, which will not be described here.

[0229] In several embodiments provided in the present application, it should be understood that the disclosed electronic device and method can be implemented in other manners. For example, the embodiments of the apparatus described above are merely schematic; the division of the units is merely logical function division; an actual implementation can be divided into different units depending on actual conditions; or a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections can be indirect couplings or communication connections through some interfaces, or can be in electrical, mechanical or other forms.

[0230] The units described as separated components can or can not be physically separated, and the components displayed as units can or can not be physical units, i.e., can be located in one place, or can be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purposes of the embodiments of the present application.

[0231] In addition, each functional unit in the embodiments of the present application can be integrated in one processing unit, or each unit can exist physically as a separate unit, or two or more units can be integrated in one unit. The integrated unit can be implemented in the form of hardware, or in the form of a software functional unit.

[0232] The above is merely specific embodiments of the present application, but the protection scope of the present application is not limited thereto; any skilled person in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method for investigating and assessing the electromagnetic environment based on vehicle-mounted electromagnetic detection equipment, characterized in that, include: Based on the spatial geographic information of the target area, determine the set of target monitoring points in the target area and their corresponding geographical coordinates; Based on the geographical coordinates of the target monitoring point set, the inspection path of the vehicle-mounted electromagnetic detection equipment is determined using a preset path planning algorithm. When controlling a vehicle equipped with electromagnetic environment detection equipment to travel along the inspection path, electromagnetic environment monitoring data is collected at each monitoring point in the target monitoring point set. The electromagnetic environment monitoring data from each monitoring point are processed to obtain an electromagnetic environment dataset. The electromagnetic environment dataset is analyzed to obtain the electromagnetic environment assessment results for the target area; The determination of the target monitoring point set for the target area based on the spatial geographic information of the target area includes: The target area is divided into spatial grids based on the target area and a preset spatial resolution threshold to generate an initial set of monitoring points. Based on the spatial geographic information of the target area, the initial set of monitoring points is filtered, and monitoring points in areas where passage is not possible are removed to obtain a filtered set of monitoring points. Based on the electromagnetic wave propagation attenuation model, the field strength distribution of a preset frequency band signal in the target area is simulated to obtain the blind zone where the field strength value is less than the detection threshold. Based on the spatial adaptive supplementation method, monitoring points are added in the blind zone to obtain a supplemented monitoring point set; Combine the selected monitoring point set and the supplementary monitoring point set into a target monitoring point set; The spatial adaptive supplementation method supplements monitoring points within the blind zone, including: The blind zone is divided into sub-regions, and the simulated field strength value at the center point of each sub-region is calculated. When the simulated field strength value of a sub-region is less than the detection threshold, the sub-region is further subdivided into smaller sub-regions. At the center point of each sub-region generated by subdivision, the difference between the simulated field strength value and the detection threshold is calculated, wherein the difference is the detection threshold minus the simulated field strength value; The center point where the difference is greater than the preset difference threshold is selected as the supplementary monitoring point; Repeat the iterative subdivision process and calculate the spatial density of the current supplementary monitoring point set, which is based on the number of monitoring points per unit area; If the spatial density is less than the preset blind zone monitoring density threshold, the sub-regions with simulated field strength values ​​less than the detection threshold are further subdivided, and the monitoring point set and spatial density value are updated and supplemented after subdivision. When the spatial density value reaches the blind zone monitoring density threshold or the number of iterations reaches the preset maximum number of iterations, the subdivision stops.

2. The electromagnetic environment investigation and assessment method based on vehicle-mounted electromagnetic detection equipment according to claim 1, characterized in that, The method for setting the preset difference threshold includes: The spatial field strength gradient statistics of the blind zone are obtained by calculating the rate of change of the simulated field strength values ​​at the center points of adjacent sub-regions within the blind zone. Adjust the preset difference threshold based on the statistical value of the spatial field strength gradient; The step of adjusting the preset difference threshold based on the spatial field strength gradient statistics includes: When the spatial field strength gradient statistical value is greater than the first statistical value threshold, the preset difference threshold is reduced by a step size. When the spatial field strength gradient statistical value is less than the second statistical value threshold, the preset difference threshold is increased by a second step.

3. The electromagnetic environment investigation and assessment method based on vehicle-mounted electromagnetic detection equipment according to claim 1, characterized in that, The process of determining the inspection path of the vehicle-mounted electromagnetic detection equipment based on the geographical coordinates of the target monitoring point set and using a preset path planning algorithm includes: Convert the geographical coordinates of the target monitoring point set into latitude and longitude coordinates; Based on the spatial geographic information of the target area and vehicle traffic constraints, feasible path nodes are generated. A weighted topology network is constructed based on the target monitoring point set and the feasible path nodes; The weighted topology network is input into the ant colony algorithm to drive the path generation of an inspection path that satisfies the vehicle traffic constraints using a multi-objective optimization function. The multi-objective optimization function includes minimizing the total travel distance of the inspection path, minimizing the vehicle energy consumption estimate based on the total travel distance of the inspection path, and minimizing the coefficient of variation of the distance between continuous monitoring points on the path.

4. The electromagnetic environment investigation and assessment method based on vehicle-mounted electromagnetic detection equipment according to claim 3, characterized in that, Also includes: When the ant colony algorithm generates multiple inspection paths, the comprehensive score of each inspection path is calculated through the path evaluation model. The inspection route with the highest overall score will be selected as the final inspection route. If the overall score difference among multiple inspection routes is less than the first difference threshold, then the inspection route with the smallest total driving distance among the multiple inspection routes is selected as the final inspection route.

5. The electromagnetic environment investigation and assessment method based on vehicle-mounted electromagnetic detection equipment according to claim 1, characterized in that, The electromagnetic environment monitoring data from each monitoring point is processed to obtain an electromagnetic environment dataset, including: The electromagnetic environment monitoring data from each monitoring point were cleaned to obtain a subset of data. The subset datasets are correlated temporally and spatially and feature extracted to generate an electromagnetic environment feature matrix. The electromagnetic environment feature matrix and the geographic information system data of the target area are fused to obtain the electromagnetic environment dataset.

6. The electromagnetic environment investigation and assessment method based on vehicle-mounted electromagnetic detection equipment according to claim 1, characterized in that, The analysis of the electromagnetic environment dataset to obtain the electromagnetic environment assessment results for the target area includes: Based on a pre-defined evaluation index system, the electromagnetic environment dataset is filtered to obtain the first electromagnetic environment dataset. The parameters of the electromagnetic environment assessment model are iteratively optimized using a genetic algorithm to obtain the target electromagnetic environment assessment model. The first electromagnetic environment dataset is input into the target electromagnetic environment assessment model to obtain the assessment result of the target area; wherein, the electromagnetic environment assessment model is an extreme gradient boosting algorithm, and the assessment result is the electromagnetic environment assessment result of the target area.

7. The electromagnetic environment investigation and assessment method based on vehicle-mounted electromagnetic detection equipment according to claim 6, characterized in that, The step of iteratively optimizing the parameters of the electromagnetic environment assessment model using a genetic algorithm to obtain the target electromagnetic environment assessment model includes: Determine the set of hyperparameters to be optimized for the extreme gradient boosting algorithm, including the number of trees, tree depth, learning rate, and minimum loss descent threshold for splitting; The fitness function is defined as the comprehensive evaluation error on the historical electromagnetic reference dataset as the fitness evaluation index, wherein the comprehensive evaluation error is the weighted sum of the normalized mean square error and the importance deviation of the spectral features; The population size is determined based on the number of frequency bands in the target region; The mutation probability is determined based on the electromagnetic field strength gradient in the electromagnetic environment dataset. The crossover probability is determined based on the spatial autocorrelation coefficient of the electromagnetic field strength in the target region. The parameters of the electromagnetic environment assessment model are iteratively optimized based on the fitness function, mutation probability, and crossover probability. The optimization is terminated when the rate of change of the continuous optimal fitness is less than the first rate of change threshold or the maximum number of iterations is reached, and the global optimal hyperparameters are obtained; the target electromagnetic environment assessment model is constructed based on the global optimal hyperparameters.

8. An electromagnetic environment survey and assessment system based on vehicle-mounted electromagnetic detection equipment, characterized in that, include: The data processing module is used to determine the set of target monitoring points and their corresponding geographical coordinates in the target area based on the spatial geographic information of the target area. The determination of the target monitoring point set for the target area based on the spatial geographic information of the target area includes: The target area is divided into spatial grids based on the target area and a preset spatial resolution threshold to generate an initial set of monitoring points. Based on the spatial geographic information of the target area, the initial set of monitoring points is filtered, and monitoring points in areas where passage is not possible are removed to obtain a filtered set of monitoring points. Based on the electromagnetic wave propagation attenuation model, the field strength distribution of a preset frequency band signal in the target area is simulated to obtain the blind zone where the field strength value is less than the detection threshold. Based on the spatial adaptive supplementation method, monitoring points are added in the blind zone to obtain a supplemented monitoring point set; Combine the selected monitoring point set and the supplementary monitoring point set into a target monitoring point set; The path planning module is used to determine the inspection path of the vehicle-mounted electromagnetic detection equipment based on the geographical coordinates of the target monitoring point set and using a preset path planning algorithm. The spatial adaptive supplementation method supplements monitoring points within the blind zone, including: The blind zone is divided into sub-regions, and the simulated field strength value at the center point of each sub-region is calculated. When the simulated field strength value of a sub-region is less than the detection threshold, the sub-region is further subdivided into smaller sub-regions. At the center point of each sub-region generated by subdivision, the difference between the simulated field strength value and the detection threshold is calculated, wherein the difference is the detection threshold minus the simulated field strength value; The center point where the difference is greater than the preset difference threshold is selected as the supplementary monitoring point; Repeat the iterative subdivision process and calculate the spatial density of the current supplementary monitoring point set, which is based on the number of monitoring points per unit area; If the spatial density is less than the preset blind zone monitoring density threshold, the sub-regions with simulated field strength values ​​less than the detection threshold are further subdivided, and the monitoring point set and spatial density value are updated and supplemented after subdivision. When the spatial density value reaches the blind zone monitoring density threshold or the number of iterations reaches the preset maximum number of iterations, the subdivision stops. The data acquisition and control module is used to control the vehicle equipped with electromagnetic environment detection equipment to collect electromagnetic environment monitoring data at each monitoring point in the target monitoring point set when it travels along the inspection path. The data acquisition and processing module is used to process the electromagnetic environment monitoring data of each monitoring point to obtain an electromagnetic environment dataset. The data analysis module is used to analyze the electromagnetic environment dataset to obtain the electromagnetic environment assessment results for the target area.

Images