Multi-radioactive-source air intelligent searching and positioning method

Abstract

The invention provides a multi-radioactive-source air intelligent searching and positioning method, which comprises the following steps of: 1, constructing an earth surface radioactivity distribution diagram based on known radioactive sources under the condition of known multi-radioactive-source nuclide types and source intensities; 2, training a dose rate distribution diagram at the 1m position of the radioactive source earth surface based on a convolutional neural network; 3, performing full-coverage search monitoring on the to-be-searched area; 4, generating a dose rate distribution diagram at the 1m position of the earth surface of the to-be-searched area; 5, analyzing the dose rate distribution diagram based on a convolutional neural network; and 6, accurately measuring the position of the radioactive source. Under the condition of unknown multi-radioactive source nuclide type and source intensity, the method comprises the following steps: A, carrying out full-coverage search monitoring on a to-be-searched area; b, clearly optimizing a monitoring area based on the dose rate distribution diagram; c, performing automatic searching based on the optimized monitoring area; and D, determining the specific position of the radioactive source based on a parameter estimation method. According to the invention, the air searching and positioning of multiple radioactive sources can be accurately determined, and the searching efficiency is maximized.

Description

Technical Field

[0001] The invention belongs to the technical field of nuclear safety, and specifically is a multi-radioactive source search and positioning method based on an intelligent optimization algorithm. Background Art

[0002] With the rapid development of nuclear and radiation technology, radioactive sources are increasingly used in industry, agriculture, scientific research and other fields. Various types of radioactive sources are used in medical diagnosis, industrial flaw detection, container inspection, irradiation breeding, food preservation and other related industries. As the number of radioactive sources used in various industries increases, although the awareness of nuclear safety and the management system of radioactive sources have developed significantly, the number of uncontrolled radioactive sources may increase. Once a radioactive source is out of control, it will pose a great threat and hazard to social public safety. When a radioactive source is lost, it is often very difficult to find and recover it. Due to the large number of possible areas involved and the large search area, the traditional carpet search by people carrying portable monitoring instruments may take several days or even longer, which is extremely inefficient. The lost radioactive source is very dangerous. The longer the search time, the greater the radiation hazard it produces. Rapid and efficient searches should be carried out based on unmanned aerial radiation monitoring platforms. Under the constraints of meeting the search needs of radioactive sources and the indicators of aerial emergency monitoring equipment, intelligent optimization algorithms should be used to scientifically plan the search path of radioactive sources. Summary of the invention

[0003] The purpose of the present invention is to provide a fast, intelligent and efficient search method for an unmanned aerial radiation monitoring platform after the radioactive source is lost. The unmanned aerial radiation monitoring platform is equipped with a radiation dose rate detector, a navigation and positioning system, an intelligent computing system, a communication system, etc., and uses an intelligent search method to realize the work of fast and efficient automatic search for radioactive sources. The present invention proposes an aerial radioactive source search method based on task planning. On the basis of estimating the scope of the radioactive source search task, the types and source strengths of known and unknown radioactive source nuclides, and on the basis of fully considering relevant constraints, relevant optimization algorithms and models are constructed to optimize the path of the radioactive source search for the unmanned aerial radiation monitoring platform, maximize the guarantee efficiency of the unmanned aerial radiation monitoring platform, shorten the task time limit, and maximize the search efficiency.

[0004] The invention is a method for searching and locating based on an unmanned aerial radiation monitoring platform and according to two conditions, namely, types of nuclides and source strength of known and unknown radioactive sources.

[0005] Assuming that the types and source strengths of multiple radioactive sources are known, the specific steps for searching and locating multiple radioactive sources are as follows:

[0006] Step 1: Construct a surface radioactivity distribution map based on known radioactive sources

[0007] For a given radionuclide, according to the basic theory of gamma radiation, the exposure rate at a given distance in the air can be expressed as:

[0008]

[0009] Where: is the exposure rate at a certain point in the air; A is the activity of the point source; r is the distance between the measurement point of the aerial nuclear radiation detector and the radiation source; μ is the linear attenuation coefficient of air; Γ is the γ irradiation rate constant, which represents the exposure rate of a radioactive source of unit activity at unit distance.

[0010] Based on the above radioactivity distribution formula, combined with parameters such as the type of radionuclides, activity, and height of nuclear radiation detectors of each radioactive source, a preliminary surface radioactivity distribution map of each radioactive source can be generated.

[0011] In air, under the same conditions, the relationship between the absorbed dose rate and the irradiation rate can be expressed as:

[0012]

[0013] Where: It is the absorbed dose rate at a certain point in the air, the unit is: uGy / h; It is the radiation rate at a certain point in the air, the unit is: R / h.

[0014] When the type and activity distribution of radioactive nuclides are known, the dose rate distribution diagram at 1m above the ground can be obtained.

[0015] Step 2: Training the dose rate distribution map 1m above the surface of the radiation source based on the convolutional neural network

[0016] For radioactive sources of different types and activities of different nuclides, various surface dose rate distribution maps are generated based on the formula in step 1; then, based on the intelligent computing system of the aerial monitoring platform, the generated surface dose rate distribution maps of each radioactive source are input into the two-dimensional convolutional neural network from three channels for network training according to the three features of RGB. The two-dimensional convolutional neural network (2D-CNN) includes three structures: convolution layer, pooling layer, and fully connected layer. The convolution layer extracts low-level features such as edges, lines, and corners, and the higher-level convolution layer extracts higher-level features. The convolution kernel slides on the input image in sequence, from left to right and from top to bottom. Each time it slides, the convolution performs a dot product calculation on the input image corresponding to its sliding window position and obtains a value, which retains the spatial characteristics of the input image. The pooling layer is a method of compressing images, mainly to accelerate the convergence process of the neural network and improve the stability during training. In the CNN structure, after multiple convolution layers and sampling layers, one or more fully connected layers are connected, and each neuron in the fully connected layer is fully connected with all neurons in the previous layer. The fully connected layer can integrate the local information with category distinction in the convolutional layer or sampling layer. Through the feature extraction (color, edge, line, texture) of the 2D-CNN layer, the weight pooling layer is used to disperse the deep network to avoid overfitting. After being processed by the fully connected layer, the dose rate distribution map of each radiation source at 1m on the surface is given through the output layer, so that the neural network has the ability to identify the dose rate distribution map of the corresponding radiation source at 1m on the surface.

[0017] Step 3: Conduct full coverage search and monitoring of the search area

[0018] The area to be searched is topologically divided into a grid area, and the search is started from the boundary of the area to be searched based on the aerial radiation monitoring platform. In the initial stage, the monitoring path is based on the "Z" type for coarse grid aerial flight nuclear radiation monitoring. During the monitoring process, the platform maintains a high flight speed at an altitude of 50 meters for monitoring. During the monitoring stage, when the aerial radiation monitoring platform detects that the dose rate value exceeds the set threshold (0.5uGy / h) during the search process, the monitoring platform reduces the flight speed to below 10 meters / second and monitors at 1 / 2 (or 1 / 3) of the original flight spacing; when the value is lower than the set threshold, the original flight speed and flight spacing are restored, and the cycle is repeated until the area to be searched is completed. During the monitoring process, the aerial radiation monitoring platform can obtain the distance from any monitoring point to a certain radiation source based on the formula in step one. Combined with the flight altitude of the monitoring platform, the dose rate value of the monitoring point at 1 meter on the equivalent surface can be inferred.

[0019] Step 4: Generate a dose rate distribution map at 1m above the surface of the area to be searched

[0020] When an aerial radiation monitoring platform performs regional flight monitoring, the nuclear radiation detectors it carries usually update the ground dose rate monitoring data once at regular intervals (usually at a frequency of 1 time per second). The data displays the longitude, latitude, altitude and dose rate value of the current monitoring point.

[0021] When the aerial radiation monitoring platform completes monitoring of the area to be searched, it will generate the location of each monitoring point on the monitoring route and the corresponding equivalent dose rate at 1m on the surface, and preliminarily generate a dose rate distribution map at 1m on the surface of the area to be searched in combination with the GIS map. In order to make the dose rate distribution map more accurate, especially the dose rate distribution map near the radiation source, the linear interpolation method can be used to generate the dose rate of each point between the two monitoring lines, and then generate the dose rate distribution map at 1m on the surface of the area to be searched. Given the positions of the two corresponding points on the two monitoring routes (x1, y1) (x2, y2) and the dose rate values ​​d1 and d2, the linear interpolation formula can be expressed as:

[0022]

[0023] Where: is the dose rate value at any point on the two monitoring routes, is the position of any point between two monitoring routes.

[0024] Step 5: Analyze the dose rate situation map based on convolutional neural network

[0025] The color, edge, line, texture and other features of the dose rate distribution map at 1 m on the surface of each suspected radiation source area are input into the trained convolutional neural network. Combined with the color, edge, line, texture and other features of the dose rate distribution map at 1 m on the surface of each known radiation source, the dose rate distribution map of each suspected radiation source area is automatically analyzed and compared to preliminarily determine the type of each radiation source. Based on the known GIS map and the location of the dose rate distribution map at 1 m on the surface, the location of each radiation source can also be preliminarily determined.

[0026] Step 6: Accurately determine the location of the radiation source

[0027] At the initially determined location of each radiation source (x', y'), the aerial radiation monitoring platform hovers for 30 seconds to monitor in order to accurately determine the dose rate value D1 at 1m above the ground at that point; then, a "loop"-shaped monitoring with a side length of 50 meters is carried out with the point (x', y') as the center point, and the platform hovers for 30 seconds at the four vertices of the "loop".

[0028] In a point source radiation field, assuming that air attenuation is ignored and the background count rate is a constant, the count rate of an aerial nuclear radiation detector can be determined according to the following formula.

[0029]

[0030] In the formula, (x i ,y i ) is the position of the airborne nuclear radiation detector, (x0, y0) is the position of the radiation source, M is the counting rate of the airborne nuclear radiation detector at a certain point in the air, A is the activity of the radiation source, ε is the efficiency of the nuclear radiation detector, r i is the distance between the aerial nuclear radiation detector and the radiation source, b is the background count rate, and h is the monitoring height of the aerial nuclear radiation detector.

[0031] Since the detector count at the measuring point is proportional to the intensity of the radiation source, the following set of equations can be formed based on the measured data of the four monitoring points.

[0032]

[0033] Solving the above equations, we can get multiple solutions of the radioactive source P1 = (x 01 ,y 01 ),P2=(x 02 ,y 02 ),P3=(x 03 ,y 03 ), P4=(x 04 ,y 04 ), and then the exact location of the radiation source can be obtained.

[0034] x0=(x 01 +x 02 +x 03 +x 04 ) / 4

[0035] y0=(y 01 +yx 02 +y 03 +y 04 ) / 4

[0036] Assuming that the nuclide types and source strengths of multiple radioactive sources are unknown, the specific steps for searching and locating multiple radioactive sources are as follows:

[0037] Step 1: Conduct a full coverage search and monitoring of the search area

[0038] The area to be searched is topologically divided into a grid area, and the search is started from the boundary of the area to be searched based on the aerial radiation monitoring platform. In the initial stage, the monitoring path is based on the "Z" type for coarse grid (two parallel lines are kept at a long distance, such as 300 meters) aerial flight nuclear radiation monitoring. During the monitoring process, the platform maintains a high flight speed (such as 50 kilometers per hour) at an altitude of 50 meters for monitoring. During the monitoring stage, when the aerial radiation monitoring platform detects that the dose rate value exceeds the set threshold (0.2uGy / h) during the search process, the monitoring platform reduces the flight speed to below 10 meters / second and monitors at 1 / 2 (1 / 3) of the original flight spacing; when the value is detected to be lower than the set threshold, the original flight speed and flight spacing are restored, and the cycle is repeated until the area to be searched is completed.

[0039] Step 2: Optimize monitoring area based on dose rate distribution map

[0040] Based on the dose rate values ​​at each location monitored by the aerial radiation monitoring platform, the interpolation method is used to generate the dose rate distribution map at 1m above the ground for each radiation source in the search area. Based on the dose rate distribution map, the approximate location of each radiation source can be preliminarily determined. In the preliminarily determined radiation source areas, the areas that exceed the set threshold are rasterized, and the length and width of each cell are set to 20 meters respectively.

[0041] Step 3: Automatic search based on optimized monitoring area

[0042] In each optimized monitoring area, monitoring starts from the first cell (1,1), and the monitoring mode is automatically searched one by one at a flight altitude of 20 meters. That is, the aerial monitoring platform hovers in the middle of each cell for 30 seconds to measure once. When the first column of cells are monitored one by one, when the dose rate of a cell is found to change from high to low, it returns from the current cell to the previous cell (x', 1); then the cells in a row corresponding to the cell are monitored one by one. When the dose rate of a cell in the row is found to change from high to low, it returns from the current cell to the previous cell in the row (x', y').

[0043] Step 4: Determine the specific location of the radiation source based on parameter estimation

[0044] Take the center point of the cell (x', y') as the "cross" monitoring point, hover and monitor for 30 seconds every 5 meters on the horizontal and vertical lines of the "cross", and select 8 monitoring points in total; from these monitoring points, select the two with the largest dose rate values, and then use the parameter estimation method to automatically determine the exact location of the radiation source. The specific formula is as follows:

[0045]

[0046] Where: D iis the dose rate of the airborne radiation monitoring platform (uGy / h); a is the background radiation constant; c is the constant determined by the radiation source; (x0, y0) is the position of the radiation source; r i is the distance between the aerial nuclear radiation detector and the radiation source.

[0047] The advantages and beneficial effects of the present invention are: the method proposed by the present invention can accurately determine the location of multiple radioactive sources in the air. The present invention can optimize the path of radioactive source search for unmanned aerial radiation monitoring platforms, maximize the security efficiency of unmanned aerial radiation monitoring platforms, shorten the mission time limit, maximize the search efficiency, and meet the needs of rapid search and location of aerial radioactive sources. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly describes the drawings required for use in the embodiments or the prior art descriptions. The drawings described below are only some embodiments of the present invention, and for those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work.

[0049] Figure 1 The present invention is a flow chart of a known multi-radioactive source aerial intelligent search and positioning method.

[0050] Figure 2 It is a flow chart of the method for intelligent aerial search and positioning of unknown multiple radioactive sources.

[0051] Figure 3 This is a schematic diagram of the estimated surface radioactivity distribution of multiple radioactive sources.

[0052] Figure 4 This is a schematic diagram of the estimated dose rate distribution at 1m above the surface for multiple radiation sources.

[0053] Figure 5 This is a schematic diagram of full coverage search and monitoring of the area to be searched.

[0054] Figure 6 This is a schematic diagram of the dose rate distribution 1m above the surface of each radioactive source in the area to be searched.

[0055] Figure 7 It is a schematic diagram of automatic search for regional optimization monitoring.

[0056] Figure 8 It is a "cross" shaped monitoring diagram. DETAILED DESCRIPTION

[0057] The implementation data of the present invention comes from the comprehensive application practice of the unmanned aerial radiation monitoring platform. According to the actual needs of the platform for performing aerial radiation source search tasks, two situations are distinguished, namely, known and unknown radioactive source nuclide types and source strengths. For the situation where the types and source strengths of multiple radioactive source nuclides are known, a surface radioactivity distribution map is constructed based on the known radioactive sources, and based on the generated distribution map, a dose rate distribution map at 1m on the surface of the radioactive source is trained based on a convolutional neural network; an unmanned aerial radiation monitoring platform is used to conduct full coverage search and monitoring of the search area, and a dose rate distribution map at 1m on the surface of the search area is generated; a comparative analysis of the dose rate distribution map is performed based on a convolutional neural network; and based on the counting rate calculation formula, local "return" type monitoring is used to accurately measure the position of the radioactive source.

[0058] For the case of unknown types and source strengths of multiple radioactive sources, full coverage search and monitoring of the search area is carried out based on the "Z"-shaped path; the optimized monitoring area is clearly defined based on the dose rate distribution map; the optimized monitoring area is automatically searched; the specific location of the radioactive source is determined based on the parameter estimation method

[0059] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be systematically described below with the drawings in the embodiments. The embodiments described in the present invention are only part of the embodiments of the present invention, rather than all the embodiments.

[0060] Figure 1 , Figure 2 The flowchart of the multi-radioactive source aerial intelligent search method of the present invention is shown below. The implementation steps are specifically described by taking an unmanned aerial radiation monitoring platform to search and locate multiple radioactive sources in the air as an example. Assuming that the types and source intensities of the multiple radioactive sources are known, the specific steps for searching and locating multiple radioactive sources are as follows:

[0061] Step 1: Preliminary generation of radioactive plume diffusion trends based on the diffusion model

[0062] Assume that the lost radioactive source is 60 Co. 137 Cs, the activities of point sources are 2.2×10 13 Bq, 2.1×10 13 Bq, the irradiation rate constant is 8.9×10 -2 C·m 2 kg -1 , 8.9×10 -2 C·m 2 kg -1 The linear attenuation coefficients of the nuclides are 0.69 and 0.7. According to the basic theory of γ radiation, the exposure rates of the two radioactive sources at different distances in the air can be obtained. The surface radioactivity distribution of the two radioactive sources is shown in Figure 3 shown.

[0063] In air, under the same conditions, the relationship between the absorbed dose rate and the irradiation rate can be expressed as:

[0064]

[0065] When the type and activity distribution of radioactive nuclides are known, the dose rate distribution at 1m above the ground can be obtained, such as Figure 4 shown.

[0066] Step 2: Training the dose rate distribution map 1m above the surface of the radiation source based on the convolutional neural network

[0067] For the generated radioactive source 60 Co. 137 The dose rate distribution map of Cs at 1m above the ground is based on the intelligent computing system of the aerial monitoring platform. According to the RGB color and edge, line, texture and other features, the two-dimensional convolutional neural network is input from different channels for network training. After training through the convolution layer, pooling layer, and fully connected layer of the two-dimensional convolutional neural network, the neural network has the ability to identify radiation sources. 60 Co. 137 Ability to map the dose rate at 1m above the surface of Cs.

[0068] Step 3: Conduct full coverage search and monitoring of the search area

[0069] The area to be searched is topologically divided into a grid area, and the search is started from the boundary of the area to be searched based on the aerial radiation monitoring platform. In the initial stage, the monitoring path is based on the "Z" type for coarse grid (the two monitoring lines are kept 300 meters apart) aerial flight nuclear radiation monitoring. During the monitoring process, the platform maintains a high flight speed at an altitude of 50 meters for monitoring. During the monitoring stage, when the aerial radiation monitoring platform detects that the dose rate value exceeds the set threshold (0.5uGy / h) during the search process, the monitoring platform reduces the flight speed to below 10 meters / second and monitors at a flight interval of 100 meters; when the monitored value is lower than the set threshold (0.5uGy / h), the original flight speed and flight interval are restored, and the cycle is repeated until the area to be searched is monitored. Figure 5 During the monitoring process, the aerial radiation monitoring platform can obtain the distance between any monitoring point and a certain radiation source according to the formula in step 1. Combined with the flight altitude of the monitoring platform, the dose rate value of the monitoring point at 1 meter above the ground can be inferred.

[0070] Step 4: Generate a dose rate distribution map at 1m above the surface of the area to be searched

[0071] When the aerial radiation monitoring platform completes monitoring of the area to be searched, it will generate the location of each monitoring point on the monitoring route and the corresponding equivalent dose rate at 1m on the surface, and preliminarily generate a dose rate distribution map at 1m on the surface of the area to be searched in combination with the GIS map.

[0072] The dose rate of each point between the two monitoring lines is generated by linear interpolation, and then the dose rate distribution map at 1m on the surface of the area to be searched is generated, such as Figure 6 As shown. Given the positions of the two corresponding points (x1, y1) (x2, y2) of the two monitoring routes and the dose rate values ​​d1 and d2, the linear interpolation formula can be expressed as:

[0073]

[0074] Step 5: Analyze the dose rate distribution map based on convolutional neural network

[0075] The color, edge, line, texture and other features of the dose rate distribution map at 1 m on the surface of each suspected radiation source area are input into the trained convolutional neural network. Combined with the color, edge, line, texture and other features of the dose rate distribution map at 1 m on the surface of each known radiation source, the dose rate distribution map of each suspected radiation source area is automatically analyzed and compared to preliminarily determine the type of each radiation source. Based on the known GIS map and the location of the dose rate distribution map at 1 m on the surface, the location of each radiation source can also be preliminarily determined.

[0076] Step 6: Accurately determine the location of the radiation source

[0077] Assume that the radiation source is initially identified 60 The coordinate position of Co is (x', y'). The aerial radiation monitoring platform hovers there for 30 seconds to accurately determine the dose rate value D1 at 1m above the ground. Then, a "loop"-shaped monitoring with a side length of 50 meters is carried out with the point (x', y') as the center point, and the platform hovers at the four vertices of the "loop" for 30 seconds. 60 The activity of the Co source is 2.2×10 13 Bq, assuming that air attenuation is ignored, the background count rate is 100CPS, the nuclear radiation detector efficiency is 1, the coordinates of the four "U"-shaped hovering monitoring points are (28500, 30450), (28500, 30500), (28450, 30500), (28450, 30450), and their count rates are 9800CPS, 9650CPS, 9500CPS, and 9700CPS respectively. The monitoring height of the aerial nuclear radiation detector is 50 meters. Substitute them into the following equations respectively.

[0078]

[0079] Multiple coordinate positions of the radiation source can be solved (28479, 30470), (28477, 30469), (28480, 30471) (28481, 30474), and then the accurate position of the radiation source can be obtained (28479, 30471).

[0080] Assuming that the nuclide types and source strengths of multiple radioactive sources are unknown, the specific steps for searching and locating multiple radioactive sources are as follows:

[0081] Step 1: Conduct a full coverage search and monitoring of the search area

[0082] The area to be searched is topologically divided into a grid area, and the search is started from the boundary of the area to be searched based on the aerial radiation monitoring platform. In the initial stage, the monitoring path is based on the "Z" type for coarse grid (the interval between two monitoring lines is 300 meters) aerial flight nuclear radiation monitoring. During the monitoring process, the platform performs coverage monitoring of the area to be searched at a flight altitude of 50 meters and a flight speed of 50 kilometers per hour (for example). During the monitoring process, when the aerial radiation monitoring platform detects that the dose rate value at 1 meter above the ground exceeds the set threshold, the flight speed of the monitoring platform is reduced to below 10 meters per second, and monitoring is carried out at a flight interval of 100 meters; when the dose rate value at 1 meter above the ground is detected to be lower than the set threshold again, it is restored to the initial flight speed and flight interval, and the cycle is repeated until the area to be searched is monitored.

[0083] Step 2: Optimize monitoring area based on dose rate distribution map

[0084] Based on the dose rate values ​​at each location monitored by the aerial radiation monitoring platform, the interpolation method is used to generate a dose rate distribution map at 1m on the surface of the area to be searched. Based on the surface dose rate distribution map, the approximate location of each radiation source can be preliminarily determined. In the preliminarily determined radiation source area, the area exceeding the set 1uGy / h is rasterized, and the length and width of each cell are set to 20 meters respectively.

[0085] Step 3: Automatic search based on optimized monitoring area

[0086] In each optimized monitoring area, monitoring starts from the first cell (1,1), and the monitoring mode is automatically searched one by one at a flight altitude of 20 meters. That is, the aerial monitoring platform hovers in the middle of each cell for 30 seconds to measure once. When the first column of cells are monitored one by one, when the dose rate of the 5th cell is found to change from high to low, it returns from the current cell to the previous cell (4,1); then the cells in the 4th row corresponding to the cell are monitored one by one. When the dose rate of the 6th cell in the row is found to change from high to low, it returns from the current cell to the previous cell in the row (4,5), such as Figure 7 shown.

[0087] Step 4: Determine the specific location of the radiation source based on parameter estimation

[0088] A "cross" monitoring is performed with the center point of the cell (4,5) as the center, such as Figure 8 As shown. A total of 8 monitoring points were selected by hovering and monitoring every 5 meters for 30 seconds on the "X"-shaped horizontal and vertical lines; from these monitoring points, the two with the largest dose rate values ​​(-5, 0, 12 cGy / h) and (0, 5, 13 cGy / h) were selected, and then the parameter estimation method was used to determine that the exact position of the radiation source was (-2.1, 3.2) in the cell (4, 5).

[0089] Although the above describes the illustrative specific embodiments of the present invention to facilitate those skilled in the art to understand the present invention, and it should be clear that the present invention is not limited to the scope of the specific embodiments, for those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations using the concept of the present invention are protected.

Claims

1. A multi-radiation source aerial intelligent search and positioning method, characterized by: Assuming that the types and source intensities of multiple radioactive sources are known, the steps for searching and locating multiple radioactive sources are as follows: Step 1: Construct a surface radioactivity distribution map based on known radioactive sources; including: estimating the exposure rate generated by the radioactive source at a given distance in the air and generating a dose rate distribution map of the radioactive source at 1m on the surface; Step 2: Training the dose rate distribution map at 1m from the surface of the radiation source based on the convolutional neural network; including: training the generated dose rate distribution map of each radiation source surface for the neural network and enabling the neural network to identify the dose rate distribution map at 1m from the surface of the radiation source; Step 3: Conduct full coverage search and monitoring of the search area, including: coarse grid aerial nuclear radiation monitoring based on the "Z" pattern, low-speed and narrow-distance monitoring by the monitoring platform when the set threshold is exceeded, and restoration of the original flight speed and flight distance by the monitoring platform when the set threshold is lower than the set threshold; Step 4: Generate a dose rate distribution map at 1m on the surface of the area to be searched; including: using the interpolation method to generate a dose rate distribution map at 1m on the surface of each radioactive source in the area to be searched based on the dose rate values ​​of each position obtained by monitoring; Step 5: Analyze the dose rate distribution map based on the convolutional neural network; including: inputting the distribution map into the trained convolutional neural network to achieve type recognition and location determination; Step 6: Accurately determine the location of the radiation source; including: performing "return" monitoring of the initial location of each radiation source and obtaining the accurate location of the radiation source based on the counting rate calculation formula.

2. The multi-radiation source aerial intelligent search and positioning method according to claim 1, characterized in that: In step 1, for a given radionuclide, based on the basic theory of gamma radiation, the exposure rate produced at a given distance in air is expressed as: Where: is the exposure rate at a certain point in the air; A is the activity of the point source; r is the distance between the measuring point of the aerial nuclear radiation detector and the radiation source; μ is the linear attenuation coefficient of air; Γ is the γ irradiation rate constant, which represents the exposure rate of a radioactive source of unit activity at a unit distance; In air, under the same conditions, the relationship between the absorbed dose rate and the irradiation rate is expressed as: Where: It is the absorbed dose rate at a certain point in the air, the unit is: uGy / h; It is the radiation rate at a certain point in the air, the unit is: R / h.

3. The multi-radiation source aerial intelligent search and positioning method according to claim 1, characterized in that: In step 2, the generated surface dose rate distribution maps of each radiation source are input into the two-dimensional convolutional neural network from three channels according to the three RGB features for network training; the two-dimensional convolutional neural network 2D-CNN includes three structures: convolution layer, pooling layer, and fully connected layer; the convolution layer extracts features, and the convolution kernel slides on the input image in sequence, from left to right and from top to bottom. Each time it slides, the convolution performs a dot product calculation on the input image corresponding to its sliding window position and obtains a value, which retains the spatial characteristics of the input image; the pooling layer compresses the image, accelerates the convergence process of the neural network and improves the stability during training; in the CNN structure, after multiple convolution layers and sampling layers, it is connected to one or more fully connected layers, and each neuron in the fully connected layer is fully connected to all neurons in the previous layer; the fully connected layer integrates local information with category distinction in the convolution layer or sampling layer.

4. The multi-radiation source aerial intelligent search and positioning method according to claim 1, characterized in that: In step 3, the area to be searched is topologically divided into a grid area, and the search is started from the boundary of the area to be searched based on the aerial radiation monitoring platform; in the initial stage, the monitoring path is based on the "Z" type to conduct coarse grid aerial flight nuclear radiation monitoring, and the platform flies at an altitude of 50 meters for monitoring during the monitoring process; in the monitoring stage, when the aerial radiation monitoring platform detects that the dose rate value exceeds the set threshold during the search process, the monitoring platform reduces the flight speed to below 10 meters / second and monitors at 1 / 2 or 1 / 3 of the original flight spacing; when the monitored value is lower than the set threshold, the original flight speed and flight spacing are restored, and the cycle is repeated until the area to be searched is monitored.

5. The multi-radiation source aerial intelligent search and positioning method according to claim 1, characterized in that: In step 4, when the aerial radiation monitoring platform performs regional flight monitoring, the nuclear radiation detector on board updates the ground dose rate monitoring data at regular intervals, and the data displays the longitude, latitude, altitude and dose rate value of the current monitoring point; When the aerial radiation monitoring platform completes monitoring of the area to be searched, it will generate the position of each monitoring point on the monitoring route and the corresponding equivalent dose rate at 1m on the surface, and preliminarily generate a dose rate distribution map at 1m on the surface of the area to be searched in combination with the GIS map; the dose rate of each point between the two monitoring lines is generated by linear interpolation, and then the dose rate distribution map at 1m on the surface of the area to be searched is generated; given the positions (x1, y1) (x2, y2) of the corresponding two points on the two monitoring routes and the dose rate values ​​d1 and d2, the linear interpolation formula is expressed as: Where: is the dose rate value at any point on the two monitoring routes, is the position of any point between two monitoring routes.

6. The multi-radiation source aerial intelligent search and positioning method according to claim 1, characterized in that: In step 6, at each of the initially determined radiation source locations (x', y'), the aerial radiation monitoring platform hovers for 30 seconds to measure the dose rate value D1 at 1m above the ground surface at that point; then, a "loop"-shaped monitoring with a side length of 50 meters is performed with the point (x', y') as the center point, and the platform hovers for 30 seconds at the four vertices of the "loop"-shaped monitoring; In the point source radiation field, air attenuation is ignored, the background count rate is a constant, and the count rate of the airborne nuclear radiation detector is determined according to the following formula; In the formula, (x i ,y i ) is the position of the airborne nuclear radiation detector, (x0, y0) is the position of the radiation source, M is the counting rate of the airborne nuclear radiation detector at a certain point in the air, A is the activity of the radiation source, ε is the efficiency of the nuclear radiation detector, r i is the distance between the aerial nuclear radiation detector and the radiation source, b is the background count rate, and h is the monitoring height of the aerial nuclear radiation detector; Since the detector count at the measurement point is proportional to the intensity of the radiation source, the following set of equations is formed based on the data measured at the four monitoring points; Solving the above equations, we can obtain multiple solutions of the radioactive source P1 = (x 01 ,y 01 ),P2=(x 02 ,y 02 ),P3=(x 03 ,y 03 ), P4=(x 04 ,y 04 ) to obtain the exact location of the radiation source; x0=(x 01 +x 02 +x 03 +x 04 ) / 4 y0=(y 01 +yx 02 +y 03 +y 04 ) / 4。 7. A multi-radiation source aerial intelligent search and positioning method, characterized by: Assuming the nuclide types and source intensities of multiple radioactive sources are unknown, the steps for searching and locating multiple radioactive sources are as follows: Step 1: Conduct full coverage search and monitoring of the search area; Step 2: Clearly optimize the monitoring area based on the dose rate distribution map; Step 3: Automatic search based on optimized monitoring area; Step 4: Determine the specific location of the radiation source based on parameter estimation method.

8. The multi-radiation source aerial intelligent search and positioning method according to claim 7, characterized in that: In step one, the area to be searched is topologically divided into a grid area, and the search is started from the boundary of the area to be searched based on the aerial radiation monitoring platform; in the initial stage, the monitoring path is based on the "Z" shape to conduct coarse grid aerial flight nuclear radiation monitoring, and the platform flies at an altitude of 50 meters for monitoring during the monitoring process; in the monitoring stage, when the aerial radiation monitoring platform detects that the dose rate value exceeds the set threshold during the search process, the monitoring platform reduces the flight speed to below 10 meters / second and monitors at 1 / 2 or 1 / 3 of the original flight spacing; when the monitored value is lower than the set threshold, the original flight speed and flight spacing are restored, and the cycle is repeated until the area to be searched is monitored.

9. The multi-radiation source aerial intelligent search and positioning method according to claim 7, characterized in that: In step two, based on the dose rate values ​​at each location obtained by the aerial radiation monitoring platform, the interpolation method is used to generate a dose rate distribution map at 1 m from the surface of each radiation source in the search area; based on the dose rate distribution map, the approximate location of each radiation source is preliminarily determined; in the preliminarily determined radiation source areas, the areas that exceed the set threshold are rasterized, and the length and width of each cell are set to 20 meters respectively.

10. The multi-radiation source aerial intelligent search and positioning method according to claim 7, characterized in that: In step three, in each optimized monitoring area, monitoring starts from the first cell (1,1), and the monitoring mode is automatically searched one by one at a flight altitude of 20 meters; that is: the aerial monitoring platform hovers in the middle of each cell for 30 seconds to measure once, and when the first column of cells are monitored one by one, when it is found that the dose rate of a cell changes from high to low, it returns from the current cell to the previous cell (x', 1); then the cells in a row corresponding to the cell are monitored one by one, and when it is found that the dose rate of a cell in the row changes from high to low, it returns from the current cell to the previous cell in the row (x', y').

11. The multi-radiation source aerial intelligent search and positioning method according to claim 7, characterized in that: In step 4, the center point of the cell (x', y') is used for "cross" monitoring. The monitoring is performed every 5 meters on the horizontal and vertical lines of the "cross" for 30 seconds, and a total of 8 monitoring points are selected; from these monitoring points, the two with the largest dose rate values ​​are selected, and then the parameter estimation method is used to automatically determine the exact location of the radiation source; the specific formula is as follows: Where: D i is the dose rate of the airborne radiation monitoring platform (uGy / h); a is the background radiation constant; c is the constant determined by the radiation source; (x0, y0) is the position of the radiation source; r i is the distance between the aerial nuclear radiation detector and the radiation source.

Images