Radioactive element on-site intelligent monitoring system based on ai

Abstract

The application relates to the technical field of radioactivity monitoring, and discloses an AI artificial intelligence-based radioactivity element on-site intelligent monitoring system, which comprises a vertical mounting support, a detection assembly and a control box, the detection assembly comprises a radioactivity detection module and an environmental parameter monitoring module, and the control box is internally provided with a data acquisition control module, a background self-adaptive analysis module and a communication and display module. The data acquisition control module collects and pre-processes radioactivity radiation data and environmental parameter data, generates a radioactivity anomaly judgment result through residual error analysis, and the communication and display module displays and uploads the monitoring result. The temperature parameter, the humidity parameter and the air pressure parameter are collected through the environmental parameter monitoring module, and corresponding virtual background radiation values are generated based on the environmental parameters, so that a dynamic correlation analysis process between the environmental parameters and the background radiation is realized.

Description

Technical Field

[0001] This invention relates to the field of radioactivity monitoring technology, and in particular to an AI-based intelligent on-site monitoring system for radioactive elements. Background Technology

[0002] On-site monitoring of radioactive elements is an important part of nuclear safety monitoring, environmental radiation monitoring, and industrial radioactivity detection. On-site monitoring equipment is usually used to continuously collect radioactive radiation levels in the monitoring area and identify radioactive anomalies based on the collection results.

[0003] In existing technologies, radioactive monitoring equipment typically uses fixed background radiation reference values ​​or fixed anomaly detection thresholds to analyze and process radioactive radiation data. This involves comparing real-time acquired radioactive radiation data with preset background radiation values, and generating an alarm when the real-time acquired radioactive radiation data exceeds the corresponding threshold. Some radioactive monitoring systems also perform filtering or time-window statistical processing on the acquired radiation data to reduce the impact of random fluctuations on the monitoring results.

[0004] However, in the process of realizing the technical solution of this application, the inventors of this application discovered that the background radiation reference value in the prior art usually adopts a fixed parameter setting method and does not dynamically adjust it in combination with the changes in environmental parameters such as temperature, humidity and air pressure, which leads to insufficient data correspondence between the background radiation analysis process and the actual environmental conditions. Summary of the Invention

[0005] To overcome the above shortcomings, this invention provides an AI-based intelligent on-site monitoring system for radioactive elements. This system aims to improve the existing technology, which does not dynamically adjust for changes in environmental parameters such as temperature, humidity, and air pressure, resulting in insufficient data correspondence between the background radiation analysis process and the actual environmental conditions.

[0006] This invention provides the following technical solution: an AI-based intelligent on-site monitoring system for radioactive elements, comprising:

[0007] A vertical mounting bracket is provided, on which a detection component and a control box are mounted; the detection component includes a radioactivity detection module and an environmental parameter monitoring module; the control box contains a data acquisition and control module, a background adaptive analysis module, and a communication and display module, and a display panel is provided on one side of the control box;

[0008] The data acquisition and control module is used to acquire the radioactive radiation data output by the radioactivity detection module and the environmental parameter data output by the environmental parameter monitoring module, and to perform digital conversion and filtering on the radioactive radiation data to generate real-time environmental characteristic data.

[0009] The background adaptive analysis module is used to acquire historical environmental parameter data and historical radioactive radiation data, and to establish a background adaptive generation model based on the historical environmental parameter data and historical radioactive radiation data. Then, real-time environmental feature data is input into the background adaptive generation model to generate corresponding virtual background radiation values, and the background adaptive generation model is updated based on the newly added environmental parameter data and newly added radioactive radiation data.

[0010] The background adaptive analysis module is also used to acquire real-time radioactive radiation data and virtual background radiation values, and generate residual data based on the difference between real-time radioactive radiation data and virtual background radiation values. Then, a radioactive anomaly judgment result is generated based on the comparison result between the residual data and a preset anomaly judgment threshold.

[0011] The communication and display module is used to display and upload the results of the radioactive anomaly determination.

[0012] Preferably, in the data acquisition and control module, the steps of digitizing and filtering the radioactive radiation data include:

[0013] The analog radiation signal output by the radioactivity detection module is processed by analog-to-digital conversion.

[0014] The digital radiation signal after analog-to-digital conversion is subjected to sliding window filtering.

[0015] Real-time radioactive radiation data is generated based on the filtered digital radiation signal.

[0016] Preferably, in the data acquisition and control module, the step of generating real-time environmental feature data includes:

[0017] The temperature, humidity, and air pressure parameters collected by the environmental parameter monitoring module are collected synchronously.

[0018] Time-correlation processing is performed on the synchronously acquired temperature, humidity, and air pressure parameters.

[0019] Real-time environmental feature data is constructed based on time-corresponding temperature, humidity, and air pressure parameters.

[0020] Preferably, in the background adaptive analysis module, the step of establishing a background adaptive generation model based on historical environmental parameter data and historical radioactive radiation data includes:

[0021] Acquire historical environmental parameter data and historical radioactive radiation data;

[0022] Sample correlation processing was performed based on historical environmental parameter data and historical radioactive radiation data.

[0023] Feature mapping training is performed based on the results of sample association processing.

[0024] A model of the correspondence between environmental parameters and background radiation is established based on the training results of feature mapping;

[0025] The correspondence model is stored as a background adaptive generation model.

[0026] Preferably, in the background adaptive analysis module, the step of generating the corresponding virtual background radiation value includes:

[0027] Acquire real-time environmental feature data;

[0028] Normalize the real-time environmental feature data;

[0029] The normalized real-time environmental feature data is input into the background adaptive generation model;

[0030] The virtual background radiation value is output based on the background adaptive generation model.

[0031] Preferably, in the background adaptive analysis module, the step of generating residual data based on the difference between real-time radioactive radiation data and virtual background radiation values ​​includes:

[0032] To acquire real-time radioactive radiation data and virtual background radiation values;

[0033] Residual data is generated based on the difference between real-time radioactive radiation data and virtual background radiation values;

[0034] The residual data is statistically processed based on a continuous time window.

[0035] Preferably, in the background adaptive analysis module, the step of generating a radioactive anomaly determination result based on the comparison result between the residual data and the preset anomaly determination threshold includes:

[0036] Obtain window statistics results of residual data;

[0037] Standardized residual indices are generated based on the window mean and window standard deviation of residual data.

[0038] The standardized residual index is compared with the preset anomaly detection threshold;

[0039] The results of the comparison are used to generate a determination of radioactive anomalies.

[0040] Preferably, in the background adaptive analysis module, the step of updating the background adaptive generation model includes:

[0041] Acquire new environmental parameter data and new radioactive radiation data;

[0042] Incremental training samples were constructed based on newly added environmental parameter data and newly added radioactive radiation data.

[0043] The parameters of the baseline adaptive generation model are updated based on the incremental training samples.

[0044] Preferably, in the communication and display module, the step of displaying and uploading the radioactive anomaly determination result includes:

[0045] Acquire real-time radioactive radiation data, virtual background radiation values, and results of radioactive anomaly assessment;

[0046] The display panel shows real-time radioactive radiation data, virtual background radiation values, and results of radioactive anomaly detection.

[0047] The real-time radioactive radiation data, virtual background radiation values, and radioactive anomaly determination results are uploaded to the remote monitoring platform via the wireless communication unit.

[0048] Preferably, the vertical mounting bracket is a column structure fixed to the ground; the detection component is disposed on the top of the vertical mounting bracket; the control box is disposed in the middle of the vertical mounting bracket; and the display panel is disposed on the outer surface of the control box.

[0049] The present invention has the following beneficial effects:

[0050] 1. This invention establishes an environmental parameter monitoring module, a data acquisition and control module, and a background adaptive analysis module. It utilizes temperature, humidity, and air pressure parameters to construct real-time environmental characteristic data and generates corresponding virtual background radiation values ​​based on the real-time environmental characteristic data. This allows the background radiation reference value to be dynamically adjusted as environmental parameters change, thereby realizing the correlation analysis process between environmental parameters and background radiation.

[0051] 2. This invention acquires real-time radioactive radiation data and virtual background radiation values ​​through a background adaptive analysis module, and generates residual data based on the difference between the real-time radioactive radiation data and the virtual background radiation values. Then, it performs statistical processing on the residual data based on a continuous time window to generate a radioactive anomaly determination result, thereby realizing a radioactive anomaly analysis process based on the residual change process.

[0052] 3. This invention utilizes the data connection between the data acquisition and control module, the background adaptive analysis module, and the communication and display module to collect, analyze, display, and upload radioactive radiation data, virtual background radiation values, and radioactive anomaly determination results. This enables a data transmission process between on-site monitoring data and the remote monitoring platform, thereby achieving remote synchronous monitoring of radioactive data. Attached Figure Description

[0053] Figure 1 This is a schematic diagram of the system framework of the AI-based intelligent on-site monitoring system for radioactive elements according to the present invention.

[0054] Figure 2 This is a schematic diagram of the elevation of the monitoring device of the present invention;

[0055] Figure 3 This is a schematic diagram of the intelligent monitoring process of the present invention.

[0056] In the diagram: 1. Radioactivity detection module; 2. Environmental parameter monitoring module; 3. Data acquisition and control module; 4. Background adaptive analysis module; 5. Communication and display module; 6. Vertical mounting bracket; 7. Detection component; 8. Control box; 9. Display panel. Detailed Implementation

[0057] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] Reference Figures 1-3 This invention provides an AI-based intelligent on-site monitoring system for radioactive elements, comprising:

[0059] A vertical mounting bracket 6 is provided, on which a detection component 7 and a control box 8 are mounted; the detection component 7 includes a radioactivity detection module 1 and an environmental parameter monitoring module 2; the control box 8 is provided with a data acquisition and control module 3, a background adaptive analysis module 4, and a communication and display module 5.

[0060] The data acquisition and control module 3 is used to acquire the radioactive radiation data output by the radioactive detection module 1 and the environmental parameter data output by the environmental parameter monitoring module 2, and to perform digital conversion and filtering on the radioactive radiation data to generate real-time environmental characteristic data.

[0061] The background adaptive analysis module 4 is used to acquire historical environmental parameter data and historical radioactive radiation data, and to establish a background adaptive generation model based on the historical environmental parameter data and historical radioactive radiation data. Then, real-time environmental feature data is input into the background adaptive generation model to generate the corresponding virtual background radiation value, and the background adaptive generation model is updated based on the newly added environmental parameter data and newly added radioactive radiation data.

[0062] The background adaptive analysis module 4 is also used to acquire real-time radioactive radiation data and virtual background radiation values, and generate residual data based on the difference between real-time radioactive radiation data and virtual background radiation values. Then, it generates radioactive anomaly judgment results based on the comparison results between the residual data and the preset anomaly judgment threshold.

[0063] The communication and display module 5 is used to display and upload the results of the radioactive anomaly determination.

[0064] Specifically, the vertical mounting bracket 6 is a column structure fixed to the ground, the detection component 7 is located at the top of the vertical mounting bracket 6, the control box 8 is located in the middle of the vertical mounting bracket 6, and the display panel 9 is located on the outer surface of the control box 8.

[0065] The radioactivity detection module 1 is used to collect radioactive rays within the monitoring area and output the corresponding radioactive radiation data. The environmental parameter monitoring module 2 is used to collect temperature, humidity, and air pressure parameters within the monitoring area and output the corresponding environmental parameter data.

[0066] The data acquisition and control module 3 receives the radioactive radiation data output by the radioactive detection module 1 and the environmental parameter data output by the environmental parameter monitoring module 2. It performs analog-to-digital conversion, filtering, and time-correspondence processing on the radioactive radiation data, and performs synchronous acquisition and normalization processing on the environmental parameter data to form corresponding real-time environmental feature data.

[0067] The data acquisition and control module 3 sends real-time environmental characteristic data to the background adaptive analysis module 4.

[0068] Background adaptive analysis module 4 receives historical environmental parameter data and historical radioactive radiation data, and performs sample association processing and feature mapping training based on the historical environmental parameter data and historical radioactive radiation data to establish a correspondence model between environmental parameters and background radiation.

[0069] The background adaptive analysis module 4 receives real-time environmental characteristic data sent by the data acquisition and control module 3, and inputs the real-time environmental characteristic data into the background adaptive generation model to generate the corresponding virtual background radiation value. Then, the background adaptive analysis module 4 acquires real-time radioactive radiation data and the virtual background radiation value, and generates corresponding residual data based on the difference between the real-time radioactive radiation data and the virtual background radiation value.

[0070] The background adaptive analysis module 4 performs statistical processing on the residual data based on a continuous time window, and generates standardized residual indices based on the statistical processing results. These standardized residual indices are then compared with a preset anomaly detection threshold to generate a radioactive anomaly detection result. The background adaptive analysis module 4 also receives newly added environmental parameter data and newly added radioactive radiation data, and constructs incremental training samples based on these data to update the parameters of the background adaptive generation model.

[0071] The communication and display module 5 receives real-time radioactive radiation data, virtual background radiation values, and radioactive anomaly determination results, and displays these data on the display panel 9. Simultaneously, it uploads the real-time radioactive radiation data, virtual background radiation values, and radioactive anomaly determination results to the remote monitoring platform via the wireless communication unit.

[0072] Furthermore, in the data acquisition and control module 3, the steps for digitizing and filtering the radioactive radiation data include:

[0073] The analog radiation signal output by the radioactivity detection module 1 is processed by analog-to-digital conversion.

[0074] The digital radiation signal after analog-to-digital conversion is subjected to sliding window filtering.

[0075] Real-time radioactive radiation data is generated based on the filtered digital radiation signal.

[0076] In the data acquisition and control module 3, the steps for generating real-time environmental characteristic data include:

[0077] The temperature, humidity, and air pressure parameters collected by the environmental parameter monitoring module 2 are collected synchronously.

[0078] Time-correlation processing is performed on the synchronously acquired temperature, humidity, and air pressure parameters.

[0079] Real-time environmental feature data is constructed based on time-corresponding temperature, humidity, and air pressure parameters.

[0080] Specifically, the analog radiation signal output by the radioactivity detection module 1 is a continuously changing voltage signal. The analog-to-digital conversion unit in the data acquisition and control module 3 samples the analog radiation signal according to a preset sampling period and converts the analog radiation signal into a digital radiation signal.

[0081] Let the original digital radiation signal corresponding to the t-th sampling time be: ;in, Indicates the first The raw digital radiation signal corresponds to each sampling time. The data acquisition and control module 3 performs sliding window filtering on the raw digital radiation signal to reduce the impact of random pulse fluctuations on real-time radioactive radiation data.

[0082] Let the length of the sliding window be... Then the filtered real-time radioactive radiation data is:

[0083] ;

[0084] in, Indicates the first Real-time radioactive radiation data corresponding to each sampling time. Indicates the length of the sliding window. Indicates the first The original digital radiation signal corresponding to each sampling time. The environmental parameter monitoring module 2 synchronously collects temperature, humidity and air pressure parameters within the monitoring area and generates corresponding environmental parameter data.

[0085] Let the first The environmental parameter vector corresponding to each sampling time is: ;in, Indicates the first The environmental parameter vector corresponding to each sampling time. Indicates temperature parameter, Indicates humidity parameter, This represents the air pressure parameter. The data acquisition and control module 3 performs time-correlation processing on the environmental parameter vector to ensure that the environmental parameter data and the real-time radioactive radiation data correspond to the same sampling time. Then, the time-correlation environmental parameter vector is normalized to generate real-time environmental feature data.

[0086] Let the normalized real-time environmental feature data be:

[0087] ;

[0088] in, This represents the normalized environmental parameter data. Indicates the first The mean value corresponding to the class environment parameters. Indicates the first The standard deviation of the class environment parameters.

[0089] The data acquisition and control module 3 sends real-time radioactive radiation data and real-time environmental characteristic data to the background adaptive analysis module 4, so that the real-time radioactive radiation data and real-time environmental characteristic data form a corresponding data input relationship for subsequent virtual background radiation value generation and radioactive anomaly judgment processing.

[0090] Furthermore, in the background adaptive analysis module 4, the steps for establishing a background adaptive generation model based on historical environmental parameter data and historical radioactive radiation data include:

[0091] Acquire historical environmental parameter data and historical radioactive radiation data;

[0092] Sample correlation processing was performed based on historical environmental parameter data and historical radioactive radiation data.

[0093] Feature mapping training is performed based on the results of sample association processing.

[0094] A model of the correspondence between environmental parameters and background radiation is established based on the training results of feature mapping;

[0095] The correspondence model is stored as the background adaptive generation model.

[0096] In the background adaptive analysis module 4, the steps for generating the corresponding virtual background radiation value include:

[0097] Acquire real-time environmental feature data;

[0098] Normalize the real-time environmental feature data;

[0099] The normalized real-time environmental feature data is input into the background adaptive generation model;

[0100] The virtual background radiation value is output based on the background adaptive generation model.

[0101] Specifically, after the background adaptive analysis module 4 acquires historical environmental parameter data and historical radioactive radiation data, it performs time correspondence processing on the historical environmental parameter data and historical radioactive radiation data to establish a correspondence between environmental parameter data and radioactive radiation data within the same time period, and constructs a historical training sample set based on the corresponding historical environmental parameter data and historical radioactive radiation data.

[0102] Let the historical training sample set be:

[0103] ;

[0104] in, Represents the set of historical training samples. This indicates the number of historical training samples. Indicates the first Environmental characteristic data corresponding to each historical sample Indicates the first Radioactive radiation data corresponding to historical samples.

[0105] The background adaptive analysis module 4 performs feature mapping training based on the historical training sample set and establishes a correspondence model between environmental parameters and background radiation. Let the background adaptive generation model be:

[0106] ;

[0107] in, Indicates the first The virtual background radiation value corresponding to each sampling time. Represents real-time environmental characteristic data. Indicates model parameters, This indicates a background-adaptive generative model.

[0108] The background adaptive analysis module 4 trains model parameters based on a historical training sample set and establishes a correspondence model between environmental parameters and background radiation based on the trained model parameters. Let the model training objective function be:

[0109] ;

[0110] in, This represents the model training objective function. Indicates the first Radioactive radiation data corresponding to each historical sample This represents the background radiation value output by the background adaptive generation model.

[0111] The background adaptive analysis module 4 stores the trained correspondence model and calls it to generate virtual background radiation values ​​during real-time monitoring. After acquiring real-time environmental feature data, the background adaptive analysis module 4 normalizes the real-time environmental feature data and inputs the normalized real-time environmental feature data into the background adaptive generation model to generate the virtual background radiation value corresponding to the current sampling time.

[0112] Let the normalized real-time environmental feature data be:

[0113] ;

[0114] in, This represents the normalized environmental parameter data. Indicates the first The original environmental parameter data corresponding to each sampling time. Indicates the first The mean value corresponding to the class environment parameters. Indicates the first The standard deviation of the class environment parameters.

[0115] The background adaptive analysis module 4 generates a corresponding virtual background radiation value based on the normalized real-time environmental feature data, and sends the virtual background radiation value to the residual analysis process, so that the environmental parameter data and the background radiation data form a corresponding data mapping relationship, thereby realizing the dynamic background radiation generation process based on the environmental parameter change process.

[0116] Furthermore, in the background adaptive analysis module 4, the step of generating residual data based on the difference between real-time radioactive radiation data and virtual background radiation values ​​includes:

[0117] To acquire real-time radioactive radiation data and virtual background radiation values;

[0118] Residual data is generated based on the difference between real-time radioactive radiation data and virtual background radiation values;

[0119] Statistical processing of residual data is performed based on continuous time windows.

[0120] In the background adaptive analysis module 4, the steps for generating radioactive anomaly determination results based on the comparison results between residual data and preset anomaly determination thresholds include:

[0121] Obtain window statistics results of residual data;

[0122] Standardized residual indices are generated based on the window mean and window standard deviation of residual data.

[0123] The standardized residual index is compared with the preset anomaly detection threshold;

[0124] The results of the comparison are used to generate a determination of radioactive anomalies.

[0125] Specifically, after acquiring real-time radioactive radiation data and virtual background radiation values, the background adaptive analysis module 4 calculates the difference between the real-time radioactive radiation data and the virtual background radiation values ​​to generate corresponding residual data. Let the first... The real-time radioactive radiation data corresponding to each sampling time is The corresponding virtual background radiation value is The corresponding residual data are:

[0126] ;

[0127] in, Indicates the first The residual data corresponding to each sampling time point This represents real-time radioactive radiation data. This represents the virtual background radiation value.

[0128] The background adaptive analysis module 4 performs statistical processing on the residual data based on continuous time windows, and arranges the residual data corresponding to multiple consecutive sampling times in a windowed manner according to the sampling time order to generate a residual window data set. Let the time window length be... The mean of the residual window within the time window is:

[0129] ;

[0130] in, Indicates the first The mean of the residual window corresponding to each sampling time. Indicates the length of the time window. Indicates the first The residual data corresponding to each sampling time point.

[0131] The corresponding residual window standard deviation is:

[0132] ;

[0133] in, Indicates the first The standard deviation of the residual window corresponding to each sampling time.

[0134] The background adaptive analysis module 4 generates standardized residual indices based on the residual window mean and standard deviation, and generates radioactive anomaly determination results based on the comparison between the standardized residual indices and the preset anomaly determination threshold. Let the standardized residual indices be:

[0135] ;

[0136] in, Indicates the first The standardized residual index corresponding to each sampling time point This represents a stable parameter.

[0137] Let the preset anomaly detection threshold be... The result of the radioactive anomaly determination is as follows:

[0138] ;

[0139] in, This indicates the result of the radioactive anomaly assessment. This indicates that the radioactive radiation data at the current sampling time meets the anomaly detection criteria. This indicates that the radioactive radiation data corresponding to the current sampling time does not meet the anomaly detection criteria.

[0140] The background adaptive analysis module 4 sends the radioactive anomaly determination results to the communication and display module 5, which then displays and uploads the results. This enables the real-time radioactive radiation data, virtual background radiation values, and residual data to form a corresponding anomaly analysis process, thereby realizing the radioactive anomaly determination process based on continuous time window statistical results.

[0141] Furthermore, in the background adaptive analysis module 4, the steps for updating the background adaptive generation model include:

[0142] Acquire new environmental parameter data and new radioactive radiation data;

[0143] Incremental training samples were constructed based on newly added environmental parameter data and newly added radioactive radiation data.

[0144] The parameters of the baseline adaptive generative model are updated based on incremental training samples.

[0145] Specifically, the background adaptive analysis module 4 continuously acquires new environmental parameter data and new radioactive radiation data during system operation, and performs time-correspondence processing on the new environmental parameter data and new radioactive radiation data according to the sampling time sequence to construct corresponding incremental training samples. Let the set of incremental training samples corresponding to the u-th model update be:

[0146] ;

[0147] in, Indicates the first The incremental training sample set corresponding to each model update This indicates the number of incremental training samples. Indicates the first Real-time environmental feature data corresponding to each incremental training sample. Indicates the first Real-time radioactive radiation data corresponding to each incremental training sample.

[0148] The background adaptive analysis module 4 normalizes the real-time environmental feature data in the incremental training sample set, and then inputs the normalized real-time environmental feature data into the background adaptive generation model to calculate the corresponding background radiation prediction value. Let the background radiation prediction value output by the background adaptive generation model be: ;in, Indicates the first The predicted background radiation value corresponding to each incremental training sample. Indicates the first Real-time environmental feature data corresponding to each incremental training sample. Indicates the first The model parameters before the next model update This indicates a background-adaptive generative model.

[0149] The background adaptive analysis module 4 calculates the model update objective function based on the difference between the predicted background radiation value and the real-time radioactive radiation data, and adjusts the model parameters based on the model update objective function. Let the model update objective function be:

[0150] ;

[0151] in, Indicates the first The model update objective function corresponding to this model update.

[0152] The background adaptive analysis module 4 updates the model parameters based on the model update objective function. Let the updated model parameters be: ;in, This represents the updated model parameters. This indicates the model update step size. This represents the gradient corresponding to the model's objective function update.

[0153] After completing the model parameter update, the background adaptive analysis module 4 writes the updated model parameters into the background adaptive generation model, and continues to generate the corresponding virtual background radiation value based on the updated background adaptive generation model. This enables the newly added environmental parameter data and the newly added radioactive radiation data to form a corresponding incremental training process, thereby realizing the background adaptive generation model update process based on the newly added monitoring data.

[0154] Furthermore, in the communication and display module 5, the steps for displaying and uploading the results of the radioactive anomaly determination include:

[0155] Acquire real-time radioactive radiation data, virtual background radiation values, and results of radioactive anomaly assessment;

[0156] The display panel 9 displays real-time radioactive radiation data, virtual background radiation values, and results of radioactive anomaly determination.

[0157] The real-time radioactive radiation data, virtual background radiation values, and radioactive anomaly determination results are uploaded to the remote monitoring platform via the wireless communication unit.

[0158] Specifically, the communication and display module 5 acquires the real-time radioactive radiation data, virtual background radiation value, and radioactive anomaly determination results output by the background adaptive analysis module 4, and performs data correspondence processing on the real-time radioactive radiation data, virtual background radiation value, and radioactive anomaly determination results according to the sampling time sequence to generate corresponding monitoring output data.

[0159] Let the first The monitoring output data corresponding to each sampling time is: ;in, Indicates the first Monitoring output data corresponding to each sampling time. This represents real-time radioactive radiation data. This represents the virtual background radiation value. This indicates the result of the radioactive anomaly assessment.

[0160] The communication and display module 5 sends the monitoring output data to the display panel 9, and displays the real-time radioactive radiation data, virtual background radiation value, and radioactive anomaly determination results based on the display panel 9. When the radioactive anomaly determination result meets the anomaly determination conditions, the communication and display module 5 controls the display panel 9 to output the corresponding anomaly status information and simultaneously generates the corresponding anomaly upload data. Let the anomaly upload data be:

[0161] ;

[0162] in, Indicates the first Abnormal uploaded data corresponding to each sampling time. Indicates the sampling time.

[0163] The communication and display module 5 encapsulates the abnormal uploaded data through the wireless communication unit and sends the abnormal uploaded data to the remote monitoring platform based on the wireless communication protocol. The wireless communication unit uploads the monitoring output data according to a preset communication cycle and performs time stamp processing on the monitoring output data during the communication process to ensure that the monitoring output data received by the remote monitoring platform corresponds to the on-site sampling time.

[0164] After receiving the monitoring output data, the remote monitoring platform stores the real-time radioactive radiation data, virtual background radiation values, and radioactive anomaly determination results, and generates corresponding monitoring record data based on the sampling time sequence.

[0165] The communication and display module 5 synchronously displays and uploads real-time radioactive radiation data, virtual background radiation values, and radioactive anomaly determination results, enabling on-site monitoring data and remote monitoring data to form a corresponding data transmission process, thereby realizing the remote synchronous transmission process of radioactive monitoring data based on wireless communication.

[0166] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An AI artificial intelligence based on-site intelligent monitoring system for radioactive elements, characterized in that, include: A vertical mounting bracket (6) is provided with a detection component (7) and a control box (8); the detection component (7) includes a radioactivity detection module (1) and an environmental parameter monitoring module (2); the control box (8) is provided with a data acquisition control module (3), a background adaptive analysis module (4) and a communication and display module (5), and a display panel (9) is provided on one side of the control box (8); The data acquisition and control module (3) is used to acquire the radioactive radiation data output by the radioactive detection module (1) and the environmental parameter data output by the environmental parameter monitoring module (2), and to perform digital conversion and filtering on the radioactive radiation data to generate real-time environmental feature data. The background adaptive analysis module (4) is used to acquire historical environmental parameter data and historical radioactive radiation data, and to establish a background adaptive generation model based on the historical environmental parameter data and historical radioactive radiation data. Then, real-time environmental feature data is input into the background adaptive generation model to generate the corresponding virtual background radiation value, and the background adaptive generation model is updated based on the newly added environmental parameter data and newly added radioactive radiation data. The background adaptive analysis module (4) is also used to acquire real-time radioactive radiation data and virtual background radiation values, and generate residual data based on the difference between real-time radioactive radiation data and virtual background radiation values, and then generate radioactive anomaly judgment results based on the comparison results between the residual data and the preset anomaly judgment threshold. The communication and display module (5) is used to display and upload the results of the radioactive anomaly determination.

2. The AI-based intelligent on-site monitoring system for radioactive elements according to claim 1, characterized in that, In the data acquisition and control module (3), the steps of digitizing and filtering the radioactive radiation data include: The analog radiation signal output by the radioactivity detection module (1) is processed by analog-to-digital conversion; The digital radiation signal after analog-to-digital conversion is subjected to sliding window filtering. Real-time radioactive radiation data is generated based on the filtered digital radiation signal.

3. The AI-based intelligent on-site monitoring system for radioactive elements according to claim 1, characterized in that, In the data acquisition and control module (3), the step of generating real-time environmental feature data includes: The temperature, humidity and air pressure parameters collected by the environmental parameter monitoring module (2) are collected synchronously; Time-correlation processing is performed on the synchronously acquired temperature, humidity, and air pressure parameters. Real-time environmental feature data is constructed based on time-corresponding temperature, humidity, and air pressure parameters.

4. The AI-based intelligent on-site monitoring system for radioactive elements according to claim 1, characterized in that, In the background adaptive analysis module (4), the steps for establishing a background adaptive generation model based on historical environmental parameter data and historical radioactive radiation data include: Acquire historical environmental parameter data and historical radioactive radiation data; Sample correlation processing was performed based on historical environmental parameter data and historical radioactive radiation data. Feature mapping training is performed based on the results of sample association processing. A model of the correspondence between environmental parameters and background radiation is established based on the training results of feature mapping; The correspondence model is stored as a background adaptive generation model.

5. The AI-based intelligent on-site monitoring system for radioactive elements according to claim 4, characterized in that, In the background adaptive analysis module (4), the step of generating the corresponding virtual background radiation value includes: Acquire real-time environmental feature data; Normalize the real-time environmental feature data; The normalized real-time environmental feature data is input into the background adaptive generation model; The virtual background radiation value is output based on the background adaptive generation model.

6. The AI-based intelligent on-site monitoring system for radioactive elements according to claim 1, characterized in that, In the background adaptive analysis module (4), the step of generating residual data based on the difference between real-time radioactive radiation data and virtual background radiation values ​​includes: To acquire real-time radioactive radiation data and virtual background radiation values; Residual data is generated based on the difference between real-time radioactive radiation data and virtual background radiation values; The residual data is statistically processed based on a continuous time window.

7. The AI-based intelligent on-site monitoring system for radioactive elements according to claim 6, characterized in that, In the background adaptive analysis module (4), the step of generating a radioactive anomaly determination result based on the comparison result between the residual data and the preset anomaly determination threshold includes: Obtain window statistics results of residual data; Standardized residual indices are generated based on the window mean and window standard deviation of residual data. The standardized residual index is compared with the preset anomaly detection threshold; The results of the comparison are used to generate a determination of radioactive anomalies.

8. The AI-based intelligent on-site monitoring system for radioactive elements according to claim 1, characterized in that, In the background adaptive analysis module (4), the step of updating the background adaptive generation model includes: Acquire new environmental parameter data and new radioactive radiation data; Incremental training samples were constructed based on newly added environmental parameter data and newly added radioactive radiation data. The parameters of the baseline adaptive generation model are updated based on the incremental training samples.

9. The AI-based intelligent on-site monitoring system for radioactive elements according to claim 1, characterized in that, In the communication and display module (5), the steps of displaying and uploading the radioactive anomaly determination result include: Acquire real-time radioactive radiation data, virtual background radiation values, and results of radioactive anomaly assessment; The display panel (9) displays real-time radioactive radiation data, virtual background radiation values, and radioactive anomaly determination results. The real-time radioactive radiation data, virtual background radiation values, and radioactive anomaly determination results are uploaded to the remote monitoring platform via the wireless communication unit.

10. The AI-based intelligent on-site monitoring system for radioactive elements according to claim 1, characterized in that, The vertical mounting bracket (6) is a column structure fixed to the ground; the detection component (7) is located on the top of the vertical mounting bracket (6); the control box (8) is located in the middle of the vertical mounting bracket (6); and the display panel (9) is located on the outer surface of the control box (8).

Images