A method and system for monitoring radioactive separation and leakage prevention liquids based on multi-sensor data.

Abstract

This invention relates to the field of radiation monitoring technology and discloses a method and system for monitoring radioactive separation and leakage prevention based on multi-sensor data. The method includes: Step 1, dividing the monitoring area and establishing a step-by-step reference model; Step 2, performing background counting measurements and baseline initialization; Step 3, simultaneously acquiring radiation, liquid level, pressure, and accumulated liquid data, determining the net radiation measurement value, and generating a unified spatiotemporal state vector; Step 4, determining the abnormal quantity of the liquid path and the liquid event judgment result; Step 5, determining the actual normalized activity distribution ratio, activity migration distortion, and activity exceeding the prohibited zone boundary, obtaining the radioactive trajectory anomaly judgment result; Step 6, obtaining the system joint state and determining the priority anomaly source segment; Step 7, determining the total measured activity of the system, the radioactive activity loss judgment result, and the monitoring result. This invention achieves automatic monitoring and location of liquid anomalies, radioactive migration anomalies, and radioactive leakage risks.

Description

Technical Field

[0001] This invention belongs to the field of radiation monitoring technology, specifically relating to a method and system for monitoring radioactive separation and leakage prevention based on multi-sensor data. Background Technology

[0002] Radionuclide separation technology is widely used in nuclear medicine isotope preparation, radiochemical analysis, and nuclear fuel reprocessing. In the radionuclide separation process, liquids containing radioactive materials typically flow between different separation units through a pipeline system, and the separation of radionuclides is completed in a specific separation medium or device. Because radioactive materials pose a high radiation hazard, any liquid leakage or abnormal migration during transport or separation can lead to equipment contamination, increased environmental radiation levels, and operational safety risks. Therefore, real-time monitoring of the liquid transport status and the distribution of radioactive materials is crucial during the operation of a radionuclide separation system.

[0003] In existing technologies, monitoring of radioactive systems typically relies primarily on radiation measurements. For example, radiation detectors are used to monitor the radiation intensity in localized areas of the system to determine if a radioactive leak is present. However, relying solely on a single radiation signal often makes it difficult to distinguish between radiation changes occurring during normal operation and those caused by abnormal leaks. Furthermore, abnormal migration of radioactive liquids within a system is usually accompanied by changes in liquid level, pressure, or abnormal liquid accumulation. Using only radiation signals for assessment can easily lead to misjudgments or missed detections.

[0004] On the other hand, in complex radioactive separation processes, radioactive materials exhibit different spatial distribution patterns in different areas of the system at different operational stages. Existing monitoring methods typically lack a comprehensive analysis of the relationship between separation process stages and radioactive distribution patterns, making it difficult to accurately identify abnormal migration paths of radioactive materials within the system. When pipeline leaks or abnormal diffusion of radioactive liquids occur, manual analysis of multiple monitoring data is often required for judgment, which limits both monitoring efficiency and reliability. Summary of the Invention

[0005] This invention provides a method and system for monitoring leak-proof liquids during radioactive separation based on multi-sensor data, which solves the technical problem in related technologies that rely solely on a single radiation measurement signal for monitoring, making it difficult to accurately identify liquid leaks and abnormal migration of radioactive materials during radioactive separation.

[0006] This invention provides a method for monitoring radioactive separation and leakage prevention based on multi-sensor data, comprising the following steps:

[0007] Step 1: Obtain the liquid path layout and process division information of the radioactive separation system, divide the monitoring area of ​​the radioactive separation system, and establish a process-based reference model for each monitoring area under each process step.

[0008] Step 2: Based on each monitoring area, perform background counting measurements and baseline initialization to obtain radiation background count values, initial liquid level baseline values, initial pressure baseline values, and initial liquid accumulation detection baseline values;

[0009] Step 3: Based on the step-by-step reference model and the current step state, simultaneously collect the original radiation measurement value, real-time liquid level value, real-time pressure value and real-time liquid accumulation detection value, determine the net radiation measurement value, and generate a unified spatiotemporal state vector;

[0010] Step 4: Based on the spatiotemporal unified state vector, the step-by-step reference model, and the initial baseline value of liquid accumulation detection, determine the abnormal quantity of the liquid path and obtain the liquid event judgment result;

[0011] Step 5: Based on the spatiotemporal unified state vector and the step-by-step reference model, determine the actual normalized activity distribution ratio, activity migration distortion, and activity overrun in the forbidden zone to obtain the radioactive trajectory anomaly judgment result.

[0012] Step 6: Based on the liquid event determination results, the radioactive trajectory anomaly determination results, the abnormal liquid path quantity, and the net radiation measurement value, obtain the system joint state and determine the priority anomaly source segment;

[0013] Step 7: Based on the system joint state, priority anomaly source segments, step-by-step reference model, and spatiotemporal unified state vector, perform automatic processing to determine the total measured activity of the system, the judgment result of radioactivity loss, and the monitoring result.

[0014] This invention also provides a radioactive separation leak prevention monitoring system based on multi-sensor data, comprising:

[0015] The reference model construction module is used to obtain information on the liquid path layout and process division of the radioactive separation system, divide the monitoring area of ​​the radioactive separation system, and establish a process-based reference model for each monitoring area under each process step.

[0016] The baseline calibration initialization module is used to perform background counting measurements and baseline initialization based on each monitoring area, and obtain radiation background count values, initial liquid level baseline values, initial pressure baseline values, and initial liquid accumulation detection baseline values.

[0017] The state vector generation module is used to synchronously collect the original radiation measurement value, real-time liquid level value, real-time pressure value and real-time liquid accumulation detection value based on the step-based reference model and the current step state, and determine the net radiation measurement value to generate a unified spatiotemporal state vector.

[0018] The liquid path anomaly determination module is used to determine the amount of liquid path anomaly based on the spatiotemporal unified state vector, the step-by-step reference model, and the initial baseline value of liquid accumulation detection, and to obtain the liquid event determination result.

[0019] The trajectory anomaly determination module is used to determine the actual normalized activity distribution ratio, activity migration distortion, and activity overrun in the forbidden zone based on the spatiotemporal unified state vector and the step-based reference model, and to obtain the radioactive trajectory anomaly determination result.

[0020] The joint analysis and localization module is used to obtain the joint state of the system and determine the priority anomaly source segment based on the liquid event judgment results, the radioactive trajectory anomaly judgment results, the liquid path anomaly quantity and the net radiation measurement value.

[0021] The disposal verification output module is used to perform automatic disposal based on the system joint state, priority abnormal source segments, step-by-step reference model and spatiotemporal unified state vector, and to determine the total measured activity of the system, the judgment result of radioactivity loss and the monitoring result.

[0022] The beneficial effects of this invention are as follows: By dividing the liquid path layout of the radioactive separation system into monitoring areas and establishing a step-by-step reference model based on the process steps of the separation process, the invention enables a unified description of the distribution of radioactivity, liquid level, and pressure at different operating stages. Based on this, multi-source monitoring data is acquired through radiation measurement units, liquid level detection units, pressure detection units, and liquid accumulation detection units, and a unified spatiotemporal state vector is constructed to uniformly characterize the system's operating state, thereby simultaneously reflecting changes in radioactivity and liquid transport status. Furthermore, this invention achieves a comprehensive judgment of liquid anomalies and radioactive migration anomalies by calculating liquid path anomalies and activity migration distortion, and identifies priority anomaly source segments through zonal location analysis, enabling rapid identification of anomaly areas. Simultaneously, by comparing the consistency between the total measured activity and the reference activity, the invention verifies whether a loss of radioactivity has occurred, thus distinguishing between general liquid anomalies and actual radioactive leaks. Overall, this invention improves the anomaly identification capability and monitoring reliability in the radioactive separation process. Attached Figure Description

[0023] Figure 1 This is a flowchart of the radioactive separation and leakage prevention monitoring method based on multi-sensor data of the present invention. Detailed Implementation

[0024] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0025] like Figure 1 As shown, the radioactive separation leak prevention monitoring method based on multi-sensor data includes the following steps:

[0026] Step 1: Obtain the liquid path layout and process division information of the radioactive separation system, divide the monitoring area of ​​the radioactive separation system, and establish a process-based reference model for each monitoring area under each process step.

[0027] Step 2: Based on each monitoring area, perform background counting measurements and baseline initialization to obtain radiation background count values, initial liquid level baseline values, initial pressure baseline values, and initial liquid accumulation detection baseline values;

[0028] Step 3: Based on the step-by-step reference model and the current step state, simultaneously collect the original radiation measurement value, real-time liquid level value, real-time pressure value and real-time liquid accumulation detection value, determine the net radiation measurement value, and generate a unified spatiotemporal state vector;

[0029] Step 4: Based on the spatiotemporal unified state vector, the step-by-step reference model, and the initial baseline value of liquid accumulation detection, determine the abnormal quantity of the liquid path and obtain the liquid event judgment result;

[0030] Step 5: Based on the spatiotemporal unified state vector and the step-by-step reference model, determine the actual normalized activity distribution ratio, activity migration distortion, and activity overrun in the forbidden zone to obtain the radioactive trajectory anomaly judgment result.

[0031] Step 6: Based on the liquid event determination results, the radioactive trajectory anomaly determination results, the abnormal liquid path quantity, and the net radiation measurement value, obtain the system joint state and determine the priority anomaly source segment;

[0032] Step 7: Based on the system joint state, priority anomaly source segments, step-by-step reference model, and spatiotemporal unified state vector, perform automatic processing to determine the total measured activity of the system, the judgment result of radioactivity loss, and the monitoring result.

[0033] In one embodiment of the present invention, the liquid path layout and process step division information of the radioactive separation system are obtained, the monitoring area of ​​the radioactive separation system is divided, and a process-based reference model of each monitoring area under each process step is established. The radioactive separation system typically consists of multiple liquid transport channels, separation structural units, and product collection units. The liquid path layout refers to the connection relationships between the liquid transport pipelines, separation units, collection containers, and outlet pipelines within the system. The process step division information refers to the separation operation stage information obtained by dividing the radioactive separation process according to the operation sequence. Each process step corresponds to a specific process operating state, and the specific process includes:

[0034] Step 11: Based on the liquid path layout of the radioactive separation system, the entire system is spatially functionally divided into multiple monitoring areas, including a sample introduction area, a separation inlet area, a separation main body area, a product collection area, a waste liquid discharge area, a containment detection area, and a non-process area within the shell. Specifically, the sample introduction area is used for radioactive samples to enter the separation system; the separation inlet area is a transitional region before the sample enters the separation device; the separation main body area is the core area where the radionuclide separation process occurs; the product collection area is used to collect target radioactive products; the waste liquid discharge area is used to discharge waste liquid generated during the separation process; the containment detection area is a leakage monitoring area located inside the system's containment structure; and the non-process area within the shell is an area inside the equipment shell that does not participate in normal process flow, used to identify abnormal radioactive migration. Through this area division, a set of monitoring areas for the separation system can be established spatially, providing a clear spatial positioning basis for subsequent multi-sensor data acquisition.

[0035] Step 12 involves configuring multiple sensor measurement units in each monitoring area, including a radiation measurement unit, a liquid level detection unit, a pressure detection unit, and a liquid accumulation detection unit. The radiation measurement unit measures the radioactive radiation count signal within the corresponding area, reflecting changes in the activity of radioactive materials. The liquid level detection unit measures changes in liquid height to reflect the liquid flow state. The pressure detection unit measures changes in internal pressure within the pipeline to reflect the liquid transport state. The liquid accumulation detection unit detects abnormal liquid accumulation within the equipment casing or containment area. These multiple sensor data sources constitute a multi-sensor data source for subsequent anomaly monitoring during the radioactive separation process.

[0036] Based on the operational flow of the radioactive separation process, the entire process is divided into multiple continuous steps, including preparation, sample loading, entry into the separation section, residence, elution, product collection, waste discharge, and final emptying. Each step corresponds to a different liquid flow state and radioactivity distribution state within the system. For example, in the sample loading step, the radioactive sample enters the separation system from the injection zone; in the residence step in the main separation zone, the radioactive material migrates and separates in the separation medium; and in the product collection step, the target radioactive product enters the product collection zone. By dividing the process flow into steps, a correspondence between the system's operational state and time series can be established.

[0037] Step 13: For each monitoring area under each process step, determine the expected activity distribution ratio, liquid level reference value, and pressure reference value for that area under normal operating conditions. The expected activity distribution ratio refers to the proportion of radioactivity in each monitoring area relative to the total activity of all monitoring areas under a given process step; this ratio reflects the spatial distribution pattern of radioactive materials in the system under normal separation conditions. The liquid level reference value refers to the normal liquid level height of the corresponding monitoring area under that process step. The pressure reference value refers to the normal operating pressure of the corresponding pipeline under that process step. All of the above reference parameters are derived from system design parameters or historical normal operating data and are stored as standard reference values.

[0038] The expected activity distribution ratio, liquid level reference value, and pressure reference value for each monitoring area under each step are uniformly linked and recorded to form a step-based reference model. This step-based reference model refers to a model structure that uses the step as an index and the monitoring area as the object, recording the reference operating state parameters of each monitoring area under different steps. This model can describe the expected spatial and temporal distribution patterns of radioactivity, liquid level, and pressure during normal radioactive separation.

[0039] This embodiment divides the radioactive separation system into multiple monitoring zones and establishes a step-by-step reference model based on process steps, enabling a clear description of the normal activity distribution pattern of the system at different operating stages. When liquid leakage, abnormal liquid accumulation, or abnormal migration of radioactive materials occurs in the system, the real-time radiation measurements, liquid level, or pressure changes in each monitoring zone will deviate from the reference model, thus allowing for timely identification. This provides a reliable data foundation for radioactive liquid leakage detection and abnormal migration identification, and improves the safety monitoring capabilities during the operation of the radioactive separation system.

[0040] In one embodiment of the present invention, after the monitoring area is divided, the sensor measurement signals in each monitoring area need to be initialized and calibrated to establish a reference baseline state before system operation. Specifically, background counting measurement and baseline initialization are performed based on each monitoring area to obtain the radiation background count value, the initial baseline value of liquid level, the initial baseline value of pressure, and the initial baseline value of liquid accumulation detection.

[0041] When the system is in a stable environment without radioactive sample introduction, background counting measurements are performed on the radiation measurement units in each monitoring area. Background counting measurement refers to the process of measuring the radiation counting signal generated by natural environmental radiation, background radiation from equipment materials, and noise from the electronic measurement system in the absence of radioactive samples. By continuously acquiring the counting signals from the radiation measurement units within a preset time window and statistically processing the acquisition results, the stable radiation count level of the monitoring area under normal background conditions can be obtained. The statistically obtained count value is defined as the radiation background count value. The radiation background count value characterizes the background radiation intensity of the monitoring area in the absence of radioactive samples. Its function is to subtract the environmental and system background effects from the original radiation measurement values ​​during subsequent real-time monitoring, thereby obtaining the effective radiation signal generated solely by radioactive materials.

[0042] While completing the radiation background counting measurement, baseline initialization is performed on the liquid level detection unit, pressure detection unit, and liquid accumulation detection unit in the system. Baseline initialization refers to recording the output signals of each sensor when the system is in a normal static state and before liquid flow begins, and using these recorded values ​​as a reference for subsequent anomaly detection. Specifically, the initial baseline value of the liquid level in the corresponding monitoring area is obtained by reading the output signals of the liquid level detection units in a stable state. The initial baseline value represents the normal initial liquid level state in the monitoring area when the system has not performed radioactive separation operations. When the real-time liquid level value deviates significantly from the initial baseline value during subsequent monitoring, possible abnormal liquid flow or liquid leakage can be identified.

[0043] By reading the output signal of the pressure detection unit when no liquid transport occurs in the system, the initial pressure baseline value of the corresponding pipeline in each monitoring area can be obtained. The initial pressure baseline value reflects the pipeline pressure state of the system under static conditions. During actual operation, if abnormal blockage, leakage, or pipeline damage occurs in liquid transport, the pipeline pressure will deviate from the initial pressure baseline value, thus providing a basis for subsequent liquid circuit anomaly identification.

[0044] Through the above steps, a complete set of multi-sensor baseline parameters can be established for each monitoring area, including radiation background count values, initial liquid level baseline values, initial pressure baseline values, and initial liquid accumulation detection baseline values. These baseline parameters serve as the initial reference state for system operation and are compared with real-time multi-sensor data during subsequent monitoring to identify changes in system state.

[0045] In one embodiment of the present invention, based on a step-based reference model and the current step state, the original radiation measurement value, real-time liquid level value, real-time pressure value, and real-time liquid accumulation detection value are simultaneously collected, and the net radiation measurement value is determined to generate a spatiotemporally unified state vector, including:

[0046] Step 21: Based on the current process step status, simultaneously collect multi-sensor measurement data from each monitoring area at the current sampling time. The current process step status refers to the stage of the radioactive separation system's operation at the current sampling time, such as preparation, sample loading, entry into the separation section, residence, elution, product collection, waste discharge, or final emptying. The current process step can be obtained in real time through the process step control system or process control logic. Subsequently, at this sampling time, multiple sensor data from each monitoring area are simultaneously read, including the raw radiation measurement value output by the radiation measurement unit, the real-time liquid level value output by the liquid level detection unit, the real-time pressure value output by the pressure detection unit, and the real-time liquid accumulation detection value output by the liquid accumulation detection unit. The raw radiation measurement value refers to the radiation count signal recorded by the radiation measurement unit per unit time. This signal reflects the intensity of radioactive radiation in the monitoring area, but it usually includes the influence of ambient background radiation and system background radiation. The real-time liquid level value represents the instantaneous state of the liquid height or volume in the corresponding monitoring area. The real-time pressure value represents the real-time pressure state in the liquid transport pipeline. The real-time liquid accumulation detection value is used to characterize whether there is abnormal liquid accumulation in the equipment enclosure area or inside the shell. By synchronously collecting the above multiple data at the same sampling time, it is possible to ensure that the data from different sensors have a consistent time reference, thereby avoiding data deviations caused by sampling time differences.

[0047] Step 22: After obtaining the original radiation measurements for each monitoring area, it is necessary to further calculate the net radiation measurement. The background radiation count refers to the stable radiation count signal generated by ambient radiation or equipment background radiation when there are no radioactive samples or abnormal leaks in the system. This background count is usually obtained through background count measurements during system initialization and saved as a baseline parameter. Therefore, in this embodiment, the net radiation measurement can be obtained by subtracting the corresponding background radiation count from the original radiation measurements for each monitoring area. The net radiation measurement reflects the effective radiation signal generated by radioactive materials within the system, thereby eliminating the influence of ambient background radiation on the measurement results and enabling subsequent analysis to more accurately reflect the true distribution of radioactive materials in each monitoring area.

[0048] Step 23: After obtaining the net radiation measurement values ​​for each monitoring area, it is necessary to integrate the multi-sensor data. In this embodiment, the current sampling time is used as a unified time identifier. The net radiation measurement values, real-time liquid level values, real-time pressure values, and real-time liquid accumulation detection values ​​of each monitoring area are integrated with the current process status to form a unified spatiotemporal state vector. The unified spatiotemporal state vector can completely describe the spatial distribution and operating status of the system at a certain moment.

[0049] This embodiment synchronously collects data from multiple sensors at a unified sampling time and constructs a unified spatiotemporal state vector containing spatial region information and process status information. This establishes a clear correspondence between radiation measurement data and process status information. This method can not only accurately reflect the distribution of radioactivity in different monitoring areas, but also combine liquid level, pressure and liquid accumulation detection information to achieve radioactive separation and leak prevention monitoring based on multi-sensor data, thereby improving the system's ability to identify radioactive leaks and abnormal migration and the reliability of safety monitoring.

[0050] In one embodiment of the present invention, based on a spatiotemporal unified state vector, a step-by-step reference model, and an initial baseline value for liquid accumulation detection, the abnormal quantity of the liquid path is determined, and a liquid event determination result is obtained, including:

[0051] Step 31: Extract the operational status parameters of each monitoring area based on the spatiotemporal unified state vector. The spatiotemporal unified state vector refers to a data set structure formed by integrating multi-sensor data from each monitoring area and system operation step information at the same sampling time. This structure simultaneously includes time information, spatial area information, and various sensor measurement values. In this embodiment, the real-time liquid level value, real-time pressure value, real-time liquid accumulation detection value, and current operation step status corresponding to each monitoring area are extracted from the spatiotemporal unified state vector. Subsequently, based on the current operation step status, the corresponding liquid level reference value and pressure reference value for each monitoring area under that operation step are read from the operation step reference model.

[0052] Step 32: After obtaining the real-time measured values ​​and reference values, calculate the liquid path anomaly for each monitoring area. Specifically, calculate the degree of difference between the real-time liquid level value and the liquid level reference value, the degree of difference between the real-time pressure value and the pressure reference value, and the degree of difference between the real-time liquid accumulation detection value and the initial baseline value of liquid accumulation detection. To reflect the magnitude of the difference, this embodiment uses the absolute value of the deviation as a quantitative indicator, i.e., obtain the absolute value of the liquid level deviation, the absolute value of the pressure deviation, and the absolute value of the liquid accumulation detection deviation. The absolute value of the deviation represents the absolute magnitude of the difference between the real-time measured value and the reference baseline value, used to characterize the degree to which the system state deviates from the normal operating state. Subsequently, multiply the above three absolute values ​​of deviation by the corresponding preset weighting coefficients, and sum the weighted results to obtain the liquid path anomaly for each monitoring area. The weighting coefficients here are used to characterize the importance of different monitoring parameters for liquid anomaly identification. For example, in some areas, pressure changes are more sensitive to anomaly identification, so the corresponding pressure deviation weighting coefficient can be set higher. This weighted summation method integrates information on liquid level, pressure, and liquid accumulation detection to form an abnormal liquid circuit quantity that can comprehensively reflect the degree of change in the liquid circuit status.

[0053] Step 33: After obtaining the abnormal liquid levels in each monitoring area, the system automatically determines liquid events by comparing them with preset abnormal liquid level judgment thresholds. These abnormal liquid level judgment thresholds are reference thresholds determined based on system design parameters, historical operating data, or experimental calibration results, used to distinguish between normal operating fluctuations and abnormal liquid states. When the abnormal liquid level in a monitoring area reaches or exceeds the corresponding abnormal liquid level judgment threshold, it indicates that the liquid level, pressure, or liquid accumulation in that area has significantly deviated from normal operating conditions, thus determining that a liquid event has occurred and marking that monitoring area as an abnormal monitoring area. When the abnormal liquid levels in all monitoring areas are below their respective judgment thresholds, it is considered that no liquid event has occurred, thus obtaining the liquid event judgment result.

[0054] Through the above steps, multi-sensor data can be used to comprehensively analyze the liquid transport status in the radioactive separation system, enabling early identification of liquid anomalies. Compared with traditional methods that rely solely on single liquid level or pressure monitoring, this embodiment combines real-time liquid level, pressure, and liquid accumulation detection values ​​with process reference status provided by a step-by-step reference model, allowing the abnormal liquid flow to more comprehensively reflect the actual operating conditions of the system. When pipeline leaks, seal failures, or abnormal liquid accumulation occur, the signals from the aforementioned multiple sensors will change simultaneously, causing the abnormal liquid flow to rise rapidly and exceed the judgment threshold, enabling timely identification of liquid events and improving the leak prevention and monitoring capabilities of the radioactive separation system.

[0055] In one embodiment of the present invention, based on a spatiotemporally unified state vector and a step-based reference model, the actual normalized activity distribution ratio, activity migration distortion, and activity exceeding the forbidden zone boundary are determined to obtain a radioactive trajectory anomaly determination result, including:

[0056] Step 41: Extract the net radiation measurement values ​​and current process status of each monitoring area based on the spatiotemporal unified state vector. According to the current process status, read the expected activity distribution ratio of each monitoring area under this process step and the set of prohibited activity monitoring areas corresponding to the current process step from the process-based reference model. The expected activity distribution ratio refers to the theoretical proportion of radioactivity in each monitoring area relative to the total activity of all monitoring areas under normal separation operation. This ratio reflects the spatial distribution pattern of radioactive materials migrating along the normal process path in the system. The set of prohibited activity monitoring areas refers to the set of monitoring areas where radioactivity should theoretically not exist under a specific process step. For example, in certain process steps, radioactive materials should only exist in the separation main area or product collection area, while radioactive signals should not be detected in the containment detection area or non-process areas within the shell.

[0057] Step 42: Compare the net radiation measurement values ​​of each monitoring area with the sum of the net radiation measurement values ​​of all monitoring areas. The ratio represents the activity proportion of that monitoring area at the current moment, and this proportion is defined as the actual normalized activity distribution proportion. The purpose of normalization is to eliminate the influence of overall system activity changes on spatial distribution analysis, so that the activity distributions of each monitoring area can be compared on a uniform scale. Subsequently, compare the actual normalized activity distribution proportion of each monitoring area with the corresponding expected activity distribution proportion, and calculate the absolute value of the deviation between the two. The absolute value of the deviation is used to characterize the degree of difference between the actual activity distribution and the theoretical distribution. To reflect the importance of different monitoring areas in radioactive migration analysis, this embodiment sets activity importance weights for each monitoring area, and multiplies the absolute value of the deviation of each monitoring area by the corresponding weight and then sums them to obtain the activity migration distortion. The activity migration distortion is used to comprehensively describe the degree of overall migration deviation of radioactive materials in the system. When radioactive materials diffuse along unexpected paths or are abnormally retained, this indicator will increase significantly.

[0058] Step 43: Based on the prohibited activity monitoring zone set of the current process step, sum the net radiation measurements of each monitoring area in the prohibited activity monitoring zone set to obtain the prohibited zone activity exceedance amount. The prohibited zone activity exceedance amount is used to represent the total radiation intensity that should appear in a non-radioactive area under the current process step. When a radioactive material leaks, a pipeline is damaged, or an abnormal migration occurs in the system, radioactive material may enter the containment detection area or the non-process area inside the shell. At this time, the prohibited zone activity exceedance amount will increase significantly.

[0059] By comparing the activity migration distortion with a preset activity migration distortion threshold, and simultaneously comparing the activity exceeding the prohibited zone limit with the prohibited zone activity exceeding the prohibited zone limit, the result of radioactive trajectory anomaly determination can be obtained. When the activity migration distortion reaches or exceeds the activity migration distortion threshold, or the activity exceeding the prohibited zone limit reaches or exceeds the prohibited zone activity exceeding the prohibited zone limit, the system determines that a radioactive trajectory anomaly has occurred; when both parameters are below their respective thresholds, it is determined that no radioactive trajectory anomaly has occurred. The radioactive trajectory anomaly indicates an abnormal state in which the spatial migration path of radioactive material between monitoring areas deviates from the normal migration path described by the process reference model during radioactive separation.

[0060] This embodiment constructs a normalized activity distribution using net radiation measurements from each monitoring area and combines this with the expected activity distribution ratio provided by a step-by-step reference model, enabling quantitative analysis of the spatial migration state of radioactive materials. Simultaneously, by setting up a set of prohibited activity monitoring zones and calculating the activity exceedance within these zones, areas where radioactive activity should theoretically be absent can be monitored in a focused manner. When a radioactive liquid leak or abnormal diffusion occurs in the system, radioactive materials will enter the abnormal area, causing changes in the activity distribution and triggering trajectory anomaly detection. This enables dynamic monitoring of radioactive migration trajectories, improving the safety monitoring capabilities during the operation of the radioactive separation system.

[0061] In one embodiment of the present invention, based on the liquid event determination result, the radioactive trajectory anomaly determination result, the liquid path anomaly quantity, and the net radiation measurement value, the system joint state is obtained, and the priority anomaly source segment is determined, including:

[0062] Step 51: A combined judgment is made based on the liquid event determination result and the radioactive trajectory anomaly determination result to determine the system's overall state. The liquid event determination result characterizes whether there are any abnormal liquid conditions in the system, such as pipeline leaks, abnormal liquid accumulation, or abnormal liquid flow. The radioactive trajectory anomaly determination result characterizes whether the spatial migration path of radioactive materials in the system deviates from the normal process path. By combining these two determination results, the overall operating state of the system can be obtained. When neither a liquid event nor a radioactive trajectory anomaly occurs, the system is in normal operation, and the overall system state is determined to be normal. When a liquid event occurs but no radioactive trajectory anomaly is detected, the system has a liquid anomaly but no radioactive migration anomaly, and the overall system state is determined to be ordinary liquid anomaly. When no liquid event occurs but a radioactive trajectory anomaly is detected, the radioactive material has undergone abnormal migration or path deviation in the system, and the overall system state is determined to be radioactive path mismigration. When both a liquid event and a radioactive trajectory anomaly occur simultaneously, the system has a risk of radioactive liquid leakage or abnormal diffusion, and the overall system state is determined to be a high-risk state for radioactive leakage. This combined judgment method can classify different types of abnormal situations in a unified manner, providing a clear basis for subsequent anomaly location and handling.

[0063] Step 52: Based on the set of prohibited activity monitoring zones for the current process step, assign a prohibited activity flag value to each monitoring zone. When a monitoring zone belongs to the set of prohibited activity monitoring zones for the current process step, the prohibited activity flag value for that monitoring zone is set to one; when the monitoring zone does not belong to the set, the flag value is set to zero. In this way, process constraint information can be introduced into the anomaly location analysis process. Subsequently, the abnormal liquid flow quantity, net radiation measurement value, and prohibited activity flag value of each monitoring zone are used as input parameters to establish a zone-by-zone location input relationship, thereby providing a unified data foundation for subsequent anomaly source location.

[0064] Step 53: Based on the abnormal liquid flow quantity, net radiation measurement value, process step prohibited activity indicator value, corresponding liquid flow abnormality threshold, and corresponding radiation threshold of each monitoring area, the ratio of the abnormal liquid flow quantity to the corresponding liquid flow abnormality threshold of each monitoring area is multiplied by the liquid flow abnormality location weight to obtain a weighted result reflecting the degree of liquid abnormality. Simultaneously, the ratio of the net radiation measurement value to the corresponding radiation threshold of each monitoring area is multiplied by the radiation abnormality location weight to obtain a weighted result reflecting the degree of radioactive abnormality. Furthermore, the process step prohibited activity indicator value of each monitoring area is multiplied by the process step prohibited activity attribute location weight to reflect the abnormal sensitivity of that area when a radioactive signal appears under the current process step. Then, the above three weighted results are summed to obtain the zonal location score for each monitoring area. This score comprehensively reflects the influence of liquid abnormality characteristics, radiation abnormality characteristics, and process constraints on abnormality location. By comparing the zonal location scores of all monitoring areas, the monitoring area with the highest score can be determined and identified as the priority abnormality source segment. The priority anomaly source segment refers to the monitoring area that is most likely to experience an anomaly or is closest to the anomaly source location under the current operating state.

[0065] The above method enables comprehensive identification of system anomalies and location of anomaly sources. Compared to traditional methods that rely solely on a single radiation signal for anomaly detection, this embodiment integrates abnormal liquid flow quantities, radiation measurement data, and process constraint information to perform multi-dimensional analysis of the system status, thereby more accurately identifying anomaly types and locating anomaly sources. This method jointly analyzes radiation measurement data with information from multiple sensors, such as liquid level, pressure, and liquid accumulation detection, and combines this with prohibited activity zones in the process flow to accurately identify abnormal migration and potential leakage risks during radioactive separation, improving the reliability and response efficiency of the radioactive separation leak prevention monitoring system based on multi-sensor data.

[0066] In one embodiment of the present invention, based on the system joint state, priority anomaly source segments, step-by-step reference model, and spatiotemporal unified state vector, automatic processing is performed to determine the total measured activity of the system, the determination result of radioactivity loss, and the monitoring results, including:

[0067] Step 61: Execute automatic response based on the system's combined state and priority anomaly source segment. The system's combined state is the system's operating status obtained by combining the liquid event determination results and the radioactive trajectory anomaly determination results, used to characterize the current system anomaly type. For example, when the system's combined state is a common liquid anomaly, it indicates that there is a liquid transport anomaly in the system but no radioactive migration anomaly has yet occurred. In this case, freezing the process and suspending subsequent liquid input can prevent the anomaly from escalating further. When the system's combined state is a radioactive path mismigration, it indicates that radioactive material has migrated abnormally within the system. In this case, pausing the current separation process switchover maintains the system in its current state for further monitoring. When the system's combined state is a high-risk radioactive leakage state, it indicates a potential risk of radioactive liquid leakage in the system. In this case, immediately stop sample injection and disconnect the upstream and downstream liquid paths associated with the priority anomaly source segment, thereby blocking possible radioactive diffusion paths. The priority anomaly source segment refers to the monitoring area most likely to experience an anomaly, obtained through anomaly location analysis. This area serves as the key control target for automatic response. Through the above automatic response measures, the system's operating state can be restricted in a timely manner when an anomaly occurs, thereby reducing the risk of radioactive material diffusion.

[0068] Step 62: Extract the net radiation measurement values ​​of each monitoring area based on the spatiotemporal unified state vector, and sum the net radiation measurement values ​​of all monitoring areas to obtain the total measured activity of the system. The total measured activity of the system refers to the total amount of radioactivity detected in all monitoring areas at the current sampling time. This parameter reflects the overall amount of radioactive material present in the system. Subsequently, retrieve the theoretical reference total activity and the preset activity loss correction value corresponding to the current step from the step-based reference model. The theoretical reference total activity refers to the total amount of radioactivity that should exist in the system at this step stage under normal separation operation conditions; the preset activity loss correction value represents the small amount of activity loss correction allowed in actual process operation. This correction value is used to compensate for the impact of measurement errors or normal system fluctuations on the activity calculation results. Based on this, subtract the total measured activity of the system and the preset activity loss correction value from the theoretical reference total activity of the current step in sequence to obtain the activity consistency difference. This parameter is used to describe the degree of difference between the theoretical activity and the actual measured activity of the system.

[0069] Step 63: By comparing the activity consistency difference with a preset activity consistency difference judgment threshold, it can be determined whether there is a real loss of radioactivity activity. When the activity consistency difference reaches or exceeds the judgment threshold, it indicates that the actual measured radioactivity activity in the system is significantly lower than the theoretical reference activity, thus determining that there is a real loss of radioactivity activity in the system. This usually means that radioactive material may have leaked or diffused from the system's process path. When the activity consistency difference is lower than the judgment threshold, it is considered that the radioactivity activity in the system is basically consistent with the theoretical reference value, thus determining that there is no real loss of radioactivity activity. After obtaining the radioactivity loss judgment result, combined with the system's combined state and the priority anomaly source segment, the final monitoring result can be formed. For example, when the system is in a normal liquid anomaly state but no radioactivity loss has occurred, it can be judged as a general liquid anomaly event; while when the system is in a high-risk state of radioactive leakage and radioactivity loss exists simultaneously, it can be judged as a radioactive leakage event.

[0070] Through the above processing steps, this embodiment enables coordinated handling of anomaly response and activity verification during the operation of the radioactive separation system. On one hand, by using a system-wide state-driven automatic response strategy, system operation can be quickly restricted upon detection of anomalies, thereby reducing the risk of radioactive material diffusion. On the other hand, by comparing the total measured activity of the system with the theoretical reference total activity, it can be determined whether radioactive material has been lost from the system, thus identifying actual radioactive leaks. This method comprehensively analyzes radiation measurement data from multiple monitoring areas to monitor changes in the overall radioactive activity of the system, and combines this with liquid anomaly detection results to form a radioactive separation leak prevention monitoring mechanism based on multi-sensor data. This mechanism can not only identify liquid anomalies and radioactive migration anomalies, but also determine whether a real leak of radioactive material has occurred through activity conservation verification, thereby improving the safety monitoring capabilities and automated handling level of the radioactive separation system.

[0071] This invention also provides a radioactive separation leak prevention monitoring system based on multi-sensor data, comprising:

[0072] The reference model construction module is used to obtain information on the liquid path layout and process division of the radioactive separation system, divide the monitoring area of ​​the radioactive separation system, and establish a process-based reference model for each monitoring area under each process step.

[0073] The baseline calibration initialization module is used to perform background counting measurements and baseline initialization based on each monitoring area, and obtain radiation background count values, initial liquid level baseline values, initial pressure baseline values, and initial liquid accumulation detection baseline values.

[0074] The state vector generation module is used to synchronously collect the original radiation measurement value, real-time liquid level value, real-time pressure value and real-time liquid accumulation detection value based on the step-based reference model and the current step state, and determine the net radiation measurement value to generate a unified spatiotemporal state vector.

[0075] The liquid path anomaly determination module is used to determine the amount of liquid path anomaly based on the spatiotemporal unified state vector, the step-by-step reference model, and the initial baseline value of liquid accumulation detection, and to obtain the liquid event determination result.

[0076] The trajectory anomaly determination module is used to determine the actual normalized activity distribution ratio, activity migration distortion, and activity overrun in the forbidden zone based on the spatiotemporal unified state vector and the step-based reference model, and to obtain the radioactive trajectory anomaly determination result.

[0077] The joint analysis and localization module is used to obtain the joint state of the system and determine the priority anomaly source segment based on the liquid event judgment results, the radioactive trajectory anomaly judgment results, the liquid path anomaly quantity and the net radiation measurement value.

[0078] The disposal verification output module is used to perform automatic disposal based on the system joint state, priority abnormal source segments, step-by-step reference model and spatiotemporal unified state vector, and to determine the total measured activity of the system, the judgment result of radioactivity loss and the monitoring result.

[0079] It should be noted that the range and threshold size are set for ease of comparison. The size of the threshold depends on the amount of sample data and the number of bases set by those skilled in the art for each set of sample data, as long as it does not affect the ratio between the parameter and the quantized value.

[0080] The embodiments of the present invention have been described above, but the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms based on the guidance of the present embodiments, all of which are within the protection scope of the present embodiments.

Claims

1. A method for monitoring radioactive separation and leakage prevention based on multi-sensor data, characterized in that, Includes the following steps: Step 1: Obtain the liquid path layout and process division information of the radioactive separation system, divide the monitoring area of ​​the radioactive separation system, and establish a process-based reference model for each monitoring area under each process step, including: Step 11: Based on the liquid path layout of the radioactive separation system, the radioactive separation system is divided into a sample introduction area, a separation inlet area, a separation main body area, a product collection area, a waste liquid outlet area, a containment detection area, and a non-process area inside the shell, thus obtaining each monitoring area; Step 12: Configure radiation measurement units, liquid level detection units, pressure detection units and liquid accumulation detection units according to each monitoring area, and divide the radioactive separation process into preparation, sample loading, entry into the separation section, residence, elution, product collection, waste liquid discharge and final emptying according to the process step division information. Step 13: For each monitoring area under each step, determine the expected activity distribution ratio, liquid level reference value and pressure reference value respectively, and associate and record the expected activity distribution ratio, liquid level reference value and pressure reference value corresponding to each monitoring area under each step to form a step-based reference model for each monitoring area under each step. Step 2: Based on each monitoring area, perform background counting measurements and baseline initialization to obtain radiation background count values, initial liquid level baseline values, initial pressure baseline values, and initial liquid accumulation detection baseline values; Step 3: Based on the step-by-step reference model and the current step state, simultaneously collect the original radiation measurement value, real-time liquid level value, real-time pressure value and real-time liquid accumulation detection value, determine the net radiation measurement value, and generate a unified spatiotemporal state vector; Step 4: Based on the spatiotemporal unified state vector, the step-by-step reference model, and the initial baseline value of liquid accumulation detection, determine the abnormal quantity of the liquid path and obtain the liquid event judgment result; Step 5: Based on the spatiotemporal unified state vector and the step-by-step reference model, determine the actual normalized activity distribution ratio, activity migration distortion, and activity overrun in the forbidden zone to obtain the radioactive trajectory anomaly judgment result. Step 6: Based on the liquid event determination results, the radioactive trajectory anomaly determination results, the abnormal liquid path quantity, and the net radiation measurement value, obtain the system joint state and determine the priority anomaly source segment; Step 7: Based on the system joint state, priority anomaly source segments, step-by-step reference model, and spatiotemporal unified state vector, perform automatic processing to determine the total measured activity of the system, the judgment result of radioactivity loss, and the monitoring result.

2. The radioactive separation and leakage prevention monitoring method based on multi-sensor data according to claim 1, characterized in that, Based on the step-by-step reference model and the current step state, the original radiation measurement values, real-time liquid level values, real-time pressure values, and real-time liquid accumulation detection values ​​are collected simultaneously, and the net radiation measurement value is determined to generate a unified spatiotemporal state vector, including: Step 21: Based on the current step status, simultaneously collect the original radiation measurement values, real-time liquid level values, real-time pressure values, and real-time liquid accumulation detection values ​​of each monitoring area at the current sampling time; where the current step status indicates the step in which the sampling is performed at the current sampling time. Step 22: Based on the original radiation measurement value and radiation background count value of each monitoring area, subtract the corresponding radiation background count value from the original radiation measurement value of each monitoring area to obtain the net radiation measurement value of each monitoring area; Step 23: Based on the current sampling time, integrate the net radiation measurement value, real-time liquid level value, real-time pressure value, real-time liquid accumulation detection value and the current working state of each monitoring area to generate a unified spatiotemporal state vector.

3. The radioactive separation leak prevention monitoring method based on multi-sensor data according to claim 1, characterized in that, Based on the spatiotemporal unified state vector, the step-by-step reference model, and the initial baseline value for liquid accumulation detection, the abnormal quantity of the liquid path is determined, and the liquid event judgment result is obtained, including: Step 31: Based on the unified spatiotemporal state vector, extract the real-time values ​​of liquid level, pressure, liquid accumulation detection, and current process status of each monitoring area, and determine the reference values ​​of liquid level and pressure of each monitoring area from the process-based reference model based on the current process status. Step 32: Based on the real-time values ​​of liquid level, pressure, liquid accumulation detection, liquid level reference, pressure reference, and initial baseline value of liquid accumulation detection in each monitoring area, determine the absolute values ​​of liquid level deviation, pressure deviation, and liquid accumulation detection deviation, and then multiply the three absolute values ​​of deviation by the corresponding weighting coefficients and sum them to obtain the abnormal liquid flow in each monitoring area. Step 33: Based on the abnormal liquid flow amount in each monitoring area and the corresponding abnormal liquid flow amount judgment threshold, compare the abnormal liquid flow amount in each monitoring area with the corresponding abnormal liquid flow amount judgment threshold; when the abnormal liquid flow amount in a certain monitoring area reaches or exceeds the corresponding abnormal liquid flow amount judgment threshold, determine that a liquid event has occurred and identify the monitoring area as an abnormal monitoring area; when the abnormal liquid flow amount in each monitoring area is lower than the corresponding abnormal liquid flow amount judgment threshold, determine that no liquid event has occurred, and obtain the liquid event judgment result.

4. The radioactive separation leak prevention monitoring method based on multi-sensor data according to claim 1, characterized in that, Based on a unified spatiotemporal state vector and a step-by-step reference model, the actual normalized activity distribution ratio, activity migration distortion, and activity exceeding the forbidden zone boundary are determined, resulting in the determination of radioactive trajectory anomalies, including: Step 41: Based on the unified spatiotemporal state vector, extract the net radiation measurement value and the current step state of each monitoring area, and determine the expected activity distribution ratio of each monitoring area and the set of prohibited activity monitoring areas for the current step from the step-based reference model based on the current step state. Step 42: The ratio of the net radiation measurement value of each monitoring area to the sum of the net radiation measurement values ​​of all monitoring areas is determined as the actual normalized activity distribution ratio of each monitoring area. The absolute value of the deviation between the actual normalized activity distribution ratio of each monitoring area and the corresponding expected activity distribution ratio is multiplied by the activity importance weight corresponding to each monitoring area and then summed to obtain the activity migration distortion. Step 43: Based on the prohibited activity monitoring area set of the current step, sum the net radiation measurement values ​​of each monitoring area in the prohibited activity monitoring area set to obtain the prohibited area activity overrun amount, and compare the activity migration distortion amount with the activity migration distortion amount judgment threshold, and compare the prohibited area activity overrun amount with the prohibited area activity overrun amount judgment threshold to obtain the radioactive trajectory anomaly judgment result.

5. The radioactive separation leak prevention monitoring method based on multi-sensor data according to claim 4, characterized in that, In step 43, when the activity migration distortion reaches or exceeds the activity migration distortion threshold, or when the activity overrun in the prohibited area reaches or exceeds the prohibited area activity overrun threshold, a radioactive trajectory anomaly is determined to have occurred; when the activity migration distortion is below the activity migration distortion threshold and the activity overrun in the prohibited area is below the prohibited area activity overrun threshold, a radioactive trajectory anomaly is determined not to have occurred.

6. The radioactive separation leak prevention monitoring method based on multi-sensor data according to claim 1, characterized in that, Based on the liquid event determination results, radioactive trajectory anomaly determination results, liquid path anomaly quantities, and net radiation measurements, the system's joint state is obtained, and priority anomaly source segments are identified, including: Step 51: Based on the liquid event determination result and the radioactive trajectory anomaly determination result, a combined determination is made. If no liquid event or radioactive trajectory anomaly occurs, the combined system state is determined to be normal. If a liquid event occurs but no radioactive trajectory anomaly occurs, the combined system state is determined to be ordinary liquid anomaly. If no liquid event occurs but a radioactive trajectory anomaly occurs, the combined system state is determined to be radioactive path mismigration. If both a liquid event and a radioactive trajectory anomaly occur, the combined system state is determined to be a high-risk state of radioactive leakage. Step 52: Based on the set of prohibited activity monitoring zones for the current step, assign a prohibited activity identifier value to each monitoring zone; when a monitoring zone belongs to the set of prohibited activity monitoring zones for the current step, set the prohibited activity identifier value of that monitoring zone to one; when a monitoring zone does not belong to the set of prohibited activity monitoring zones for the current step, set the prohibited activity identifier value of that monitoring zone to zero, and establish a zone-by-zone positioning input relationship based on the abnormal liquid flow, net radiation measurement value, and prohibited activity identifier value of each monitoring zone. Step 53: Based on the abnormal liquid flow quantity, net radiation measurement value, step prohibited activity indicator value, corresponding liquid flow abnormality threshold and corresponding radiation threshold of each monitoring area, multiply the ratio of the abnormal liquid flow quantity to the corresponding liquid flow abnormality threshold of each monitoring area by the liquid flow abnormality location weight, multiply the ratio of the net radiation measurement value to the corresponding radiation threshold of each monitoring area by the radiation abnormality location weight, multiply the step prohibited activity indicator value by the step prohibited activity attribute location weight, and sum the above three results to obtain the zoning location score of each monitoring area, and determine the monitoring area with the largest zoning location score as the priority abnormality source segment.

7. The radioactive separation leak prevention monitoring method based on multi-sensor data according to claim 1, characterized in that, Based on the system's joint state, priority anomaly source segments, step-by-step reference model, and spatiotemporal unified state vector, automatic processing is performed to determine the system's total measured activity, radioactivity activity loss assessment results, and monitoring results, including: Step 61: Perform automatic handling based on the system joint status and priority anomaly source segment. When the system joint status is ordinary liquid anomaly, freeze the progress of the process and suspend subsequent liquid input. When the system joint status is radioactive path mismigration, suspend the switching of the current separation process. When the system joint status is high risk of radioactive leakage, stop the injection and cut off the upstream and downstream liquid paths associated with the priority anomaly source segment. Step 62: Extract the net radiation measurement values ​​of each monitoring area based on the spatiotemporal unified state vector, and sum the net radiation measurement values ​​of each monitoring area to obtain the total measured activity of the system; call the theoretical reference total activity of the current step and the preset activity loss correction value of the current step based on the step-based reference model, and subtract the total measured activity of the system and the preset activity loss correction value of the current step from the theoretical reference total activity of the current step in turn to obtain the activity consistency difference; Step 63: Based on the comparison between the activity consistency difference and the activity consistency difference judgment threshold, when the activity consistency difference reaches or exceeds the activity consistency difference judgment threshold, it is determined that there is a real loss of radioactivity; when the activity consistency difference is lower than the activity consistency difference judgment threshold, it is determined that there is no real loss of radioactivity. The monitoring results are obtained by combining the system joint status, priority abnormal source segment and radioactivity loss judgment results.

8. A radioactive separation leak prevention monitoring system based on multi-sensor data, characterized in that, The radioactive separation leak prevention monitoring method based on multi-sensor data as described in any one of claims 2-7 includes: The reference model construction module is used to obtain information on the liquid path layout and process division of the radioactive separation system, divide the monitoring area of ​​the radioactive separation system, and establish a process-based reference model for each monitoring area under each process step. The baseline calibration initialization module is used to perform background counting measurements and baseline initialization based on each monitoring area, and obtain radiation background count values, initial liquid level baseline values, initial pressure baseline values, and initial liquid accumulation detection baseline values. The state vector generation module is used to synchronously collect the original radiation measurement value, real-time liquid level value, real-time pressure value and real-time liquid accumulation detection value based on the step-based reference model and the current step state, and determine the net radiation measurement value to generate a unified spatiotemporal state vector. The liquid path anomaly determination module is used to determine the amount of liquid path anomaly based on the spatiotemporal unified state vector, the step-by-step reference model, and the initial baseline value of liquid accumulation detection, and to obtain the liquid event determination result. The trajectory anomaly determination module is used to determine the actual normalized activity distribution ratio, activity migration distortion, and activity overrun in the forbidden zone based on the spatiotemporal unified state vector and the step-based reference model, and to obtain the radioactive trajectory anomaly determination result. The joint analysis and localization module is used to obtain the joint state of the system and determine the priority anomaly source segment based on the liquid event judgment results, the radioactive trajectory anomaly judgment results, the liquid path anomaly quantity and the net radiation measurement value. The disposal verification output module is used to perform automatic disposal based on the system joint state, priority abnormal source segments, step-by-step reference model and spatiotemporal unified state vector, and to determine the total measured activity of the system, the judgment result of radioactivity loss and the monitoring result.

Images