Real-time data-based at-211 separation system start-stop adaptive control method and system

Abstract

The application provides an At-211 separation system start-stop adaptive control method and system based on real-time data, and relates to the technical field of data processing. The method comprises the following steps: determining the migration start time and migration convergence time of a nuclide by positioning activity change data, extracting the activity change amplitude and evaporation change amplitude, calculating the matching degree between the activity change amplitude and the evaporation change amplitude in sections, obtaining the migration stability judgment quantity, marking the data section in the activity change data as a deviation fluctuation section, deleting the deviation fluctuation section, extracting the continuous convergence length and the switching node jump amplitude, calculating the synchronization degree between the channel switching time and the migration convergence state, obtaining the termination locking quantity, determining the stop action node and the start action node, and obtaining the adaptive control sequence. The application can accurately control the start and stop of the separation system and effectively improve the stability and accuracy of the separation process.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to an adaptive start-stop control method and system for the At-211 separation system based on real-time data. Background Technology

[0002] In the At-211 separation process for nuclear medicine preparation, existing technologies typically achieve start-up and shutdown control by real-time acquisition of separation system operating data, such as evaporation chamber temperature, transport gas flow rate, trapping end pressure, and radioactivity monitoring values. Based on the acquired real-time status parameters, the heating power, valve opening and closing sequence, and transport time are dynamically adjusted to establish nuclide migration conditions during the start-up phase and to switch the trapping path and shut down the system during the shutdown phase, thus completing the adaptive start-up and shutdown control of the At-211 separation process.

[0003] However, in actual radiopharmaceutical preparation scenarios, existing start-stop control systems typically rely on data from a single sensor as the key criterion for judgment. For example, the stop operation may be triggered solely based on the rising activity trend at the capture end. When the target loading is high or gas remains in the pipeline, the activity signal may fluctuate briefly. The system may misjudge the separation as complete and prematurely shut down the transport pathway, which may result in some At-211 remaining in the vaporization chamber or pipeline before it has fully migrated, leading to the problem of radioactive residues during the start-stop phase being difficult to recover. Summary of the Invention

[0004] The purpose of this invention is to provide an adaptive start-stop control method and system for the At-211 separation system based on real-time data, aiming to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0006] Firstly, an adaptive start-stop control method for the At-211 separation system based on real-time data, the method comprising:

[0007] Acquire activity change data, evaporation status data, and pathway switching data of the separation system;

[0008] By locating the boundary positions of continuous rising segments and continuous flat segments in the activity change data, the migration start time and migration convergence time of the nuclide are determined, and the migration response segment data are obtained.

[0009] Based on migration response segment data and evaporation state data, the activity change amplitude and evaporation change amplitude are extracted, and the two are combined into a response correlation sequence in time order to obtain migration-driven correlation data.

[0010] Based on migration-driven correlation data, the degree of matching between the activity change range and the evaporation change range is calculated segment by segment to obtain the migration stability factor.

[0011] Based on the migration stability threshold, data segments in the activity change data whose migration stability threshold is lower than the preset stability threshold are marked as offset fluctuation segments, and the offset fluctuation segments are deleted to obtain the purification migration data.

[0012] Based on the purification migration data and pathway switching data, the continuous convergence length and switching node jump amplitude are extracted, the synchronization degree between the pathway switching time and the migration convergence state is calculated, and the termination lock amount is obtained.

[0013] An adaptive control sequence is obtained by determining the node whose termination locking amount meets the preset locking threshold as the stop action node and the initial node of the offset fluctuation segment as the start action node.

[0014] Furthermore, by locating the boundary positions of continuously rising and continuously flat segments in the activity change data, the start and convergence times of nuclide migration are determined, yielding migration response segment data, including:

[0015] Based on the activity change data, the difference between the activity values ​​at adjacent sampling times is calculated and arranged in chronological order to form an activity change difference sequence, thus obtaining the differential activity data;

[0016] By filtering out the intervals of continuously positive sampling points in the differential activity data and merging the temporally continuous positive differential intervals, a set of continuously rising segments is obtained.

[0017] By filtering the sampling point intervals in the differential activity data whose variation amplitude remains within a preset smooth range, and merging the temporally continuous smooth intervals, a set of continuous smooth segments is obtained.

[0018] Based on the set of continuously rising segments and the set of continuously flat segments, the adjacent connection positions of the continuously rising segments and the continuously flat segments on the time axis are determined, and the boundary nodes are extracted as the time intervals for the start time and the convergence time of the migration, thus obtaining the migration response segment data.

[0019] Furthermore, based on the migration response segment data and evaporation state data, the activity change amplitude and evaporation change amplitude are extracted, and the two are combined in chronological order to form a response correlation sequence, resulting in migration-driven correlation data, including:

[0020] Based on the migration response segment data, the activity time range of the migration response segment is identified, and activity change data and evaporation state data are acquired and synchronized within the activity time range to obtain the synchronization time segment dataset.

[0021] Based on the synchronous time segment dataset, the difference between the activity change and the evaporation state change between adjacent sampling points is calculated to obtain the activity amplitude change sequence and the evaporation amplitude change sequence.

[0022] Based on the activity amplitude change sequence and the evaporation amplitude change sequence, compare the ratio of the two amplitude changes at each time point, and calculate the ratio of the two amplitude changes to obtain the amplitude correlation score sequence.

[0023] Based on the magnitude correlation score sequence, identify the time segments with high correlation in the magnitude correlation score sequence to obtain migration-driven association data.

[0024] Furthermore, based on migration-driven correlation data, the degree of matching between the activity change amplitude and the evaporation change amplitude is calculated segment by segment to obtain migration stability parameters, including:

[0025] Based on migration-driven correlation data, the time derivative of activity change is calculated to obtain the activity change rate term; the time derivative of evaporation state is calculated to obtain the evaporation change rate term.

[0026] Based on the activity change data and evaporation state data, the degree of synergy between the two during the migration process is calculated to obtain the proportional relationship term; the degree of synergy between the activity change rate term and the evaporation change rate term is calculated to obtain the matching degree term.

[0027] The matching degree item and the proportional relationship item are fused together to calculate the overall stability score and obtain the stability score item. Based on the stability score item, the matching degree and stability of the activity and evaporation state of the system in different time periods are calculated to obtain the migration stability factor.

[0028] Furthermore, based on the migration stabilization threshold, data segments in the activity change data with migration stabilization thresholds lower than the preset stabilization zone threshold are marked as offset fluctuation segments, and these offset fluctuation segments are deleted to obtain purified migration data, including:

[0029] By comparing the migration stability assessment value with a preset stability assessment threshold, data segments with migration stability assessment values ​​lower than the stability assessment threshold are filtered out and marked as low-stability segments.

[0030] By examining the temporal relationship between low-stability segments, adjacent low-stability segments are merged to obtain a set of offset fluctuation segments.

[0031] The fluctuation intensity value is obtained by calculating the duration and amplitude of the offset fluctuation segment, and high-intensity fluctuation segments are selected based on the fluctuation intensity value to obtain a set of high offset fluctuation segments;

[0032] Based on the high-offset fluctuation segment set, the corresponding time period data is deleted from the activity change data, and the stable migration process data segment is retained to obtain the purification migration data.

[0033] Furthermore, based on the purification migration data and pathway switching data, the continuous convergence length and switching node jump amplitude are extracted, and the synchronization degree between the pathway switching time and the migration convergence state is calculated to obtain the termination lock amount, including:

[0034] Based on the purification migration data, the duration of stability maintained by the system after reaching a steady state during the migration process is calculated, resulting in the convergence duration term; based on the path switching data, the degree of drastic change experienced by the system at the switching node is calculated, resulting in the jump amplitude term.

[0035] Based on the convergence duration term and the jump amplitude term, the time synchronization degree between the convergence state and the switching moment is calculated, and the time synchronization term is obtained.

[0036] Based on the time synchronization term, the degree of time synchronization between all convergence segments and switching nodes is calculated, and the timing of stopping during the migration process is comprehensively evaluated to obtain the termination lock amount.

[0037] Furthermore, by determining the node whose termination lock amount meets the preset lock threshold as the stop action node, and determining the initial node of the offset fluctuation segment as the start action node, an adaptive control sequence is obtained, including:

[0038] By comparing the termination lock amount with a preset lock threshold, nodes whose termination lock amount is higher than the preset lock threshold during the migration process are filtered out and marked as stop action nodes;

[0039] Based on the offset fluctuation segment, identify the initial node of the offset fluctuation segment and mark the initial node as the initiation action node;

[0040] Sort the nodes by time based on the stop action nodes and start action nodes, with the stop action nodes first and the start action nodes last, to obtain the time sequence node set;

[0041] Based on the time sequence node set, calculate the time interval between adjacent start action nodes and stop action nodes, and count the number of each time interval value within a fixed time interval to obtain the node interval distribution set;

[0042] Based on the node interval distribution set, calculate the ratio of the number of nodes in each fixed time interval to the length of the fixed time interval to obtain the node density value, and arrange the node density values ​​in descending order to form a node execution order list;

[0043] Based on the node execution order table, the fixed time interval with the highest node density value is used as the base interval. The start action nodes and stop action nodes in the base interval are grouped, and the execution time of the stop action node in each group is limited to be later than the execution time of the start action node, thus obtaining the grouped node set.

[0044] An adaptive control sequence is obtained by sequentially outputting start and stop control signals according to the time order of each group in the group node set, and by smoothly correcting the time interval between adjacent groups.

[0045] Secondly, an adaptive start-stop control system for the At-211 separation system based on real-time data, the system comprising:

[0046] The data module is used to acquire activity change data, evaporation status data, and pathway switching data of the separation system;

[0047] The migration response module is used to determine the start and convergence times of nuclide migration by locating the boundary positions of continuous rising segments and continuous flat segments in the activity change data, thereby obtaining migration response segment data.

[0048] The driving correlation module is used to extract the activity change amplitude and evaporation change amplitude based on the migration response segment data and evaporation state data, and to combine the two into a response correlation sequence in chronological order to obtain migration driving correlation data;

[0049] The migration stability determination module is used to calculate the degree of matching between the activity change range and the evaporation change range segment by segment based on migration-driven correlation data, and obtain the migration stability determination quantity.

[0050] The purification migration module is used to mark data segments in the activity change data whose migration stability judgment value is lower than the preset stability judgment zone threshold as offset fluctuation segments based on the migration stability judgment value, and delete the offset fluctuation segments to obtain purification migration data.

[0051] The termination lock module is used to extract the continuous convergence length and the jump amplitude of the switching node based on the purification migration data and the pathway switching data, calculate the degree of synchronization between the pathway switching time and the migration convergence state, and obtain the termination lock amount.

[0052] The control module is used to determine the node whose termination locking amount meets the preset locking threshold as the stop action node and the initial node of the offset fluctuation segment as the start action node to obtain an adaptive control sequence.

[0053] The above-described solution of the present invention has at least the following beneficial effects:

[0054] This invention locates the boundary positions of continuously rising and continuously flat segments in activity change data, determining the start and convergence times of migration. This transforms the original activity time series into segmented data objects with clear start and end boundaries, enabling the control system to map continuous monitoring signals to process stage identifiers in data processing. All subsequent calculations are confined to the migration response segment, thus forming stage constraints in the start-stop control logic. This prevents data from non-migration stages from being included in the same judgment criteria. At the same time, determining the start and convergence times of migration allows the system to output time node parameters corresponding to the nuclide migration process, providing a staged index for subsequent correlation sequence construction, stability calculation, and stop action node generation.

[0055] This invention extracts the activity change amplitude and evaporation change amplitude based on migration response segment data and evaporation state data, and assembles them into a response correlation sequence in chronological order. It establishes a correspondence between driving variables and response variables at the data processing layer, so that activity change is no longer treated as an isolated output signal, but forms a synchronous amplitude comparison sequence with evaporation state change. A data structure reflecting the coupling relationship between migration drive and migration response is generated within the control system. This structure provides the input basis for subsequent matching degree calculation and enables the control system to express the correlation trajectory between evaporation drive change and nuclide migration output on the time axis.

[0056] This invention obtains the migration stability judgment quantity by calculating the matching degree between the activity change amplitude and the evaporation change amplitude segment by segment. The migration-driven correlation data is further converted into segmented stability measurement parameters, so that the control system can obtain a quantitative judgment basis for local segments of the migration process. This migration stability judgment quantity serves as the sole input condition for subsequent fluctuation segment identification, enabling the start-stop control logic to achieve the mapping from correlation sequence to judgment index in data processing. At the same time, the segmented calculation mechanism enables different stages of the migration process to output corresponding stability judgment results respectively.

[0057] This invention marks data segments with migration stability judgment values ​​below a preset stability judgment threshold as offset fluctuation segments and deletes these segments. This forms a segment filtering and abnormal segment removal operation in the data processing link, so that segments in the original activity sequence that are inconsistent with the stability judgment conditions are marked separately and removed from the subsequent calculation dataset, generating purified migration data that only contains the migration stability stage. This purified migration data serves as the input for subsequent termination locking calculation, avoiding interference from offset fluctuation segments in the termination node determination. At the same time, this step realizes the structured cleaning and reconstruction of the process quantity sequence within the control system.

[0058] This invention extracts the continuous convergence length and the jump amplitude of the switching node based on purification migration data and path switching data, calculates the degree of synchronization, and obtains the termination lock value. At the data processing layer, the migration convergence state parameters and the execution path event parameters are synchronized and analyzed, so that the stopping judgment criterion is expanded from a single activity convergence trend to a time consistency measure between the convergence duration characteristics and the path switching event. The termination lock value serves as the input of the stopping action node, enabling the control system to output a termination judgment index corresponding to the migration convergence state. At the same time, the extraction of the continuous convergence length and jump amplitude enables the system to express the synchronization relationship between the process convergence segment and the execution event at the data layer. Attached Figure Description

[0059] Figure 1 This is a flowchart of the At-211 separation system start-stop adaptive control method based on real-time data provided in an embodiment of the present invention. Detailed Implementation

[0060] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0061] like Figure 1 As shown, embodiments of the present invention propose an adaptive start-stop control method for the At-211 separation system based on real-time data, the method comprising:

[0062] Acquire activity change data, evaporation status data, and pathway switching data of the separation system;

[0063] By locating the boundary positions of continuous rising segments and continuous flat segments in the activity change data, the migration start time and migration convergence time of the nuclide are determined, and the migration response segment data are obtained.

[0064] Based on migration response segment data and evaporation state data, the activity change amplitude and evaporation change amplitude are extracted, and the two are combined into a response correlation sequence in time order to obtain migration-driven correlation data.

[0065] Based on migration-driven correlation data, the degree of matching between the activity change range and the evaporation change range is calculated segment by segment to obtain the migration stability factor.

[0066] Based on the migration stability threshold, data segments in the activity change data whose migration stability threshold is lower than the preset stability threshold are marked as offset fluctuation segments, and the offset fluctuation segments are deleted to obtain the purification migration data.

[0067] Based on the purification migration data and pathway switching data, the continuous convergence length and switching node jump amplitude are extracted, the synchronization degree between the pathway switching time and the migration convergence state is calculated, and the termination lock amount is obtained.

[0068] An adaptive control sequence is obtained by determining the node whose termination locking amount meets the preset locking threshold as the stop action node and the initial node of the offset fluctuation segment as the start action node.

[0069] In this embodiment of the invention, activity change data, evaporation state data, and pathway switching data of the separation system are acquired to avoid misjudgment and bias caused by relying on a single data source, ensuring that the influence of multiple factors on the separation process can be comprehensively considered. By locating the boundary positions of continuous rising segments and continuous flat segments in the activity change data, the migration start time and migration convergence time of the nuclide are determined, and migration response segment data are obtained, effectively eliminating the interference of data noise in the non-migration stage on subsequent steps, ensuring more accurate monitoring of the migration process. Based on the migration response segment data and evaporation state data, the activity change amplitude and evaporation change amplitude are extracted, and the two are combined into a response correlation sequence in chronological order to obtain migration-driven correlation data, which can capture the driving effect of the evaporation process on nuclide migration, avoiding control errors caused by relying on only a single monitoring quantity. Based on the migration-driven correlation data, the matching degree between the activity change amplitude and the evaporation change amplitude is calculated segment by segment to obtain the migration stability judgment quantity, ensuring that the system can sensitively reflect the changing trend in the migration process and avoid problems such as false stop.

[0070] Based on the migration stability threshold, data segments in the activity change data where the migration stability threshold is lower than the preset stability threshold are marked as offset fluctuation segments and deleted to obtain purified migration data. This preserves valid migration process data and avoids control errors caused by noise interference. Based on the purified migration data and path switching data, the continuous convergence length and switching node jump amplitude are extracted. The synchronization degree between the path switching time and the migration convergence state is calculated to obtain the termination lock value. This avoids erroneous shutdowns caused by asynchrony between the switching event and the migration convergence state, ensuring accurate selection of the stop action node and preventing premature or delayed stop operations. By determining the node whose termination lock value meets the preset lock threshold as the stop action node and the initial node of the offset fluctuation segment as the start action node, an adaptive control sequence is obtained. This sequence can flexibly respond to changes under different operating conditions, improve the system's control accuracy and response capability, and ensure the smooth operation of the At-211 separation process.

[0071] This includes acquiring activity change data, evaporation state data, and pathway switching data of the separation system, specifically including:

[0072] The system acquires multidimensional data related to the At-211 separation process in real time through precise sensor configuration and data acquisition modules to ensure comprehensive monitoring and accurate control of the separation process. First, activity change data is acquired using high-precision radionuclide detectors. These detectors are deployed at key locations in the separation system, such as the evaporation chamber inlet, outlet, and trapping end. The detectors collect radioactivity signals from various regions of the system in real time and convert them into digital signals, which are then input into the control system. This data reflects the nuclide migration during the separation process; by monitoring this activity data in real time, it can be determined whether At-211 is migrating as expected.

[0073] Evaporation status data is acquired through multiple temperature and pressure sensors. The temperature and pressure within the evaporation chamber are critical parameters in the nuclide migration process, especially temperature changes, which directly affect the evaporation efficiency and separation rate of At-211. Therefore, sensors installed around the evaporation chamber continuously monitor internal temperature and pressure changes and record this data in real time. This real-time data allows the system to determine whether the evaporation chamber is operating normally, thus supporting subsequent control decisions. Temperature sensors are typically placed at multiple locations within the evaporation chamber to obtain accurate temperature distribution in that area, while pressure sensors monitor pressure changes within the chamber to ensure that the evaporation process is not affected by excessively high or low pressures that could impair nuclide migration.

[0074] The path switching data is provided by multiple gas flow sensors and valve position sensors in the control system. These sensors monitor changes in the airflow path and valve opening / closing status in the separation system in real time. Specifically, the system records the open / closed position of each valve and changes in gas flow rate during separation. For example, if a separation path needs to be switched or the flow rate adjusted during migration, the system will adjust the flow rate and direction based on feedback signals from the valve sensors. Changes in gas flow rate directly affect not only separation efficiency but also the migration path of the At-211; therefore, precise monitoring and control of path switching are required at every critical moment throughout the process.

[0075] The real-time acquisition and precise transmission of this sensor data ensures that the system can obtain a complete picture of the separation process at any time, providing a reliable basis for subsequent data analysis and control decisions. The precision and real-time performance of the data acquisition system are the foundation for the efficient start-stop adaptive control of this invention. It enables the system to dynamically adjust heating power, valve opening and closing sequence, and gas flow rate during processing, thereby ensuring the accuracy and safety of At-211 separation.

[0076] In a preferred embodiment of the present invention, by locating the boundary positions of continuously rising segments and continuously flat segments in the activity change data, the migration start time and migration convergence time of the nuclide are determined, and migration response segment data are obtained, including:

[0077] Based on the activity change data, the difference between the activity values ​​at adjacent sampling times is calculated and arranged in chronological order to form an activity change difference sequence, thus obtaining the differential activity data;

[0078] By filtering out the intervals of continuously positive sampling points in the differential activity data and merging the temporally continuous positive differential intervals, a set of continuously rising segments is obtained.

[0079] By filtering the sampling point intervals in the differential activity data whose variation amplitude remains within a preset smooth range, and merging the temporally continuous smooth intervals, a set of continuous smooth segments is obtained.

[0080] Based on the set of continuously rising segments and the set of continuously flat segments, the adjacent connection positions of the continuously rising segments and the continuously flat segments on the time axis are determined, and the boundary nodes are extracted as the time intervals for the start time and the convergence time of the migration, thus obtaining the migration response segment data.

[0081] In this embodiment of the invention, based on activity change data, the difference between activity values ​​at adjacent sampling times is calculated and arranged in chronological order to obtain a differential activity change sequence, resulting in differential activity data. This allows for a focused capture of instantaneous changes in activity, ensuring more targeted analysis of signal fluctuations. By filtering out continuously positive sampling point intervals in the differential activity data and merging temporally continuous positive difference intervals, a continuously rising segment set is obtained, avoiding overly fine segment division due to excessively small fluctuations and ensuring data stability and coherence. Furthermore, by filtering out differential activity data where the change amplitude remains within a preset smooth range... The sampling point intervals are divided into continuous and smooth intervals to obtain a set of continuous smooth segments. This avoids including stable segments unrelated to the nuclide migration process in the judgment range and ensures that the stopping judgment is based on a stable migration process rather than occasional background fluctuations. Based on the set of continuous rising segments and the set of continuous smooth segments, the adjacent connection positions of the continuous rising segments and the continuous smooth segments on the time axis are determined, and the boundary nodes are extracted as the time intervals for the start and convergence times of migration to obtain migration response segment data. This ensures that the data segments that truly reflect the dynamics of nuclide migration can be extracted from the activity change sequence, guaranteeing the accurate execution of the control logic.

[0082] Specifically, by filtering sampling point intervals in the differential activity data whose variation amplitude remains within a preset smooth range, and merging temporally continuous smooth intervals, a set of continuous smooth segments is obtained, including:

[0083] First, the system calculates differential activity data from the activity change data. This dataset represents the change in activity values ​​between adjacent sampling points. Next, the system filters this differential data according to a preset flattening range. A flattening range is typically defined as a small fluctuation range, indicating that activity changes do not fluctuate significantly within a certain time period. The system iterates through the differential activity data, identifying data points with small fluctuations within the flattening range, and groups these data points together to form flattening intervals. For data segments that continuously meet the flattening criteria, the system automatically merges them into a large flattening segment. This means that if the differential values ​​at multiple time points are within the flattening range, they will be considered a continuous flattening interval. The merging operation is based on time order; that is, if the fluctuation range within a certain time period consistently remains within the set flattening range, these time periods will be merged into a single set of flattening segments.

[0084] Specifically, based on the sets of continuously rising segments and continuously flat segments, the adjacent connection positions of continuously rising segments and continuously flat segments on the time axis are determined, and boundary nodes are extracted as time intervals to define the migration start time and migration convergence time, thus obtaining migration response segment data, which specifically includes:

[0085] The system needs to determine the effective time interval of the nuclide migration process by analyzing the temporal relationship between consecutive rising segments and consecutive flat segments. First, based on the set of consecutive rising segments obtained in the previous stage, the system determines which rising segments' start times are adjacent to the end times of flat segments. Rising segments typically represent the initiation phase of the nuclide migration process, while flat segments represent the end or plateau phase. The system determines the connection points between consecutive rising and flat segments by examining their adjacent positions on the time axis. Extracting these connection points provides clear boundaries for subsequent migration start and convergence times. The migration start time typically corresponds to the start time of the first rising segment, while the migration convergence time is typically the end time of the last flat segment. These two time nodes define the effective time interval of the migration process, marking the beginning and end of nuclide migration. By extracting these boundary nodes, the system distinguishes the effective time period from irrelevant phases in the migration process, ensuring that subsequent control operations focus only on the actual process of nuclide migration and are not interfered with by other flat or irrelevant data. This operation avoids including irrelevant stable phases in the analysis, providing a clear time frame and basis for subsequent process control, judgment, and optimization.

[0086] In a preferred embodiment of the present invention, based on migration response segment data and evaporation state data, the activity change amplitude and evaporation change amplitude are extracted, and the two are arranged in chronological order to form a response correlation sequence, thereby obtaining migration-driven correlation data, including:

[0087] Based on the migration response segment data, the activity time range of the migration response segment is identified, and activity change data and evaporation state data are acquired and synchronized within the activity time range to obtain the synchronization time segment dataset.

[0088] Based on the synchronous time segment dataset, the difference between the activity change and the evaporation state change between adjacent sampling points is calculated to obtain the activity amplitude change sequence and the evaporation amplitude change sequence.

[0089] Based on the activity amplitude change sequence and the evaporation amplitude change sequence, compare the ratio of the two amplitude changes at each time point, and calculate the ratio of the two amplitude changes to obtain the amplitude correlation score sequence.

[0090] Based on the magnitude correlation score sequence, identify the time segments with high correlation in the magnitude correlation score sequence to obtain migration-driven association data.

[0091] In this embodiment of the invention, based on the migration response segment data, the active time range of the migration response segment is identified, and activity change data and evaporation state data are acquired and synchronized within the active time range to obtain a synchronized time segment dataset. This ensures the time alignment of activity change and evaporation state data, preventing inaccurate calculation results due to deviations between different data sources. Based on the synchronized time segment dataset, the difference between the activity change and evaporation state change between adjacent sampling points is calculated to obtain an activity amplitude change sequence and an evaporation amplitude change sequence. This clearly reflects the changing trend between activity and evaporation state during nuclide migration, providing effective input data for subsequent processing. Based on the activity amplitude change sequence and the evaporation amplitude change sequence, the ratio of their change amplitudes at each time point is compared, and the ratio of the two change amplitude ratios is calculated to obtain an amplitude correlation score sequence. This reveals the correlation strength between the two, quantifies the driver-response relationship during migration, and identifies the effectiveness and stability of the driving process. Based on the amplitude correlation score sequence, highly correlated time segments in the amplitude correlation score sequence are identified to obtain migration-driven correlation data. This effectively removes irrelevant fluctuation data that may be caused by system noise or interference, ensuring the accuracy and reliability of the system's judgment criteria.

[0092] Specifically, based on the synchronous time segment dataset, the differences between the activity change and the evaporation state change between adjacent sampling points are calculated to obtain the activity amplitude change sequence and the evaporation amplitude change sequence, which include:

[0093] First, activity change data and evaporation state data are compared point-by-point in chronological order, and the differences between adjacent sampling points are calculated. During this process, for each pair of adjacent sampling points, the system calculates the difference in activity change and evaporation state change between these two points. Activity change refers to the difference in activity values ​​between adjacent sampling points, and evaporation state change refers to the difference in evaporation state values ​​between adjacent sampling points. In calculating these changes, the system uses a precise difference calculation method, that is, obtaining the change at each time point by subtracting the value of the previous sampling point. The system stores these changes as two independent sequences: an activity amplitude change sequence and an evaporation amplitude change sequence. These two sequences represent the magnitude and trend of activity change and evaporation state change, respectively. Through these data sequences, the system can clearly depict the specific changes in activity and evaporation state over time during nuclide migration, providing basic data for subsequent data processing steps.

[0094] Specifically, based on the activity amplitude change sequence and the evaporation amplitude change sequence, the ratio of the two amplitude changes at each time point is compared, and the ratio of the two amplitude changes is calculated to obtain the amplitude correlation score sequence, which includes:

[0095] The system divides the change in activity amplitude by the change in evaporation amplitude at each time point to obtain the ratio value for that time point. Through this calculation, the system can derive a ratio reflecting the relative relationship between activity change and evaporation state change. The ratio can be calculated by dividing the activity change amplitude by the evaporation change amplitude, or vice versa, depending on the system's preset calculation method. During the calculation, a larger ratio indicates that the activity change is more significant relative to the evaporation state change; conversely, a smaller ratio may indicate that the evaporation change has a greater impact on nuclide migration. The ratios generated at each time point are recorded, generating an amplitude ratio sequence. This sequence will reflect the relative strength between activity change and evaporation state change, as well as their consistency or differences throughout the migration process.

[0096] The system further analyzes the changes in these ratio values ​​and generates an amplitude correlation score sequence for each time point. The amplitude correlation score sequence reflects the strength of the relationship between changes in activity and evaporation state. This score sequence is generated based on the trend and stability of the ratio values. A high amplitude correlation score indicates a strong coupling between the driving and responding factors in the migration process when the amplitude ratios of activity and evaporation state maintain high consistency. Conversely, a low amplitude correlation score indicates a weak relationship or significant interference between the two factors when the ratio values ​​fluctuate significantly. This calculation process quantifies the synchronicity and consistency between activity and evaporation state during nuclide migration.

[0097] In a preferred embodiment of the present invention, based on migration-driven correlation data, the degree of matching between the activity change amplitude and the evaporation change amplitude is calculated segment by segment to obtain the migration stability determination factor, including:

[0098] Based on migration-driven correlation data, the time derivative of activity change is calculated to obtain the activity change rate term; the time derivative of evaporation state is calculated to obtain the evaporation change rate term.

[0099] Based on the activity change data and evaporation state data, the degree of synergy between the two during the migration process is calculated to obtain the proportional relationship term; the degree of synergy between the activity change rate term and the evaporation change rate term is calculated to obtain the matching degree term.

[0100] The matching degree item and the proportional relationship item are fused together to calculate the overall stability score and obtain the stability score item. Based on the stability score item, the matching degree and stability of the activity and evaporation state of the system in different time periods are calculated to obtain the migration stability factor.

[0101] In this embodiment of the invention, based on migration-driven correlation data, the time derivative of activity change is calculated to obtain the activity change rate term, which clearly reflects the actual state of the migration process and avoids misjudgment due to sudden data fluctuations or short-term changes. The time derivative of evaporation state is calculated to obtain the evaporation change rate term, reflecting the fluctuations in temperature and flow rate and their impact on the migration process, distinguishing the relationship between evaporation changes caused by equipment fluctuations and the actual nuclide migration process. Based on activity change data and evaporation state data, the degree of synergy between the two in the migration process is calculated to obtain a proportional relationship term, revealing the synchronicity between the two, helping to identify the true migration pattern and potential interference, and ensuring... The accuracy of start-up and shutdown operations is ensured; the degree of coordinated change between the activity change rate and evaporation change rate terms is calculated to obtain the matching degree term, revealing the dynamic matching effect during the migration process and effectively avoiding misjudgments caused by random fluctuations or equipment errors; the matching degree term and the proportional relationship term are integrated to calculate the overall stability score, obtaining the stability score term, clearly distinguishing the stable and unstable segments during the migration process, and improving the system's adaptability to dynamic changes; based on the stability score term, the matching degree and stability of the system's activity and evaporation state in different time periods are calculated to obtain the migration stability judgment quantity, identify the convergence and stable states during the migration process, and ensure the accurate determination of the stopping node.

[0102] In a preferred embodiment of the present invention, based on the migration stability assessment value, data segments in the activity change data whose migration stability assessment value is lower than a preset stability assessment threshold are marked as offset fluctuation segments, and the offset fluctuation segments are deleted to obtain purified migration data, including:

[0103] By comparing the migration stability assessment value with a preset stability assessment threshold, data segments with migration stability assessment values ​​lower than the stability assessment threshold are filtered out and marked as low-stability segments.

[0104] By examining the temporal relationship between low-stability segments, adjacent low-stability segments are merged to obtain a set of offset fluctuation segments.

[0105] The fluctuation intensity value is obtained by calculating the duration and amplitude of the offset fluctuation segment, and high-intensity fluctuation segments are selected based on the fluctuation intensity value to obtain a set of high offset fluctuation segments;

[0106] Based on the high-offset fluctuation segment set, the corresponding time period data is deleted from the activity change data, and the stable migration process data segment is retained to obtain the purification migration data.

[0107] In this embodiment of the invention, by comparing the migration stability assessment value with a preset stability assessment threshold, data segments with migration stability assessment values ​​lower than the stability assessment threshold are filtered out and marked as low-stability segments. This effectively identifies fluctuating or unstable data segments that may exist during system operation, avoiding interference from abnormal fluctuations in the final decision to stop the system. By examining the temporal relationship between low-stability segments, adjacent low-stability segments are merged to obtain a set of offset fluctuation segments. This prevents local small fluctuations from misleading the judgment of the overall migration status and ensures the accuracy of key data segments during the migration process. By calculating the duration and amplitude of the offset fluctuation segments, the fluctuation intensity value is obtained, and high-intensity fluctuation segments are filtered out based on the fluctuation intensity value to obtain a set of high-offset fluctuation segments. This effectively distinguishes significant fluctuations that have an important impact on the migration process, avoiding the impact of weak and insignificant fluctuations on system decisions. Based on the set of high-offset fluctuation segments, the corresponding time period data is deleted from the activity change data, retaining stable migration process data segments to obtain purified migration data. This ensures that the data processed by the system only contains valid migration process data, making subsequent judgments more accurate and reliable.

[0108] Specifically, by examining the temporal relationships between low-stability segments, adjacent low-stability segments are merged to obtain a set of offset fluctuation segments, which includes:

[0109] First, the activity change data acquired by the system needs to be carefully analyzed. The migration stability indicators calculated by the system reflect the migration stability within each time period. When these stability indicators are lower than a preset stability threshold, they are considered low-stability segments. Next, the system will detect the temporal relationship between these low-stability segments, that is, evaluate the time interval between adjacent segments. By setting a predetermined time interval threshold, the system will check the time difference between adjacent low-stability segments. If the time interval between adjacent low-stability segments is less than the preset threshold, the system will determine that the two segments are related and can be classified into a whole offset fluctuation segment. Once the temporal relationship between low-stability segments is checked, the system will automatically merge these adjacent low-stability segments into an offset fluctuation segment. The offset fluctuation segment set contains these merged data segments, representing the abnormal fluctuations that occurred during the migration process. The core of this process is to judge the characteristics of fluctuations through time intervals and continuity, avoiding misjudging small fluctuations as independent anomalies, and instead merging them into larger fluctuation segments, thereby reducing the complexity of data processing and improving the efficiency of data processing.

[0110] Specifically, the fluctuation intensity value is obtained by calculating the duration and amplitude of the offset fluctuation segment, and high-intensity fluctuation segments are selected based on the fluctuation intensity value to obtain a set of high-offset fluctuation segments, which specifically includes:

[0111] The system calculates the fluctuation intensity value by measuring the duration and amplitude of these fluctuation segments. It measures the span of each offset fluctuation segment along the time axis based on its duration and combines this with the amplitude changes in the activity data to arrive at a numerical value for fluctuation intensity. The fluctuation intensity value is an indicator that integrates the time dimension and amplitude changes, reflecting the severity of each fluctuation segment. A larger fluctuation intensity value indicates a more significant impact of the fluctuation segment on the migration process. Based on the fluctuation intensity value, the system filters all offset fluctuation segments, extracting those with higher intensity to form a high-offset fluctuation segment set. To filter high-intensity fluctuation segments, the system sets a fluctuation intensity threshold; all fluctuation segments with intensity values ​​greater than this threshold are considered high-offset fluctuation segments. These high-offset fluctuation segments typically indicate some abnormal state during the migration process, such as large-amplitude fluctuations caused by equipment failure, gas stagnation, or other system anomalies. The final high-offset fluctuation segment set is a collection containing all significant abnormal fluctuation data.

[0112] In a preferred embodiment of the present invention, based on the purification migration data and the path switching data, the continuous convergence length and the switching node jump amplitude are extracted, the synchronization degree between the path switching time and the migration convergence state is calculated, and the termination lock amount is obtained, including:

[0113] Based on the purification migration data, the duration of stability maintained by the system after reaching a steady state during the migration process is calculated, resulting in the convergence duration term; based on the path switching data, the degree of drastic change experienced by the system at the switching node is calculated, resulting in the jump amplitude term.

[0114] Based on the convergence duration term and the jump amplitude term, the time synchronization degree between the convergence state and the switching moment is calculated, and the time synchronization term is obtained.

[0115] Based on the time synchronization term, the degree of time synchronization between all convergence segments and switching nodes is calculated, and the timing of stopping during the migration process is comprehensively evaluated to obtain the termination lock amount.

[0116] In this embodiment of the invention, based on purification migration data, the duration for which the system remains stable after reaching a steady state during the migration process is calculated, resulting in a convergence duration term. This accurately identifies and calculates the length of time the system remains stable in a steady state, preventing the system from being mistakenly judged as convergent before fully reaching a steady state, thus avoiding unnecessary stop operations. Based on path switching data, the degree of drastic change experienced by the system at the switching node is calculated, resulting in a jump amplitude term. This accurately captures the drastic changes in the system caused by the switching node, ensuring the accuracy of the switching operation and the assessment of its impact on process control, avoiding accidental stops or unnecessary interventions. Based on the convergence duration term and the jump amplitude term... The time synchronization degree between the convergence state and the switching moment is calculated to obtain the time synchronization term. This quantifies the time synchronization relationship between the convergence state and the switching event, providing a quantitative standard for determining whether a stop operation should be performed. This ensures that the stop operation is triggered at the correct time point, preventing the system from erroneously stopping or continuing to run due to time lag or prematureness. Based on the time synchronization term, the time synchronization degree between all convergence segments and the switching node is calculated. The timing of the stop during the migration process is comprehensively evaluated to obtain the termination lock quantity. This ensures the accuracy of the stop timing and avoids premature or delayed stop due to poor synchronization, ensuring the efficient and reliable completion of the At-211 separation process.

[0117] In a preferred embodiment of the present invention, by determining the node whose termination locking amount meets a preset locking threshold as the stop action node, and determining the initial node of the offset fluctuation segment as the start action node, an adaptive control sequence is obtained, including:

[0118] By comparing the termination lock amount with a preset lock threshold, nodes whose termination lock amount is higher than the preset lock threshold during the migration process are filtered out and marked as stop action nodes;

[0119] Based on the offset fluctuation segment, identify the initial node of the offset fluctuation segment and mark the initial node as the initiation action node;

[0120] Sort the nodes by time based on the stop action nodes and start action nodes, with the stop action nodes first and the start action nodes last, to obtain the time sequence node set;

[0121] Based on the time sequence node set, calculate the time interval between adjacent start action nodes and stop action nodes, and count the number of each time interval value within a fixed time interval to obtain the node interval distribution set;

[0122] Based on the node interval distribution set, calculate the ratio of the number of nodes in each fixed time interval to the length of the fixed time interval to obtain the node density value, and arrange the node density values ​​in descending order to form a node execution order list;

[0123] Based on the node execution order table, the fixed time interval with the highest node density value is used as the base interval. The start action nodes and stop action nodes in the base interval are grouped, and the execution time of the stop action node in each group is limited to be later than the execution time of the start action node, thus obtaining the grouped node set.

[0124] An adaptive control sequence is obtained by sequentially outputting start and stop control signals according to the time order of each group in the group node set, and by smoothly correcting the time interval between adjacent groups.

[0125] In this embodiment of the invention, by comparing the termination lock value with a preset lock threshold, nodes whose termination lock value exceeds the preset lock threshold during the migration process are selected and marked as stop action nodes. This accurately identifies that the migration process has completely stabilized and converged, effectively avoiding premature stopping and ensuring accurate stopping timing, ensuring that the stop action is executed after the migration is completed. Based on the offset fluctuation segment, the initial node of the offset fluctuation segment is identified and marked as the start action node, ensuring that the system can start in time when offset or abnormal fluctuation is detected, avoiding premature shutdown due to misjudgment of migration completion, and avoiding automatic shutdown of the system before the migration is fully completed. Based on the time order of the stop action nodes and start action nodes, the stop action nodes are arranged first and the start action nodes are arranged last, resulting in a time sequence node set, ensuring that the operations in the control sequence have a clear and accurate time sequence relationship, avoiding time misalignment between nodes. Based on the time sequence node set, the time interval value between adjacent start action nodes and stop action nodes is calculated, and the number of each time interval value within a fixed time interval is counted to obtain a node interval distribution set, avoiding resource waste and system burden caused by frequent start and stop actions, ensuring a smoother and more efficient control process. The system calculates the node density value by dividing the number of nodes in each fixed time interval by the length of that interval. This node density value is then arranged in descending order to form a node execution sequence list. This allows for the rational scheduling of start and stop operations within each time interval, prioritizing high-density intervals, avoiding chaotic node execution, and preventing resource waste caused by frequent control actions in certain time periods. Based on the node execution sequence list, the fixed time interval with the highest node density value is used as the baseline interval. Start and stop action nodes within the baseline interval are grouped, with the execution time of stop action nodes in each group being later than that of start action nodes. This results in a grouped node set, avoiding time conflicts or errors between start and stop operations. Furthermore, smooth corrections are performed between adjacent operations to prevent unnecessary resource consumption caused by excessive start-stop switching, improving system control stability and response accuracy. By sequentially outputting start and stop control signals according to the time order of each group in the grouped node set, and smoothing the time interval between adjacent groups, an adaptive control sequence is obtained. This enables the system to maintain precise and efficient operation under dynamically changing working conditions, maximizing the use of system resources and ensuring smooth transitions and eventual stable operation.

[0126] Specifically, based on the node interval distribution set, the ratio of the number of nodes in each fixed time interval to the length of that fixed time interval is calculated to obtain the node density value. The node density values ​​are then arranged in descending order to form a node execution order list, which includes:

[0127] First, the system calculates the time intervals for all sorted start and stop action nodes. These nodes are arranged in chronological order and paired one by one, and the time interval between each pair of adjacent nodes is calculated. This time difference is recorded and organized into a node interval distribution set. The node interval distribution set reflects the temporal relationship between the nodes, providing basic data for further analysis. After obtaining the node interval distribution set, the system calculates the ratio of the number of nodes in each fixed time interval to the length of that time interval. This ratio is the node density value, indicating the frequency of start or stop control actions within that interval. The higher the node density value, the more frequent the control actions within that time period; the lower the node density value, the fewer the control actions within that time period. Then, the system sorts the nodes according to their density values ​​in descending order, generating a node execution sequence table. This node execution sequence table sorts the fixed time intervals according to their node density values, providing a basis for subsequent control signal output.

[0128] Specifically, based on the node execution order table, a fixed time interval with the highest node density value is used as the baseline interval. Start-up and stop-up nodes within the baseline interval are grouped, and the execution time of the stop-up nodes within each group is limited to be later than the execution time of the start-up nodes, resulting in a grouped node set, which specifically includes:

[0129] The system selects the time interval with the highest node density as the baseline interval and groups nodes within this interval. All start and stop action nodes within the baseline interval are extracted and grouped according to their chronological order. Within each group, the system ensures that the execution time of stop action nodes is later than that of start action nodes, and that control operations within the same group follow the order of start-then-stop. It also ensures that start and stop actions are not executed in reverse order within each time period, thus avoiding system control logic errors. In this way, all start and stop action nodes are assigned to different time periods, ensuring the correct control order within each time period. After node grouping, the system obtains a set of grouped nodes containing start and stop nodes from multiple time intervals. These nodes have been allocated and organized chronologically and meet the logical requirements of start and stop operations.

[0130] Specifically, by sequentially outputting start and stop control signals according to the time order of each group in the grouped node set, and smoothing the time interval between adjacent groups, an adaptive control sequence is obtained, which includes:

[0131] The system outputs start and stop control signals from the grouped node sets sequentially according to time. For each group, a start control signal is output first, followed by a stop control signal at an appropriate time. This process ensures that each start signal is output before its corresponding stop signal, maintaining the consistency and order of the control logic. Next, the system smooths the time intervals between adjacent groups to ensure a smoother transition between control signals. Specifically, the purpose of smoothing is to adjust for excessively large or small time intervals between adjacent control signals, preventing the system from reacting too quickly or too slowly, leading to instability or unnecessary operational fluctuations. Smoothing is typically performed using smoothing algorithms such as weighted averaging and interpolation. These algorithms help adjust uneven time intervals, enabling the system to execute control tasks within a reasonable and stable time sequence. Through this smoothing, the generated adaptive control sequence becomes more stable and fluid, ensuring that each control action of the system is performed at an appropriate rhythm and sequence, improving the accuracy and stability of the entire control process.

[0132] Embodiments of the present invention also provide an adaptive start-stop control system for the At-211 separation system based on real-time data, the system comprising:

[0133] The data module is used to acquire activity change data, evaporation status data, and pathway switching data of the separation system;

[0134] The migration response module is used to determine the start and convergence times of nuclide migration by locating the boundary positions of continuous rising segments and continuous flat segments in the activity change data, thereby obtaining migration response segment data.

[0135] The driving correlation module is used to extract the activity change amplitude and evaporation change amplitude based on the migration response segment data and evaporation state data, and to combine the two into a response correlation sequence in chronological order to obtain migration driving correlation data;

[0136] The migration stability determination module is used to calculate the degree of matching between the activity change range and the evaporation change range segment by segment based on migration-driven correlation data, and obtain the migration stability determination quantity.

[0137] The purification migration module is used to mark data segments in the activity change data whose migration stability judgment value is lower than the preset stability judgment zone threshold as offset fluctuation segments based on the migration stability judgment value, and delete the offset fluctuation segments to obtain purification migration data.

[0138] The termination lock module is used to extract the continuous convergence length and the jump amplitude of the switching node based on the purification migration data and the pathway switching data, calculate the degree of synchronization between the pathway switching time and the migration convergence state, and obtain the termination lock amount.

[0139] The control module is used to determine the node whose termination locking amount meets the preset locking threshold as the stop action node and the initial node of the offset fluctuation segment as the start action node to obtain an adaptive control sequence.

[0140] It should be noted that this system is a system corresponding to the above method. All implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0141] Embodiments of the present invention also provide a computing device, including: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, performs the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0142] Embodiments of the present invention also provide a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to perform the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0143] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. An adaptive start-stop control method for the At-211 separation system based on real-time data, characterized in that, The method includes: Acquire activity change data, evaporation status data, and pathway switching data of the separation system; By locating the boundary positions of continuous rising segments and continuous flat segments in the activity change data, the migration start time and migration convergence time of the nuclide are determined, and the migration response segment data are obtained. Based on migration response segment data and evaporation state data, the activity change amplitude and evaporation change amplitude are extracted, and the two are combined into a response correlation sequence in time order to obtain migration-driven correlation data. Based on migration-driven correlation data, the degree of matching between the activity change range and the evaporation change range is calculated segment by segment to obtain the migration stability determination quantity; Based on the migration stabilization threshold, data segments in the activity change data with migration stabilization thresholds lower than the preset stabilization zone threshold are marked as offset fluctuation segments, and these offset fluctuation segments are deleted to obtain the purified migration data, including: By comparing the migration stability assessment value with a preset stability assessment threshold, data segments with migration stability assessment values ​​lower than the stability assessment threshold are filtered out and marked as low-stability segments. By examining the temporal relationship between low-stability segments, adjacent low-stability segments are merged to obtain a set of offset fluctuation segments. The fluctuation intensity value is obtained by calculating the duration and amplitude of the offset fluctuation segment, and high-intensity fluctuation segments are selected based on the fluctuation intensity value to obtain a set of high offset fluctuation segments; Based on the high-offset fluctuation segment set, the corresponding time period data is deleted from the activity change data, and the stable migration process data segment is retained to obtain the purification migration data. Based on the purification migration data and pathway switching data, the continuous convergence length and switching node jump amplitude are extracted. The synchronization degree between the pathway switching time and the migration convergence state is calculated to obtain the termination lock value, including: Based on the purification migration data, the duration of stability maintained by the system after reaching a steady state during the migration process is calculated, resulting in the convergence duration term; based on the path switching data, the degree of drastic change experienced by the system at the switching node is calculated, resulting in the jump amplitude term. Based on the convergence duration term and the jump amplitude term, the time synchronization degree between the convergence state and the switching moment is calculated, and the time synchronization term is obtained. Based on the time synchronization term, calculate the degree of time synchronization between all convergence segments and switching nodes, comprehensively evaluate the stopping timing during the migration process, and obtain the termination lock amount; An adaptive control sequence is obtained by determining the node whose termination locking amount meets the preset locking threshold as the stop action node and the initial node of the offset fluctuation segment as the start action node.

2. The adaptive start-stop control method for the At-211 separation system based on real-time data according to claim 1, characterized in that, By locating the boundary positions of continuously rising and continuously level segments in the activity change data, the migration start and convergence times of the nuclide are determined, yielding migration response segment data, including: Based on the activity change data, the difference between the activity values ​​at adjacent sampling times is calculated and arranged in chronological order to form an activity change difference sequence, thus obtaining the differential activity data; By filtering out the intervals of continuously positive sampling points in the differential activity data and merging the temporally continuous positive differential intervals, a set of continuously rising segments is obtained. By filtering the sampling point intervals in the differential activity data whose variation amplitude remains within a preset smooth range, and merging the temporally continuous smooth intervals, a set of continuous smooth segments is obtained. Based on the set of continuously rising segments and the set of continuously flat segments, the adjacent connection positions of the continuously rising segments and the continuously flat segments on the time axis are determined, and the boundary nodes are extracted as the time intervals for the start time and the convergence time of the migration, thus obtaining the migration response segment data.

3. The adaptive start-stop control method for the At-211 separation system based on real-time data according to claim 2, characterized in that, Based on migration response segment data and evaporation state data, the activity change amplitude and evaporation change amplitude are extracted, and the two are combined in chronological order to form a response correlation sequence, resulting in migration-driven correlation data, including: Based on the migration response segment data, the activity time range of the migration response segment is identified, and activity change data and evaporation state data are acquired and synchronized within the activity time range to obtain the synchronization time segment dataset. Based on the synchronous time segment dataset, the difference between the activity change and the evaporation state change between adjacent sampling points is calculated to obtain the activity amplitude change sequence and the evaporation amplitude change sequence. Based on the activity amplitude change sequence and the evaporation amplitude change sequence, compare the ratio of the two amplitude changes at each time point, and calculate the ratio of the two amplitude changes to obtain the amplitude correlation score sequence. Based on the magnitude correlation score sequence, identify the time segments with high correlation in the magnitude correlation score sequence to obtain migration-driven association data.

4. The adaptive start-stop control method for the At-211 separation system based on real-time data according to claim 3, characterized in that, Based on migration-driven correlation data, the degree of matching between the activity change amplitude and the evaporation change amplitude is calculated segment by segment to obtain migration stability parameters, including: Based on migration-driven correlation data, the time derivative of activity change is calculated to obtain the activity change rate term; the time derivative of evaporation state is calculated to obtain the evaporation change rate term. Based on the activity change data and evaporation state data, the degree of synergy between the two during the migration process is calculated to obtain the proportional relationship term; the degree of synergy between the activity change rate term and the evaporation change rate term is calculated to obtain the matching degree term. The matching degree item and the proportional relationship item are fused together to calculate the overall stability score and obtain the stability score item. Based on the stability score item, the matching degree and stability of the activity and evaporation state of the system in different time periods are calculated to obtain the migration stability factor.

5. The adaptive start-stop control method for the At-211 separation system based on real-time data according to claim 4, characterized in that, By defining the node whose termination lock value meets the preset lock threshold as the stop action node and defining the initial node of the offset fluctuation segment as the start action node, an adaptive control sequence is obtained, including: By comparing the termination lock amount with a preset lock threshold, nodes whose termination lock amount is higher than the preset lock threshold during the migration process are filtered out and marked as stop action nodes; Based on the offset fluctuation segment, identify the initial node of the offset fluctuation segment and mark the initial node as the initiation action node; Sort the nodes by time based on the stop action nodes and start action nodes, with the stop action nodes first and the start action nodes last, to obtain the time sequence node set; Based on the time sequence node set, calculate the time interval between adjacent start action nodes and stop action nodes, and count the number of each time interval value within a fixed time interval to obtain the node interval distribution set; Based on the node interval distribution set, calculate the ratio of the number of nodes in each fixed time interval to the length of the fixed time interval to obtain the node density value, and arrange the node density values ​​in descending order to form a node execution order list; Based on the node execution order table, the fixed time interval with the highest node density value is used as the base interval. The start action nodes and stop action nodes in the base interval are grouped, and the execution time of the stop action node in each group is limited to be later than the execution time of the start action node, thus obtaining the grouped node set. An adaptive control sequence is obtained by sequentially outputting start and stop control signals according to the time order of each group in the group node set, and by smoothly correcting the time interval between adjacent groups.

6. An adaptive start-stop control system for the At-211 separation system based on real-time data, characterized in that, The system is used to perform the method as described in any one of claims 1 to 5, the system comprising: The data module is used to acquire activity change data, evaporation status data, and pathway switching data of the separation system; The migration response module is used to determine the start and convergence times of nuclide migration by locating the boundary positions of continuous rising segments and continuous flat segments in the activity change data, thereby obtaining migration response segment data. The driving correlation module is used to extract the activity change amplitude and evaporation change amplitude based on the migration response segment data and evaporation state data, and to combine the two into a response correlation sequence in chronological order to obtain migration driving correlation data; The migration stability determination module is used to calculate the degree of matching between the activity change range and the evaporation change range segment by segment based on migration-driven correlation data, and obtain the migration stability determination quantity. The purification migration module is used to mark data segments in the activity change data whose migration stability judgment value is lower than the preset stability judgment zone threshold as offset fluctuation segments based on the migration stability judgment value, and delete the offset fluctuation segments to obtain purification migration data. The termination lock module is used to extract the continuous convergence length and the jump amplitude of the switching node based on the purification migration data and the pathway switching data, calculate the degree of synchronization between the pathway switching time and the migration convergence state, and obtain the termination lock amount. The control module is used to determine the node whose termination locking amount meets the preset locking threshold as the stop action node and the initial node of the offset fluctuation segment as the start action node to obtain an adaptive control sequence.

7. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 5.

Images