At-211 online separation process simulation method and platform based on digital twinning

Abstract

The invention provides an At-211 online separation process simulation method and platform based on digital twinning, relates to the technical field of data processing, and identifies a separation behavior evolution mode based on a data change association relationship in a continuous time interval by collecting process monitoring data reflecting a material distribution state and separation result change in a separation process. And dividing the separation process into a plurality of operation segments. And aiming at different separation behavior evolution modes, respectively constructing digital twin simulation sub-models with different state variable composition modes and state change constraint conditions. In the online simulation process, the corresponding relation between the current separation behavior evolution mode and the adopted digital twinborn simulation sub-model is continuously judged, continuous simulation is maintained when the current separation behavior evolution mode is consistent with the adopted digital twinborn simulation sub-model, existing state continuation is terminated and switched to the matched digital twinborn simulation sub-model when the current separation behavior evolution mode is inconsistent with the adopted digital twinborn simulation sub-model, and the subsequent simulation evolution process is reestablished. Therefore, simulation of the multi-stage operation state of the At-211 online separation process is realized.

Description

A Simulation Method and Platform for the Online Separation Process of At-211 Based on Digital Twins Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a simulation method and platform for the At-211 online separation process based on digital twins. Background Technology

[0002] In existing technologies, digital twin technology has been widely applied to online simulation and optimization of industrial processes, especially in the modeling and real-time monitoring of continuous or discrete processes in chemical engineering and manufacturing. By constructing a simulation model in digital space corresponding to the physical device or process flow, and combining it with historical data and real-time sensor data, a synchronous mapping of the actual operating state can be achieved, thereby enabling process monitoring, operational prediction, and decision support. In the field of separation processes, for processes such as solvent extraction and adsorption separation, there are already digital twin-based systems used for visual monitoring and parameter simulation of separation processes to support operators in analyzing and managing the status of separation processes.

[0003] However, when applying the aforementioned digital twin online simulation technology to continuous separation processes, existing simulation models often struggle to effectively depict the actual separation behavior throughout the entire process, as the material state and separation behavior change with each stage of operation. Particularly when feed composition changes, separation medium gradually becomes saturated, or elution conditions switch, the evolutionary characteristics of the separation process undergo phased transformations. Existing digital twin systems typically continue to simulate the state evolution path formed in the previous separation stage, leading to the continuation of historical states no longer applicable to the current separation behavior characteristics into subsequent simulations. This results in systematic simulation biases at the boundaries between separation stages, affecting the accuracy of the simulation results in reflecting the actual separation process. Summary of the Invention

[0004] The purpose of this invention is to provide a simulation method and platform for the online separation process of At-211 based on digital twins, in order 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: Firstly, a simulation method for the At-211 online separation process based on digital twins, the method comprising: collecting process monitoring data used to characterize the material distribution state and changes in separation results during the At-211 online separation process; analyzing the state evolution characteristics of the separation process based on the correlation of changes in the process monitoring data within a continuous time interval; identifying separation behavior evolution patterns characterizing the evolution law of the separation process's operating state; dividing the At-211 online separation process into multiple separation process operation segments according to different separation behavior evolution patterns, each separation process operation segment corresponding to a separation behavior evolution pattern that remains relatively stable within a predetermined time range; and constructing an At-211 simulation method to describe the separation behavior evolution pattern according to each separation behavior evolution pattern. 11. Digital twin simulation sub-models for separating process state changes. Different digital twin simulation sub-models differ in the composition of process state variables and / or state change constraints. During online digital twin simulation, based on real-time collected process monitoring data, it is continuously determined whether the current separation behavior evolution mode corresponds to the running digital twin simulation sub-model. When they are consistent, the continuous simulation of the separation process state by the digital twin simulation sub-model is maintained. When they are inconsistent, the continuation of the existing process state in the current digital twin simulation sub-model is terminated, and the model is switched to the digital twin simulation sub-model corresponding to the newly determined separation behavior evolution mode. Based on the process state variables and state change constraints in the digital twin simulation sub-model, the process state of the subsequent separation process is re-deduced.

[0006] Preferably, based on the correlation of changes in process monitoring data within a continuous time interval, the state evolution characteristics of the separation process are analyzed, and separation behavior evolution patterns characterizing the evolution law of the separation process's operating state are identified. This includes: collecting multiple process monitoring parameters used to characterize the material distribution state and changes in separation results during the separation process according to a preset sampling period, and sorting the process monitoring parameters according to the collection time to form multi-parameter time series data reflecting the continuous operating state of the separation process; processing the direction, magnitude, and synchronicity of change of each process monitoring parameter within a continuous time interval based on the multi-parameter time series data to obtain joint change characteristics characterizing the coordinated change relationship of each process monitoring parameter within the corresponding time interval; analyzing the state evolution trajectory formed by the change of process state over time during the separation process based on the joint change characteristics to generate a set of state evolution characteristics characterizing the stability and consistency of the state evolution trajectory; distinguishing the state evolution trajectories in different time intervals based on the set of state evolution characteristics, and merging state evolution trajectories with the same state evolution characteristics into the same separation behavior evolution pattern to form a separation behavior evolution pattern identification result characterizing the evolution law of the separation process's operating state.

[0007] Preferably, for each separation behavior evolution mode, a digital twin simulation sub-model is constructed to describe the state change relationship of the At-211 separation process under that separation behavior evolution mode. Different digital twin simulation sub-models differ in the composition of process state variables and / or the constraints on state changes. This includes: based on a identified separation behavior evolution mode, analyzing the participation of each process monitoring parameter in the state evolution trajectory corresponding to that separation behavior evolution mode on the state changes of the separation process; selecting process monitoring parameters that match that separation behavior evolution mode; and mapping the selected process monitoring parameters to corresponding process state variables to form a process state variable. A set of process state variables is generated. Based on the set of process state variables, the process state variables are organized according to their roles and mutual influences in the separation process, and a state variable organization structure is constructed to describe the state transfer relationships between process state variables. Based on the state variable organization structure, state change constraints that limit the order of change, change boundaries, and mutual constraints of process state variables are determined, forming state change rules applicable to the evolution mode of this separation behavior. A digital twin simulation sub-model corresponding to the evolution mode of this separation behavior is constructed from the set of process state variables, the state variable organization structure, and the state change constraints.

[0008] Preferably, when the two are inconsistent, the continuation of the existing process state in the current digital twin simulation sub-model is terminated, and the model is switched to the digital twin simulation sub-model corresponding to the newly determined separation behavior evolution mode. Based on the process state variables and their state change constraints in the digital twin simulation sub-model, the process state of the subsequent separation process is re-deduced, including: after determining that the separation behavior evolution mode corresponding to the current separation process is inconsistent with the running digital twin simulation sub-model, the continuation of the deduction of the separation process state of the subsequent time step based on the digital twin simulation sub-model is stopped, and the transmission of the process state already formed in the digital twin simulation sub-model to the subsequent simulation process is blocked; according to the newly determined separation behavior evolution mode, the digital twin simulation sub-model corresponding to the separation behavior evolution mode is selected as the digital twin simulation sub-model used for the online simulation of the subsequent separation process; based on the real-time collected process monitoring data, the process state variables in the selected digital twin simulation sub-model are initialized, and under the constraints of the corresponding state change, the process state of the subsequent separation process is deduced step by step, thereby re-deducing the subsequent separation process.

[0009] Preferably, based on the joint change characteristics, the state evolution trajectory formed by the change of process state over time during the separation process is analyzed to generate a set of state evolution features to characterize the stability and consistency of the state evolution trajectory. This includes: connecting the process states in continuous time intervals according to the joint change characteristics in chronological order to form a state evolution trajectory composed of multiple continuous process state points, where each process state point corresponds to the value result of the joint change characteristics in the corresponding time interval; analyzing the change relationship between adjacent process state points segment by segment according to the state evolution trajectory to determine the degree of continuity of the change of process state in adjacent time intervals, generating a continuity analysis result to characterize the local continuity of the state evolution trajectory; analyzing the overall change direction and degree of consistency of the state evolution trajectory in multiple continuous time intervals according to the continuity analysis result, generating a trend consistency analysis result to characterize the overall change law of the state evolution trajectory; and generating stability features to characterize the stability of the state evolution trajectory and consistency features to characterize the consistency of the state evolution trajectory according to the continuity analysis result and the trend consistency analysis result, and combining the stability features and consistency features to form a set of state evolution features.

[0010] Preferably, based on the set of state evolution features, the state evolution trajectories within different time intervals are distinguished, and state evolution trajectories with the same state evolution features are grouped into the same separation behavior evolution pattern, forming a separation behavior evolution pattern recognition result used to characterize the state evolution law of the separation process. This includes: aligning the stability and consistency features of each state evolution trajectory according to the corresponding set of state evolution features within different time intervals to form a feature description result used to characterize the feature composition of each state evolution trajectory; comparing the stability and consistency features between different state evolution trajectories one by one according to the feature description result to determine the consistency relationship in feature composition between each state evolution trajectory, generating a feature consistency judgment result between state evolution trajectories; grouping state evolution trajectories with the same feature composition into the same trajectory set according to the feature consistency judgment result, with each trajectory set corresponding to a state evolution law; determining each trajectory set as a separation behavior evolution pattern, and establishing a correspondence between the separation behavior evolution pattern and the corresponding set of state evolution features to form a separation behavior evolution pattern recognition result.

[0011] Preferably, based on the identified separation behavior evolution pattern, the participation of each process monitoring parameter in the state evolution trajectory corresponding to the separation behavior evolution pattern on the state changes of the separation process is analyzed. Process monitoring parameters matching the separation behavior evolution pattern are selected, and the selected process monitoring parameters are mapped to corresponding process state variables to form a set of process state variables. This includes: extracting the parameter change trajectory of each process monitoring parameter during the formation of the state evolution trajectory according to the state evolution trajectory corresponding to the separation behavior evolution pattern, forming parameter change description results corresponding to each process monitoring parameter; analyzing the correspondence between each process monitoring parameter and the state changes of the separation process during the formation of the state evolution trajectory according to the parameter change description results, generating participation analysis results to characterize the degree of participation of each process monitoring parameter in the state changes of the separation process; screening process monitoring parameters that play a major characterizing role in the state changes of the separation process under the separation behavior evolution pattern according to the participation analysis results, forming a set of candidate process monitoring parameters; and mapping each process monitoring parameter in the candidate process monitoring parameter set to corresponding process state variables according to its physical meaning and functional role in the separation process, forming a set of process state variables corresponding to the separation behavior evolution pattern.

[0012] Secondly, a digital twin-based At-211 online separation process simulation platform includes: a separation behavior evolution pattern recognition module, used to collect process monitoring data characterizing the material distribution state and separation result changes during the At-211 online separation process, and analyze the state evolution characteristics of the separation process based on the correlation of changes in process monitoring data within a continuous time interval, identifying separation behavior evolution patterns characterizing the evolution law of the separation process operation state; a running segment division module, used to divide the At-211 online separation process into multiple separation process running segments according to different separation behavior evolution patterns, each separation process running segment corresponding to a separation behavior evolution pattern that remains relatively stable within a predetermined time range; and a digital twin simulation sub-model construction module, used to construct a model describing the At-211 under each separation behavior evolution pattern. The system comprises several digital twin simulation sub-models for separating process state changes. Different digital twin simulation sub-models differ in the composition of process state variables and / or state change constraints. A separation behavior evolution mode determination module, based on real-time process monitoring data collected during online digital twin simulation, continuously determines whether the current separation behavior evolution mode corresponds to the running digital twin simulation sub-model. A simulation control module, when consistent, maintains continuous simulation of the separation process state by the digital twin simulation sub-model; when inconsistent, it terminates the continuation of the existing process state in the current digital twin simulation sub-model, switches to the digital twin simulation sub-model corresponding to the newly determined separation behavior evolution mode, and re-deduces the process state of subsequent separation processes based on the process state variables and state change constraints in this digital twin simulation sub-model.

[0013] The above-mentioned solution of the present invention includes at least the following beneficial effects: First, by introducing a separation behavior evolution pattern recognition method based on the correlation of process monitoring data changes, the digital twin simulation no longer relies solely on a single continuous state model to describe the separation process, but can start from the separation process operation data itself to characterize the evolution characteristics of the material distribution state and the separation result over time during the separation process, thereby providing a basis for distinguishing the states of different operating stages in the continuous separation process.

[0014] Secondly, by dividing the online separation process into multiple separation process segments according to different separation behavior evolution modes, each segment corresponds to a separation behavior feature that remains relatively stable within a predetermined time range. This avoids mixing and processing inapplicable state evolution features from different operation stages, which helps improve the simulation process's ability to adapt to the phased changes in the continuous separation process.

[0015] Furthermore, by constructing digital twin simulation sub-models with different process state variable composition methods and state change constraints for different separation behavior evolution modes, the digital twin simulation sub-models can match the corresponding separation behavior characteristics at the state description level, avoiding the deviation caused by using a single simulation model to uniformly characterize multiple separation operation states, and improving the rationality of expressing the state change relationship of the separation process.

[0016] Furthermore, during the online simulation of the digital twin, by continuously determining the correspondence between the current separation behavior evolution mode and the running digital twin simulation sub-model, when the separation behavior evolution mode changes, the continuation of the existing process state is terminated and switched to the matching digital twin simulation sub-model, and the subsequent simulation evolution process is re-established. This avoids directly inheriting the state evolution path formed in the previous separation stage to the new separation stage, which helps to reduce the systematic simulation deviation at the boundary of the separation stages.

[0017] Through the aforementioned progressive processing method, the online digital twin simulation process can dynamically adjust the simulation evolution process as the separation process operation status changes, thereby improving the simulation results' ability to reflect the actual continuous separation process operation status and facilitating the analysis and understanding of the separation process. Attached Figure Description

[0018] Figure 1 is a flowchart of the At-211 online separation process simulation method based on digital twin provided by an embodiment of the present invention. Detailed Implementation

[0019] 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.

[0020] As shown in Figure 1, an embodiment of the present invention proposes a simulation method for the At-211 online separation process based on digital twins. The method includes: collecting process monitoring data during the At-211 online separation process to characterize the material distribution state and changes in the separation results; analyzing the state evolution characteristics of the separation process based on the correlation of changes in the process monitoring data over a continuous time interval; identifying separation behavior evolution patterns that characterize the evolution law of the separation process's operating state; dividing the At-211 online separation process into multiple separation process operation segments according to different separation behavior evolution patterns, with each segment corresponding to a relatively stable separation behavior evolution pattern within a predetermined time range; and constructing a method to describe the At-211 separation process under each separation behavior evolution pattern. The digital twin simulation sub-models for state change relationships differ from each other in the composition of process state variables and / or state change constraints. During the online digital twin simulation, based on real-time collected process monitoring data, it is continuously determined whether the evolution mode of the separation behavior corresponding to the current separation process corresponds to the running digital twin simulation sub-model. When they are consistent, the continuous simulation of the separation process state by the digital twin simulation sub-model is maintained. When they are inconsistent, the continuation of the existing process state in the current digital twin simulation sub-model is terminated, and the model is switched to the digital twin simulation sub-model corresponding to the newly determined separation behavior evolution mode. Based on the process state variables and their state change constraints in the digital twin simulation sub-model, the process state of the subsequent separation process is re-deduced.

[0021] In this embodiment of the invention, by introducing continuous perception and analysis of the operating status of the At-211 online separation process in the digital space, the simulation process no longer relies on a single continuous state evolution path. Instead, it can characterize the state evolution characteristics of the separation process based on the changes in material distribution and separation results over time, thereby identifying the separation behavior evolution pattern that reflects the evolution law of the separation process's operating status. This enables the simulation process to identify the stage-by-stage changes in the separation process, providing a foundation for subsequent segmented processing in the simulation process.

[0022] Based on the identified different separation behavior evolution patterns, the At-211 online separation process is divided into multiple separation process segments, with each segment corresponding to a relatively stable separation behavior evolution pattern within a predetermined time range. In this way, the originally continuous separation process, which involves internal state transitions, is broken down into multiple operating intervals with consistent evolutionary characteristics. This avoids carrying over historical evolutionary characteristics that are not applicable to the current operating state into subsequent simulations, thereby improving the adaptability of the simulation process to the actual separation process operating state.

[0023] For different separation behavior evolution modes, corresponding digital twin simulation sub-models are constructed to match the corresponding separation behavior evolution mode in terms of the composition of process state variables and state change constraints. By introducing multiple digital twin simulation sub-models with different structures, the simulation process can adopt different state description methods and evolution rules for different separation behavior evolution characteristics, thereby avoiding simulation bias caused by using a single model to uniformly describe all operating states.

[0024] During the online digital twin simulation, the evolution mode of the separation behavior corresponding to the current separation process is continuously determined based on the real-time collected process monitoring data. When the determination result indicates that the current separation process is still in a separation behavior evolution mode that matches the running digital twin simulation sub-model, the existing digital twin simulation sub-model is maintained to continuously simulate the state of the separation process, so that the simulation process can smoothly reflect the state evolution of the separation process within this running segment.

[0025] When the determination result indicates that the evolutionary pattern of the separation behavior corresponding to the current separation process has changed and is inconsistent with the currently running digital twin simulation sub-model, the continuation of the existing process state in the current digital twin simulation sub-model is terminated, and the system switches to the digital twin simulation sub-model corresponding to the newly determined separation behavior evolutionary pattern. Simultaneously, the simulation evolution process for subsequent separation processes is re-established. By reconstructing the simulation state when the separation behavior evolutionary pattern changes, the system avoids directly inheriting the state evolution path formed in the previous running segment to the new running segment, thereby reducing simulation deviations at the boundary of separation stages.

[0026] For example, in the continuous separation process of At-211, when the feed composition changes or the separation medium gradually reaches saturation, the material distribution and separation results exhibit different characteristics over time compared to the initial operating stage. In this embodiment of the invention, through continuous analysis of process monitoring data, the change in the separation behavior evolution mode corresponding to this change can be identified, and the separation process can be divided into different operating segments. In the initial operating segment, a digital twin simulation sub-model matching the initial separation behavior evolution mode is used for state evolution simulation; when the separation behavior evolution mode changes, the corresponding new digital twin simulation sub-model is switched in a timely manner and the simulation evolution process is re-established, so that the entire online simulation process can maintain consistency with the actual operating state of the At-211 separation process at different operating stages, facilitating the analysis and understanding of the separation process.

[0027] In a preferred embodiment of the present invention, the At-211 online separation process is divided into multiple separation process operation segments according to different separation behavior evolution modes. Each separation process operation segment corresponds to a separation behavior evolution mode that remains relatively stable within a predetermined time range. This includes: after identifying the separation behavior evolution mode, scanning the time axis of continuous operation in the online separation process, and using the separation behavior evolution mode corresponding to each moment as the operation status identifier for that moment; merging time intervals in which the separation behavior evolution mode remains consistent across multiple consecutive time points according to time order to form a continuous time period; defining the continuous time period as a separation process operation segment, wherein the separation behavior evolution mode remains unchanged within the separation process operation segment; when a change in the separation behavior evolution mode is detected, ending the current separation process operation segment, and using the time point corresponding to the new separation behavior evolution mode as the starting point of the next separation process operation segment; through the above method, the entire At-211 online separation process is divided into multiple separation process operation segments arranged in chronological order, so that each operation segment corresponds to a stable separation behavior evolution mode.

[0028] In a preferred embodiment of the present invention, during the online digital twin simulation process, based on real-time collected process monitoring data, it is continuously determined whether the evolution mode of the separation behavior corresponding to the current separation process corresponds to the running digital twin simulation sub-model. This includes: during the online digital twin simulation, acquiring the process monitoring data at the current moment in real time according to a preset update cycle; based on the process monitoring data at the current moment, determining the evolution mode of the separation behavior corresponding to the current separation process according to the identification method of the evolution mode of the separation behavior; acquiring the evolution mode of the separation behavior corresponding to the currently running digital twin simulation sub-model; comparing the evolution mode of the separation behavior corresponding to the current separation process with the evolution mode of the separation behavior corresponding to the current digital twin simulation sub-model; when the two are consistent, determining that the current digital twin simulation sub-model is still applicable to the current separation process running state; when the two are inconsistent, determining that the current digital twin simulation sub-model is no longer applicable to the current separation process running state, and triggering subsequent simulation switching processing.

[0029] In a preferred embodiment of the present invention, the predetermined time range is a time interval for determining whether the separation behavior evolution pattern remains stable, including: during the separation behavior evolution pattern identification process, in order to avoid the influence of instantaneous data fluctuations on the separation behavior determination, a minimum duration for determining the stability of the separation behavior evolution pattern is preset; when a certain separation behavior evolution pattern is identified as the same pattern in multiple consecutive time points and the duration reaches the minimum duration, it is determined that the separation behavior evolution pattern remains stable in the corresponding time interval; the time interval in which the duration meets the stability requirement is used as the predetermined time range to form a separation process running segment; when the separation behavior evolution pattern changes and the duration does not reach the minimum duration, a new separation process running segment is not immediately divided, but the determination of subsequent time points continues until the stability requirement is met before the running segment is divided; through the above method, the division of the separation process running segment is based on the stable maintenance of the separation behavior evolution pattern within a certain time range, thereby avoiding frequent switching of running segments due to short-term fluctuations.

[0030] In a preferred embodiment of the present invention, based on the correlation of changes in process monitoring data within a continuous time interval, the state evolution characteristics of the separation process are analyzed, and separation behavior evolution patterns characterizing the evolution law of the separation process's operating state are identified. This includes: collecting multiple process monitoring parameters during the separation process according to a preset sampling period to characterize the changes in material distribution and separation results; sorting the process monitoring parameters according to the collection time to form multi-parameter time series data reflecting the continuous operating state of the separation process; processing the direction, magnitude, and synchronicity of change of each process monitoring parameter within a continuous time interval based on the multi-parameter time series data to obtain joint change characteristics characterizing the coordinated change relationship of each process monitoring parameter within the corresponding time interval; analyzing the state evolution trajectory formed by the change of process state over time during the separation process based on the joint change characteristics to generate a set of state evolution characteristics characterizing the stability and consistency of the state evolution trajectory; and distinguishing the state evolution trajectories within different time intervals based on the set of state evolution characteristics, merging state evolution trajectories with the same state evolution characteristics into the same separation behavior evolution pattern to form a separation behavior evolution pattern identification result characterizing the evolution law of the separation process's operating state.

[0031] In this embodiment of the invention, the state evolution characteristics of the separation process are analyzed layer by layer based on the correlation of changes in process monitoring data over continuous time intervals. This allows the identification of the separation process's operating state to no longer rely on preset process stages or manual experience divisions, but rather to be naturally formed by the synergistic changes of process monitoring parameters over time. By constructing a set of multi-parameter time series, joint change characteristics, and state evolution characteristics, the state evolution trajectories of the separation process in different time intervals can be distinguished and classified, thereby obtaining a separation behavior evolution pattern that reflects the evolution law of the separation process's operating state. This approach allows the identification results of separation behavior to directly originate from the operating data of the separation process itself, improving the ability to characterize the state change characteristics of the separation process.

[0032] In a preferred embodiment of the present invention, multiple process monitoring parameters used to characterize the material distribution state and changes in separation results are collected during the separation process according to a preset sampling period. These process monitoring parameters are then sorted according to the collection time to form multi-parameter time series data reflecting the continuous operation of the separation process. This includes: during the operation of the At-211 online separation process, determining multiple process monitoring parameters needed to characterize the separation process operation state, including at least operational parameters that can reflect changes in material distribution state and separation results; synchronously acquiring the numerical information corresponding to each process monitoring parameter at each sampling moment according to the set sampling period; grouping multiple process monitoring parameters acquired at the same sampling moment into a set of process monitoring data and labeling the corresponding collection time for this set of process monitoring data; arranging the multiple sets of process monitoring data obtained at consecutive sampling moments according to the order of collection time, so that each process monitoring parameter forms a continuous data record in the time dimension; through the above method, multi-parameter time series data reflecting the changes of each process monitoring parameter over time during the continuous operation of the separation process is formed, providing a data foundation for subsequent change correlation analysis.

[0033] In a preferred embodiment of the present invention, based on multi-parameter time series data, the direction, magnitude, and synchronicity of change of each process monitoring parameter are processed interval by interval within a continuous time interval to obtain joint change characteristics characterizing the coordinated change relationship of each process monitoring parameter within the corresponding time interval. This includes: dividing the multi-parameter time series data into several consecutive time intervals according to time sequence, with each time interval containing multiple adjacent sampling results; comparing the numerical changes of each process monitoring parameter at the start and end times of each interval within each time interval to determine the direction of change of that process monitoring parameter within that time interval; analyzing the degree of numerical change of each process monitoring parameter within the time interval based on the determined direction of change to characterize the magnitude of change of that process monitoring parameter within that time interval; comparing and analyzing the direction and magnitude of change of multiple process monitoring parameters within the same time interval to determine whether each process monitoring parameter exhibits a mutually correlated change trend within the same time interval, thereby obtaining the synchronicity of change reflecting the coordinated change of multiple process monitoring parameters; and combining the direction, magnitude, and synchronicity of change of each process monitoring parameter within each time interval to form joint change characteristics characterizing the coordinated change relationship of each process monitoring parameter within that time interval.

[0034] In a preferred embodiment of the present invention, the preset sampling period is a time interval used to determine the frequency of process monitoring parameter acquisition, including: before the start of the At-211 online separation process, determining the sampling time interval for acquiring process monitoring parameters based on the separation process running speed and the time scale of process monitoring parameter changes; ensuring that the sampling time interval can cover the time range of changes in material distribution and separation results during the separation process, thereby ensuring that continuous sampling data can reflect the changing trend of the separation process state; during the separation process operation, periodically acquiring process monitoring parameters according to the sampling time interval, so that the process monitoring data between adjacent sampling times remains continuous in time; by fixing the sampling period, the resulting multi-parameter time series data has a consistent interval structure in the time dimension, which facilitates subsequent analysis and processing of the relationship between process monitoring parameter changes in continuous time intervals.

[0035] In a preferred embodiment of the present invention, a digital twin simulation sub-model is constructed to describe the state change relationship of the At-211 separation process under each separation behavior evolution mode. The different digital twin simulation sub-models differ in the composition of process state variables and / or the constraints on state changes. This includes: analyzing the participation of each process monitoring parameter in the state evolution trajectory corresponding to the identified separation behavior evolution mode on the state changes of the separation process; selecting process monitoring parameters that match the separation behavior evolution mode; and mapping the selected process monitoring parameters to corresponding process state variables. A set of process state variables is formed. Based on the set of process state variables, the process state variables are organized according to their roles and mutual influences in the separation process, and a state variable organization structure is constructed to describe the state transfer relationship between process state variables. Based on the state variable organization structure, state change constraints that limit the change order, change boundary, and mutual constraint relationship of process state variables are determined, and state change rules applicable to the evolution mode of separation behavior are formed. A digital twin simulation sub-model corresponding to the evolution mode of separation behavior is constructed from the set of process state variables, the state variable organization structure, and the state change constraints.

[0036] In this embodiment of the invention, digital twin simulation sub-models are constructed according to different separation behavior evolution modes, ensuring that the composition of state variables and the constraints of state changes in the digital twin simulation sub-models match the corresponding separation behavior evolution characteristics. By analyzing the participation of process monitoring parameters in the state changes of the separation process under each separation behavior evolution mode, and determining the process state variables and their organizational structure accordingly, different digital twin simulation sub-models exhibit differences in state description, thereby avoiding the use of a single simulation model to uniformly describe all separation operation states. This approach enables the simulation model to reflect the state change relationships of the separation process separately for different separation behavior evolution modes, improving the adaptability of the simulation process to the multi-stage operational characteristics of the separation process.

[0037] In a preferred embodiment of the present invention, based on the set of process state variables, the process state variables are organized according to their roles and mutual influences in the separation process to construct a state variable organization structure for describing the state transmission relationships between process state variables. This includes: after obtaining a set of process state variables that matches a certain separation behavior evolution pattern, analyzing the functional roles of each process state variable in the separation process, and clarifying the roles of each process state variable in material transfer, separation, and result formation; determining the state transmission direction between process state variables based on the order of their actions in the separation process, so that state changes can be transmitted according to the actual operating logic of the separation process; further analyzing whether there are mutual constraints or mutual influences between the process state variables, and associating process state variables with related relationships; and based on the above analysis results, organizing the process state variables in layers or groups to form a state variable organization structure that reflects the state transmission paths and mutual relationships between process state variables, used to describe the overall composition of the separation process state.

[0038] In a preferred embodiment of the present invention, based on the state variable organization structure, state change constraints are determined to limit the order of changes, boundaries, and mutual constraints of process state variables, forming state change rules applicable to the separation behavior evolution mode. This includes: determining the allowed order of changes for each process state variable during the separation process based on the constructed state variable organization structure, ensuring that state changes conform to the actual operation flow of the separation process; determining the allowed range of values ​​for each process state variable under the corresponding separation behavior evolution mode, based on the operating conditions and technological limitations of the separation process, thereby limiting the boundaries of process state variables; determining the constraints on other process state variables when one process state variable changes, based on the mutual influence relationships between process state variables, to avoid conflicts between state changes that do not conform to the actual separation process; and unifying and organizing the change order, boundaries, and mutual constraints to form state change rules for limiting the change methods of process state variables.

[0039] In a preferred embodiment of the present invention, a digital twin simulation sub-model corresponding to the separation behavior evolution mode is constructed from the set of process state variables, the organizational structure of state variables, and the constraints of state changes. This includes: establishing a state description framework based on the set of process state variables to characterize the operating state of the separation process, so that each process state variable serves as a state description element in the digital twin simulation sub-model; introducing the organizational structure of state variables into the state description framework to determine the state transmission path and interrelationships between each process state variable; loading state change rules into the digital twin simulation sub-model to constrain the change mode and sequence of each process state variable during the simulation; and constructing a digital twin simulation sub-model matching the specific separation behavior evolution mode, so that the digital twin simulation sub-model can reflect the changing relationships of the separation process states under the separation behavior evolution mode.

[0040] In a preferred embodiment of the present invention, when the two are inconsistent, the continuation of the existing process state in the current digital twin simulation sub-model is terminated, and the system switches to a digital twin simulation sub-model corresponding to the newly determined separation behavior evolution mode. Based on the process state variables and their state change constraints in the digital twin simulation sub-model, the process state of the subsequent separation process is re-deduced, including: after determining that the separation behavior evolution mode corresponding to the current separation process is inconsistent with the running digital twin simulation sub-model, the system stops the continued deduction of the separation process state of the subsequent time step based on the digital twin simulation sub-model and blocks the transmission of the process state already formed in the digital twin simulation sub-model to the subsequent simulation process; according to the newly determined separation behavior evolution mode, a digital twin simulation sub-model corresponding to the separation behavior evolution mode is selected as the digital twin simulation sub-model used for the online simulation of the subsequent separation process; based on the real-time collected process monitoring data, the process state variables in the selected digital twin simulation sub-model are initialized, and under the constraints of the corresponding state change, the process state of the subsequent separation process is gradually deduced, thereby re-deducing the subsequent separation process.

[0041] In this embodiment of the invention, by continuously determining the correspondence between the current separation behavior evolution mode and the running digital twin simulation sub-model during the online digital twin simulation, the simulation process can promptly perceive changes in the separation process's operating state. When the separation behavior evolution mode changes, the continuation of the existing process state is terminated and switched to the corresponding digital twin simulation sub-model, while simultaneously re-establishing the simulation evolution process. This avoids directly carrying over the state evolution relationship formed in the previous operating state to the new operating state. This method enables the simulation process to re-match the new state evolution rules when the separation behavior evolution mode switches, thereby reducing simulation deviations caused by state mismatches at the boundary of separation stages.

[0042] In a preferred embodiment of the present invention, after determining that the separation behavior evolution mode corresponding to the current separation process is inconsistent with the running digital twin simulation sub-model, the continued deduction of the separation process state for subsequent time steps based on the digital twin simulation sub-model is stopped, and the transmission of the process state already formed in the digital twin simulation sub-model to the subsequent simulation process is blocked. This includes: during the online simulation of the digital twin, when the current separation behavior evolution mode identified based on real-time process monitoring data is inconsistent with the separation behavior evolution mode corresponding to the currently running digital twin simulation sub-model, a simulation switching trigger signal is generated; upon receiving the simulation switching trigger signal, the state deduction operation of the current digital twin simulation sub-model for subsequent time points is immediately terminated, so that the current digital twin simulation sub-model no longer generates new simulation states based on existing states; at the same time, the process states already formed in the current digital twin simulation sub-model are frozen, so that the process states are no longer used as the initial state or reference state for subsequent simulation calculations; through the above methods, the state continuation path of the current digital twin simulation sub-model is interrupted, and the state evolution results formed under the previous separation behavior evolution mode are prevented from being brought into the subsequent mismatched simulation process.

[0043] In a preferred embodiment of the present invention, a digital twin simulation sub-model corresponding to the newly determined separation behavior evolution mode is selected as the digital twin simulation sub-model used for subsequent online simulation of the separation process, based on the newly determined separation behavior evolution mode. This includes: after completing the re-determination of the current separation behavior evolution mode, using the identified separation behavior evolution mode as the basis for selecting the simulation model; retrieving a digital twin simulation sub-model that corresponds to the separation behavior evolution mode from a pre-established set of digital twin simulation sub-models; when a digital twin simulation sub-model matching the separation behavior evolution mode is found, setting that digital twin simulation sub-model as the digital twin simulation sub-model used for subsequent online simulation; through the above method, the digital twin simulation sub-model used in the subsequent simulation process is kept consistent with the current separation behavior evolution mode in terms of the composition of state variables and the constraints of state changes.

[0044] In a preferred embodiment of the present invention, based on real-time acquired process monitoring data, the process state variables in the selected digital twin simulation sub-model are initialized, and under the constraints of corresponding state changes, the process state of the subsequent separation process is progressively deduced, thereby re-deducing the subsequent separation process. This includes: after selecting a new digital twin simulation sub-model, acquiring real-time process monitoring data corresponding to the current time point; mapping the monitoring data corresponding to the process state variables in the digital twin simulation sub-model to initial state values ​​based on the real-time process monitoring data, which are used to initialize each process state variable in the digital twin simulation sub-model; after completing the state initialization, progressively deducing the separation process state at subsequent time points according to the pre-set state variable organization structure and state change constraints in the digital twin simulation sub-model; through the above method, the state evolution path of the separation process is re-established under the new separation behavior evolution mode, so that the simulation process can reflect the state change relationship under the current separation process operation state.

[0045] In a preferred embodiment of the present invention, based on joint change characteristics, the state evolution trajectory formed by the change of process state over time during the separation process is analyzed to generate a set of state evolution features characterizing the stability and consistency of the state evolution trajectory. This includes: connecting process states within continuous time intervals according to the joint change characteristics in chronological order to form a state evolution trajectory composed of multiple continuous process state points, where each process state point corresponds to the value of the joint change characteristics within the corresponding time interval; analyzing the change relationship between adjacent process state points segment by segment based on the state evolution trajectory to determine the degree of continuity of process state changes within adjacent time intervals, generating a continuity analysis result characterizing the local continuity of the state evolution trajectory; analyzing the overall change direction and degree of consistency of the state evolution trajectory within multiple continuous time intervals based on the continuity analysis result, generating a trend consistency analysis result characterizing the overall change law of the state evolution trajectory; and generating stability features characterizing the stability of the state evolution trajectory and consistency features characterizing the consistency of the state evolution trajectory based on the continuity analysis result and the trend consistency analysis result, respectively, and combining the stability features and consistency features to form a set of state evolution features.

[0046] In this embodiment of the invention, the state evolution trajectory formed by the change of process state over time during the separation process is analyzed based on joint change characteristics. State evolution characteristics are extracted from both local continuity and overall change patterns, so that the changes in the operating state of the separation process are no longer merely manifested as discrete parameter fluctuations, but are transformed into a set of trajectory features that reflect the characteristics of state evolution. By characterizing the stability and consistency of the state evolution trajectory, the inherent laws of the state evolution of the separation process within different time intervals are clearly depicted, thus providing a structured state feature basis for subsequent differentiation and determination of the separation behavior evolution pattern.

[0047] In a preferred embodiment of the present invention, based on the state evolution trajectory, the relationship between adjacent process state points is analyzed segment by segment to determine the degree of continuity of process state changes within adjacent time intervals, generating a continuity analysis result to characterize the local continuity of the state evolution trajectory. This includes: after obtaining the state evolution trajectory composed of multiple continuous process state points, selecting two adjacent process state points sequentially as an analysis unit according to time order; comparing the value changes of corresponding process state variables in the two adjacent process state points to analyze whether each process state variable maintains a stable change within the adjacent time interval; when the direction and degree of change of most process state variables remain consistent within the adjacent time interval, the adjacent time interval is determined as a time interval of continuous state change; performing the above analysis sequentially on all adjacent time intervals in the state evolution trajectory, and summarizing the results to obtain a continuity analysis result reflecting the continuous change of the state evolution trajectory within a local time range, which is used to describe the local continuity characteristics of the state evolution trajectory.

[0048] In a preferred embodiment of the present invention, based on the continuity analysis results, the overall direction and consistency of change of the state evolution trajectory within multiple consecutive time intervals are analyzed to generate a trend consistency analysis result characterizing the overall change law of the state evolution trajectory. This includes: after obtaining the continuity analysis results corresponding to the state evolution trajectory, performing a combined analysis on the continuity determination results within multiple consecutive time intervals; based on the combined analysis, determining whether the state evolution trajectory exhibits a stable change trend over a longer time period, i.e., determining whether the process state continues to evolve along the same change direction; when the state evolution trajectory maintains a consistent change direction within multiple consecutive time intervals, the state evolution trajectory within that time interval is determined to have a consistent overall change trend; through the above method, a trend consistency analysis result characterizing the consistency of the change direction of the state evolution trajectory on the overall time scale is formed to reflect the overall change law of the state evolution trajectory.

[0049] In a preferred embodiment of the present invention, based on the results of continuity analysis and trend consistency analysis, stability features for characterizing the stability of the state evolution trajectory and consistency features for characterizing the consistency of the state evolution trajectory are generated respectively. The stability features and consistency features are then combined to form a set of state evolution features, including: based on the continuity analysis results, summarizing cases where the state evolution trajectory maintains continuous change within adjacent time intervals to form stability features describing the smoothness of change of the state evolution trajectory at local time scales; based on the trend consistency analysis results, summarizing cases where the state evolution trajectory maintains a consistent direction of change within multiple consecutive time intervals to form consistency features describing the consistency of the direction of change of the state evolution trajectory at the overall time scale; combining stability features and consistency features as two dimensions describing the characteristics of the state evolution trajectory to form a set of state evolution features that comprehensively reflects the changing characteristics of the state evolution trajectory. Through the above method, the local change characteristics and overall change patterns of the state evolution trajectory can be uniformly characterized, providing a basis for the subsequent differentiation and judgment of separation behavior evolution patterns.

[0050] In a preferred embodiment of the present invention, based on the set of state evolution features, state evolution trajectories within different time intervals are distinguished, and state evolution trajectories with the same state evolution features are grouped into the same separation behavior evolution pattern to form a separation behavior evolution pattern identification result for characterizing the evolution law of the separation process. This includes: aligning the stability and consistency features of each state evolution trajectory according to the corresponding set of state evolution features within different time intervals to form a feature description result characterizing the feature composition of each state evolution trajectory; comparing the stability and consistency features between different state evolution trajectories one by one according to the feature description result to determine the consistency relationship in feature composition between each state evolution trajectory, generating a feature consistency determination result between state evolution trajectories; grouping state evolution trajectories with the same feature composition into the same trajectory set according to the feature consistency determination result, with each trajectory set corresponding to a state evolution law; determining each trajectory set as a separation behavior evolution pattern, and establishing a correspondence between the separation behavior evolution pattern and the corresponding set of state evolution features to form a separation behavior evolution pattern identification result.

[0051] In this embodiment of the invention, by aligning, comparing, and merging state evolution trajectories within different time intervals based on a set of state evolution features, the state evolution patterns exhibited by the separation process in different time intervals can be systematically distinguished. By merging state evolution trajectories with the same state evolution features into the same separation behavior evolution pattern, the formation process of the separation behavior evolution pattern is directly based on the consistency of state evolution features, thereby avoiding the need to artificially segment the separation process by relying on fixed time divisions or operating condition markers. This method enables the identification results of the separation behavior evolution pattern to truly reflect the evolutionary patterns of the separation process's operating state.

[0052] In a preferred embodiment of the present invention, based on the feature description results, the stability features and consistency features between different state evolution trajectories are compared one by one to determine the consistency relationship in feature composition between each state evolution trajectory, and a feature consistency determination result between state evolution trajectories is generated. This includes: after obtaining the feature description results corresponding to different state evolution trajectories, selecting any two state evolution trajectories as comparison objects; for the two state evolution trajectories, extracting their corresponding stability features and consistency features respectively, and comparing the corresponding features item by item; determining whether the two state evolution trajectories exhibit the same degree of stability in terms of stability features, and whether they exhibit the same overall direction of change in terms of consistency features; when the two state evolution trajectories are consistent in both stability features and consistency features, it is determined that the two state evolution trajectories have a consistency relationship in feature composition; by performing the above comparison processing on different state evolution trajectories in sequence, a feature consistency determination result reflecting the feature consistency relationship between each state evolution trajectory is formed for subsequent trajectory merging processing.

[0053] In a preferred embodiment of the present invention, based on an identified separation behavior evolution pattern, the participation of each process monitoring parameter in the state evolution trajectory corresponding to the separation behavior evolution pattern on the state changes of the separation process is analyzed. Process monitoring parameters matching the separation behavior evolution pattern are selected, and the selected process monitoring parameters are mapped to corresponding process state variables to form a set of process state variables. This includes: extracting the parameter change trajectory of each process monitoring parameter during the formation of the state evolution trajectory according to the state evolution trajectory corresponding to the separation behavior evolution pattern, forming a parameter change description result corresponding to each process monitoring parameter; analyzing the correspondence between each process monitoring parameter and the state changes of the separation process during the formation of the state evolution trajectory according to the parameter change description result, generating a participation analysis result to characterize the degree of participation of each process monitoring parameter in the state changes of the separation process; screening process monitoring parameters that play a major characterizing role in the state changes of the separation process under the separation behavior evolution pattern according to the participation analysis result, forming a set of candidate process monitoring parameters; and mapping each process monitoring parameter in the candidate process monitoring parameter set to a corresponding process state variable according to its physical meaning and functional role in the separation process, forming a set of process state variables corresponding to the separation behavior evolution pattern.

[0054] In this embodiment of the invention, for a specific separation behavior evolution mode, the change trajectories of each process monitoring parameter in its corresponding state evolution trajectory are analyzed. Based on the correspondence between each process monitoring parameter and the state changes of the separation process, process monitoring parameters are selected, thus establishing a correlation between the determination process of process state variables and the specific separation behavior evolution mode. By mapping the selected process monitoring parameters to process state variables, the set of process state variables used under different separation behavior evolution modes becomes targeted, thereby providing a foundation for constructing a digital twin simulation sub-model that matches the separation behavior evolution mode, ensuring that the simulation model maintains consistency with the corresponding separation process operating characteristics at the state description level.

[0055] In a preferred embodiment of the present invention, based on the state evolution trajectory corresponding to the separation behavior evolution mode, the parameter change trajectory of each process monitoring parameter during the formation of the state evolution trajectory is extracted to form a parameter change description result corresponding to each process monitoring parameter. This includes: after determining a certain separation behavior evolution mode, obtaining the state evolution trajectory corresponding to the separation behavior evolution mode; extracting the monitoring data corresponding to each process monitoring parameter at each time point in chronological order for the time interval covered by the state evolution trajectory; connecting the monitoring data of the same process monitoring parameter changing with time within the time interval to form a parameter change trajectory reflecting the change process of the process monitoring parameter; performing the above processing on different process monitoring parameters respectively to form a parameter change description result corresponding to each process monitoring parameter for subsequent participation analysis.

[0056] In a preferred embodiment of the present invention, based on the parameter change description results, the correspondence between each process monitoring parameter and the state change of the separation process during the formation of the state evolution trajectory is analyzed to generate participation analysis results characterizing the degree of participation of each process monitoring parameter in the state change of the separation process. This includes: after obtaining the parameter change description results corresponding to each process monitoring parameter, aligning the change trajectory of each parameter with the state evolution trajectory corresponding to the separation process in time; analyzing whether each process monitoring parameter changes synchronously within the time interval of the state evolution trajectory change, and whether the change trend remains consistent; when the change trajectory of a certain process monitoring parameter maintains a synchronous change relationship with the state evolution trajectory in multiple state change intervals, it is determined that the process monitoring parameter has a high degree of participation in the state change of the separation process; when the change trajectory of a certain process monitoring parameter lacks a synchronous relationship with the state evolution trajectory, it is determined that the process monitoring parameter has a low degree of participation in the state change of the separation process; by performing the above analysis on each process monitoring parameter in sequence, participation analysis results characterizing the degree of participation of each process monitoring parameter in the state change of the separation process are formed, providing a basis for subsequent parameter selection.

[0057] In a preferred embodiment of the present invention, based on the participation analysis results, process monitoring parameters that play a major characterizing role in the state changes of the separation process under the separation behavior evolution mode are screened to form a candidate process monitoring parameter set. This includes: after obtaining the participation analysis results corresponding to each process monitoring parameter, organizing all process monitoring parameters according to their degree of participation in the state changes of the separation process; based on the participation analysis results, identifying process monitoring parameters that can continuously reflect the characteristics of the state changes of the separation process during the state evolution trajectory change process; determining the process monitoring parameters that can reflect the characteristics of the state changes of the separation process as candidate process monitoring parameters, and distinguishing them from process monitoring parameters with lower participation and weaker correlation with the state changes of the separation process; through the above screening process, a candidate process monitoring parameter set is formed to characterize the state changes of the separation process under the current separation behavior evolution mode, providing a basis for the determination of subsequent process state variables.

[0058] In a preferred embodiment of the present invention, each process monitoring parameter in the candidate process monitoring parameter set is mapped to a corresponding process state variable based on its physical meaning and functional role in the separation process, forming a set of process state variables corresponding to the separation behavior evolution mode. This includes: after obtaining the candidate process monitoring parameter set, analyzing the physical meaning reflected by each candidate process monitoring parameter and its functional role in the separation process, clarifying the separation process state meaning corresponding to each process monitoring parameter; based on the physical meaning and functional role, mapping each candidate process monitoring parameter to a specific state description element in the separation process, so that the process monitoring parameter can be used to characterize the changes in the separation process state; converting the process monitoring parameters with the established correspondence into process state variables and uniformly incorporating them into the separation process state description system; through the above method, a set of process state variables matching the current separation behavior evolution mode is formed, so that the subsequently constructed digital twin simulation sub-model can reflect the operating characteristics of the separation process under the separation behavior evolution mode at the state description level.

[0059] Embodiments of the present invention also provide an At-211 online separation process simulation platform based on digital twins. The platform includes: a separation behavior evolution pattern recognition module, used to collect process monitoring data characterizing the material distribution state and changes in separation results during the At-211 online separation process, and analyze the state evolution characteristics of the separation process based on the correlation of changes in process monitoring data over a continuous time interval, identifying separation behavior evolution patterns characterizing the evolution law of the separation process's operating state; a running segment division module, used to divide the At-211 online separation process into multiple separation process running segments according to different separation behavior evolution patterns, each running segment corresponding to a separation behavior evolution pattern that remains relatively stable within a predetermined time range; and a digital twin simulation sub-model construction module, used to construct a model describing the At-211 separation behavior evolution pattern according to each separation behavior evolution pattern. 211. A digital twin simulation sub-model for the state change relationship of the separation process. Different digital twin simulation sub-models differ in the composition of process state variables and / or state change constraints. A separation behavior evolution mode determination module is used to continuously determine whether the separation behavior evolution mode corresponding to the current separation process corresponds to the running digital twin simulation sub-model based on real-time collected process monitoring data during the online digital twin simulation. A simulation control module is used to maintain the continuous simulation of the separation process state by the digital twin simulation sub-model when the two are consistent; when the two are inconsistent, the continuation of the existing process state in the current digital twin simulation sub-model is terminated, and the model is switched to the digital twin simulation sub-model corresponding to the newly determined separation behavior evolution mode. Based on the process state variables and state change constraints in the digital twin simulation sub-model, the process state of the subsequent separation process is re-deduced.

[0060] 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.

[0061] 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.

[0062] 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.

[0063] 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. A simulation method for the online separation process of At-211 based on digital twins, characterized in that, The method includes: collecting process monitoring data to characterize the material distribution state and changes in separation results during the At-211 online separation process; analyzing the state evolution characteristics of the separation process based on the correlation of changes in the process monitoring data within a continuous time interval; identifying separation behavior evolution patterns that characterize the evolution law of the separation process's operating state; dividing the At-211 online separation process into multiple separation process operation segments according to different separation behavior evolution patterns, with each separation process operation segment corresponding to a separation behavior evolution pattern that remains relatively stable within a predetermined time range; and constructing a digital twin simulation sub-model to describe the state change relationship of the At-211 separation process under each separation behavior evolution pattern. The digital twin simulation sub-models differ from each other in the composition of process state variables and / or the constraints of state changes. During the online simulation of the digital twin, based on the real-time collected process monitoring data, it is continuously determined whether the evolution mode of the separation behavior corresponding to the current separation process corresponds to the running digital twin simulation sub-model. When they are consistent, the continuous simulation of the separation process state by the digital twin simulation sub-model is maintained. When they are inconsistent, the continuation of the existing process state in the current digital twin simulation sub-model is terminated, and the model is switched to the digital twin simulation sub-model corresponding to the newly determined evolution mode of separation behavior. Based on the process state variables and their state change constraints in the digital twin simulation sub-model, the process state of the subsequent separation process is re-deduced.

2. The simulation method for the At-211 online separation process based on digital twins according to claim 1, characterized in that, Based on the correlation of changes in process monitoring data over a continuous time interval, this study analyzes the state evolution characteristics of the separation process and identifies separation behavior evolution patterns that characterize the evolution of the separation process's operating state. This includes: collecting multiple process monitoring parameters to characterize material distribution and separation results changes during the separation process according to a preset sampling period; sorting these parameters based on the collection time to form multi-parameter time series data reflecting the continuous operating state of the separation process; processing the direction, magnitude, and synchronicity of change of each process monitoring parameter within a continuous time interval based on the multi-parameter time series data to obtain joint change characteristics characterizing the coordinated change relationship of each process monitoring parameter within the corresponding time interval; analyzing the state evolution trajectory formed by the change of process state over time during the separation process based on the joint change characteristics to generate a set of state evolution characteristics characterizing the stability and consistency of the state evolution trajectory; and distinguishing the state evolution trajectories within different time intervals based on the set of state evolution characteristics, merging state evolution trajectories with the same state evolution characteristics into the same separation behavior evolution pattern to form a separation behavior evolution pattern identification result characterizing the evolution of the separation process's operating state.

3. The simulation method for the At-211 online separation process based on digital twins according to claim 1, characterized in that, Based on each separation behavior evolution mode, a digital twin simulation sub-model is constructed to describe the state change relationship of the At-211 separation process under that separation behavior evolution mode. Different digital twin simulation sub-models differ in the composition of process state variables and / or state change constraints. This includes: analyzing the participation of each process monitoring parameter in the state evolution trajectory corresponding to a given separation behavior evolution mode in the state change change of the separation process; selecting process monitoring parameters that match the separation behavior evolution mode and mapping them to corresponding process state variables to form a set of process state variables; organizing the process state variables according to their roles and mutual influences in the separation process, constructing a state variable organization structure to describe the state transmission relationship between process state variables; determining state change constraints that limit the sequence, boundaries, and mutual constraints of process state variable changes, forming state change rules applicable to the separation behavior evolution mode; and constructing a digital twin simulation sub-model corresponding to the separation behavior evolution mode from the set of process state variables, the state variable organization structure, and the state change constraints.

4. The simulation method for the At-211 online separation process based on digital twins according to claim 1, characterized in that, When the two are inconsistent, the continuation of the existing process state in the current digital twin simulation sub-model is terminated, and the system switches to the digital twin simulation sub-model corresponding to the newly determined separation behavior evolution mode. Based on the process state variables and their state change constraints in the digital twin simulation sub-model, the process state of the subsequent separation process is re-deduced, including: after determining that the separation behavior evolution mode corresponding to the current separation process is inconsistent with the running digital twin simulation sub-model, the system stops the continued deduction of the separation process state of subsequent time steps based on the digital twin simulation sub-model and blocks the transmission of the process state already formed in the digital twin simulation sub-model to the subsequent simulation process; according to the newly determined separation behavior evolution mode, the system selects the digital twin simulation sub-model corresponding to the separation behavior evolution mode as the digital twin simulation sub-model used for the online simulation of the subsequent separation process; based on the real-time collected process monitoring data, the system initializes the process state variables in the selected digital twin simulation sub-model and, under the constraints of the corresponding state change, gradually deduces the process state of the subsequent separation process.

5. The simulation method for the At-211 online separation process based on digital twins according to claim 2, characterized in that, Based on the joint change characteristics, the state evolution trajectory formed by the change of process state over time during the separation process is analyzed, generating a set of state evolution features to characterize the stability and consistency of the state evolution trajectory. This includes: connecting process states in continuous time intervals according to the joint change characteristics in chronological order to form a state evolution trajectory composed of multiple continuous process state points, where each process state point corresponds to the value of the joint change characteristics in the corresponding time interval; analyzing the change relationship between adjacent process state points segment by segment based on the state evolution trajectory to determine the degree of continuity of process state changes in adjacent time intervals, generating a continuity analysis result to characterize the local continuity of the state evolution trajectory; analyzing the overall change direction and degree of consistency of the state evolution trajectory in multiple continuous time intervals based on the continuity analysis result, generating a trend consistency analysis result to characterize the overall change law of the state evolution trajectory; and generating stability features to characterize the stability of the state evolution trajectory and consistency features to characterize the consistency of the state evolution trajectory based on the continuity analysis result and the trend consistency analysis result, respectively, and combining the stability features and consistency features to form a set of state evolution features.

6. The simulation method for the At-211 online separation process based on digital twins according to claim 2, characterized in that, Based on the set of state evolution features, state evolution trajectories within different time intervals are distinguished. Trajectories with the same state evolution features are grouped into the same separation behavior evolution pattern, forming a separation behavior evolution pattern recognition result to characterize the state evolution law of the separation process. This includes: aligning the stability and consistency features of each state evolution trajectory according to the corresponding set of state evolution features within different time intervals to form a feature description result characterizing the feature composition of each state evolution trajectory; comparing the stability and consistency features between different state evolution trajectories one by one based on the feature description result to determine the consistency relationship in feature composition between each state evolution trajectory, generating a feature consistency judgment result between state evolution trajectories; grouping state evolution trajectories with the same feature composition into the same trajectory set, with each trajectory set corresponding to a state evolution law; determining each trajectory set as a separation behavior evolution pattern and establishing a correspondence between the separation behavior evolution pattern and the corresponding set of state evolution features, forming a separation behavior evolution pattern recognition result.

7. The simulation method for the At-211 online separation process based on digital twins according to claim 3, characterized in that, Based on an identified separation behavior evolution pattern, the participation of each process monitoring parameter in the state evolution trajectory corresponding to this pattern on the state changes of the separation process is analyzed. Process monitoring parameters matching this separation behavior evolution pattern are selected, and these parameters are mapped to corresponding process state variables, forming a set of process state variables. This includes: extracting the parameter change trajectory of each process monitoring parameter during the formation of the state evolution trajectory, based on the state evolution trajectory corresponding to the separation behavior evolution pattern, and generating parameter change description results for each process monitoring parameter; analyzing the correspondence between each process monitoring parameter and the state changes of the separation process during the formation of the state evolution trajectory, generating participation analysis results to characterize the degree of participation of each process monitoring parameter in the state changes of the separation process; screening process monitoring parameters that play a major characterizing role in the state changes of the separation process under the separation behavior evolution pattern, forming a set of candidate process monitoring parameters; and mapping each process monitoring parameter in the candidate set to corresponding process state variables based on its physical meaning and functional role in the separation process, forming a set of process state variables corresponding to this separation behavior evolution pattern.

8. A digital twin-based online separation process simulation platform for At-211, characterized in that, Applied to the method described in any one of claims 1 to 7, the platform comprises: a separation behavior evolution pattern recognition module, used to collect process monitoring data characterizing the material distribution state and changes in separation results during the At-211 online separation process, and based on the correlation of changes in process monitoring data within a continuous time interval, analyze the state evolution characteristics of the separation process and identify separation behavior evolution patterns characterizing the evolution law of the separation process operation state; a running segment division module, used to divide the At-211 online separation process into multiple separation process running segments according to different separation behavior evolution patterns, each separation process running segment corresponding to a separation behavior evolution pattern that remains relatively stable within a predetermined time range; and a digital twin simulation sub-model construction module, used to construct a model to describe the At-211 separation process under each separation behavior evolution pattern. The system comprises several digital twin simulation sub-models, each with different process state variables and / or state change constraints. A separation behavior evolution mode determination module, based on real-time process monitoring data, continuously determines whether the current separation behavior evolution mode corresponds to the running digital twin simulation sub-model during online digital twin simulation. A simulation control module maintains continuous simulation of the separation process state using the digital twin simulation sub-model when both are consistent; otherwise, it terminates the continuation of the existing process state in the current digital twin simulation sub-model, switches to the digital twin simulation sub-model corresponding to the newly determined separation behavior evolution mode, and re-deduces the process state of subsequent separation processes based on the process state variables and state change constraints in that digital twin simulation sub-model.

9. 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 7.

10. 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 7.

Images