Enterprise twin compliance control method and system based on spatiotemporal knowledge evolution and non-equilibrium state alignment

Abstract

The application provides an enterprise twin compliance control method and system based on space-time knowledge evolution and non-equilibrium state alignment, comprising: acquiring transmission link delay distribution and collection cycle of each business subsystem, extracting delay absolute time length arrangement and beat frequency interference waveform of each link; according to the delay absolute time length arrangement and the beat frequency interference waveform, analyzing the collection rhythm between the multi-source data streams, determining the occurrence interval of the delay resonance and the time stamp dislocation distance of the periodic distortion component; jointly analyzing the filtering results of the deviation degree and the collection time sequence dislocation event to obtain a comprehensive deviation evaluation value of each business subsystem, mapping the business subsystems with the comprehensive deviation evaluation value exceeding the preset allowable range to corresponding frequency adjustment signals; updating the collection frequency of the business subsystems with deviation to obtain an optimized data collection frequency configuration.

Description

Technical Field

[0001] This invention relates to the field of information technology, and in particular to an enterprise twin compliance control method and system based on spatiotemporal knowledge evolution and non-equilibrium state alignment. Background Technology

[0002] In the context of enterprise management and digital transformation, building enterprise twin systems based on spatiotemporal knowledge has become an important direction for improving business compliance and operational efficiency. This field strives to achieve a precise mapping between virtual models and actual business operations through real-time monitoring and analysis of business data, thereby ensuring compliant operation of enterprises in complex environments. However, the application of this technology is directly related to the accuracy of enterprise decision-making and risk control capabilities, and its importance is self-evident. Currently, although many solutions aim to enhance the real-time performance of systems by increasing the frequency of data collection, these methods often overlook the complex interactions of multi-source data over time. Simply increasing the collection frequency cannot completely solve the data synchronization problem; on the contrary, inconsistent collection rhythms from different data sources may cause new interference, leading to data alignment chaos in the system at certain times, affecting the reliability of compliance judgments. The mutual interference between multi-source data collection cycles is the beat interference phenomenon caused by the mismatch in the temporal rhythms of different data sources. This phenomenon refers to the fact that due to slight differences in the collection time intervals of various data sources, the timestamps of the data cannot be accurately aligned during certain time periods, thus affecting the accuracy of the system's judgment of business status. This time misalignment not only undermines data integrity but can also cause systems to misinterpret normal business fluctuations as anomalies, leading to erroneous compliance warnings. Specifically, in daily business operations, such as supply chain management scenarios, inventory, logistics, and sales data are often collected separately by different systems. If the collection intervals for these data are inconsistent—for example, inventory updates hourly while logistics information updates every half hour—then at certain points in time, the system may miscalculate the matching degree between inventory and logistics due to the time misalignment, leading to misjudgments of supply chain anomalies and unnecessary business adjustments or resource waste. Therefore, in the complex environment of multi-source data collection, resolving the beat interference problem caused by mismatched collection rhythms, ensuring the accuracy of data time alignment, and avoiding resulting compliance misjudgments have become key issues in the field of enterprise twin compliance control. Summary of the Invention

[0003] This invention provides an enterprise twin compliance control method based on spatiotemporal knowledge evolution and non-equilibrium state alignment, mainly including: Obtain the transmission link delay distribution and collection period of each business subsystem, and form a correspondence between link delay and collection period; Based on the aforementioned correspondence, the acquisition rhythm between multiple data streams is analyzed to determine the interval where time delay resonance occurs; The occurrence range of the time delay resonance is evaluated, resonance ranges exceeding the preset range are identified, and distortion range identifiers are generated; Based on the distortion interval identifier, extract the component data stream within the corresponding time window, match the component data stream with the distortion interval identifier, and determine the corrected deviation and the filtering result of the acquisition timing misalignment event. By jointly analyzing the filtering results of deviation and timing misalignment events, a comprehensive deviation evaluation value for each business subsystem is obtained. Business subsystems whose comprehensive deviation evaluation values ​​exceed the preset allowable range are mapped to corresponding frequency adjustment signals. The sampling frequency of each business subsystem is adjusted according to the frequency adjustment signal. The sampling interval and data reporting cycle of each business subsystem after adjustment are obtained. The comprehensive deviation evaluation value is recalculated. If the updated comprehensive deviation evaluation value still exceeds the preset convergence threshold, it is determined that there is a deviation in the business subsystem. Update the data collection frequency of the business subsystems with deviations to obtain an optimized data collection frequency configuration.

[0004] Furthermore, the step of obtaining the transmission link delay distribution and collection period of each service subsystem to form a correspondence between link delay and collection period includes: Obtain the data reporting nodes and transmission path identifiers of each business subsystem; For each transmission path identifier, collect the transmission timestamp of the data packet from the source node to the sink node; Calculate the one-way transmission delay of each link based on the transmission timestamp; The unidirectional transmission delay is grouped and aggregated according to the business subsystem to obtain the transmission link delay distribution of each business subsystem. Read the preset collection period of each business subsystem, establish an association between the collection period and the corresponding transmission link delay distribution, and form a correspondence table between link delay and collection period.

[0005] Furthermore, based on the aforementioned correspondence, the acquisition rhythm between multi-source data streams is analyzed to determine the occurrence interval of time delay resonance, including: calculating the time delay difference between adjacent service subsystems for the sorted time delay values; If the time delay difference is less than a preset threshold, it is marked as a potential interference segment; Based on the potential interference segment, a beat frequency interference waveform is generated; By scanning the beat frequency interference waveform, the peak and trough points of the waveform are identified, and the rhythm synchronization period is determined. If the rhythm synchronization period is less than a preset threshold, it is determined that there is a time delay resonance phenomenon, and the corresponding time range is marked as the time delay resonance occurrence interval.

[0006] Furthermore, the evaluation of the occurrence interval of the time-delay resonance, the identification of resonance intervals exceeding a preset range, and the generation of distortion interval identifiers include: Extract the start and end time coordinates of each time-delay resonance occurrence interval and calculate the duration; If the duration exceeds a preset threshold, it is marked as an over-limit resonance interval; For the aforementioned over-limit resonance range, read the associated beat frequency interference waveform segments and calculate the waveform kurtosis index; The degree of deviation is calculated based on the waveform kurtosis index and the timestamp misalignment interval. The time alignment window size is determined based on the timestamp misalignment spacing. The interval boundary, deviation value, and time alignment window scale of the over-limit resonance interval are encapsulated as a distortion interval identifier.

[0007] Furthermore, the step of extracting the component data stream within the corresponding time window based on the distortion interval identifier, matching the component data stream with the distortion interval identifier, and determining the corrected deviation and the filtering result of the acquisition timing misalignment event includes: Read the start and end time coordinates and time alignment window scale from the distortion interval identifier; For the business subsystems corresponding to the distortion interval identifiers, extract the original data records within the start and end time coordinate range to form component data streams; Record the time offset sequence of each data record in the component data stream; If the offset in the time offset sequence is less than the time alignment window scale, it is marked as data within the window; If the offset is greater than or equal to the time alignment window scale, it is marked as a timing misalignment event. The percentage of data within the statistical window is used to determine the deviation after correction.

[0008] Furthermore, the filtering results of deviation and timing misalignment events are jointly analyzed to obtain a comprehensive deviation evaluation value for each business subsystem. Business subsystems whose comprehensive deviation evaluation values ​​exceed a preset allowable range are mapped to corresponding frequency adjustment signals, including: For each business subsystem, extract the deviation value of the corresponding distortion interval identifier and calculate the average deviation. Count the total number of data collection timing misalignment events corresponding to the aforementioned business subsystem; Calculate the comprehensive deviation assessment value based on the average deviation and the total number of events; If the comprehensive deviation evaluation value exceeds the preset range threshold, it is marked as a subsystem to be adjusted; For the subsystem to be adjusted, the adjustment level is determined, and a frequency adjustment signal is generated.

[0009] Furthermore, the step of adjusting the acquisition frequency of each service subsystem according to the frequency adjustment signal, obtaining the sampling interval duration and data reporting cycle of each service subsystem after adjustment, recalculating the comprehensive deviation evaluation value, and determining the service subsystem with deviation if the updated comprehensive deviation evaluation value still exceeds the preset convergence threshold, includes: Based on the frequency adjustment signal, read the adjustment level of the subsystem to be adjusted; The current acquisition frequency is adjusted incrementally according to the adjustment level described above; Calculate the sampling interval duration based on the adjusted sampling frequency; The data reporting cycle is determined based on the sampling interval duration; Re-collect the data stream according to the adjusted collection frequency and data reporting cycle; Calculate the updated comprehensive deviation assessment value based on the re-acquired data stream; If the updated comprehensive deviation assessment value exceeds the preset threshold, it is determined that the business subsystem has a deviation.

[0010] Furthermore, the updated data collection frequency of the service subsystem with deviations is used to obtain an optimized data collection frequency configuration, including: Obtain the comprehensive deviation assessment value of the business subsystems that exhibit deviations; The frequency increment amplitude is determined based on the difference between the comprehensive deviation evaluation value and the preset threshold. The frequency increment is increased by the current sampling frequency to obtain the sampling frequency after secondary adjustment. Based on the adjusted acquisition frequency, the data stream is reacquired and the comprehensive deviation evaluation value is calculated. If the overall deviation assessment value still exceeds the preset threshold, the frequency increment adjustment will be repeated. If the comprehensive deviation evaluation value falls within the preset threshold range, the acquisition frequency value is recorded, and an optimized data acquisition frequency configuration is generated.

[0011] Furthermore, the method also includes: under the optimized data acquisition frequency, retrieving the twin mirror operation trajectory and sampling interval duration boundary of each business subsystem, extracting the data reporting cycle and delay resonance suppression performance after frequency adjustment in the twin environment, evaluating the consistency deviation between the twin mirror and the physical link in data acquisition rhythm, and obtaining the data acquisition frequency optimization result verified by twin compliance.

[0012] This invention also provides an enterprise twin compliance control system based on spatiotemporal knowledge evolution and non-equilibrium state alignment, the system comprising: The link delay distribution and waveform extraction module is used to obtain the transmission link delay distribution and acquisition period of each service subsystem, and form a correspondence between link delay and acquisition period; The time delay resonance and misalignment analysis module is used to analyze the acquisition rhythm between multiple data streams based on the correspondence and determine the occurrence interval of time delay resonance. The resonance interval evaluation and fusion module is used to evaluate the occurrence interval of the time delay resonance, identify resonance intervals that exceed the preset range, and generate distortion interval identifiers. The component data stream extraction and matching processing module is used to extract the component data stream within the corresponding time window according to the distortion interval identifier, match the component data stream with the distortion interval identifier, and determine the corrected deviation and the filtering result of the acquisition timing misalignment event. The deviation and filtering result joint analysis module is used to jointly analyze the deviation and the filtering results of the acquisition timing misalignment event to obtain the comprehensive deviation evaluation value of each business subsystem. The business subsystem whose comprehensive deviation evaluation value exceeds the preset allowable range is mapped to the corresponding frequency adjustment signal. The frequency adjustment signal mapping module is used to adjust the acquisition frequency of each service subsystem according to the frequency adjustment signal, obtain the sampling interval duration and data reporting cycle of each service subsystem after adjustment, recalculate the comprehensive deviation evaluation value, and if the updated comprehensive deviation evaluation value still exceeds the preset convergence threshold, it is determined that there is a deviation in the service subsystem. The data acquisition frequency adjustment and optimization module is used to update the acquisition frequency of business subsystems with deviations, thereby obtaining an optimized data acquisition frequency configuration.

[0013] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects: This invention discloses an enterprise twin compliance control method and system based on spatiotemporal knowledge evolution and non-equilibrium alignment. It proposes a complete solution to problems such as latency resonance, misalignment of acquisition rhythm, and frequency adjustment deviation in multi-source data stream acquisition links. This invention accurately identifies latency resonance intervals and timestamp misalignments through latency-aware sampling and beat frequency interference detection. By fusing time alignment window scales and distortion interval identifiers, it corrects data stream deviations and achieves precise calibration of acquisition timing. Simultaneously, it dynamically adjusts the acquisition frequency based on a comprehensive deviation evaluation value and verifies the optimization results through twin mirror operation trajectory verification, ensuring consistency in data acquisition rhythm between the physical link and the twin environment. The core innovation of this invention lies in utilizing spatiotemporal knowledge evolution and non-equilibrium alignment technology to achieve stable suppression of latency resonance in multi-source data streams and dynamic optimization of frequency configuration, ultimately improving the compliance and stability of enterprise data acquisition and providing an efficient control method for complex business systems. Attached Figure Description

[0014] Figure 1 This is a flowchart of an enterprise twin compliance control method based on spatiotemporal knowledge evolution and non-equilibrium state alignment according to the present invention.

[0015] Figure 2 This is a schematic diagram of an enterprise twin compliance control method based on spatiotemporal knowledge evolution and non-equilibrium state alignment according to the present invention.

[0016] Figure 3 This is another schematic diagram of an enterprise twin compliance control method based on spatiotemporal knowledge evolution and non-equilibrium state alignment according to the present invention.

[0017] Figure 4 This is a schematic diagram of the structure of an enterprise twin compliance control system based on spatiotemporal knowledge evolution and non-equilibrium state alignment according to the present invention. Detailed Implementation

[0018] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] like Figures 1-4 This embodiment of an enterprise twin compliance control method and system based on spatiotemporal knowledge evolution and non-equilibrium state alignment may specifically include: Step S101: Obtain the transmission link delay distribution and acquisition period of each service subsystem, and extract the absolute delay duration arrangement and beat frequency interference waveform of each link.

[0020] By acquiring data from multiple data streams, the data reporting nodes and transmission path identifiers of each business subsystem are obtained. For each transmission path identifier, the transmission timestamp of the data packet from the source node to the aggregation node is collected. The one-way transmission delay of each link is calculated based on the transmission timestamp. The one-way transmission delays are grouped and aggregated according to the business subsystem to obtain the transmission link delay distribution of each business subsystem. At the same time, the preset collection period of each business subsystem is read, and an association storage is established between the collection period and the corresponding transmission link delay distribution to form a correspondence table between link delay and collection period. According to the correspondence table, the transmission link delay values ​​of each business subsystem are extracted and sorted in ascending order of delay value to form an absolute delay duration arrangement. For the delay difference between adjacent business subsystems in the absolute delay duration arrangement, if the difference is less than a preset delay difference threshold, it is marked as a potential interference segment. The start and end positions of each potential interference segment are bound to the corresponding business subsystem identifier to obtain an absolute delay duration arrangement with interference segment markings. Based on the interference segment markers, the acquisition cycle values ​​of the corresponding service subsystems are read, and the acquisition frequency is obtained by taking the reciprocal of each acquisition cycle value. The difference between the acquisition frequencies of adjacent service subsystems is calculated as the beat frequency. A corresponding sine waveform is generated for each beat frequency, and the amplitude is accumulated point by point to obtain the beat frequency signal curve of each interference segment. The beat frequency signal curve is discretely sampled, and the amplitude change sequence of the sampling points is extracted. A beat frequency interference waveform is generated based on the amplitude change sequence and stored in correspondence with the absolute time delay.

[0021] In one implementation, the specific process involves using an amplitude variation sequence as input, such as the sequence [1.2, 1.5, 1.3, 1.4], and generating a continuous waveform using a linear interpolation algorithm with an interpolation interval of 0.1 seconds. A multi-source data stream acquisition link refers to the physical or logical channel through which various business subsystems in an enterprise twin environment transmit business data to the data aggregation platform. Each acquisition link corresponds to an independent business subsystem, which generates data packets within a preset acquisition period and sends them to the aggregation node. The data reporting node is the logical unit in the business subsystem responsible for data encapsulation and transmission. The transmission path identifier is used to uniquely identify the transmission route from the data reporting node to the aggregation node.

[0022] Specifically, for each transmission path identifier, the sending timestamp carried by the data packet and the receiving timestamp recorded by the aggregation node when receiving the data packet are collected. The receiving timestamp is subtracted from the sending timestamp to obtain the one-way transmission delay consumed by the data packet during transmission. In the supply chain management scenario, the inventory subsystem, logistics subsystem, and sales subsystem report data through their respective collection links. Due to differences in network topology and transmission distance, the one-way transmission delays of each link differ. Multiple one-way transmission delays collected by the same business subsystem within a preset time window are statistically aggregated to obtain the transmission link delay distribution of the business subsystem. This delay distribution is stored in the form of delay value ranges and corresponding frequencies of occurrence.

[0023] In one embodiment, the process of forming the absolute delay length arrangement is to extract the mean delay from the transmission link delay distribution of each service subsystem and sort them from low to high according to the numerical value.

[0024] For example, if the average latency of the inventory subsystem is 15 milliseconds, the logistics subsystem is 20 milliseconds, and the sales subsystem is 18 milliseconds, then the order of arrangement is inventory subsystem, sales subsystem, and logistics subsystem. The latency differences between adjacent business subsystems after the arrangement are compared one by one. If the difference is less than a preset latency difference threshold, it indicates that the data arrival times of these two business subsystems are close, and timestamp overlap or misalignment is likely to occur during data aggregation. This adjacent position is marked as a potential interference segment.

[0025] It should be noted that the generation of the beat frequency interference waveform is based on the difference in acquisition frequencies between various service subsystems. The acquisition frequency is the reciprocal of the acquisition period. When there is a slight difference in the acquisition frequencies of two service subsystems, their data reporting rhythm will exhibit a periodic alternation of synchronization and misalignment. The difference in acquisition frequencies between adjacent service subsystems is used as the beat frequency. A sine waveform is generated using this beat frequency as a parameter. The amplitude of the sine waveforms corresponding to multiple beat frequencies within the same interference segment is accumulated point by point to form a beat frequency signal curve. The beat frequency signal curve is discretely sampled at fixed intervals, and the amplitude of each sampling point is extracted. The amplitudes are arranged in chronological order to form an amplitude change sequence. The beat frequency interference waveform is generated based on the peak and trough distribution characteristics of the amplitude change sequence to characterize the rhythmic interference state of multi-source data streams in the time dimension.

[0026] Step S102: Based on the absolute duration of the time delay and the beat frequency interference waveform, analyze the acquisition rhythm between the multi-source data streams to determine the occurrence interval of time delay resonance and the timestamp misalignment distance of the periodic distortion components.

[0027] Based on the absolute duration of the delay, the delay values ​​and their order corresponding to each service subsystem are extracted. For the delay values ​​of adjacent service subsystems in the arrangement, the difference between adjacent delay values ​​is calculated to form a delay difference sequence. Each difference in the delay difference sequence is traversed, and adjacent positions with differences less than a preset delay difference threshold are marked. The service subsystems corresponding to the marked positions are defined as rhythm offset groups. The start and end service subsystem identifiers of each rhythm offset group are recorded to obtain a rhythm offset distribution record. Based on the rhythm offset distribution record, the beat frequency interference waveforms corresponding to each rhythm offset group are read. The beat frequency interference waveforms are scanned point by point. The position where the amplitude changes from rising to falling in the waveform is identified as the waveform peak point, and the position where the amplitude changes from falling to rising is identified as the waveform valley point. The time interval between adjacent waveform peak points is defined as the rhythm synchronization period, and the time interval between the waveform peak point and the adjacent waveform valley point is defined as the rhythm misalignment half period, forming the acquisition rhythm feature sequence of each rhythm offset group. Based on the acquired rhythm feature sequence, the rhythm synchronization period of each rhythm offset group is extracted. If the rhythm synchronization period is less than a preset resonance period threshold, it is determined that the rhythm offset group has a time delay resonance phenomenon. The time range corresponding to the rhythm offset group is marked as the time delay resonance occurrence interval. The start time coordinate and end time coordinate of each time delay resonance occurrence interval are recorded to form a time delay resonance occurrence interval list, and the corresponding beat frequency interference waveform segment is associated with it. For each occurrence interval in the time delay resonance occurrence interval list, the associated beat frequency interference waveform segment is read, and the waveform segment is spectrally decomposed. The frequency component whose amplitude exceeds the preset distortion amplitude threshold is extracted as the periodic distortion component. For each periodic distortion component, the reciprocal of its frequency value is taken to obtain the corresponding period duration. The period duration is compared with the rhythm misalignment half-cycle to obtain the timestamp misalignment spacing corresponding to the periodic distortion component.

[0028] In one implementation, the specific comparison method involves calculating the absolute value of the difference between the two, i.e., the timestamp misalignment interval equals the absolute difference between the cycle duration and the rhythm misalignment half-cycle. If the difference is less than 0.1 seconds, it is considered a low misalignment; otherwise, it is a high misalignment. A rhythm offset group refers to a combination of business subsystems formed when the latency difference between adjacent business subsystems is less than a preset latency difference threshold in the absolute latency duration arrangement. In the supply chain management scenario of an enterprise twin environment, if the transmission latency difference between the inventory subsystem and the sales subsystem is lower than this threshold, they are grouped into the same rhythm offset group, indicating a risk of overlapping time windows for their data to arrive at the aggregation node.

[0029] Specifically, the beat frequency interferometry detection unit is a processing unit that performs point-by-point scanning and feature extraction on the beat frequency interferometry waveform. This detection unit receives the beat frequency interferometry waveforms corresponding to each rhythm offset group as input and traverses the amplitude sequence of the waveform according to a preset time step. During the traversal, the detection unit records the amplitude of each sampling point and compares it with the amplitudes of adjacent sampling points. When the amplitude of a sampling point is greater than both the preceding and following sampling points, that sampling point is identified as a waveform peak; when the amplitude of a sampling point is less than both the preceding and following sampling points, that sampling point is identified as a waveform valley. Through this point-by-point comparison method, the detection unit can accurately locate all extreme points in the beat frequency interferometry waveform.

[0030] For example, in the analysis of the collection rhythm of logistics data and sales data, the beat frequency interference waveform exhibits periodic fluctuations. The peak point of the waveform corresponds to the moment when the data reporting times of the two subsystems are closest, and the valley point of the waveform corresponds to the moment when the data reporting times of the two subsystems differ the most.

[0031] It's important to note that the rhythm synchronization period and the rhythm misalignment half-cycle are two core metrics describing the relationship between the collection rhythms of multi-source data streams. The rhythm synchronization period is defined as the time interval between two adjacent waveform peaks, reflecting the complete time required for the data reporting rhythms of two business subsystems to transition from a synchronized state to a synchronized state and back to a synchronized state. In a supply chain management scenario, if the inventory subsystem collects data every 60 minutes and the logistics subsystem collects data every 55 minutes, then their rhythm synchronization period is approximately 660 minutes, meaning that the data reporting times of the two subsystems will approach each other again every 660 minutes. The rhythm misalignment half-cycle is defined as the time interval between a waveform peak and an adjacent waveform trough, reflecting the time elapsed from a synchronized state to the state of maximum misalignment. This metric is used to assess the severity of data timestamp misalignment.

[0032] In one embodiment, the collection rhythm feature sequence of each rhythm offset group includes the rhythm synchronization period and rhythm misalignment half-cycle values ​​corresponding to all business subsystem combinations within the group, stored in a list and associated with the corresponding business subsystem identifiers. Further, the determination of the time delay resonance phenomenon is based on a comparison between the rhythm synchronization period and a preset resonance period threshold. When the rhythm synchronization period is less than the resonance period threshold, it indicates that the collection rhythm differences between the two business subsystems are small, and their data reporting times are frequently close, easily leading to timestamp confusion during data aggregation. In the compliance monitoring scenario of enterprise twins, if the rhythm synchronization period of inventory data and sales data is too short, data timestamps will repeatedly appear close within a short period, affecting the accurate judgment of inventory-sales matching.

[0033] Understandably, the start and end time coordinates of the time delay resonance occurrence interval are used to mark which time periods the time delay resonance phenomenon exists, and this interval list provides a location basis for subsequent data repair.

[0034] In one possible implementation, spectral decomposition is the process of separating the frequency components of a beat frequency interference waveform segment within the time-delay resonance interval. By applying a Fourier transform to the waveform segment, the time-domain waveform is converted into a frequency-domain representation, obtaining each frequency component and its corresponding amplitude. In the frequency-domain representation, frequency components with amplitudes exceeding a preset distortion amplitude threshold are extracted as periodic distortion components. These components represent rhythmic differences that interfere with data time alignment at specific frequencies. For each periodic distortion component, the reciprocal of its frequency value is taken to obtain the corresponding period duration. The difference between this period duration and the rhythmic misalignment half-cycle is calculated to obtain the timestamp misalignment interval, which is used to quantify the degree of influence of periodic distortion on the data timestamp offset.

[0035] Preferably, the timestamp misalignment interval is stored in milliseconds and bound to the corresponding time delay resonance occurrence interval to form a complete association record between time delay resonance and timestamp misalignment. This record is used to guide the setting of the data time alignment window.

[0036] Step S103: Evaluate the occurrence range of time delay resonance, identify resonance occurrence ranges that exceed the preset allowable range, and fuse the beat frequency interference waveform of the resonance occurrence range with the timestamp misalignment spacing to obtain the time alignment window scale and distortion range identifier.

[0037] Based on the list of time delay resonance occurrence intervals, the start and end time coordinates of each occurrence interval are extracted, the duration of each occurrence interval is calculated, and the duration is compared with a preset allowable duration threshold. If the duration exceeds the allowable duration threshold, the occurrence interval is marked as an over-limit resonance interval. The interval boundary and the corresponding business subsystem identifier of each over-limit resonance interval are recorded. The business subsystem identifier is derived from the association field of each interval in the time delay resonance occurrence interval list, forming an over-limit resonance interval set. For each over-limit resonance interval in the set of over-limit resonance intervals, the associated beat frequency interference waveform segment is read. This waveform segment is obtained from the waveform database by matching the service subsystem identifier and interval boundary. Amplitude statistics are performed on the waveform segment, and the maximum amplitude and average amplitude are extracted. The ratio of the maximum amplitude to the average amplitude is defined as the waveform peak-to-average power ratio (PAPR). Simultaneously, the timestamp misalignment distance corresponding to the over-limit resonance interval is read. This timestamp misalignment distance is obtained by calculating the average difference of the timestamp sequences within the interval. The PAPR is multiplied by the timestamp misalignment distance to obtain the deviation value of each over-limit resonance interval. For each over-limit resonance interval, the timestamp misalignment distance is multiplied by a preset window expansion coefficient to obtain the time alignment window scale corresponding to the over-limit resonance interval. For each over-limit resonance interval, the interval boundary, deviation value, and time alignment window scale are associated and encapsulated to generate a distortion interval identifier for that over-limit resonance interval. The distortion interval identifier includes the start and end time coordinates of the over-limit resonance interval, the identifier of the business subsystem involved, and the corresponding time alignment window scale. Each distortion interval identifier is stored in the distortion interval identifier list.

[0038] In one implementation, the over-limit resonance interval refers to the interval in the list of time delay resonance occurrence intervals whose duration exceeds a preset allowable duration threshold. The duration is calculated as the difference between the ending time coordinate and the starting time coordinate. In the enterprise twin supply chain management scenario, if the time delay resonance between the inventory subsystem and the logistics subsystem lasts for too long, it indicates that the data timestamps of the two subsystems are frequently close within that time period, which can easily lead to data alignment confusion.

[0039] Specifically, the allowable duration threshold is set based on the data collection cycle of the business subsystem and the timeliness requirements of compliance judgment. In the context of supply chain compliance monitoring, if the data collection cycle is thirty minutes, the allowable duration threshold is usually set as a multiple of the collection cycle to filter out long-term resonance intervals that have a significant impact on compliance judgment.

[0040] It should be noted that the waveform kurtosis index is used to quantify the severity of amplitude fluctuations in beat frequency interference waveforms within the over-limit resonance range. The calculation process involves iterating through the amplitude values ​​of all sampling points within the beat frequency interference waveform segment corresponding to the over-limit resonance range, recording the maximum amplitude, and simultaneously calculating the arithmetic mean of the amplitude values ​​at all sampling points to obtain the average amplitude. Dividing the maximum amplitude by the average amplitude yields the waveform kurtosis index. A large waveform kurtosis index indicates the presence of moments with significant amplitude abrupt changes within the over-limit resonance range, i.e., a steep transition point where the data reporting rhythm rapidly switches from a synchronized state to a misaligned state. In an enterprise twin environment, such a steep rhythm switch can easily lead to significant deviations in compliance judgments within a short period.

[0041] For example, in the analysis of the collection rhythm of inventory and sales data, if a waveform segment in a certain over-limit resonance interval exhibits a sharp peak shape, it indicates that the data reporting times of the two subsystems underwent a rapid change from high synchronization to significant misalignment within that interval, and the waveform kurtosis index corresponds to a high value. Furthermore, the deviation value is the product of the waveform kurtosis index and the timestamp misalignment distance, used to comprehensively assess the severity of the over-limit resonance interval. The waveform kurtosis index reflects the severity of amplitude fluctuations, while the timestamp misalignment distance reflects the magnitude of the data timestamp offset. The deviation value obtained by multiplying the two can simultaneously capture the steepness of the rhythm change and the magnitude of the time offset.

[0042] In one embodiment, if the waveform kurtosis index of a certain over-limit resonance interval is two and the timestamp misalignment interval is fifteen milliseconds, then the deviation value is thirty. The larger the value, the more serious the interference of the interval on data time alignment.

[0043] In one possible implementation, the time alignment window size is determined based on the timestamp misalignment distance and a preset window expansion coefficient. The window expansion coefficient is a value greater than one, used to expand the time window to both sides of the timestamp misalignment distance to form a time window with permissible offset. The time alignment window size equals the timestamp misalignment distance multiplied by the window expansion coefficient, representing the maximum allowed time offset range when performing data time alignment. In a supply chain management scenario, if the timestamp misalignment distance is ten milliseconds and the window expansion coefficient is 1.5, then the time alignment window size is fifteen milliseconds. That is, within this over-limit resonance range, data with timestamp deviations within fifteen milliseconds are allowed to be treated as data from the same moment and aligned accordingly.

[0044] Understandably, the value of the window expansion factor is related to the data sensitivity of the business subsystem. For business subsystems with high requirements for compliance judgment accuracy, a smaller window expansion factor is used to reduce the time alignment window scale; for business subsystems with strong fault tolerance, a larger window expansion factor is used to expand the time alignment window scale.

[0045] Preferably, the distortion interval identifier is the associative encapsulation result of various features of the over-limit resonance interval, including the start and end time coordinates of the over-limit resonance interval, the identifiers of the involved business subsystems, the deviation degree value, and the time alignment window scale. In an enterprise twin environment, the distortion interval identifiers are stored in a distortion interval identifier list in the form of structured data, with each distortion interval identifier corresponding to an over-limit resonance interval. By reading the start and end time coordinates in the distortion interval identifier, the time period requiring data repair can be located; by reading the business subsystem identifier, the data source involved can be determined; and by reading the time alignment window scale, the allowable range for data timestamp correction can be obtained.

[0046] In one embodiment, the list of distortion interval identifiers is stored and arranged according to the degree of deviation. Distortion interval identifiers with larger deviation values ​​are arranged first, which facilitates prioritizing the processing of over-limit resonance intervals that have a greater impact on compliance judgment, thereby realizing hierarchical identification and location of latency resonance problems in multi-source data streams.

[0047] Step S104: Based on the time alignment window scale and the distortion interval identifier, extract the component data stream within the corresponding time window, match the component data stream with the distortion interval identifier, and determine the corrected deviation and the filtering result of the acquisition timing misalignment event.

[0048] Based on the list of distortion interval identifiers, the start and end time coordinates and time alignment window scale of each distortion interval identifier are read. For the business subsystem corresponding to each distortion interval identifier, the original data records reported by that business subsystem within the start and end time coordinate range are extracted. The original data records are arranged in timestamp order to form component data streams. At the same time, the offset of the timestamp of each data record in the component data stream from the start time coordinate of the distortion interval is recorded to obtain the time offset sequence of each data record. For each data record in the component data stream, the offset in its time offset sequence is read and compared with the time alignment window scale. If the offset is less than the time alignment window scale, the data record is marked as data within the window. If the offset is greater than or equal to the time alignment window scale, the data record is marked as a data acquisition timing misalignment event. The number of data acquisition timing misalignment events corresponding to each distortion interval identifier is counted, and the timestamp and offset of each data acquisition timing misalignment event are recorded to form the filtering result of data acquisition timing misalignment events. Based on the filtering results, for each distortion interval identifier, the ratio of the number of data in the statistical window to the total data volume of the component data stream is calculated. This ratio is defined as the corrected deviation. Each distortion interval identifier is associated with and stored in conjunction with the corresponding corrected deviation and the filtering results of the acquisition timing misalignment event.

[0049] In one implementation, a component data stream refers to a local subset of data extracted from a multi-source data stream according to the time range identified by the distortion interval and the business subsystem identifier. In a supply chain management scenario, if a distortion interval identifier involves the inventory subsystem, then all data records within the start and end time coordinate range of that distortion interval are extracted from the complete data stream of the inventory subsystem. These data records are arranged in timestamp order to form the component data stream corresponding to that distortion interval.

[0050] Specifically, the time offset sequence is used to record the offset of the timestamp of each data record in the component data stream relative to the starting time coordinate of the distortion interval. The offset is calculated by subtracting the starting time coordinate of the distortion interval from the timestamp of each data record to obtain the relative time position of the data record within the distortion interval. In an enterprise twin environment, the time offset sequence is stored as a list of values, where each value corresponds to the offset of a data record in the component data stream.

[0051] It's important to note that the determination of data within the window and data acquisition time-series misalignment events is based on a comparison of the offset and the time alignment window scale. The time alignment window scale defines the maximum allowed time offset range when performing data time alignment. When the offset of a data record is less than the time alignment window scale, it indicates that the timestamp of the data record is within the allowable offset range, and the data record is marked as data within the window. When the offset is greater than or equal to the time alignment window scale, it indicates that the timestamp of the data record exceeds the allowable offset range, and the data record is marked as a data acquisition time-series misalignment event. In supply chain compliance monitoring scenarios, data acquisition time-series misalignment events indicate a significant deviation between the reporting time of inventory or logistics data and the expected synchronization time. If such events are not identified, compliance judgments can easily misjudge normal business fluctuations as abnormalities.

[0052] For example, if the time alignment window scale for a certain distortion interval is 20 milliseconds, and the offset of a data record in the component data stream is 25 milliseconds, then this data record is marked as a timing misalignment event, and its timestamp and offset are recorded. Further, the corrected deviation is the ratio of the number of data records within the window to the total amount of data in the component data stream. This ratio reflects the completeness of data alignment within the distortion interval after applying time alignment window filtering. The closer the corrected deviation is to one, the more likely the vast majority of data records fall within the time alignment window, indicating better consistency of data timestamps; the smaller the corrected deviation, the more likely a large number of data records are identified as timing misalignment events, indicating a more severe data time misalignment problem in this distortion interval.

[0053] Step S105: Perform joint analysis on the filtering results of deviation and timing misalignment events to obtain the comprehensive deviation evaluation value of each business subsystem, and map the business subsystem whose comprehensive deviation evaluation value exceeds the preset allowable range to the corresponding frequency adjustment signal.

[0054] Based on the corrected deviation and the filtering results of timing misalignment events, for each business subsystem, the deviation values ​​of all distortion interval identifiers corresponding to that subsystem are extracted. The arithmetic mean of these deviation values ​​is calculated to obtain the average deviation. Simultaneously, the total number of timing misalignment events corresponding to that subsystem is counted. The reciprocal of the average deviation is then added to the total number of events to obtain the comprehensive deviation evaluation value for each business subsystem. For each comprehensive deviation evaluation value, it is compared with a preset allowable threshold. If the comprehensive deviation evaluation value exceeds the allowable threshold, the subsystem is marked as a subsystem to be adjusted. The identifier of each subsystem to be adjusted and its corresponding comprehensive deviation evaluation value are recorded, forming a list of subsystems to be adjusted. Based on this list, for each subsystem to be adjusted, the difference between its comprehensive deviation evaluation value and the allowable threshold is read. The adjustment level is determined based on the magnitude of the difference; the larger the difference, the higher the adjustment level. The adjustment level and the identifier of the subsystem to be adjusted are encapsulated into a frequency adjustment signal.

[0055] In one implementation, the comprehensive deviation assessment value is a comprehensive indicator used to quantify the data time alignment status of each business subsystem. For each business subsystem, deviation values ​​are extracted from all corresponding distortion interval identifiers, and these deviation values ​​are arithmetically averaged to obtain the average deviation. The average deviation reflects the average completeness of data alignment for that business subsystem across all distortion intervals; the closer the value is to one, the better the alignment. The reciprocal of the average deviation is then added to the total number of data collection time-series misalignment events to obtain the comprehensive deviation assessment value. In a supply chain management scenario, if the average deviation of the inventory subsystem is 0.8 and the total number of data collection time-series misalignment events is five, then the comprehensive deviation assessment value is 1.25 plus five, i.e., 6.25.

[0056] Specifically, the allowable range threshold is a numerical boundary pre-set based on business compliance requirements. In an enterprise twin environment, different business subsystems have different requirements for the accuracy of data time alignment, and the setting of the allowable range threshold needs to be adjusted according to the characteristics of the business.

[0057] It should be noted that the list of subsystems to be adjusted records all business subsystems whose comprehensive deviation assessment values ​​exceed the allowable threshold, along with their corresponding assessment values. In a supply chain compliance monitoring scenario, if the comprehensive deviation assessment values ​​of both the inventory subsystem and the logistics subsystem exceed the threshold, both will be included in the list of subsystems to be adjusted. This indicates that the data collection frequency of these two subsystems differs significantly from that of other subsystems, which could easily lead to confusion in data timestamp alignment.

[0058] In one embodiment, the adjustment magnitude level is determined based on the difference between the comprehensive deviation assessment value and the allowable range threshold. A larger difference indicates a more severe data time deviation in the business subsystem, and the corresponding adjustment magnitude level is higher. The adjustment magnitude level is expressed numerically, with a larger value indicating a stronger adjustment. The frequency adjustment signal encapsulates the adjustment magnitude level with the identifier of the subsystem to be adjusted, forming structured data for subsequent processing, which guides the specific adjustment direction and magnitude of the business subsystem's acquisition frequency.

[0059] Step S106: Adjust the acquisition frequency of each service subsystem according to the frequency adjustment signal, obtain the sampling interval duration and data reporting cycle of each service subsystem after adjustment, recalculate the comprehensive deviation evaluation value, and if the updated comprehensive deviation evaluation value still exceeds the preset convergence threshold, it is determined that there is a deviation in the service subsystem.

[0060] Based on the frequency adjustment signal, the adjustment magnitude level and subsystem identifier of each subsystem to be adjusted are read. For each subsystem to be adjusted, its current acquisition frequency is incrementally adjusted according to the adjustment magnitude level. The higher the adjustment magnitude level, the larger the increment of the acquisition frequency, thus obtaining the adjusted acquisition frequency of each subsystem to be adjusted. The sampling interval duration is obtained by taking the reciprocal of the adjusted acquisition frequency. At the same time, the sampling interval duration is multiplied by a preset reporting multiple of 5 to obtain the data reporting period, forming the adjusted acquisition configuration of each subsystem to be adjusted. According to the adjusted acquisition configuration, for each subsystem to be adjusted, the data stream is re-acquired according to the adjusted acquisition frequency and data reporting period. The distortion interval identifier and deviation are extracted from the re-acquired data stream. The number of acquisition timing misalignment events is counted. The reciprocal of the deviation is added to the number of events to obtain the updated comprehensive deviation evaluation value of each subsystem to be adjusted. In one implementation, the extraction process employs a time-aligned window algorithm. The input consists of a data stream sequence and a baseline time series. The start and end coordinates of the distortion interval are calculated as identifiers. The deviation D is calculated using the formula D=sum(|xi-yi|) / n, where xi is the acquired value, yi is the baseline value, and n is the number of points. An example output is identifier [10, 20] and D=0.3. For the updated comprehensive deviation evaluation value, the updated comprehensive deviation evaluation value of each subsystem to be adjusted is compared with a preset convergence threshold 1. If the updated comprehensive deviation evaluation value still exceeds the convergence threshold, the subsystem to be adjusted is determined to be a business subsystem with deviation, and its subsystem identifier, sampling interval duration, data reporting cycle, and updated comprehensive deviation evaluation value are recorded.

[0061] The frequency adjustment signal includes the identifier of the subsystem to be adjusted and the adjustment magnitude level. The adjustment magnitude level is expressed numerically, with a larger value indicating a stronger adjustment. The specific process for incrementally adjusting the current acquisition frequency according to the adjustment magnitude level for each subsystem to be adjusted is as follows: read the current acquisition frequency value of the subsystem, multiply the adjustment magnitude level by a preset unit increment value to obtain the frequency increment, and add the frequency increment to the current acquisition frequency to obtain the adjusted acquisition frequency.

[0062] Specifically, the sampling interval is an inverse of the sampling frequency. In a supply chain management scenario, if the adjusted sampling frequency of the inventory subsystem is twice per minute, then the sampling interval is thirty seconds. The data reporting cycle is the product of the sampling interval and the preset reporting multiple, where the reporting multiple indicates how many sampling cycles data is reported to the aggregation node once.

[0063] In one embodiment, if the sampling interval is 30 seconds and the reporting multiplier is 2, then the data reporting period is 60 seconds.

[0064] It should be noted that the adjusted acquisition configuration is stored in structured data format, including the identifier of the subsystem to be adjusted, the adjusted acquisition frequency, sampling interval duration, and data reporting cycle. Furthermore, after re-acquiring the data stream according to the adjusted acquisition frequency and data reporting cycle, the re-acquisitioned data stream undergoes the same processing procedure as the initial evaluation, including extracting distortion interval identifiers, calculating the deviation, counting the number of acquisition timing misalignment events, and adding the reciprocal of the deviation to the number of events to obtain the updated comprehensive deviation evaluation value.

[0065] In one possible implementation, the convergence threshold is a pre-defined lower limit of the comprehensive deviation assessment value, used to determine whether the data time alignment status after frequency adjustment has reached the expected level. The convergence threshold is typically lower than the allowable range threshold, indicating that the comprehensive deviation assessment value should be reduced to a more stringent range after frequency adjustment. If the updated comprehensive deviation assessment value still exceeds the convergence threshold, it indicates that a single frequency adjustment has failed to bring the data time alignment status of this business subsystem to the expected convergence level. This subsystem is identified as a business subsystem with deviation, and its subsystem identifier, sampling interval duration, data reporting cycle, and updated comprehensive deviation assessment value are fully recorded to facilitate subsequent, more in-depth frequency stability control.

[0066] Step S107: Update the acquisition frequency of the business subsystem with deviation to obtain the optimized data acquisition frequency configuration.

[0067] The process involves acquiring the identifier of the business subsystem exhibiting deviation and the updated comprehensive deviation assessment value. Based on the difference between the comprehensive deviation assessment value and the convergence threshold, the frequency increment is determined; a larger difference results in a larger increment. This increment is then added to the current acquisition frequency of the business subsystem to obtain the second-adjusted acquisition frequency. Using this second-adjusted acquisition frequency, the data stream is re-acquired, and the comprehensive deviation assessment value is recalculated. This comprehensive deviation assessment value is compared again with the convergence threshold. If it still exceeds the convergence threshold, the frequency increment adjustment is repeated until the comprehensive deviation assessment value falls within the convergence threshold range. The acquisition frequency value at this point is recorded. The sampling interval duration is obtained by taking the reciprocal of the acquisition frequency value. The sampling interval duration is then multiplied by a preset reporting multiplier to obtain the data reporting cycle. The acquisition frequency value, sampling interval duration, and data reporting cycle are encapsulated into an optimized data acquisition frequency configuration. This optimized data acquisition frequency configuration is then distributed to the corresponding business subsystem to achieve stable suppression of latency resonance in multi-source data streams.

[0068] In one implementation, the system receives the identifier of the service subsystem with the deviation and the corresponding comprehensive deviation assessment value as input. The frequency increment magnitude is determined based on the difference between the comprehensive deviation assessment value and the convergence threshold. A larger difference indicates a greater gap between the data time alignment status of the service subsystem and the expected convergence level, and the corresponding frequency increment magnitude is larger.

[0069] Specifically, the secondary adjustment process is a further adjustment made after the initial frequency adjustment failed to bring the overall deviation assessment value to converge. In a supply chain management scenario, if the overall deviation assessment value of the inventory subsystem still exceeds the convergence threshold after the initial frequency adjustment, the frequency stabilization control unit increases the frequency increment of its current acquisition frequency to obtain the acquisition frequency after the secondary adjustment. Data streams are then re-acquired according to the secondary adjustment acquisition frequency, and the same deviation calculation and event statistics process as described above is executed to obtain the updated overall deviation assessment value. If this overall deviation assessment value still exceeds the convergence threshold, frequency increment adjustment continues until the overall deviation assessment value falls within the convergence threshold range.

[0070] It should be noted that the frequency increment during the iterative adjustment process is not a fixed value, but gradually decreases as the number of iterations increases.

[0071] In one embodiment, the frequency increment amplitude of the previous iteration is multiplied by a preset attenuation coefficient to obtain the frequency increment amplitude of the current iteration, thereby achieving fine adjustment of the sampling frequency. Further, when the comprehensive deviation evaluation value falls within the convergence threshold range, the sampling frequency value at this time is recorded, the reciprocal of this value is taken to obtain the sampling interval duration, and the sampling interval duration is multiplied by a preset reporting multiple to obtain the data reporting period.

[0072] Preferably, the optimized data acquisition frequency configuration is encapsulated in structured data format, including the business subsystem identifier, acquisition frequency value, sampling interval duration, and data reporting cycle. After this configuration is distributed to the corresponding business subsystem, the subsystem performs data acquisition and reporting according to the optimized acquisition frequency and data reporting cycle, achieving coordination of data rhythm with other business subsystems, thereby stably suppressing the latency resonance phenomenon between multi-source data streams.

[0073] Under the optimized data acquisition frequency, the running trajectory and sampling interval of the twin mirror of each business subsystem are retrieved, the data reporting cycle and latency resonance suppression performance after frequency adjustment in the twin environment are extracted, the consistency deviation between the twin mirror and the physical link in data acquisition rhythm is evaluated, and the optimized data acquisition frequency result after twin compliance verification is obtained.

[0074] Based on the optimized data acquisition frequency configuration, the corresponding twin images of each business subsystem are retrieved, and the running trajectory data of the twin images within a preset simulation period is read. This running trajectory data includes the timestamp sequence of the twin images simulating data acquisition at the optimized acquisition frequency. Simultaneously, the preset sampling interval duration boundaries of each business subsystem are obtained, forming a correlation record between the twin image running trajectory and the sampling boundaries. For each business subsystem in the correlation record, the data reporting cycle after frequency adjustment is extracted from the twin image running trajectory data. The number of times and duration of time delay resonance occur in the twin image within the simulation period are counted. A time delay resonance suppression performance score is obtained based on the ratio of the number of resonances and their duration to the total simulation period. Simultaneously, the data reporting cycle of the corresponding business subsystem's physical link is read. Based on the time delay resonance suppression performance score, the absolute value of the difference between the twin image's data reporting cycle and the physical link's data reporting cycle is calculated. This absolute value of the difference is compared with the sampling interval duration boundaries to obtain the consistency deviation rate. The consistency deviation rate and time delay resonance suppression performance score are recorded for each business subsystem, forming a set of twin compliance verification indicators. For each business subsystem in the set of twin compliance verification indicators, if the consistency deviation rate is lower than the preset deviation tolerance threshold and the delay resonance suppression performance score is higher than the preset suppression score threshold, then the data acquisition frequency configuration of the business subsystem is determined to have passed the twin compliance verification, and the configuration that has passed the verification is marked as the data acquisition frequency optimization result that has passed the twin compliance verification.

[0075] In one implementation, a twin mirror refers to a virtual simulation model established for each business subsystem within an enterprise twin environment. This virtual simulation model can simulate the data acquisition and reporting process according to a specified acquisition frequency, and its operational trajectory data records the timestamp and data volume of each data acquisition during the simulation. In a supply chain management scenario, the twin mirror of the inventory subsystem simulates the inventory data acquisition process in a virtual environment according to an optimized acquisition frequency, forming a series of timestamped simulation data records.

[0076] Specifically, the sampling interval duration boundary is a pre-defined allowable range for sampling intervals for each business subsystem. This boundary includes the minimum and maximum values ​​of the sampling interval, used to constrain the sampling frequency of the business subsystem within a reasonable range.

[0077] In one embodiment, if the sampling interval of the inventory subsystem is between 20 and 60 seconds, it indicates that the sampling interval of the subsystem should not be less than 20 seconds or more than 60 seconds, and the sampling frequency configuration exceeding this boundary range is considered to be non-compliant with business specifications.

[0078] It should be noted that the association record between the twin mirror's running trajectory and sampling boundaries binds and stores the timestamp sequence of the twin mirror within the simulation period with the sampling interval duration boundary of the corresponding business subsystem. This association record allows for boundary compliance checks on the twin mirror's running trajectory, determining whether the sampling interval during the simulation falls within the allowable range. Furthermore, the latency resonance suppression performance score quantifies the degree to which the twin mirror suppresses latency resonance phenomena within the simulation period. In an enterprise twin environment, when the twin mirror runs at the optimized sampling frequency, if the data reporting times of twin mirrors from multiple business subsystems no longer frequently appear close, it indicates that latency resonance phenomena have been effectively suppressed. The latency resonance suppression performance score is calculated based on two indicators: first, the ratio of the number of resonances to the total simulation period duration (a smaller ratio indicates a lower frequency of resonance); second, the ratio of the resonance duration to the total simulation period duration (a smaller ratio indicates a shorter duration of each resonance). The weighted sum of these two ratios is then taken as the reciprocal to obtain the latency resonance suppression performance score; a higher score indicates better suppression performance.

[0079] For example, in a supply chain compliance monitoring scenario, if the twin mirror images of the inventory subsystem and the logistics subsystem only experience two time-delay resonances within a 60-minute simulation period, with each resonance lasting for 30 seconds, then the resonance frequency ratio is low, the resonance duration ratio is also low, and the corresponding time-delay resonance suppression performance score is high.

[0080] In one possible implementation, the consistency deviation rate is used to evaluate the degree of matching between the twin mirror and the physical link in terms of data acquisition rhythm. The data reporting cycle of the twin mirror is extracted from the simulation run trajectory, while the data reporting cycle of the physical link is read from the optimized data acquisition frequency configuration. The consistency deviation rate is calculated as follows: the difference between the data reporting cycle of the twin mirror and the data reporting cycle of the physical link is calculated, and the absolute value of the difference is taken. This absolute value is then compared with the range value of the sampling interval duration boundary to obtain the proportion of the deviation to the boundary range. The smaller this proportion, the better the consistency between the twin mirror and the physical link in terms of data acquisition rhythm.

[0081] Understandably, the twin compliance verification indicator set stores the consistency deviation rate and latency resonance suppression performance score of each business subsystem in a structured data format. For each business subsystem, the set records its subsystem identifier, consistency deviation rate value, and latency resonance suppression performance score value, forming a complete verification indicator record.

[0082] Preferably, the twin compliance verification adopts a dual-threshold mechanism, that is, simultaneously setting threshold conditions for both the consistency deviation rate and the latency resonance suppression performance score. The deviation tolerance threshold defines the upper limit of the consistency deviation rate; when the consistency deviation rate is below this threshold, it indicates that the rhythm difference between the twin mirror and the physical link is within an acceptable range. The suppression score threshold defines the lower limit of the latency resonance suppression performance score; when the score is above this threshold, it indicates that the latency resonance phenomenon has been sufficiently suppressed. Only when the business subsystem simultaneously meets both threshold conditions is its data acquisition frequency configuration determined to have passed the twin compliance verification and marked as the optimized data acquisition frequency result that has passed the twin compliance verification.

[0083] In one embodiment, the optimized data acquisition frequency result, verified by the twin compliance, is output in structured data format, including the business subsystem identifier, the optimized acquisition frequency value, the sampling interval duration, the data reporting cycle, the consistency deviation rate, and the latency resonance suppression performance score. This optimization result can be distributed to the physical business subsystem for execution, achieving stable suppression of latency resonance in multi-source data streams, and reducing the risk of physical link deployment through pre-verification in the twin environment.

[0084] This invention provides an enterprise twin compliance control system based on spatiotemporal knowledge evolution and non-equilibrium state alignment, mainly comprising: The link delay distribution and waveform extraction module is used to obtain the transmission link delay distribution and acquisition period of each service subsystem, and extract the absolute delay duration arrangement and beat frequency interference waveform of each link; The time delay resonance and misalignment analysis module is used to analyze the acquisition rhythm between multi-source data streams based on the absolute time delay duration arrangement and beat frequency interference waveform, and to determine the occurrence interval of time delay resonance and the timestamp misalignment spacing of periodic distortion components. The resonance interval evaluation and fusion module is used to evaluate the occurrence interval of time delay resonance, identify resonance occurrence intervals that exceed the preset allowable range, and fuse the beat frequency interference waveform of the resonance occurrence interval with the timestamp misalignment spacing to obtain the time alignment window scale and distortion interval identifier. The component data stream extraction and matching processing module is used to extract the component data stream within the corresponding time window based on the time alignment window scale and the distortion interval identifier, match the component data stream with the distortion interval identifier, and determine the corrected deviation and the filtering result of the acquisition timing misalignment event. The deviation and filtering result joint analysis module is used to jointly analyze the deviation and the filtering results of the acquisition timing misalignment event to obtain the comprehensive deviation evaluation value of each business subsystem. The business subsystem whose comprehensive deviation evaluation value exceeds the preset allowable range is mapped to the corresponding frequency adjustment signal. The frequency adjustment signal mapping module is used to adjust the acquisition frequency of each service subsystem according to the frequency adjustment signal, obtain the sampling interval duration and data reporting cycle of each service subsystem after adjustment, recalculate the comprehensive deviation evaluation value, and if the updated comprehensive deviation evaluation value still exceeds the preset convergence threshold, it is determined that there is a deviation in the service subsystem. The data acquisition frequency adjustment and optimization module is used to update the acquisition frequency of business subsystems with deviations, thereby obtaining an optimized data acquisition frequency configuration.

[0085] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A corporate twin compliance control method based on spatiotemporal knowledge evolution and non-equilibrium state alignment, characterized in that, The method includes: Obtain the transmission link delay distribution and collection period of each business subsystem, and form a correspondence between link delay and collection period; Based on the aforementioned correspondence, the acquisition rhythm between multiple data streams is analyzed to determine the interval where time delay resonance occurs; The occurrence range of the time delay resonance is evaluated, resonance ranges exceeding the preset range are identified, and distortion range identifiers are generated; Based on the distortion interval identifier, extract the component data stream within the corresponding time window, match the component data stream with the distortion interval identifier, and determine the corrected deviation and the filtering result of the acquisition timing misalignment event. By jointly analyzing the filtering results of deviation and timing misalignment events, a comprehensive deviation evaluation value for each business subsystem is obtained. Business subsystems whose comprehensive deviation evaluation values ​​exceed the preset allowable range are mapped to corresponding frequency adjustment signals. The sampling frequency of each business subsystem is adjusted according to the frequency adjustment signal. The sampling interval and data reporting cycle of each business subsystem after adjustment are obtained. The comprehensive deviation evaluation value is recalculated. If the updated comprehensive deviation evaluation value still exceeds the preset convergence threshold, it is determined that there is a deviation in the business subsystem. Update the data collection frequency of the business subsystems with deviations to obtain an optimized data collection frequency configuration.

2. The enterprise twin compliance control method based on spatiotemporal knowledge evolution and non-equilibrium state alignment according to claim 1, characterized in that, The process of obtaining the transmission link delay distribution and collection period of each service subsystem to form a correspondence between link delay and collection period includes: Obtain the data reporting nodes and transmission path identifiers of each business subsystem; For each transmission path identifier, collect the transmission timestamp of the data packet from the source node to the sink node; Calculate the one-way transmission delay of each link based on the transmission timestamp; The unidirectional transmission delay is grouped and aggregated according to the business subsystem to obtain the transmission link delay distribution of each business subsystem. Read the preset collection period of each business subsystem, establish an association between the collection period and the corresponding transmission link delay distribution, and form a correspondence table between link delay and collection period.

3. The enterprise twin compliance control method based on spatiotemporal knowledge evolution and non-equilibrium state alignment according to claim 1, characterized in that, Based on the aforementioned correspondence, the acquisition rhythm between multiple data streams is analyzed to determine the interval of time delay resonance, including: calculating the time delay difference between adjacent business subsystems for the sorted time delay values. If the time delay difference is less than a preset threshold, it is marked as a potential interference segment; Based on the potential interference segment, a beat frequency interference waveform is generated; By scanning the beat frequency interference waveform, the peak and trough points of the waveform are identified, and the rhythm synchronization period is determined. If the rhythm synchronization period is less than a preset threshold, it is determined that there is a time delay resonance phenomenon, and the corresponding time range is marked as the time delay resonance occurrence interval.

4. The enterprise twin compliance control method based on spatiotemporal knowledge evolution and non-equilibrium state alignment according to claim 1, characterized in that, The evaluation of the occurrence interval of the time-delay resonance, identification of resonance intervals exceeding a preset range, and generation of distortion interval identifiers include: Extract the start and end time coordinates of each time-delay resonance occurrence interval and calculate the duration; If the duration exceeds a preset threshold, it is marked as an over-limit resonance interval; For the aforementioned over-limit resonance range, read the associated beat frequency interference waveform segments and calculate the waveform kurtosis index; The degree of deviation is calculated based on the waveform kurtosis index and the timestamp misalignment interval. The time alignment window size is determined based on the timestamp misalignment spacing. The interval boundary, deviation value, and time alignment window scale of the over-limit resonance interval are encapsulated as a distortion interval identifier.

5. The enterprise twin compliance control method based on spatiotemporal knowledge evolution and non-equilibrium state alignment according to claim 1, characterized in that, The step of extracting component data streams within the corresponding time window based on the distortion interval identifier, matching the component data streams with the distortion interval identifiers, and determining the corrected deviation and the filtering results of the acquisition timing misalignment events includes: Read the start and end time coordinates and time alignment window scale from the distortion interval identifier; For the business subsystems corresponding to the distortion interval identifiers, extract the original data records within the start and end time coordinate range to form component data streams; Record the time offset sequence of each data record in the component data stream; If the offset in the time offset sequence is less than the time alignment window scale, it is marked as data within the window; If the offset is greater than or equal to the time alignment window scale, it is marked as a timing misalignment event. The percentage of data within the statistical window is used to determine the deviation after correction.

6. The enterprise twin compliance control method based on spatiotemporal knowledge evolution and non-equilibrium state alignment according to claim 1, characterized in that, The filtering results of deviation and timing misalignment events are jointly analyzed to obtain a comprehensive deviation evaluation value for each business subsystem. Business subsystems whose comprehensive deviation evaluation values ​​exceed a preset allowable range are mapped to corresponding frequency adjustment signals, including: For each business subsystem, extract the deviation value of the corresponding distortion interval identifier and calculate the average deviation. Count the total number of data collection timing misalignment events corresponding to the aforementioned business subsystem; Calculate the comprehensive deviation assessment value based on the average deviation and the total number of events; If the comprehensive deviation evaluation value exceeds the preset range threshold, it is marked as a subsystem to be adjusted; For the subsystem to be adjusted, the adjustment level is determined, and a frequency adjustment signal is generated.

7. The enterprise twin compliance control method based on spatiotemporal knowledge evolution and non-equilibrium state alignment according to claim 1, characterized in that, The process involves adjusting the acquisition frequency of each service subsystem based on the frequency adjustment signal, obtaining the adjusted sampling interval and data reporting cycle of each service subsystem, recalculating the comprehensive deviation evaluation value, and determining the service subsystem with deviation if the updated comprehensive deviation evaluation value still exceeds the preset convergence threshold. This includes: Based on the frequency adjustment signal, read the adjustment level of the subsystem to be adjusted; The current acquisition frequency is adjusted incrementally according to the adjustment level described above; Calculate the sampling interval duration based on the adjusted sampling frequency; The data reporting cycle is determined based on the sampling interval duration; Re-collect the data stream according to the adjusted collection frequency and data reporting cycle; Calculate the updated comprehensive deviation assessment value based on the re-acquired data stream; If the updated comprehensive deviation assessment value exceeds the preset threshold, it is determined that the business subsystem has a deviation.

8. The enterprise twin compliance control method based on spatiotemporal knowledge evolution and non-equilibrium state alignment according to claim 1, characterized in that, The updated data collection frequency of the service subsystem with discrepancies is used to obtain an optimized data collection frequency configuration, including: Obtain the comprehensive deviation assessment value of the business subsystems that exhibit deviations; The frequency increment amplitude is determined based on the difference between the comprehensive deviation evaluation value and the preset threshold. The frequency increment is increased by the current sampling frequency to obtain the sampling frequency after secondary adjustment. Based on the adjusted acquisition frequency, the data stream is reacquired and the comprehensive deviation evaluation value is calculated. If the overall deviation assessment value still exceeds the preset threshold, the frequency increment adjustment will be repeated. If the comprehensive deviation evaluation value falls within the preset threshold range, the acquisition frequency value is recorded, and an optimized data acquisition frequency configuration is generated.

9. The enterprise twin compliance control method and system based on spatiotemporal knowledge evolution and non-equilibrium state alignment according to claim 1, characterized in that, The method further includes: under the optimized data acquisition frequency, retrieving the twin mirror operation trajectory and sampling interval duration boundary of each business subsystem, extracting the data reporting cycle and delay resonance suppression performance after frequency adjustment in the twin environment, evaluating the consistency deviation between the twin mirror and the physical link in data acquisition rhythm, and obtaining the data acquisition frequency optimization result verified by twin compliance.

10. A corporate twin compliance control system based on spatiotemporal knowledge evolution and non-equilibrium state alignment, characterized in that, The system includes: The link delay distribution and waveform extraction module is used to obtain the transmission link delay distribution and acquisition period of each service subsystem, and form a correspondence between link delay and acquisition period; The time delay resonance and misalignment analysis module is used to analyze the acquisition rhythm between multiple data streams based on the correspondence and determine the occurrence interval of time delay resonance. The resonance interval evaluation and fusion module is used to evaluate the occurrence interval of the time delay resonance, identify resonance intervals that exceed the preset range, and generate distortion interval identifiers. The component data stream extraction and matching processing module is used to extract the component data stream within the corresponding time window according to the distortion interval identifier, match the component data stream with the distortion interval identifier, and determine the corrected deviation and the filtering result of the acquisition timing misalignment event. The deviation and filtering result joint analysis module is used to jointly analyze the deviation and the filtering results of the acquisition timing misalignment event to obtain the comprehensive deviation evaluation value of each business subsystem. The business subsystem whose comprehensive deviation evaluation value exceeds the preset allowable range is mapped to the corresponding frequency adjustment signal. The frequency adjustment signal mapping module is used to adjust the acquisition frequency of each service subsystem according to the frequency adjustment signal, obtain the sampling interval duration and data reporting cycle of each service subsystem after adjustment, recalculate the comprehensive deviation evaluation value, and if the updated comprehensive deviation evaluation value still exceeds the preset convergence threshold, it is determined that there is a deviation in the service subsystem. The data acquisition frequency adjustment and optimization module is used to update the acquisition frequency of business subsystems with deviations, thereby obtaining an optimized data acquisition frequency configuration.

Images