A dynamic updating method of a digital twin substrate

Abstract

The application discloses a dynamic updating method of a digital twin bottom plate, and relates to the technical field of data processing. The method constructs a periodic influence intensity field according to the influence intensity of periodic indexes of associated services on data structure changes of the bottom plate, constructs a disturbance vector according to the influence intensity of external event trigger factors of the associated services on the data structure changes of the bottom plate, determines a data structure change rate of the digital twin bottom plate based on a historical data structure change sequence of the digital twin bottom plate, constructs a structure gradient tensor according to the data structure change rate, inputs the structure gradient tensor, the periodic influence intensity field and the disturbance vector into a preset tensor equation as variable, determines a change prediction result through eigenvalue decomposition of the tensor equation, generates a new data structure of the bottom plate according to the change prediction result, and migrates data in a current data structure of the bottom plate to the new data structure of the bottom plate, thereby improving change real-time performance.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, in particular to a dynamic updating method of digital twin bottom plate. BACKGROUND

[0002] Digital twin technology is widely used in intelligent manufacturing, smart city, intelligent transportation and other fields. Its core is to realize the mapping, analysis and prediction of real physical entities through real-time data driving virtual models. In this system, the digital twin bottom plate is the basic carrier supporting twin modeling, data access, state presentation and feedback control. Its data structure and updating mechanism are crucial.

[0003] In related digital twin bottom plate systems, static or semi-dynamic bottom plate structures are mostly used. Through pre-set data model structures, a small number of field supplements or static replacements are made after initialization to realize the updating of the bottom plate structure, which lacks structure change prediction capability, resulting in lagging response of the bottom plate structure updating. SUMMARY

[0004] The main purpose of the present application is to provide a dynamic updating method of digital twin bottom plate, aiming to solve the technical problem of how to improve the real-time performance of digital twin bottom plate updating.

[0005] To achieve the above purpose, the present application provides a dynamic updating method of digital twin bottom plate, which comprises:

[0006] According to the influence intensity of the periodic index of the associated business of the digital twin bottom plate on the data structure change of the bottom plate, a periodic influence intensity field is constructed;

[0007] According to the influence intensity of the external event trigger factor of the associated business on the data structure change of the bottom plate, a disturbance vector is constructed;

[0008] Based on the historical data structure change sequence of the digital twin bottom plate, the data structure change rate of the digital twin bottom plate is determined, and according to the data structure change rate, a structure gradient tensor is constructed;

[0009] The structure gradient tensor, the periodic influence intensity field and the disturbance vector are input into a pre-set tensor equation as variables of the tensor equation, and the eigenvalue decomposition result is determined through the calculation result of the tensor equation;

[0010] Based on the eigenvalue decomposition result, the change prediction result of the current bottom plate data structure is determined;

[0011] According to the change prediction result, a new bottom plate data structure is generated, and the bottom plate data in the current bottom plate data structure is migrated to the new bottom plate data structure.

[0012] In an embodiment, before the step of constructing a cycle influence intensity field according to the influence intensity of the periodic indicators of the business associated with the digital twin substrate on the data structure changes of the substrate, the method comprises:

[0013] The time series data in the digital twin substrate is separated by a preset time series decomposition algorithm to obtain a periodic term of the time series data;

[0014] The peak-valley difference of the periodic term within the cycle length is determined as the cycle amplitude of the periodic term;

[0015] The periodic term is subjected to Fourier transform to obtain a cycle frequency of the periodic term;

[0016] The cycle amplitude and the cycle frequency are taken as the periodic indicators.

[0017] In an embodiment, the step of constructing a cycle influence intensity field according to the influence intensity of the periodic indicators of the business associated with the digital twin substrate on the data structure changes of the substrate comprises:

[0018] Based on each data structure field in the current substrate data structure, a node set is constructed;

[0019] Based on the change association relationship between each data structure field in the historical data structure change sequence and the node set, a directed edge set is constructed;

[0020] According to the node set and the directed edge set, a node topology graph is constructed;

[0021] According to the node topology graph, the association influence intensity between each node is determined, and the association influence intensity represents the dependence intensity between nodes;

[0022] The association influence intensity, the cycle amplitude and the cycle frequency are taken as variables of a preset cycle influence intensity field equation to construct the cycle influence intensity field.

[0023] In an embodiment, the step of determining the association influence intensity between each node according to the node topology graph, and the association influence intensity representing the dependence intensity between nodes, comprises:

[0024] Based on the historical data structure change sequence, each associated node of a current node is determined, and the current node points to the associated node via a directed edge;

[0025] According to the number of times that the current node drives the associated node to change, the weight of each directed edge of the current node is determined;

[0026] Determine an associated influence strength of the current node according to a weight sum of each directed edge of the current node, wherein the associated influence strength is proportional to the weight sum.

[0027] In an embodiment, the step of determining an influence strength of the bottom plate data structure change according to the external event trigger factor of the associated business includes:

[0028] The spatial position coordinates of the node associated with each external event trigger factor are taken as the vector direction of the disturbance vector;

[0029] Based on the historical data structure change sequence, the number and change frequency of the node associated with each external event trigger factor are determined to determine the vector size of the disturbance vector.

[0030] In an embodiment, the step of determining the data structure change rate of the digital twin bottom plate based on the historical data structure change sequence of the digital twin bottom plate includes:

[0031] Divide the time period corresponding to the historical data structure change sequence into two or more time windows;

[0032] Construct a time change sequence according to the number of data structure changes in each time window;

[0033] Map the change field in the historical data structure change sequence to the node topology graph to determine a change node, and take the number of changes of the change field as the number of changes of the change node to construct a spatial change sequence;

[0034] Determine the structure change rate of the digital twin bottom plate based on the time change sequence and the spatial change sequence.

[0035] In an embodiment, the eigenvalue decomposition result includes an eigenvalue and a feature vector corresponding to the eigenvalue, and the step of determining a change prediction result of the current bottom plate data structure based on the eigenvalue decomposition result includes:

[0036] Determine the change strength of the data structure field corresponding to the eigenvalue according to the eigenvalue;

[0037] Determine the change probability of different change types of the data structure field corresponding to the eigenvalue according to the feature vector corresponding to the eigenvalue.

[0038] In an embodiment, the step of generating a new bottom plate data structure according to the change prediction result includes:

[0039] Generate a candidate bottom plate data structure according to the change prediction result;

[0040] determine an evaluation score of each of the candidate bottom board data structure based on the target optimization parameter;

[0041] determine the new bottom board data structure from the candidate bottom board data structure according to the evaluation score.

[0042] In an embodiment, after the step of migrating the bottom board data in the current bottom board data structure to the new bottom board data structure, the method further comprises:

[0043] construct a source hash tree according to hash values of the bottom board data in the previous bottom board data structure, and determine a root node of the source hash tree;

[0044] construct a new hash tree according to hash values of the bottom board data in the new bottom board data structure, and determine a root node of the new hash tree;

[0045] if the root nodes of the source hash tree and the new hash tree are consistent, determine that the digital twin bottom board updating is completed.

[0046] In an embodiment, after the step of constructing the new hash tree according to the hash values of the bottom board data in the new bottom board data structure, and determining the root node of the new hash tree, the method further comprises:

[0047] if the root nodes of the source hash tree and the new hash tree are inconsistent, compare intermediate nodes of the source hash tree and the new hash tree, and determine a to-be-modified node according to a comparison result;

[0048] re-migrate the bottom board data corresponding to the to-be-modified node in the previous bottom board data structure to the new bottom board data structure.

[0049] The application provides a dynamic updating method of a digital twin bottom board, constructs a periodic influence intensity field according to an influence intensity of a periodic index of a business associated with the digital twin bottom board on a bottom board data structure change;

[0050] constructs a disturbance vector according to an influence intensity of an external event trigger factor of the business on the bottom board data structure change, determines a data structure change rate of the digital twin bottom board based on a historical data structure change sequence of the digital twin bottom board, and constructs a structure gradient tensor according to the data structure change rate; inputs the structure gradient tensor, the periodic influence intensity field and the disturbance vector as variables of a preset tensor equation, determines an eigenvalue decomposition result through a calculation result of the tensor equation; determines a change prediction result of a current bottom board data structure based on the eigenvalue decomposition result; generates a new bottom board data structure according to the change prediction result, and migrates bottom board data in the current bottom board data structure to the new bottom board data structure.

[0051] The method quantifies the influence distribution of the periodic operation of the strength field on the data structure change by constructing a cycle, captures the direction and amplitude of the non-business interference of external events on the data structure by a disturbance vector, and constructs a structure gradient tensor based on a historical change sequence to quantify the spatiotemporal gradient of the data structure change, and then integrates the three into a tensor equation, extracts the dominant change mode by using eigenvalue decomposition, outputs the change prediction result, guides the generation of a new bottom plate structure that meets the dynamic demand, realizes the change from passive response to active prediction, and significantly improves the real-time performance of the digital twin bottom plate update. BRIEF DESCRIPTION OF DRAWINGS

[0052] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and serve to explain the principles of the present application together with the specification.

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, for those skilled in the art, other drawings can also be obtained without creative labor based on these drawings.

[0054] Figure 1 The flowchart provided for the first embodiment of the dynamic updating method of the digital twin bottom plate of the present application;

[0055] Figure 2 The flowchart provided for the second embodiment of the dynamic updating method of the digital twin bottom plate of the present application;

[0056] Figure 3 The flowchart provided for the third embodiment of the dynamic updating method of the digital twin bottom plate of the present application;

[0057] Figure 4 The node topology diagram construction schematic provided for the third embodiment of the dynamic updating method of the digital twin bottom plate of the present application;

[0058] Figure 5 The flowchart provided for the fourth embodiment of the dynamic updating method of the digital twin bottom plate of the present application;

[0059] Figure 6 The flowchart provided for the sixth embodiment of the dynamic updating method of the digital twin bottom plate of the present application;

[0060] Figure 7 The system architecture block diagram of the dynamic updating method of the digital twin bottom plate of the present application;

[0061] Figure 8 The flowchart of the dynamic updating method of the digital twin bottom plate of the present application;

[0062] Figure 9A device structure schematic diagram of a hardware running environment involved in a dynamic updating method of a digital twin bottom plate in the embodiments of the present application.

[0063] The object implementation, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0064] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and do not limit the present application.

[0065] In order to better understand the technical solutions of the present application, the following will be described in detail in combination with the drawings and specific embodiments in the specification. It should be noted that all the actions of obtaining signals, information or data in the present application are carried out under the premise of complying with the corresponding data protection regulations and policies of the place, and with the authorization of the corresponding device owner.

[0066] Digital twin technology is widely used in intelligent manufacturing, smart city, intelligent transportation and other fields. Its core is to realize the mapping, analysis and prediction of real physical entities through real-time data driving virtual models. In this system, the digital twin bottom plate is a basic carrier supporting twin modeling, data access, state presentation and feedback control. Its data structure and updating mechanism are crucial.

[0067] In related digital twin bottom plate systems, static or semi-dynamic bottom plate structures are mostly used. Through pre-set data model structure, a small amount of field supplement or static replacement is carried out after initialization to realize the updating of the bottom plate structure, which lacks structure change prediction ability, resulting in lag of the bottom plate structure updating response.

[0068] In view of the above problems, the present application proposes a dynamic updating method of a digital twin bottom plate. According to the influence intensity of the periodic index of the associated business on the data structure change of the digital twin bottom plate, a periodic influence intensity field is constructed. According to the influence intensity of the external event trigger factor of the associated business on the data structure change of the digital twin bottom plate, a disturbance vector is constructed. Based on the historical data structure change sequence of the digital twin bottom plate, the data structure change rate of the digital twin bottom plate is determined, and according to the data structure change rate, a structure gradient tensor is constructed. The structure gradient tensor, the periodic influence intensity field and the disturbance vector are input into a pre-set tensor equation as variables, and the eigenvalue decomposition result is determined through the calculation result of the tensor equation. Based on the eigenvalue decomposition result, the change prediction result of the current bottom plate data structure is determined. According to the change prediction result, a new bottom plate data structure is generated, and the bottom plate data in the current bottom plate data structure is migrated to the new bottom plate data structure.

[0069] The method quantifies the influence distribution of the periodic operation of the business on the data structure change by constructing a cycle influence intensity field, captures the direction and amplitude of the non-business interference of external events on the data structure by a disturbance vector, and constructs a structure gradient tensor based on a historical change sequence to quantify the space-time gradient of the data structure change, and then integrates the three into a tensor equation, extracts the dominant change mode by eigenvalue decomposition, outputs the change prediction result, guides the generation of a new bottom plate structure that meets the dynamic demand, realizes the change from passive response to active prediction, and significantly improves the real-time performance of the digital twin bottom plate update.

[0070] It should be noted that the execution subject of the embodiment can be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, etc., or an electronic device, a system, etc. that can realize the above functions.

[0071] Based on this, the first embodiment of the present application provides a dynamic updating method of a digital twin bottom plate, which is described with reference to Figure 1 In the embodiment, the dynamic updating method of the digital twin bottom plate includes steps S10-S50:

[0072] Step S10, according to the influence intensity of the periodic index of the digital twin bottom plate associated business on the bottom plate data structure change, construct a cycle influence intensity field.

[0073] It should be noted that the periodic index refers to the inherent regular fluctuation characteristics in the business activities associated with the digital twin bottom plate. For example, the shift cycle of the manufacturing production line, the morning and evening peak period of the smart city traffic flow, etc. The periodic index can quantify the business activity rules to provide periodic change reference data for dynamically updating the bottom plate data structure. The cycle influence intensity field is a continuous space-time field function obtained by extracting the mapping relationship between the periodic index and the bottom plate data structure change through a physical field driving model, which aims to map discrete business cycle data such as peak period, trough period, etc. to energy field intensity values with spatial distribution characteristics. The energy field intensity value refers to the change intensity of the data structure field in the digital twin bottom plate, including the number of data structure field expansion or reduction, the change size of the data structure field value, etc.

[0074] Optionally, all bottom plate data in the digital twin bottom plate is acquired, a time series analysis algorithm such as Fourier transform or wavelet analysis is applied, periodic data in the historical data such as daily, weekly, monthly, etc. is identified, and the amplitude and phase of each periodic data is calculated to represent the regular periodic characteristics of business activities. According to the above periodic characteristics and the historical data structure change sequence associated with the digital twin bottom plate, the historical data structure change intensity corresponding to each periodic characteristic is determined to reflect the potential influence intensity of business period on the bottom plate data structure change. According to the mapping relationship between the above periodic characteristics and the historical data structure change intensity, a continuous influence intensity curve or surface is constructed to form a spatial and temporal distribution of periodic influence intensity field. Wherein, the periodic characteristics and the business influence intensity value can be input into the preset physical field driving model, and the preset physical field driving model learns the mapping relationship according to the input periodic characteristics and historical data structure change intensity value, and outputs the periodic influence intensity field.

[0075] Step S20, according to the influence intensity of the bottom plate data structure change of the external event trigger factor of the associated business, a disturbance vector is constructed.

[0076] The external event trigger factor refers to a non-periodic, sudden external interference factor, including device failure, external instruction and other unpredictable events. Such factors have irregularity and high dynamics, which can be obtained by real-time monitoring and included in the physical field driving model for calculation. The disturbance vector is a vector parameter obtained by quantizing the mapping relationship between the external event trigger factor and the bottom plate data structure change through mathematical conversion, and the purpose is to represent the change intensity of the external interference on the data structure field in the digital twin bottom plate, and to provide a basis for predicting the bottom plate data structure change.

[0077] Optionally, according to the historical data structure change sequence associated with the digital twin bottom plate, the historical external event trigger factor associated with the historical data structure change is determined. According to the change field corresponding to each historical external event trigger factor, the vector direction of the disturbance vector is determined; according to the data structure change frequency corresponding to each historical external event trigger factor, the vector size of the disturbance vector is determined. Wherein, the data structure change frequency can be represented as the change frequency in each vector direction, i.e. the change frequency of each change field.

[0078] Step S30, based on the historical data structure change sequence of the digital twin bottom plate, the data structure change rate of the digital twin bottom plate is determined, and a structure gradient tensor is constructed according to the data structure change rate.

[0079] It should be noted that the historical data structure change sequence is an ordered change event set of the data structure of the digital twin substrate evolving over time during the historical running process, including change events such as data structure field addition, deletion, and modification. The structure gradient tensor refers to the change rate of the substrate data structure change in the time and space dimensions by performing spatiotemporal gradient analysis on the historical data structure change sequence, forming a quantitative representation of the substrate data structure change.

[0080] Exemplarily, the historical data structure change record is obtained from the update log of the digital twin substrate, sorted by timestamp, and forms a historical data structure change sequence. The historical data structure change sequence is divided into multiple subsequences by using a sliding window algorithm, and each subsequence covers a fixed time range such as a week, a month, etc. For each subsequence, the change type, change frequency, change strength, and other change characteristics of each data structure field in the historical data structure are counted, and the change rate of the above change characteristics is taken as the structure gradient tensor.

[0081] Optionally, when the change type is field value change, the change strength represents the change size of the field value.

[0082] Exemplarily, when the data structure field type is numerical, the change strength can be represented as the difference between the data structure field values before and after the change. For example, if a data structure field represents the number of people, in a data structure change, its field value is changed from 10 to 100, and the change strength can be represented as 90; when the data structure field type is non-numerical, the change strength can be determined according to the semantic similarity of the data structure field characters before and after the change, wherein the semantic similarity is inversely proportional to the change strength.

[0083] Optionally, when the change type is data structure field expansion or deletion, the change strength represents the number of data structure field expansion or deletion. Wherein, the number of data structure field expansion or deletion is normalized as the change strength.

[0084] Optionally, when the change type is data field type change such as from numerical to character, the change strength can be represented by a preset numerical value.

[0085] In step S40, the structure gradient tensor, the periodic influence strength field, and the perturbation vector are input as variables of a preset tensor equation, and the eigenvalue decomposition result is determined by the calculation result of the tensor equation.

[0086] It can be understood that the above structure gradient tensor can represent the gradient change of the bottom plate data structure in the time and space dimensions. The periodic influence intensity field can reflect the influence distribution of different periodic business operations on the data structure change. The disturbance vector describes the non-business influence direction and amplitude of external events on the data structure. By integrating the structure gradient tensor, the periodic influence intensity field and the disturbance vector into a unified mathematical model, the dynamic change mechanism of the bottom plate data structure can be quantified.

[0087] Optionally, the tensor equation can adopt a linear combination or a nonlinear coupling form. Preferably, the expression of the tensor T(x, y, t) can refer to the following formula:

[0088]

[0089] wherein, represents the tensor corresponding to the data structure field corresponding to the spatial position (x, y) at time t. Wherein, x represents the data table row, and y represents the data table column. represents the structure gradient tensor, represents the disturbance vector, represents the periodic influence intensity field, represents the unit tensor, and a, β and γ are weight coefficients. The above coupling structure gradient tensor and periodic influence intensity field to reflect how the business periodic operation amplifies or suppresses the bottom plate data structure change; the cumulative effect of external interference is expressed by a quadratic form; is used to ensure that the tensor equation remains static balance when there is no input. It should be noted that the expression of the preset tensor equation can be quantified by learning the mapping relationship between the learned tensor and the structure gradient tensor, the periodic influence intensity field and the disturbance vector through the preset physical field driving model according to the historical data structure change sequence data. Step S50, determining the change prediction result of the current bottom plate data structure based on the eigenvalue decomposition result.

[0090] After constructing the tensor equation, the dominant mode and intensity of the bottom plate data structure change can be extracted by analyzing the eigenvalue decomposition result of the tensor equation.

[0091] Exemplarily, the spatial position (x, y) of the bottom plate data structure is discretized into a grid, and the time t is discretized into a time step Δt. At each grid point (xi, yj) and time step tk, the specific value of the tensor is calculated. The eigenvalue decomposition is performed on the tensor

[0092] of each spatial position (xi, yj): ​​​

[0093]

[0094] wherein, is an eigenvalue, representing the change intensity and direction of the data structure field corresponding to the spatial position, and a positive eigenvalue represents that the data structure field tends to change, and a negative eigenvalue represents that the field tends to be stable or rollback. is an eigenvector, representing the change probability corresponding to each change type.

[0095] In order to better understand the scheme given in this example, the example is further described in combination with a specific application scenario.

[0096] Suppose that after eigenvalue decomposition by a tensor equation, the following two eigenvalues and their corresponding eigenvectors are obtained: = 0.8, = [0.9, -0.1], = 0.2, = [0.3, 0.9]. Wherein, if the eigenvector represents the probabilities corresponding to "add" and "field value change", then the eigenvalue = 0.8 represents that the change intensity value of the data structure field corresponding to the spatial position (1, 3) at the 3rd time step is 0.8, and the probability of change is large; and the eigenvector = [0.9, -0.1] represents that the probability of the change type of the data structure field corresponding to the spatial position (1, 3) being "add" is 0.9, and the probability of "field value change" is -0.1. Similarly, the eigenvalue = 0.2 represents that the change intensity value of the data structure field corresponding to the spatial position (1, 2) at the 3rd time step is 0.2, and the probability of change is small; and the eigenvector = [0.3, 0.9] represents that the probability of the change type of the data structure field corresponding to the spatial position (1, 2) being "add" is 0.3, and the probability of "field value change" is 0.9.

[0097] Step S60, generating a new bottom plate data structure according to the change prediction result, and migrating the bottom plate data in the current bottom plate data structure to the new bottom plate data structure.

[0098] After obtaining the above change prediction result, a new bottom plate data structure can be generated according to the change prediction structure. For example, a threshold value of the change intensity and the change probability corresponding to each change type is respectively preset, and when the change prediction structure result of any data structure field exceeds the corresponding threshold value, a new bottom plate data is generated according to the change prediction structure, the data in the current bottom plate data is migrated to the new bottom plate data structure, and the change of the digital twin bottom plate is completed.

[0099] It can be understood that when the change type is data structure field reduction, the corresponding structure field will be deleted in the new bottom plate data structure. Therefore, when data migration is performed, the corresponding field value in the bottom plate data structure before the change is found according to the data structure field in the new bottom plate data structure, and is migrated to the bottom plate data structure.

[0100] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as the above embodiment one can refer to the above introduction, and the following will not be repeated. On this basis, please refer to Figure 2 , before step S10, the dynamic updating method of the digital twin bottom plate further includes steps S70-S100:

[0101] Step S70, separate the time series data in the digital twin bottom plate by a preset time series decomposition algorithm to obtain a periodic term of the time series data.

[0102] Optionally, the time series data X(t) is obtained from the data interface of the digital twin bottom plate in real time or in batches, and the corresponding timestamp sequence x covers the business operation period. The periodic term S(t) is separated from the time series data X(t) by a seasonal-trend decomposition algorithm. Wherein, the seasonal-trend decomposition (Seasonal and Trend decomposition using Loess, STL) is a time series decomposition method, which is used to decompose time series data into trend term (Trend), seasonal term (Seasonal) and residual term (Residual), in this embodiment, the seasonal parameter in the seasonal-trend decomposition algorithm can be modified by a periodic index, and then the periodic term is obtained by decomposition.

[0103] Optionally, the periodic term S(t) is separated from the time series data X(t) by an Empirical Mode Decomposition (EMD) algorithm. The Empirical Mode Decomposition can decompose the time series data X(t) into a plurality of Intrinsic Mode Functions (IMFs), each of which represents an oscillation mode of different time scales. Specifically, first, step one is performed: initialize the residual r(t) = X(t), and set the IMF set to be empty. Then, step two is performed: extract the kth IMF, determine all local maxima and minima of r(t), and generate the upper envelope and the lower envelope by interpolation. Then, the average envelope m(t) is calculated, and the candidate IMF h(t) = r(t) - m(t) is obtained. Repeat the above steps until h(t) satisfies the preset IMF condition, such as symmetry or zero mean. After h(t) satisfies the preset IMF condition, step three is performed: add h(t) to the IMF set, and update the residual r(t) = r(t) - h(t). Repeat steps two and three until the residual r(t) is a monotonic function or a constant. The final IMF is taken as the periodic term S(t).

[0104] In step S80, the peak-valley difference of the periodic term in the period length is determined as the periodic amplitude of the periodic term.

[0105] In step S90, the periodic term is subjected to Fourier transform to obtain the periodic frequency of the periodic term.

[0106] In step S100, the periodic amplitude and the periodic frequency are taken as the periodicity index.

[0107] For example, in each period length Lk of the periodic term S(t), the maximum value Smax and the minimum value Smin of the periodic term are identified, the difference between Smax and Smin is taken as the peak-valley difference, and the periodic amplitude is determined. Then, the periodic term S(t) is subjected to Fourier transform to convert the time domain signal to the frequency domain, and the main peak frequency in the frequency spectrum, i.e., the frequency value corresponding to the maximum amplitude, is taken as the periodic frequency. Finally, the periodic amplitude and the periodic frequency are taken as the periodicity index.

[0108] Based on the above embodiments of the present application, in the third embodiment of the present application, the same or similar contents as the above embodiments can be referred to the above description, and will not be described in detail. On this basis, please refer to Figure 3 , step S10 includes steps S11-S15:

[0109] In step S11, a node set is constructed based on each data structure field in the current data structure.

[0110] In step S12, a directed edge set is constructed based on the change association relationship between each data structure field in the historical data structure change sequence and the node set.

[0111] Step S13, constructing a node topology graph according to the node set and the directed edge set.

[0112] Parsing the current digital twin substrate data structure, taking each data structure field as a node to generate a node set , each node can include field metadata such as field value, field type, etc. Then, according to the historical data structure change sequence, according to the change association relationship between the data structure fields corresponding to each node in the historical data structure change sequence, a directed edge set E={eij} is constructed. Among them, the change association relationship between the data structure fields refers to the dynamic dependency or driving relationship between different data structure fields, i.e. nodes, in the process of historical data structure change. Specifically, when a data structure field changes, it will directly trigger or require another data structure field to change accordingly. For example, if the change of node vi directly drives the change of node vj, a directed edge eij from node vi to node vj is established. According to the node set and the directed edge set, a node topology graph is constructed.

[0113] For example, if the value of the "device temperature" field in the digital twin substrate is modified, it may be necessary to adjust the "temperature risk value" field simultaneously. At this time, the change of the "device temperature" field directly drives the change of the "temperature risk value", and is constructed into a directed edge from the "device temperature" field to the "temperature risk value" field.

[0114] Referring to Figure 4 , Figure 4 On the left is a data table, which includes data structure fields: "Oeder_id" (V1), "Varchar" (V2), "ORD202310010001" (V3), "status" (V4), "Enum" (V5), "paid" (V6), "Pay_time" (V7), "Datetime" (V8), "2023-10-0114:30:00" (V9). Among them, "Varchar" (V2), "Enum" and "Datetime" (V8) are field types, and the change of their field values will directly trigger the change of the field types of "ORD202310010001" (V3), "paid" (V6) and "2023-10-0114:30:00" (V9); and "paid" (V6) indicates the order status, which will directly trigger the change of the field value of the payment time "2023-10-0114:30:00" (V9). Therefore, in the generated node topology graph, directed edges will be generated between V2 and V3, V5 and V6, V8 and V9, and V6 and V9.

[0115] The node topology diagram described above transforms the abstract field change relationship into a graph representation of data structure changes. When a data structure field changes, the topology diagram can quickly locate other data structure fields that it directly drives and the data structure fields that subsequently pass on the impact, thereby predicting the scope of the data structure change's influence.

[0116] Step S14: Determine the correlation influence strength between each node based on the node topology graph, where the correlation influence strength represents the dependence strength between nodes.

[0117] Optionally, for each directed edge connecting the source node and the target node, the weight of the directed edge is determined based on the number of times the source node drives changes in the target node, and this weight is used to represent the dependency strength between the source node and the target node. Then, based on the weight of the directed edge, the strength of the association influence between the source node and the target node is determined.

[0118] In one feasible implementation, step S14 includes steps S141 to S143:

[0119] Step S141: Determine each associated node of the current node based on the historical data structure change sequence, wherein the current node points to the associated node via directed edges.

[0120] Step S142: Determine the weight of each directed edge of the current node based on the number of times the current node drives changes in the associated nodes.

[0121] Step S143: Determine the association influence strength of the current node based on the sum of the weights of each directed edge of the current node, wherein the association influence strength is proportional to the sum of the weights.

[0122] For example, suppose the current node Associated nodes Connected by directed edges. The current node is determined based on the historical data structure change sequence. Drive associated nodes Number of changes The frequency of interaction between the two nodes is used as a metric. Then, based on this frequency, the weight of the directed edge between the two nodes is determined. The weight of this directed edge is directly proportional to the frequency of interaction between the two nodes. Finally, based on the sum of the weights of this directed edge, the current node is determined. The strength of the correlation influence.

[0123] It is understandable that the stronger the association influence of a node, the more changes in the data structure fields corresponding to that node will cause changes in other data structure fields. In other words, the changes in the twin base data structure are more significant. The strength of the association influence can reveal the impact of the association between data structure fields on changes in the data structure.

[0124] Step S15, input the correlation influence strength, the periodic amplitude and the periodic frequency as variables of the periodic influence strength field equation into the preset periodic influence strength field equation, and construct the periodic influence strength field.

[0125] For example, the periodic influence strength field equation can refer to the following formula:

[0126]

[0127] wherein, represents the periodic influence strength field value corresponding to the data structure field at time t and spatial position (x, y). Wherein, x represents the data table row, and y represents the data table column. is the decay coefficient, represents the periodic amplitude of the kth period, represents the periodic frequency of the kth period, is the correlation influence strength of the node, i.e. the data structure field. The isolated node can be defaulted . .

[0128] Based on the above embodiments of the present application, in the fourth embodiment of the present application, the same or similar contents as the above embodiments can refer to the above introduction, and will not be described in detail. On this basis, please refer to Figure 5 , step S30 further includes steps S31-S34:

[0129] Step S31, divide the time period corresponding to the historical data structure change sequence into two or more time windows.

[0130] Step S32, construct a time change sequence according to the number of data structure changes in each time window.

[0131] For example, the historical data structure change sequence ΔS of the digital twin bottom plate is obtained as {Δ , i = 1, 2, …, m}, wherein, a single change event is defined as a four-tuple: Δ ( , ). is the change occurrence timestamp, is the change node identifier such as field ID, is the operation type code, for example, “0” represents adding, “1” represents deleting, and “2” represents modifying the field type, is the change strength. A preset time period is divided into time windows with a fixed period, and preferably, it can be divided according to the periodic index. Then, by counting the number of changes in each time window, time series data is generated to quantify the frequency change of data structure changes, and a time change sequence is obtained​ .

[0132] Step S33, mapping the change field in the historical data structure change sequence to the node topology graph, determining the change node, and taking the change number of the change field as the change number of the change node, to construct the spatial change sequence.

[0133] Exemplarily, according to the historical data structure change sequence, each change field that has changed is determined, the change field is mapped to the node topology graph, and according to the quotient of the change number of the change node corresponding to each change field and the change number of all change nodes in the node topology graph, the spatial change sequence is determined. For example, the spatial change sequence may be represented by the following formula:

[0134]

[0135] wherein, represents the change number of the change node corresponding to the spatial position (x, y), represents the change number of all change nodes in the node topology graph. Step S34, determining the structure change rate of the digital twin substrate based on the time change sequence and the spatial change sequence.

[0136] Exemplarily, according to the time change sequence and the spatial change sequence, a structure change equation

[0137] is constructed. Wherein, represents the average value of the change intensity of all change fields.

[0138] Then, according to the above structure change equation, the structure change rate of the digital twin substrate is determined, and a structure gradient tensor is generated:

[0139]

[0140] It can be understood that the traditional change prediction may only rely on time series such as “change number in the last 1 hour” or spatial distribution such as “a field changes most frequently”, while the structure gradient tensor jointly models time and space, simultaneously characterizing “when the change occurs” and “where the change occurs”, thereby more comprehensively capturing the global pattern of the change. By combining the directed edges in the node topology graph, i.e., the change association relationship, the structure gradient tensor can reflect the change conduction logic between different data structure fields, so that the prediction result is more consistent with the actual business scenario.

[0141] ​Based on the above embodiments of the application, in the fifth embodiment of the application, the same or similar contents as the above embodiments can be referred to the above introduction, and the subsequent will not be described. On this basis, step S20 further comprises steps S21-S22:

[0142] Step S21, the spatial position coordinates of the node associated with each of the external event trigger factors are taken as the vector direction of the perturbation vector.

[0143] Step S22, based on the history data structure change sequence, the vector size of the perturbation vector is determined according to the number and change frequency of the node associated with each of the external event trigger factors.

[0144] Illustratively, query the history data structure change sequence, identify the data structure field changed due to the external event trigger factor, that is, the associated node, and take the spatial position coordinates (x, y) of the node as the vector direction of the perturbation vector.

[0145] Then, according to the history data structure change sequence, the change frequency of the data structure field changed due to each external event trigger factor and the number of the data structure field changed due to each external event trigger factor are counted.

[0146] Optionally, the sum of the change frequency of the changed data structure field and the number of the changed data structure field is taken as the vector intensity of the perturbation vector, or the weighted sum of the change frequency and the number is taken as the vector intensity of the perturbation vector.

[0147] Based on the above embodiments of the application, in the sixth embodiment of the application, the same or similar contents as the above embodiments can be referred to the above introduction, and the subsequent will not be described. On this basis, please refer to Figure 6 , step S60 further comprises steps S61-S63:

[0148] Step S61, generating a candidate bottom plate data structure according to the change prediction result.

[0149] The prediction result includes change position, change type and change probability, wherein the change position refers to the node in the node topology graph, that is, the data structure field corresponding to the spatial position coordinates (x, y) of the bottom plate data structure. According to the above prediction change result, a new candidate bottom plate data structure is generated.

[0150] Optionally, the prediction results are prioritized based on the change probability, the first k change positions corresponding to the to-be-changed nodes are selected according to the sorting result, and the to-be-changed data structure field corresponding to the to-be-changed node is changed according to the change type.

[0151] Exemplarily, if the change type is "add", the to-be-changed data structure field is added in the to-be-selected bottom board data structure, and its attributes such as field type, default value, constraint condition and the like are initialized. At the same time, it can be checked whether the parent node of the to-be-changed node, i.e. the upstream node of the to-be-changed node in the topology graph, needs to add a dependent relationship. For example, the index value of the to-be-changed data structure field is added in the "sub-field list" of the parent node.

[0152] Exemplarily, if the change type is "delete", the to-be-changed data structure field is removed in the to-be-selected bottom board data structure; if the change type is "modify", the attributes of the to-be-changed data structure field are adjusted, such as changing the field type from "string" to "integer" or modifying the default value.

[0153] Optionally, the prediction results are prioritized based on the change probability, and according to the sorting results, the change is propagated from the to-be-changed node with the highest change probability along the directed edge in the node topology graph, and after adjusting the current to-be-changed node according to the change type, the structure of the downstream nodes thereof is recursively adjusted, so as to ensure that the cascade effect of the change is correctly captured, and a new to-be-selected bottom board data structure conforming to the topology logic is generated.

[0154] In step S62, the evaluation score of each to-be-selected bottom board data structure is determined based on the target optimization parameter.

[0155] In step S63, the new bottom board data structure is determined in the to-be-selected bottom board data structure according to the evaluation score.

[0156] Optionally, the target optimization parameter can include a performance optimization parameter, a cost optimization parameter, a stability optimization parameter and other preset optimization parameters. The performance optimization parameter can be obtained by calculating the query efficiency or computational complexity of the to-be-selected bottom board data structure; the cost optimization parameter can be obtained by calculating the storage space or computing resource occupation of the to-be-selected bottom board data structure; and the stability optimization parameter can be obtained by calculating the abnormal rate after the field change. The weighted sum of each target optimization parameter is taken as the evaluation score of the to-be-selected bottom board data structure, and the one with the highest evaluation score is taken as the new bottom board data structure.

[0157] Optionally, the target optimization parameter is regarded as a multi-objective optimization problem, and a non-dominated solution is screened out from the to-be-selected bottom board data structure through the Pareto front, i.e. the to-be-selected bottom board data structure in which one target optimization parameter cannot be further optimized without damaging other target optimization parameters, is taken as the new bottom board data structure.

[0158] Exemplarily, a target function is defined for each target optimization parameter, and a score of each candidate bottom plate data structure on all target functions is calculated to form a solution set. An optimal solution is selected from the solution set according to a preset screening condition of a Pareto frontier.

[0159] Based on the above embodiments of the present application, in the seventh embodiment of the present application, the same or similar contents as the above embodiments can be referred to the above introduction, and the subsequent will not be described in detail. On this basis, after step S60, the dynamic updating method of the digital twin bottom plate further includes steps S110-S130:

[0160] Step S110, a source hash tree is constructed according to the hash values of the bottom plate data in the changed bottom plate data structure, and the root node of the source hash tree is determined.

[0161] Step S120. A new hash tree is constructed according to the hash values of the bottom plate data in the new bottom plate data structure, and the root node of the new hash tree is determined.

[0162] Step S130, if the root nodes of the source hash tree and the new hash tree are consistent, it is determined that the digital twin bottom plate updating is completed.

[0163] Exemplarily, the bottom plate data in the changed bottom plate data structure is divided into data blocks of a preset size, the hash value of each data block is calculated, the leaf node is generated, then the hash values of two adjacent leaf nodes are combined to generate a parent node and calculate the hash of the parent node. Repeat the above merging process until the last root node is left. After data migration is completed, a new hash tree is constructed according to the above method for the bottom plate data in the new bottom plate data structure, and the root node of the new hash tree is obtained. The root nodes of the source hash tree and the new hash tree are compared, and if the comparison result is consistent, it means that the data migration is successful.

[0164] Based on the above embodiments of the present application, in the seventh embodiment of the present application, the same or similar contents as the above embodiments can be referred to the above introduction, and the subsequent will not be described in detail. On this basis, after step S60, the dynamic updating method of the digital twin bottom plate further includes steps S110-S130:

[0165] Step S140, if the root nodes of the source hash tree and the new hash tree are inconsistent, the intermediate nodes of the source hash tree and the new hash tree are compared, and the to-be-modified node is determined according to the comparison result.

[0166] Step S150, the bottom plate data corresponding to the to-be-modified node in the changed bottom plate data structure is re-migrated to the new bottom plate data structure.

[0167] Exemplarily, if the root nodes of the source hash tree and the new hash tree are inconsistent, the error range of the to-be-modified node can be further narrowed down quickly by comparing the intermediate node hash values of the source hash tree and the new hash tree. For example, assuming that the second layer nodes of the source hash tree are H1 (H(B1||B2)) and H2 (H(B3||B4)). The second layer nodes of the new hash tree are H1' and H2'. If H1≠H1', but H2=H2', the error range is narrowed down to B1 or B2. Further comparison of H(B1) and H(B1') is performed: if inconsistent, B1 is wrong; if consistent, B2 is wrong. The corresponding baseboard data of the to-be-modified node in the baseboard data structure before the change is re-migrated to the new baseboard data structure.

[0168] Exemplarily, in order to help understand the implementation process of the dynamic updating method of the digital twin baseboard obtained after the above embodiment, please refer to Figure 7 , Figure 7 A system architecture block diagram of a dynamic updating method of a digital twin baseboard is provided, specifically:

[0169] The system architecture includes a data entity layer, an intelligent prediction layer, a self-adaptive scheduling layer, an execution verification layer, and a twin baseboard layer.

[0170] The data entity layer, as the data cornerstone of the entire architecture, is responsible for collecting and integrating the historical data structure change sequence in the digital twin substrate, including substrate data such as sensor data, business data, and environmental data in each data structure. These data provide a rich source of information for subsequent intelligent analysis and decision-making, ensuring that the system can operate based on real and comprehensive data. The intelligent prediction layer, based on the data provided by the data entity layer, performs substrate data structure change prediction, trend analysis, and pattern recognition. By presetting tensor equations, it mines the potential laws and change trends hidden in the data, providing forward-looking prediction information for the system's adaptive adjustment, enabling the system to anticipate possible changes in advance. The adaptive scheduling layer, based on the results of the intelligent prediction layer, conducts multi-objective optimization operations. It can determine the priority of each candidate substrate data structure according to different business requirements and target optimization parameters, to achieve efficient use of system resources and optimal overall performance, ensuring that the system can respond flexibly and reasonably to various changes. The execution verification layer, based on the decision of the adaptive scheduling layer, performs consistency checking, gradual rollback, and intelligent diagnosis operations. It verifies and monitors the results of data migration execution; when abnormal situations occur, it can recover to a stable state through gradual rollback and other mechanisms, and find the root cause of the problem through intelligent diagnosis. The twin substrate layer, as the final presentation layer of the system, realizes functions such as dynamic substrate real-time synchronization and version management. It integrates and displays the results processed and verified by the previous layers, and the twin substrate can timely and accurately reflect the latest state and changes of the system, while recording the evolution process of the system through version management, facilitating subsequent queries and backtracking.

[0171] Please refer to Figure 8 , Figure 8 A flowchart of a dynamic updating method of a digital twin substrate is provided, specifically:

[0172] First, enter the "data monitoring" phase, the system real-time monitoring of various data sources, access to the latest sensor data, business data and environmental data, etc. Then "feature extraction", from the massive original data extraction key feature information, for subsequent analysis to provide a simplified and effective data basis. Then through the "pattern recognition", the extracted features are analyzed, and the patterns and rules in the data are identified. Based on the results of pattern recognition, the system performs "change prediction", predicts the possible changes in the data structure, and generates multiple selected base data structures. Then enter the "multi-objective optimization" link, consider the performance, resource utilization, stability and other multiple objectives of each selected base data structure. Then through "priority evaluation", determine the priority of each selected base data structure, and finally "strategy selection" to develop a selected base data structure suitable for the current situation. According to the selected new base data structure "execute update" operation, the digital twin base is actually changed. After execution, enter the "consistency check whether passed" judgment node. If the check fails, the system will perform "progressive rollback", gradually restore to the stable state before updating, and perform "exception handling", troubleshoot and solve the problem that causes the check to fail; if the check passes, continue the subsequent process. After the consistency check passes, the system "completes the update", at which time "performance monitoring" is performed to monitor the running performance of the system after updating. At the same time, through "model feedback optimization", according to the performance monitoring results and actual running situation, the model is fed back and optimized, and the performance and accuracy of the system are continuously improved.

[0173] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the dynamic updating method of the digital twin base of the present application. Further simple transformations based on this technical concept are within the scope of protection of the present application.

[0174] The present application provides a dynamic updating device of a digital twin base, which comprises a processor and a memory in communication connection with the processor; wherein the memory stores instructions executable by the processor, and the instructions are executed by the processor to enable the processor to perform the dynamic updating method of the digital twin base in Embodiment I.

[0175] Reference will now be made to Figure 9 which shows a structural schematic diagram of a dynamic updating device of a digital twin base suitable for implementing embodiments of the present application. The dynamic updating device of the digital twin base in the embodiments of the present application can include but is not limited to mobile terminals such as notebook computers, tablet computers (PAD, Portable Application Description) and the like, and fixed terminals such as desktop computers and the like. Figure 9The dynamic updating device of the illustrated digital twin substrate is merely an example and should not bring any limitation to the function and use range of the embodiments of the present application.

[0176] As shown in Figure 9 The dynamic updating device of the digital twin substrate can include a processing device 1001 (e.g., a central processing unit, a graphics processing unit, etc.) that can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. In the random access memory 1004, various programs and data required for the operation of the dynamic updating device of the digital twin substrate are also stored. The processing device 1001, the read-only memory 1002, and the random access memory 1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; the storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the dynamic updating device of the digital twin substrate to communicate with other devices wirelessly or by wire to exchange data. Although the dynamic updating device of the digital twin substrate with various systems is shown in the figure, it should be understood that all the systems shown are not required to be implemented or possessed. More or fewer systems can be alternatively implemented or possessed.

[0177] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program codes for executing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network through the communication device, or installed from the storage device 1003, or installed from the read-only memory 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the methods of the embodiments disclosed in the present application are performed.

[0178] The dynamic updating device of the digital twin bottom plate provided by the application adopts the dynamic updating method of the digital twin bottom plate in the above embodiment, and can solve the technical problem of how to improve the real-time performance of the digital twin bottom plate updating. Compared with the prior art, the beneficial effects of the dynamic updating device of the digital twin bottom plate provided by the application are the same as those of the dynamic updating method of the digital twin bottom plate provided by the above embodiment, and other technical features in the dynamic updating device of the digital twin bottom plate are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0179] It should be understood that parts of the present application can be realized by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0180] The above is merely specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical scope disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0181] The present application provides a computer readable storage medium having computer readable program instructions (i.e. computer programs) stored thereon, the computer readable program instructions being used to execute the dynamic updating method of the digital twin bottom plate in the above embodiment.

[0182] The computer readable storage medium provided in the present application may, for example, be a U disk, but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, system, or device, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more conductive wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM) or a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present embodiment, the computer readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer readable storage medium can be transmitted in any suitable medium, including but not limited to an electrical wire, an optical cable, a radio frequency (RF), and the like, or any suitable combination of the above.

[0183] The above computer readable storage medium can be included in the dynamic updating device of the digital twin substrate, or can exist separately and not be assembled into the dynamic updating device of the digital twin substrate.

[0184] The above computer readable storage medium carries one or more programs, which, when executed by the dynamic updating device of the digital twin substrate, enable the dynamic updating device of the digital twin substrate to write computer program code in one or more programming languages or combinations thereof for performing the operations of the present application. The programming languages include object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as the "C" language or similar programming languages. The program code can be executed entirely on a user computer, partially on a user computer, or as a separate software package, partially on a user computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, through the Internet using an Internet service provider).

[0185] The flowcharts and block diagrams in the drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowcharts or block diagrams can represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently or the blocks can sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and / or flowchart illustrations, and combinations thereof, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or combinations of hardware and software.

[0186] The modules involved in the embodiments of the present application can be implemented in software or in hardware. In some cases, the names of the modules do not limit the modules themselves.

[0187] The readable storage medium provided by the present application is a computer readable storage medium, which stores computer readable program instructions (i.e., a computer program) for executing the dynamic updating method of the digital twin bottom plate, and can solve the technical problem of how to improve the real-time performance of the digital twin bottom plate updating. Compared with the prior art, the computer readable storage medium provided by the present application has the same beneficial effects as the dynamic updating method of the digital twin bottom plate provided by the above-mentioned embodiments, which will not be repeated here.

[0188] The present application also provides a computer program product comprising a computer program, which, when executed by a processor, implements the steps of the dynamic updating method of the digital twin bottom plate as described above.

[0189] The computer program product provided by the present application can solve the technical problem of how to improve the real-time performance of the digital twin bottom plate updating. Compared with the prior art, the computer program product provided by the present application has the same beneficial effects as the dynamic updating method of the digital twin bottom plate provided by the above-mentioned embodiments, which will not be repeated here.

[0190] The above only describes some embodiments of the present application, and does not limit the patent scope of the present application, and any equivalent structural transformation made by using the content of the present application specification and drawings, or direct / indirect application in other related technical fields is included in the patent protection scope of the present application.

Claims

1. A method for dynamically updating a digital twin baseboard, characterized in that, The dynamic update method for the digital twin baseplate includes: The time-series data in the digital twin baseboard is separated by a preset time-series decomposition algorithm to obtain the periodic terms of the time-series data; The peak-to-valley difference within the period length of the periodic term is determined as the periodic amplitude of the periodic term; Perform a Fourier transform on the periodic term to obtain the periodic frequency of the periodic term; The periodic amplitude and the periodic frequency are used as periodic indicators of the digital twin baseboard associated services; Construct a node set based on the data structure fields in the current base plate data structure; Based on the change relationships between various data structure fields in the historical data structure change sequence of the digital twin base and the node set, a directed edge set is constructed; Construct a node topology graph based on the set of nodes and the set of directed edges; The strength of the association influence between each node is determined based on the node topology graph, where the strength of the association influence represents the strength of the dependency between nodes. The correlation intensity, the periodic amplitude, and the periodic frequency are used as variables in the periodic influence intensity field equation and input into the preset periodic influence intensity field equation to construct the periodic influence intensity field. Based on the impact strength of the external event triggering factors of the related business on the changes in the baseboard data structure, a disturbance vector is constructed; Based on the historical data structure change sequence, the data structure change rate of the digital twin baseboard is determined, and a structural gradient tensor is constructed according to the data structure change rate. The structural gradient tensor, the periodic influence intensity field, and the perturbation vector are used as variables in the tensor equation and input into a preset tensor equation. The eigenvalue decomposition result is determined by the calculation result of the tensor equation. Based on the eigenvalue decomposition results, the change prediction result of the current base plate data structure is determined; A new base plate data structure is generated based on the change prediction results, and the base plate data in the current base plate data structure is migrated to the new base plate data structure.

2. The dynamic update method for a digital twin baseboard as described in claim 1, characterized in that, The step of determining the correlation strength between nodes based on the node topology graph, wherein the correlation strength represents the dependency strength between nodes, includes: Based on the historical data structure change sequence, the associated nodes of the current node are determined, and the current node points to the associated nodes via directed edges; The weights of each directed edge of the current node are determined based on the number of times the current node drives changes in the associated nodes. The strength of the association influence of the current node is determined based on the sum of the weights of each directed edge of the current node, wherein the strength of the association influence is proportional to the sum of the weights.

3. The dynamic update method for a digital twin baseboard as described in claim 1, characterized in that, The step of constructing the disturbance vector based on the impact strength of the external event triggering factors of the related business on the baseboard data structure change includes: The spatial coordinates of the nodes associated with each of the external event triggering factors are used as the vector direction of the disturbance vector; Based on the historical data structure change sequence, the vector size of the disturbance vector is determined according to the number of nodes associated with each external event triggering factor and the change frequency.

4. The dynamic update method for a digital twin baseboard as described in claim 1, characterized in that, The step of determining the data structure change rate of the digital twin baseboard based on the historical data structure change sequence includes: Divide the time period corresponding to the historical data structure change sequence into two or more time windows; Construct a time change sequence based on the number of data structure changes within each time window; Map the changed fields in the historical data structure change sequence to the node topology graph to determine the changed nodes, and use the number of changes of the changed fields as the number of changes of the changed nodes to construct a spatial change sequence; Based on the time change sequence and the spatial change sequence, the structural change rate of the digital twin base plate is determined.

5. The dynamic update method for a digital twin baseboard as described in claim 1, characterized in that, The eigenvalue decomposition result includes eigenvalues ​​and eigenvectors corresponding to the eigenvalues. The step of determining the change prediction result of the current base plate data structure based on the eigenvalue decomposition result includes: determining the change intensity of the data structure field corresponding to the eigenvalue according to the eigenvalue. Based on the feature vector corresponding to the feature value, determine the change probability of different change types corresponding to the data structure field.

6. The dynamic update method for a digital twin baseboard as described in claim 1, characterized in that, The step of generating a new base plate data structure based on the change prediction results includes: Based on the change prediction results, a candidate base plate data structure is generated; Based on the target optimization parameters, the evaluation score of each of the candidate base plate data structures is determined; Based on the evaluation score, the new base plate data structure is determined from the candidate base plate data structures.

7. The dynamic update method for a digital twin baseboard as described in claim 1, characterized in that, After the step of migrating the base plate data in the current base plate data structure to the new base plate data structure, the method further includes: Construct a source hash tree based on the hash values ​​of the base plate data in the base plate data structure before the change, and determine the root node of the source hash tree; A new hash tree is constructed based on the hash values ​​of the base plate data in the new base plate data structure, and the root node of the new hash tree is determined. If the root node of the source hash tree is the same as that of the new hash tree, then the digital twin baseboard update is considered complete.

8. The dynamic update method for a digital twin baseboard as described in claim 7, characterized in that, After the step of constructing a new hash tree based on the hash values ​​of the base plate data in the new base plate data structure and determining the root node of the new hash tree, the method further includes: If the root node of the source hash tree is inconsistent with that of the new hash tree, then the intermediate nodes of the source hash tree and the new hash tree are compared, and the node to be modified is determined based on the comparison result. The base plate data corresponding to the node to be modified in the base plate data structure before the change will be migrated to the new base plate data structure.

Images