Device twin data simulation system and device twin data simulation method

Abstract

The present invention provides a simulation system for device twin data and a simulation method for device twin data. The simulation system for device twin data includes a storage device and a processor. The storage device stores a simulation configuration data group and a plurality of modules. The processor is coupled to the processor and is used to execute: inputting the simulation configuration data group into the initial data generation module to obtain an initial data group; performing statistical calculation on the initial data group through the data calculation module to obtain a statistical feature data group; obtaining a simulation data group based on the simulation configuration data group, the initial data group and the statistical feature data group through the data generation module; performing abnormal data judgment based on the simulation data group through the abnormal data detection module, and outputting a corresponding output data group / warning information based on the judgment result of the abnormal data judgment.

Description

Technical Field

[0001] The present invention relates to a device twin data simulation technology, in particular to a device twin data simulation system and a device twin data simulation method. Background Art

[0002] Among the various data of the simulated equipment, since the parameters / data required by the equipment are not a single-line data set, multiple simulation data sets need to be generated during the simulation of the equipment twin data. In other words, the data required and associated in the operation of the production / manufacturing equipment are multi-dimensional data and are assembled into data sets / groups. Therefore, when simulating, evaluating, and optimizing product production in a simulated digital factory from the perspective of generating data twins, it is necessary to provide a multi-dimensional simulation data set and a system and method that can simultaneously monitor and simulate multiple sets of equipment parameters for abnormalities. Summary of the Invention

[0003] The present invention is directed to a simulation system and a simulation method for device twin data, which can generate a simulation data group containing multiple parameters according to different set values.

[0004] According to an embodiment of the present invention, the simulation system of the device twin data of the present invention includes a storage device and a processor. The storage device stores multiple modules, and the multiple modules include an initial data generation module, a data calculation module, a data generation module, an abnormal data detection module and a data output module. The processor is coupled to the storage device. The processor executes the initial data generation module to generate an initial data group according to the simulation configuration data group through the initial data generation module. And the processor executes the data calculation module to perform statistical calculations on the initial data group through the data calculation module, and then generates multiple statistical feature data groups respectively. The processor executes the data generation module to generate a simulation data group according to the simulation configuration data group, the initial data group and the statistical feature data group through the data generation module. And the data generation module inputs the simulation data group into the database. The processor performs abnormal data judgment on the simulation data group according to the abnormal data detection module, and decides to continue executing or stop the data generation module to generate the next set of simulation data groups based on the judgment result of the abnormal data judgment.

[0005] According to an embodiment of the present invention, the simulation method of device twin data of the present invention includes the following steps: generating an initial data group according to a simulation configuration data group through an initial data generation module; performing statistical calculations on the initial data group through a data calculation module to generate multiple statistical feature data groups respectively; generating a simulation data group according to the simulation configuration data group, the initial data group and multiple statistical feature data groups through a data generation module, and inputting the simulation data group into a database; performing abnormal data judgment according to the simulation data group stored in the database through an abnormal data detection module, and deciding to continue executing or stop the data generation module to generate the next set of simulation data groups according to the judgment result of the abnormal data judgment.

[0006] Based on the above, the device twin data simulation system and device twin data simulation method of the present invention can generate simulation data sets corresponding to different set values. The simulation data set contains multiple parameters / data related to the equipment. In this way, the device twin data simulation system and method achieve a higher degree of simulation / realism during the operation of the digital factory simulation. At the same time, it provides comprehensive and real-time abnormality judgment data and processing to more closely resemble the various situations of the equipment in actual operation.

[0007] In order to make the above features and advantages of the present invention more clearly understood, embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 is a schematic diagram of a simulation system for device twin data according to an embodiment of the present invention;

[0009] Figure 2 is a flowchart of a method for simulating device twin data according to an embodiment of the present invention;

[0010] Figure 3 It is a flowchart of a simulation system for device twin data according to an embodiment of the present invention.

[0011] Description of Reference Numerals

[0012] 100: Simulation system of device twin data;

[0013] 110: processor;

[0014] 120: storage device;

[0015] 121: initial data generation module;

[0016] 122: data calculation module;

[0017] 123: data generation module;

[0018] 1231: normal data generation unit;

[0019] 1232: abnormal data generation unit;

[0020] 124: Abnormal data detection module;

[0021] 125: database;

[0022] 126: data output module;

[0023] 127: Abnormal probability trigger module;

[0024] 131: simulation configuration data;

[0025] 132: Abnormal data pattern library;

[0026] 133: Data pattern library;

[0027] S210~S260: steps. DETAILED DESCRIPTION

[0028] Reference will now be made in detail to exemplary embodiments of the present invention, examples of which are illustrated in the accompanying drawings. Whenever possible, the same reference numerals are used in the drawings and the description to refer to the same or like parts.

[0029] Figure 1 Schematic diagram of a device twin data simulation system according to an embodiment of the present invention. Figure 1 , the simulation system 100 of device twin data includes a processor 110 and a storage device 120. The processor 110 is coupled to the storage device 120. The storage device 120 can store a simulation configuration data set, a database 125, and a plurality of modules. The processor 110 can access the data in the storage device 120 to execute a plurality of modules. In this embodiment, the plurality of modules include an initial data generation module 121, a data calculation module 122, a data generation module 123, an abnormal data detection module 124, and a data output module 126. For example, the initial data generation module 121 may be, for example, an initializer to generate initial data according to the initial settings in the simulation configuration data set.

[0030] Figure 2 This is a flow chart of a method for simulating device twin data according to an embodiment of the present invention. Figure 1 as well as Figure 2, the simulation system 100 of the device twin data of this embodiment can execute the following steps S210 to S260 to generate a simulation data group corresponding to the simulation configuration data group, and perform abnormality detection on each data. In step S210, the processor 110 inputs the simulation configuration data group into the initial data generation module 121, and then generates an initial data group according to the simulation configuration data through the initial data generation module 121. For example, the processor 110 inputs the simulation configuration data group pre-set / selected by the user into the initial data generation module 121, so that the initial data generation module 121 generates an initial data group. In this embodiment, the initial data group is the first group of multiple data in the simulation data. For example, the initial data group may include items such as cooling water, tool and screw, all with serial numbers 1 and values ​​20, 100 and 0 respectively.

[0031] In step S220, the processor 110 may perform statistical calculations on the initial data set via the data calculation module 122 to generate a statistical feature data set. For example, the processor 110 executes the data calculation module 122 to perform statistical calculations on the initial data set based on different project / device parameter types to generate corresponding multiple statistical feature data sets. In step S230, the processor 110 generates a simulation data set based on the simulation configuration data set, the initial data set, and the statistical feature data set via the data generation module 123.

[0032] In this way, the data generation module 123 generates corresponding simulation data according to the multi-dimensional data in the simulation configuration data group, the multi-dimensional data type (for example, different growth trends / data trends of multiple data), the relationship between multiple data, and the multiple set values ​​corresponding to the multiple data (for example, upper and lower bounds, average values, starting values, variances, the amount of generated data, the type of abnormality, and the probability of abnormality, etc.), and the multiple simulation data form a simulation data group. For example, multi-dimensional data are, for example, multiple data that are related to each other, so the related multiple data can be represented by a plane data graph or a multi-dimensional graph. In other words, the simulation configuration data group includes not only multiple data related to the device, but also the relationship between multiple data, so that the simulation system 100 of the device twin device and the method thereof can provide a high-simulation, high-realism simulation data group.

[0033] In step S240, the processor 110 inputs the simulated data set into the database 125 via the data generation module 123. The database 125 is, for example, a memory block in the storage device 120 for storing the initial data set and the simulated data set. In step S250, the abnormal data detection module 124 performs abnormal data determination based on the simulated data set. In this embodiment, the processor 110 executes the abnormal data detection module 124 to perform abnormal data determination based on the simulated data set stored in the database 125.

[0034] In step S260, according to the judgment result of the abnormal data judgment, a warning message / simulation data group is output. In the present embodiment, the processor 110 determines to continue to execute or stop the data generation module 123 to generate the next set of simulation data groups based on the judgment result of the abnormal data judgment (i.e., the judgment performed by the abnormal data detection module 124). In one embodiment, the processor 110 determines to continue to execute or stop the data generation module 126 to generate the next set of simulation data groups based on the judgment result of the abnormal data judgment, and further comprises: when the judgment result of the abnormal data judgment is that there is abnormal data, the abnormal data detection module 124 outputs a warning message, and stops the data generation module 123 from generating the next set of simulation data groups, and when the judgment result of the abnormal data judgment is that there is no abnormal data, the abnormal data detection module 124 outputs the simulation data group through the data output module 126, and the processor 110 continues / executes the data generation module 123 again to generate the next set of simulation data groups.

[0035] That is to say, the simulation system 100 and method of device twin data stores each simulation data set into the database 125, so that the abnormal data detection module 124 connects to the database 125 to access any data in the database 125. In this way, the abnormal data detection module 124 can detect and determine in real time whether any data value of the simulation data set has a data abnormality (for example, the value is greater than / less than a pre-set normal range of data). If so, the data generation module 123 is stopped from generating the next set of simulation data sets, and a warning message is output to the external system / user's device. If not, the data generation module 123 is executed to continue generating the next set of simulation data sets.

[0036] In one embodiment, the processor 110 may also be coupled to an external database or an internal database 125 to read / write the simulation configuration data group in the external database or the internal database 125 and the simulation configuration data group pre-set by the user. In another embodiment, the simulation system 100 of device twin data may also include a transceiver to receive / transmit the data group / setting data group transmitted by the client. In this way, the simulation system 100 of device twin data of the present invention can generate corresponding simulation data groups based on the selected data type, multiple parameter setting values ​​and multi-dimensional data, and at the same time perform abnormal data judgment on each simulation data group according to the abnormal data detection module 124. In this way, the simulation data group generated by the simulation system 100 of device twin data and its method has the advantage of high authenticity, and has the function of real-time judgment of data anomalies for different setting ranges in the data.

[0037] In this embodiment, the processor 110 may include, for example, a central processing unit (CPU), or other programmable general-purpose or special-purpose microprocessor, digital signal processor (DSP), application-specific integrated circuit (ASIC), programmable logic device (PLD), other similar processing circuits, or a combination of these devices. The storage device 120 may include memory and / or a database. In one embodiment, the storage device 120 may store simulation configuration data sets, a data pattern library, an abnormal data pattern library, and a database 125. Furthermore, the storage device 120 may be, for example, a non-volatile memory (NVM). The storage device 120 may store relevant programs, modules, systems, or algorithms for implementing various embodiments of the present invention, for access and execution by the processor 110 to implement the relevant functions and operations described in various embodiments of the present invention.

[0038] In one embodiment, the initial data generation module 121, the data calculation module 122, the data generation module 123, the abnormal data detection module 124, the database 125, the data output module 126 and the abnormal probability trigger module 127 can be implemented in a programming language such as JSON (JavaScript Object Notation), Extensible Markup Language (XML) or YAML, but the present invention is not limited to this. In this embodiment, the simulation system 100 of device twin data can be specifically implemented in a personal computer (PC), a local server (Server) or a cloud server, and the present invention is not limited to this. In one embodiment, the simulation system 100 of device twin data can also be integrated into an enterprise resource planning (ERP) system to provide highly realistic device twin data (simulation data group) according to pre-set data types.

[0039] In the present invention, the data type in the simulation configuration data group can be at least one of a fluctuating type, an increasing type, an increasing level type, a decreasing type, a decreasing level type, an exponential type, or a linear rule type. In addition, the processor 110 uses the data generation module 123 and the data calculation module 122 to make the simulation data of the next group in the simulation data group conform to the corresponding data type with the previous group of simulation data. In this way, the user can set the corresponding data type according to the various parameter types of the equipment to be simulated / simulated. For example, the simulated production equipment data includes tool wear data and water temperature data. In this way, the user selects a decreasing data type to simulate tool wear and selects an increasing row data type to simulate the gradual increase in water temperature. In this way, multiple data in the simulation data group can be simulated and generated according to the corresponding data type, so that the simulation data group generated by the simulation system 100 of equipment twin data and its method has high simulation and high applicability.

[0040] Figure 3 This is a flow chart of a device twin data simulation system according to an embodiment of the present invention. Figures 1 to 3 In this embodiment, the multiple modules further include a data output module 126 and an abnormal probability trigger module 127. Furthermore, the storage device 120 further stores an abnormal data pattern library 132 and a data pattern library 133. In this embodiment, the data generation module 123 further includes a normal data generation unit 1231 and an abnormal data generation unit 1232.

[0041] In one embodiment, the step of generating a simulated data set based on the simulated configuration data set (simulated configuration data 131), the initial data set, and multiple statistical feature data sets by the data generation module 123 includes: a normal data generation unit 1231 generating normal data based on the simulated configuration data set (simulated configuration data 131), the initial data set, and the multiple statistical feature data sets to generate a normal pattern data set, wherein the simulated data set (i.e., the current simulated data set) includes the initial data set and the normal pattern data set. Furthermore, the normal data generation unit 1231 inputs the simulated data set (i.e., the current simulated data set) into the abnormal data generation unit 1232. Subsequently, the abnormal data generation unit 1232 generates an abnormal pattern data set based on multiple abnormality occurrence probabilities in the simulated configuration data set (simulated configuration data 131). When the abnormal data generation unit 1232 generates the abnormal pattern data set, the simulated data set (i.e., the current simulated data set) also includes the abnormal pattern data set.

[0042] It is worth noting that the normal mode data group mentioned in the present invention is a data group generated according to the setting of the normal mode, that is, even the data group generated in the normal mode may be greater than the upper critical value and the lower critical value (i.e., the upper / lower limit) in the simulation configuration data 131. For example, if the production equipment to be simulated is a wear-type equipment, even if any abnormality probability is not triggered / no abnormality occurs. However, according to the actual operation scenario and the normal mode data group of the present invention, after a certain number of uses / times of wear-type production equipment, the data may still exceed the critical value (for example, the number of uses of the wear-type production equipment has reached the critical value and the equipment temperature is too high). In this way, the simulation system 100 of the device twin data of the present invention and the method thereof can not only be used to generate a device simulation data group with high authenticity, but can also be used to simulate the estimated operating time of each parameter / each component of the device, so that users can obtain device twin data through the simulation system.

[0043] In one embodiment, when the abnormal data generation unit 1232 generates an abnormal pattern data group based on multiple abnormal occurrence probabilities, the abnormal data generation unit 1232 performs abnormal data generation to generate an abnormal pattern data group based on the simulation configuration data group (simulation configuration data 131), the initial data group and the current simulation data group (that is, the normal pattern data group and the initial data group generated by the normal data generation unit 1231 in the previous step), and the abnormal data generation unit 1232 uses the abnormal pattern data group, the initial data group and the current simulation data group as the current simulation data group, and inputs them into the data calculation module 122.

[0044] On the other hand, when the abnormal data generation unit 1232 does not generate an abnormal pattern data set based on the multiple abnormal occurrence probabilities (i.e., none of the abnormal occurrence probabilities are triggered / no abnormality occurs), the abnormal data generation unit 1232 uses the initial data set and the simulated data set as the current simulated data set and inputs them to the data calculation module 122. Next, the data calculation module 122 performs statistical calculations based on the current simulated data set to generate multiple current statistical feature data sets, and the processor 110 repeatedly performs normal data generation and abnormal data generation via the normal data generation unit 1231 and the abnormal data generation unit 1232, respectively, to obtain the next set of simulated data sets until the processor 110 stops the normal data generation unit 1231 and the abnormal data generation unit 1232 from generating a new simulated data set via the abnormal data detection module 124.

[0045] For example, the simulation configuration data 131 may store at least one of a normal data type, a starting value, an average value, a maximum value, a minimum value, an abnormal data type, a data type (such as a normal data type), an abnormal occurrence probability, and a single data volume. It is worth noting that the simulation configuration data group of the present invention (i.e., the simulation configuration data 131) also includes multiple abnormal data types and corresponding abnormal probabilities for a single generated data item. In other words, the abnormal pattern data group generated by the abnormal data generation unit 1232 may include different abnormal probabilities corresponding to multiple abnormal occurrences / data types. For example, the abnormal state of the screw vibration in the production equipment may include multiple abnormal states such as position abnormality, speed abnormality, or power abnormality. The simulation configuration data 131 can set different abnormal probabilities (for example, one in a thousand, one in a hundred thousand) for different abnormal states, so that the abnormal pattern data group has high simulation.

[0046] The simulation configuration data 131 may be as shown in Table 1 below.

[0047]

[0048]

[0049] Table 1

[0050] Next, the initial data generation module 121 generates an initial data set based on the simulated configuration data set (step S210). The initial data set may be as shown in Table 2 below. The initial data generation module 121 inputs the initial data set into the database 125 so that the abnormal data detection module 124 can detect the initial data set stored in the database 125. In this embodiment, the abnormal data detection module 124 determines that there is no abnormal data. Therefore, the processor 110 inputs the initial data set into the data calculation module 122 via the database 125. The data calculation module 122 performs statistical calculations on the initial data set to generate a corresponding statistical feature data set (step S220). The initial data set may be as shown in Table 3 below.

[0051] project Serial number Numerical cooling water 1 20 Knives 1 100 screw 1 0

[0052] Table 2

[0053]

[0054]

[0055] Table 3

[0056] In this example, the processor 110 then generates a simulation data set (i.e., a normal mode data set and an abnormal mode data set) based on the simulation configuration data set, the initial data set, and the statistical feature data set, respectively, through the normal data generation unit 1231 and the abnormal data generation unit 1232 in the data generation module 123 (step S230). In this example, the normal mode data set may be as shown in Table 4 below, and the abnormal mode data set may be as shown in Table 5 below. The current simulation data set may be as shown in Table 6 below.

[0057]

[0058] Table 4

[0059]

[0060] Table 5

[0061] project Serial number Numerical cooling water 1 20 cooling water 2 23 Knives 1 100 Knives 2 99.9998 screw 1 0 screw 2 4950

[0062] Table 6

[0063] In one embodiment, the data calculation module 122 performs a statistical calculation group based on the simulation data group and the current simulation data group stored in the database 125 to generate a statistical feature data group corresponding to each data. That is, the simulation system 100 of the device twin data and the simulation method of the device twin data will repeatedly generate a normal mode data group and an abnormal mode data group through the normal data generation unit 1231 and the abnormal data generation unit 1232 in the data generation module 123. Moreover, each time the normal mode data group and the abnormal mode data group are generated, the statistical feature data of the current simulation data (i.e., the current normal mode data, the current abnormal mode data and the initial data) are used as reference data to make the simulation data group highly consistent with the set value in the simulation configuration data group.

[0064] For example, after the data generation module 123 generates a first set of simulation data, it stores the first set of simulation data in the database 125. Then, the data calculation module 122 reads the first set of simulation data and the initial data to perform statistical calculations, thereby generating a statistical feature data set corresponding to the first set of simulation data and the initial data. Next, the data generation module 123 performs data generation again based on the statistical feature data, the initial data, and the first set of simulation data to generate a second set of simulation data (i.e., including the initial data, the first simulation data, and the second simulation data). Similarly, the data generation module 123 stores the second set of simulation data in the database 125 and repeats the above statistical feature analysis and data generation until the abnormal data detection module 124 detects abnormal data and outputs an abnormality determination result to stop the data generation module 123 from generating the next set of simulation data.

[0065] In this example, the processor 110 inputs the normal pattern data set and the abnormal pattern data set into the database 125, so that the abnormal data detection module 124 reads the data in the database 125 and generates a corresponding abnormal data judgment result (step S250). In this example, the abnormal data judgment result is that there is no abnormal data. Therefore, the processor 110 continues to execute the data calculation module 122 and the data generation module 123 through the abnormal data detection module 124 based on the abnormal data judgment result (step S260) to repeatedly generate the next set of statistical feature data sets, the next set of normal pattern data sets, and the next set of abnormal pattern data sets until the abnormal data generated by the abnormal data detection module 124 is judged to have abnormal data. In this embodiment, the next set of statistical feature data sets can be shown in Table 7 below, the next set of normal pattern data sets can be shown in Table 8 below, and the next set of abnormal pattern data sets can be shown in Table 9 below.

[0066] project Statistical characteristics Numerical cooling water Total data 2 average value 21.5 Maximum 23 Minimum 20 Knives Total data 2 average value 99.9999 Maximum 100 Minimum 99.9998 screw Total data 2 average value 2475 Maximum 4950 Minimum 0

[0067] Table 7

[0068]

[0069]

[0070] Table 8

[0071]

[0072] Table 9

[0073] In one embodiment, the data pattern library 133 stores normal data patterns. Thus, the normal data generation unit 1231 generates a normal pattern data set based on the normal data patterns. The normal pattern data set includes multiple device data types and corresponding sets of normal device data. The abnormal data pattern library 132 stores abnormal data patterns. Thus, the abnormal data generation unit 1232 generates an abnormal pattern data set based on the abnormal data patterns. The abnormal pattern data set includes multiple device data types and corresponding sets of abnormal device data.

[0074] For example, a normal data pattern can be at least one of a fluctuating pattern, an increasing pattern, an increasing level pattern, a decreasing pattern, a decreasing level pattern, an exponential pattern, and a linear regular pattern. An abnormal data pattern can be at least one of a single-point abnormality, a multi-point continuous abnormality, an alarm-type abnormality, and a data-free abnormality, but the present invention is not limited thereto. Specifically, the two storage blocks divided in storage device 120, data pattern library 133 and abnormal data pattern library 132, are used to increase the speed at which normal data generation unit 1231 and abnormal data generation unit 1232 read normal data patterns and abnormal data patterns from storage device 120.

[0075] In one embodiment, the data output module 126 is used to output simulated data sets and / or warning information to an external electronic device. The output simulated data sets are user-specified. The data output module 126 can be, for example, a data exchange interface or an output module communicatively connected to a data transceiver. Thus, the data output module 126 is used to output simulated data to an external electronic device. Examples of such electronic devices include servers, databases, laptop computers, and desktop computers.

[0076] In one embodiment, the database 125 can store the initial data set, the normal mode data set, the abnormal mode data set, the current simulation data set, and the simulation data set in a list format. The result data set 125 is coupled to the data output module 126 to output the designated data via the data output module 126. For example, a user or relevant personnel transmits a data designation instruction via a transceiver and the data output module 126. In this manner, the processor 110 outputs the designated data to the user's electronic device according to the data designation instruction.

[0077] In one embodiment, the abnormal probability trigger module 127 triggers the abnormal data generation unit 1232 to generate an abnormal pattern data group based on multiple abnormal occurrence probabilities, and the multiple abnormal occurrence probabilities include multiple abnormal states and corresponding multiple abnormal probabilities. The simulation configuration data group (i.e., the simulation configuration data 131) stores multiple abnormal occurrence probabilities, and corresponds to multiple abnormal states (as shown in Table 1 above). For example, abnormal states are, for example, cooling water blockage, power failure, and excessive water temperature, but the present invention is not limited to this. The abnormal occurrence probabilities corresponding to the multiple abnormal states may be, for example, 1 in 500,000, 1 in 100, and 1 in 1,000, respectively. Moreover, each abnormal state corresponds to an abnormal data pattern (i.e., a data type). In this way, the abnormal data generation unit 1232 generated by the simulation system 100 of device twin data and the simulation method of device twin data can not only cover multiple abnormal states, but also the simulation configuration data group (i.e., the simulation configuration data 131) can include different abnormal occurrence probabilities according to different abnormal state settings. Therefore, the device twin data simulation system 100 and method thereof have the advantages of being applicable to different production equipment and generating abnormal data accordingly, and thus having the advantages of being flexible and comprehensively applicable to different production equipment.

[0078] In summary, the simulation system 100 of the equipment twin data and the simulation method of the equipment twin data of the present invention can generate simulation data groups according to different set values. The simulation data group contains multiple parameters / data related to the equipment. In this way, the simulation system of the equipment twin data and the method thereof achieve a higher degree of simulation / realism during the simulation operation of the digital factory. At the same time, the simulation system 100 of the equipment twin data and the simulation method of the equipment twin data of the present invention provide comprehensive and real-time abnormal data judgment and processing through the abnormal data detection module 124. Moreover, by simulating the configuration data group including multi-dimensional / multi-item data combinations and settings to simulate the comprehensive status data of the equipment, it is closer to various situations of the equipment under the actual operating state.

[0079] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A device twin data simulation system, characterized in that: include: a storage device for storing a simulation configuration data set, a database, and a plurality of modules, wherein the plurality of modules include an initial data generation module, a data calculation module, a data generation module, an abnormal data detection module, and a data output module; as well as a processor coupled to the storage device, wherein the processor executes the initial data generation module and inputs the simulation configuration data group into the initial data generation module so as to generate an initial data group through the initial data generation module; wherein the processor executes the data calculation module to perform statistical calculations on the initial data set to generate a plurality of statistical feature data sets respectively; wherein the processor executes the data generation module to generate a simulation data set according to the simulation configuration data set, the initial data set, and the statistical feature data set through the data generation module, and the data generation module inputs the simulation data set into the database; wherein the processor executes the abnormal data detection module to perform abnormal data judgment based on the simulation data group stored in the database, and the processor decides to continue or stop the data generation module to generate the next set of simulation data groups based on the judgment result of the abnormal data judgment, The simulation configuration data group includes: normal data type, starting value, average value, maximum value, minimum value, abnormal data type, multiple data types, abnormal occurrence probability, data generation volume and at least one of single data volume.

2. The device twin data simulation system according to claim 1, characterized in that: When the abnormal data judgment result is that there is abnormal data, the abnormal data detection module outputs a warning message and stops executing the data generation module to generate the next set of simulation data groups. When the determination result of the abnormal data determination is that there is no abnormal data, the abnormal data detection module outputs the simulation data set through the data output module, and the processor continues to execute the data generation module to generate the next set of simulation data sets.

3. The device twin data simulation system according to claim 1, characterized in that: The data generation module generates the simulation data group according to the multiple data types in the simulation configuration data group, and the multiple data types include at least one of the fluctuation type, increasing type, increasing level type, decreasing type, decreasing level type, exponential type and linear rule type, and the processor uses the data generation module and the data calculation module to make the data relationship between the next group of data groups in the simulation data group and the previous group of data groups conform to the corresponding multiple data types.

4. The device twin data simulation system according to claim 1, characterized in that: The data generation module also includes a normal data generation unit and an abnormal data generation unit. wherein the normal data generating unit performs normal data generation according to the simulation configuration data group, the initial data group, and the plurality of statistical feature data groups to generate a normal pattern data group, and the simulation data group includes the initial data group and the normal pattern data group, The abnormal data generating unit generates an abnormal pattern data group based on a plurality of abnormal occurrence probabilities in the simulation configuration data group, and when the abnormal data generating unit generates the abnormal pattern data group, the simulation data group also includes the abnormal pattern data group.

5. The device twin data simulation system according to claim 4, characterized in that: When the abnormal data generating unit generates the abnormal pattern data group based on the multiple abnormal occurrence probabilities, the abnormal data generating unit performs abnormal data generation to generate the abnormal pattern data group according to the simulation configuration data group, the initial data group and the simulation data group, and the abnormal data generating unit uses the abnormal pattern data group, the initial data group and the simulation data group as the current simulation data group and inputs them into the data calculation module, When the abnormal data generating unit does not generate the abnormal pattern data group based on the multiple abnormal occurrence probabilities, the abnormal data generating unit uses the initial data group and the simulation data group as the current simulation data group and inputs them into the data calculation module, The data calculation module performs the statistical calculation based on the current simulation data group to generate multiple current statistical feature data groups, and the processor repeatedly performs the normal data generation and the abnormal data generation through the normal data generation unit and the abnormal data generation unit to obtain the next set of the simulation data groups until the processor stops the normal data generation unit and the abnormal data generation unit from generating the simulation data groups through the abnormal data detection module.

6. The device twin data simulation system according to claim 5, characterized in that: The storage device also stores a data pattern library and an abnormal data pattern library. The data pattern library stores normal data patterns, and the normal data generation unit generates the normal pattern data group based on the normal data pattern, wherein the normal pattern data group includes multiple device data types and corresponding multiple groups of device normal data. The abnormal data pattern library stores abnormal data patterns, and the abnormal data generation unit generates the abnormal pattern data group based on the abnormal data pattern, wherein the abnormal pattern data group includes multiple device data types and corresponding multiple groups of device abnormal data. The database stores the initial data set, the normal mode data set, the abnormal mode data set, the current simulation data set, and the simulation data set in a column-by-column manner.

7. The device twin data simulation system according to claim 5, characterized in that: The multiple modules also include an abnormal probability trigger module, which triggers the abnormal data generation unit to generate the abnormal pattern data group based on the multiple abnormal occurrence probabilities, and the multiple abnormal occurrence probabilities include multiple abnormal states and corresponding multiple abnormal probabilities.

8. The device twin data simulation system according to claim 1, characterized in that: The data output module is used to output the simulation data group and warning information to an external electronic device.

9. The device twin data simulation system according to claim 5, characterized in that: The data calculation module performs the statistical calculation according to the simulation data set stored in the database and the current simulation data set to generate a statistical feature data set corresponding to each data.

10. A method for simulating device twin data, characterized in that: include: Generate an initial data set according to the simulation configuration data set by an initial data generation module; Performing statistical calculations on the initial data set by a data calculation module to generate a plurality of statistical feature data sets respectively; Generate a simulation data set according to the simulation configuration data set, the initial data set and the plurality of statistical feature data sets by a data generation module, and input the simulation data set into a database; as well as The abnormal data detection module performs abnormal data judgment based on the simulation data group stored in the database, and decides to continue or stop the data generation module to generate the next set of simulation data groups based on the judgment result of the abnormal data judgment. The simulation configuration data group includes: normal data type, starting value, average value, maximum value, minimum value, abnormal data type, multiple data types, abnormal occurrence probability, data generation volume and at least one of single data volume.

11. The device twin data simulation method according to claim 10, characterized in that: Also includes: When the abnormal data determination result is that there is abnormal data, the abnormal data detection module outputs a warning message and stops executing the data generation module to generate the next set of simulation data groups. When the abnormal data determination result is that there is no abnormal data, the abnormal data detection module and the data output module output the simulation data set, and the data generation module generates the next set of simulation data sets.

12. The device twin data simulation method according to claim 10, characterized in that: The simulation data set is generated according to a plurality of data types in the simulation configuration data set, and the plurality of data types include at least one of a fluctuation type, an increasing type, an increasing level type, a decreasing type, a decreasing level type, an exponential type, and a linear rule type; The simulation method further comprises: The data generating module and the data calculating module are used to make the data relationship between the next data group in the simulation data group and the previous data group conform to the corresponding multiple data types.

13. The device twin data simulation method according to claim 10, characterized in that: Also includes: Normal data generation is performed by a normal data generation unit according to the simulation configuration data group, the initial data group and the plurality of statistical feature data groups to generate a normal pattern data group, wherein the simulation data group includes the initial data group and the normal pattern data group. The abnormal data generating unit generates an abnormal pattern data group based on a plurality of abnormality occurrence probabilities in the simulation configuration data group, and when the abnormal data generating unit generates the abnormal pattern data group, the simulation data group also includes the abnormal pattern data group.

14. The device twin data simulation method according to claim 13, characterized in that: Also includes: When the abnormal data generating unit generates the abnormal pattern data group based on the multiple abnormal occurrence probabilities, the abnormal data generating unit performs abnormal data generation to generate the abnormal pattern data group according to the simulation configuration data group, the initial data group and the simulation data group, and the abnormal data generating unit uses the abnormal pattern data group, the initial data group and the simulation data group as the current simulation data group and inputs them into the data calculation module, When the abnormal data generating unit does not generate the abnormal pattern data group based on the multiple abnormal occurrence probabilities, the initial data group and the simulation data group are used as the current simulation data group by the abnormal data generating unit and input into the data calculation module, The data calculation module performs the statistical calculation based on the current simulation data group to generate multiple current statistical feature data groups, and the normal data generation unit and the abnormal data generation unit repeatedly perform the normal data generation and the abnormal data generation respectively to obtain the next set of the simulation data groups until the abnormal data detection module stops the normal data generation unit and the abnormal data generation unit from generating the simulation data groups.

15. The device twin data simulation method according to claim 14, characterized in that: Also includes: The normal data pattern is stored in a data pattern library, and the normal data generation unit generates the normal pattern data group based on the normal data pattern, wherein the normal pattern data group includes multiple device data types and corresponding multiple groups of device normal data. The abnormal data pattern is stored in an abnormal data pattern library, and the abnormal data generation unit generates the abnormal pattern data group based on the abnormal data pattern, wherein the abnormal pattern data group includes multiple device data types and corresponding multiple groups of device abnormal data. The database stores the initial data set, the normal mode data set, the abnormal mode data set, the current simulation data set, and the simulation data set in a column-by-column manner.

16. The device twin data simulation method according to claim 14, characterized in that: Also includes: The abnormal data generation unit is triggered by an abnormal probability trigger module to generate the abnormal pattern data group based on the multiple abnormal occurrence probabilities, and the multiple abnormal occurrence probabilities include multiple abnormal states and corresponding multiple abnormal probabilities.

17. The device twin data simulation method according to claim 10, characterized in that: The data output module is used to output the simulation data group and the warning information to an external electronic device.

18. The device twin data simulation method according to claim 14, characterized in that: The statistical calculation is performed by the data calculation module according to the simulation data set stored in the database and the current simulation data set to generate a statistical feature data set corresponding to each data.

Images