An automatic correction method, apparatus, device, and medium for point binding anomalies

Abstract

This invention discloses an automatic correction method, apparatus, device, and medium for point binding anomalies. The method includes: receiving a point data packet from a third-party IoT subsystem containing the target third-party point identifier and actual data values; determining matching virtual object instances and attribute types based on a point binding relationship table; searching for corresponding attribute value ranges through an industry knowledge base; if the actual data value does not fall into the range, marking it as an abnormal binding point; filtering candidate attribute value ranges into which the actual data value can fall, and determining candidate virtual object attribute types; constructing a correction reference factor by combining the original description information and candidate attributes, inputting it into a pre-trained point recognition model for identification, and correcting the abnormal binding relationship accordingly. This invention, through an automated process combined with an industry knowledge base and a point recognition model, solves the problems of low efficiency, high error rate, and difficulty in anomaly detection associated with manual point binding, improving the efficiency, accuracy, and data quality of point binding on digital twin platforms.

Description

Technical Field

[0001] This invention relates to the field of digital twin technology, and in particular to an automatic correction method, apparatus, device, and medium for point binding anomalies. Background Technology

[0002] In the field of digital twins, IoT data acquisition is a core component. Current digital twin platforms require manual identification of point lists (i.e., third-party point tables) from third-party IoT subsystems, and manual establishment of mapping relationships between the platform and subsystem information points to complete data acquisition. Furthermore, after data access, acquisition errors caused by manual mapping are difficult to detect during the acquisition phase, and mostly need to be passively discovered in subsequent data applications through problem-driven approaches. The existing technology's method of manually identifying and binding points has the following drawbacks:

[0003] The current IoT data acquisition solutions of digital twin platforms have two major defects: (1) Manual identification and manual binding are inefficient and have a high error rate. Manually establishing mappings is time-consuming and labor-intensive; and due to staff fatigue and insufficient industry knowledge, binding errors occur frequently and are difficult to control. (2) Data acquisition errors cannot be effectively investigated. Only some data acquisition failures can be found. Problems such as incorrect or erroneous acquisition caused by manual mapping are difficult to investigate and are often exposed only in the data application stage, resulting in a large amount of historical garbage data and reducing data quality. Summary of the Invention

[0004] Based on this, the present invention provides an automatic correction method, device, equipment and medium for point binding anomalies, so as to solve the problems of low efficiency, high error rate and difficulty in actively detecting and correcting binding anomalies in the prior art.

[0005] In a first aspect, embodiments of the present invention provide an automatic correction method for point binding anomalies, including:

[0006] Receive a target location data packet from a third-party IoT subsystem, wherein the target location data packet carries at least the target third-party location identifier and the target actual data value;

[0007] Based on the location binding relationship table pre-stored in the digital twin platform, the target virtual object instance and target virtual object attribute type that match the target third-party location identifier are determined; wherein, the location binding relationship table records the association relationship between the third-party location identifier, the virtual object instance name, the virtual object attribute type, and the original description information;

[0008] The target attribute value range corresponding to the target virtual object attribute type is found by searching the industry knowledge base. If the actual data value of the target does not fall into the target attribute value range, the target third-party point is marked as a binding abnormal point.

[0009] In the industry knowledge base, filter all candidate attribute value ranges that the actual data value of the target can fall into. Based on the attribute value ranges corresponding to each virtual object attribute type pre-stored in the industry knowledge base, determine all candidate virtual object attribute types corresponding to the actual value of the target.

[0010] By combining the original description information and the attributes of each candidate virtual object, a correction reference factor corresponding to the binding abnormal point is constructed. The correction reference factor is input into the pre-trained point recognition model to identify the original description information, and the abnormal binding relationship of the point is corrected according to the recognition result.

[0011] Secondly, embodiments of the present invention also provide an automatic correction device for point binding anomalies, comprising:

[0012] The point data packet receiving module is used to receive target point data packets from a third-party IoT subsystem, wherein the target point data packets carry at least the target third-party point identifier and the target actual data value;

[0013] The point binding relationship acquisition module is used to determine the target virtual object instance and the target virtual object attribute type that match the target third-party point identifier based on the point binding relationship table pre-stored in the digital twin platform; wherein, the point binding relationship table records the association relationship between the third-party point identifier, the virtual object instance name, the virtual object attribute type, and the original description information;

[0014] The point binding anomaly determination module is used to search for the target attribute value range corresponding to the target virtual object attribute type through the industry knowledge base. If the actual data value of the target does not fall into the target attribute value range, the target third-party point is marked as a binding anomaly point.

[0015] The virtual object attribute type filtering module is used to filter all candidate attribute value ranges that the actual data value of the target can fall into in the industry knowledge base. Based on the attribute value ranges corresponding to each virtual object attribute type pre-stored in the industry knowledge base, it determines all candidate virtual object attribute types corresponding to the actual value of the target.

[0016] The point correction module is used to combine the original description information and the attributes of each candidate virtual object to construct a correction reference factor corresponding to the binding abnormal point. The correction reference factor is input into the pre-trained point recognition model to identify the original description information, and the abnormal binding relationship of the point is corrected according to the recognition result.

[0017] Thirdly, embodiments of the present invention also provide an electronic device, the electronic device comprising:

[0018] At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform an automatic correction method for point binding anomalies as described in any embodiment of the present invention.

[0019] Fourthly, embodiments of the present invention also provide a computer-readable storage medium storing computer instructions, which are used to cause a processor to execute an automatic correction method for point binding anomalies as described in any embodiment of the present invention.

[0020] This invention eliminates the need for manual identification of location lists or manual mapping. It automates the identification and correction of location binding anomalies, significantly reducing binding processing time and lowering manual operation costs. By relying on standard attribute value ranges from an industry knowledge base to identify anomalies and combining this with a pre-trained model to accurately match correct binding relationships, it avoids misbinding issues caused by personnel fatigue or insufficient industry knowledge, improving the stability and accuracy of binding results. Furthermore, it detects binding anomalies during the data collection phase, rather than passively identifying them during data application, avoiding the accumulation of large amounts of historical junk data and directly improving the overall data quality of the digital twin platform.

[0021] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a flowchart of an automatic correction method for point binding anomalies provided in Embodiment 1 of the present invention;

[0024] Figure 2 This is a flowchart of another automatic correction method for point binding anomalies provided in Embodiment 2 of the present invention;

[0025] Figure 3 This is a schematic diagram of an automatic correction device for point binding anomalies provided in Embodiment 3 of the present invention;

[0026] Figure 4This is a schematic diagram of the structure of an electronic device that implements an automatic correction method for point binding anomalies according to an embodiment of the present invention. Detailed Implementation

[0027] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] Example 1

[0030] Figure 1 This is a flowchart of an automatic correction method for point binding anomalies provided in Embodiment 1 of the present invention. This embodiment is applicable to point binding and anomaly correction scenarios between a digital twin platform and a third-party IoT subsystem. The method can be executed by an automatic correction device for point binding anomalies, which can be implemented in hardware and / or software and can be configured within the digital twin platform. Figure 1 As shown, the method includes:

[0031] S110. Receive target location data packets from a third-party IoT subsystem, wherein the target location data packets carry at least the target third-party location identifier and the target actual data value.

[0032] A third-party IoT subsystem refers to an external IoT system (such as a sensor network or equipment monitoring system) that interacts with the digital twin platform, responsible for collecting real-time data from physical entities and sending it to the digital twin platform. A target location data packet refers to an information packet containing data from a specific location sent by the third-party IoT subsystem; it is the basic unit of data transmission. The target third-party location identifier is a unique identifier assigned to each collection point in the third-party IoT subsystem to distinguish different locations (e.g., "Sensor_302" represents sensor number 2 in workshop 3). The target actual data value is the real-time data of the physical entity collected by the target third-party location (e.g., temperature 25℃, pressure 3MPa), which is the raw data reflecting the entity's state. The digital twin platform receives the target location data packet sent by the third-party IoT subsystem through a pre-defined communication interface and extracts key information from it. The target third-party location identifier is used to locate the location, and the target actual data value is used for subsequent anomaly detection.

[0033] S120. Based on the location binding relationship table pre-stored in the digital twin platform, determine the target virtual object instance and target virtual object attribute type that match the target third-party location identifier; wherein, the location binding relationship table records the association relationship between the third-party location identifier, the virtual object instance name, the virtual object attribute type, and the original description information.

[0034] The digital twin platform is a platform for constructing a digital mapping of the physical world. It includes virtual object instances, twin data, and various functional modules for simulating, monitoring, and managing physical entities. The point-to-point binding relationship table is a pre-stored relational database table within the digital twin platform, used to record the mapping relationship between third-party points and virtual objects within the platform; it is the core carrier of the binding relationship. A virtual object instance refers to a digital replica of a physical entity within the digital twin platform, such as the virtual mapping of water pump A in workshop 3, possessing attributes and states corresponding to the physical entity. Virtual object attribute types refer to the specific attribute categories contained in the virtual object instance, such as pressure, temperature, and rotational speed, used to describe the characteristics or state of the virtual object. Original description information refers to the inherent descriptive information of the third-party point, such as the outlet pressure sensor of the water pump in workshop 3, used to describe the physical location and monitoring object of the point, typically entered when the point is connected. The point-to-point binding relationship table establishes the association between third-party points and virtual objects on the digital twin platform. Based on the target third-party point identifier, a matching record is queried in the point-to-point binding relationship table to obtain the corresponding target virtual object instance and target virtual object attribute type. For example, the query shows that “Sensor_302” is bound to the “outlet pressure” attribute of “Workshop 3 Water Pump A”, and the original description information is “Workshop 3 Water Pump Outlet Pressure Sensor”.

[0035] S130. Search for the target attribute value range corresponding to the target virtual object attribute type through the industry knowledge base. If the actual data value of the target does not fall into the target attribute value range, mark the target third-party point as a binding abnormal point.

[0036] The industry knowledge base is a database storing standard data ranges and association rules for various virtual object attributes within a specific industry (e.g., "the normal range for pump outlet pressure is 100-200 kPa"). The target attribute value range is the standard data range in the industry knowledge base corresponding to the target virtual object attribute type, used to determine the reasonableness of actual data (e.g., the range for outlet pressure is [100, 200] kPa). Abnormal binding points refer to points where the binding relationship between a third-party point and a virtual object attribute is incorrect. First, the target attribute value range corresponding to the target virtual object attribute type is retrieved from the industry knowledge base. Then, the actual target data value is compared with this range. If the actual data value exceeds the range, the current binding relationship is determined to be incorrect, and the point is marked as an abnormal binding point.

[0037] Optionally, the target attribute value range corresponding to the target virtual object's attribute type can be found through an industry knowledge base. If the actual target data value does not fall within the target attribute value range, the target third-party point is marked as an abnormal binding point, which may include:

[0038] Retrieve the attribute value range associated with the attribute type of the target virtual object from the industry knowledge base, and determine the attribute value range as the target attribute value range;

[0039] Extract the actual data value of the target from the target location data packet, and compare the actual data value of the target with the target attribute value range;

[0040] If the actual target data value is less than the minimum value of the target attribute value range or greater than the maximum value of the target attribute value range, it is determined that the actual target data value does not fall within the target attribute value range, and the target third-party point corresponding to the actual target data value is marked as a binding abnormal point.

[0041] Associative storage refers to a method in which virtual object attribute types and attribute value ranges in the industry knowledge base are stored in a one-to-one correspondence according to preset rules. For example, the temperature attribute corresponds to [-50, 150℃], and the voltage attribute corresponds to [200, 240V], ensuring that each attribute type has a clear standard data range. Using the target virtual object attribute type as a search keyword, the system queries the industry knowledge base for the associated attribute value range and directly determines this range as the target attribute value range for comparison. For example, if the target virtual object attribute type is motor speed, the standard range [1000, 3000 r / min] corresponding to motor speed is retrieved from the industry knowledge base as the target attribute value range. First, the actual target data value is extracted from the target location data package. Then, this value is compared with the target attribute value range, specifically determining whether the actual data value is greater than the minimum value and less than the maximum value of the range. For example, comparing 280 kPa with [100, 200 kPa] shows that the actual data exceeds the upper limit of the range.

[0042] S140. Filter all candidate attribute value ranges that the actual data value of the target can fall into in the industry knowledge base, and determine all candidate virtual object attribute types corresponding to the actual value of the target based on the attribute value ranges corresponding to each virtual object attribute type pre-stored in the industry knowledge base.

[0043] The candidate attribute value range is any range of attribute values ​​in the industry knowledge base that can contain the actual data value of the target (for example, the actual value of the target 280 kPa can fall into the [200, 300] kPa range of "pipeline pressure" and the [250, 350] kPa range of "fan pressure," etc.). The candidate virtual object attribute type refers to the virtual object attribute type corresponding to the candidate attribute value range. By traversing the attribute value ranges of all virtual object attribute types in the industry knowledge base, candidate attribute value ranges that can contain the actual data value of the target are filtered out. Then, based on the correspondence between "attribute type - range," the candidate virtual object attribute type is obtained. For example, after filtering, the candidate attributes corresponding to 280 kPa are determined to be "pipeline pressure" and "fan pressure."

[0044] Furthermore, the industry knowledge base is used to filter all candidate attribute value ranges into which the target actual data value can fall. Based on the attribute value ranges corresponding to each virtual object attribute type pre-stored in the industry knowledge base, all candidate virtual object attribute types corresponding to the target actual value are determined, which may include:

[0045] Retrieve the attribute range association list from the industry knowledge base, which describes the correspondence between the attribute types and attribute value ranges of virtual objects;

[0046] Traverse the attribute range association list and filter out all target attribute value ranges that the actual target data value can fall into, as candidate attribute value ranges;

[0047] Based on the attribute range association list, determine the attribute type of each candidate virtual object corresponding to each candidate attribute value range, and obtain all candidate virtual object attribute types.

[0048] The attribute interval association list is a structured data table stored in the industry knowledge base. It records the one-to-one correspondence between virtual object attribute types and attribute value ranges in the form of entries, serving as the core carrier for quickly retrieving the correspondence between attributes and ranges. The digital twin platform retrieves a pre-organized attribute interval association list from the industry knowledge base. This list covers all virtual object attribute types and their corresponding standard value ranges within the current industry, providing a complete comparison basis for subsequent filtering. For example, the retrieved list contains attribute types such as "temperature," "pressure," and "speed," along with their corresponding range data. The attribute interval association list is traversed sequentially. For each entry's attribute value range, it is determined whether the target actual data value falls within that range. If so, the range is included in the candidate range as a candidate attribute value range. Based on the correspondence recorded in the attribute interval association list, for each selected candidate attribute value range, its associated virtual object attribute type is queried in reverse. After summarizing these attribute types, all candidate virtual object attribute types matching the target actual data value are obtained.

[0049] S150. Combining the original description information and the attributes of each candidate virtual object, construct a correction reference factor corresponding to the binding abnormal point. Input the correction reference factor into the pre-trained point recognition model to identify the original description information, and correct the abnormal binding relationship of the point according to the recognition result.

[0050] The correction reference factor is structured data that integrates the original description information and the attributes of candidate virtual objects, used to assist the model in accurately identifying the correct binding relationship. The pre-trained point recognition model is an artificial intelligence model trained based on industry knowledge base and historical binding data, capable of analyzing the correlation between the original description information and candidate attributes. First, the original description information and candidate virtual object attributes are integrated into the correction reference factor; then, the correction reference factor is input into the point recognition model, and the model identifies the most matching virtual object attribute type and the corresponding virtual object instance through semantic analysis; finally, the point binding relationship table is updated according to the model output results to complete the automatic correction of abnormal bindings, such as rebinding "Sensor_302" to the "pipeline pressure" attribute of "main pipeline B of workshop 3"). In a specific example, the correction reference factor can be: {Original description: water pump outlet pressure sensor of workshop 3, candidate attributes: [pipeline pressure, fan pressure]}). After semantic analysis, the model determines that the correlation between the water pump outlet and the pipeline is higher than that between the fan and the pipeline, so the most matching virtual object attribute type is pipeline pressure, and the corresponding virtual object instance is main pipeline B of workshop 3.

[0051] This invention eliminates the need for manual identification of location lists or manual mapping. It automates the identification and correction of location binding anomalies, significantly reducing binding processing time and lowering manual operation costs. By relying on standard attribute value ranges from an industry knowledge base to identify anomalies and combining this with a pre-trained model to accurately match correct binding relationships, it avoids misbinding issues caused by personnel fatigue or insufficient industry knowledge, improving the stability and accuracy of binding results. Furthermore, it detects binding anomalies during the data collection phase, rather than passively identifying them during data application, avoiding the accumulation of large amounts of historical junk data and directly improving the overall data quality of the digital twin platform.

[0052] Example 2

[0053] Figure 2 This is a flowchart of another automatic correction method for point binding anomalies provided in Embodiment 2 of the present invention. This embodiment is a refinement based on Embodiment 1, specifically as follows: Figure 2 As shown, the method includes:

[0054] S210. Receive target location data packets from a third-party IoT subsystem, wherein the target location data packets carry at least the target third-party location identifier and the target actual data value.

[0055] S220. Based on the location binding relationship table pre-stored in the digital twin platform, determine the target virtual object instance and target virtual object attribute type that match the target third-party location identifier; wherein, the location binding relationship table records the association relationship between the third-party location identifier, the virtual object instance name, the virtual object attribute type, and the original description information.

[0056] Optionally, before determining the target virtual object instance and target virtual object attribute type that match the target third-party location identifier based on the location binding relationship table pre-stored in the digital twin platform, the following may also be included:

[0057] Receive a list of third-party locations provided by a third-party IoT subsystem; wherein the list of third-party locations includes the third-party location identifier and original description information corresponding to each third-party location;

[0058] The list of third-party locations is input into a location recognition model pre-trained based on an industry knowledge base. By reading the original description information corresponding to each third-party location, the object class and information point class of each third-party location are determined.

[0059] Based on the object class to which the current third-party location belongs, a set of candidate virtual object instances is selected from the twin data of the digital twin platform, and the original description information is matched with the set of candidate virtual object instances to obtain a virtual object instance that matches the current third-party location.

[0060] Based on the virtual object instance, search all candidate virtual object attribute types contained in the virtual object instance in the digital twin platform, and determine the virtual object attribute type from multiple candidate virtual object attribute types based on the information point class to which the current third-party location belongs;

[0061] Establish the association between the current third-party location identifier, virtual object instance, virtual object attribute type and original description information, and generate a location binding relationship table.

[0062] The third-party location list refers to a list provided by a third-party IoT subsystem that records basic information about all locations to be connected. It serves as the foundational data source for location access to the digital twin platform. The digital twin platform receives the third-party location list from the third-party IoT subsystem via a data interface. Each record in the list contains two core pieces of information: the third-party location identifier and the corresponding original description information. The object class refers to the category of the physical entity monitored by the third-party location, used to define the scope of the monitoring object. The information point class refers to the category of the data type collected by the third-party location, used to define the scope of the monitoring attributes of the location. The location recognition model, pre-trained based on an industry knowledge base, is a semantic recognition model trained using labeled data of "original description-object class-information point class" in the industry knowledge base, possessing the ability to extract key categories from descriptive text.

[0063] The third-party list of locations is input into a pre-trained location recognition model. The model reads the original description information of each location and extracts key information through semantic analysis. For example, it identifies "water pump" as the object class to which the location belongs and "pressure" as the information point class. Finally, it outputs the corresponding object class and information point class for each location, providing a basis for subsequent matching of virtual object instances and attribute types. Twin data refers to the metadata of all virtual object instances stored in the digital twin platform; it is a digital archive of virtual object instances. The candidate virtual object instance set refers to all virtual object instances in the digital twin platform that have the same object class as the current location. Similarity matching refers to comparing the degree of matching between the original description information and the metadata of the virtual object instance using a text similarity algorithm; the higher the score, the stronger the association.

[0064] The information determined in the above steps (third-party location identifier, matched virtual object instance, matched virtual object attribute type, and original description information) is stored in a data table in the form of association relationships, forming a location binding relationship table. For example, the association relationship established for "Sensor_302" is: {"Third-party location identifier": "Sensor_302", "Virtual object instance": "Workshop 3 water pump A", "Virtual object attribute type": "Outlet pressure", "Original description information": "Workshop 3 water pump outlet pressure sensor"}.

[0065] S230. Search for the target attribute value range corresponding to the target virtual object attribute type through the industry knowledge base. If the actual data value of the target does not fall into the target attribute value range, mark the target third-party point as a binding abnormal point.

[0066] S240. Filter all candidate attribute value ranges that the actual data value of the target can fall into in the industry knowledge base, and determine all candidate virtual object attribute types corresponding to the actual value of the target based on the attribute value ranges corresponding to each virtual object attribute type pre-stored in the industry knowledge base.

[0067] S250. Based on the target third-party location identifier, query the corresponding original description information from the location binding relationship table.

[0068] The original description information refers to the inherent information describing the physical characteristics of a location, stored in the location binding relationship table and associated with the target third-party location identifier. This information does not change with binding anomalies and serves as the location's identity label. Since the location binding relationship table pre-stores the association between the third-party location identifier and the original description information, the corresponding original description information can be directly queried and extracted from the table using the target third-party location identifier as the search keyword. This ensures that there is clear basic location information when constructing and correcting reference factors later.

[0069] S260. Combine the original description information with the attribute types of all candidate virtual objects to generate corrected description information, which serves as a correction reference factor.

[0070] The corrected description information is structured information formed by combining the original description information with the attribute types of all candidate virtual objects, and it serves as the concrete carrier of the corrected reference factors. The corrected reference factors are the core data used as input to the pre-trained point recognition model, and are composed of both the original description information and the attributes of the candidate virtual objects. By associating and combining the original description information with the attribute types of all candidate virtual objects, the corrected description information is formed. This corrected description information directly serves as the corrected reference factors, ensuring that the factors simultaneously encompass the inherent description of the point and the attribute range suitable for data adaptation, providing a complete basis for model recognition.

[0071] S270. The correction reference factor is analyzed through the point identification model, and a new object class and a new information point class corresponding to the target third-party point identifier are output.

[0072] The location identification model is a pre-trained model used to identify the association categories of locations. The new object class is the object category corresponding to the target third-party location identifier, redefined after the location identification model analyzes and corrects the reference factors. The new information point class is the information point category corresponding to the target third-party location identifier, redefined after the location identification model analyzes and corrects the reference factors. The corrected reference factors are input into the pre-trained location identification model, which comprehensively analyzes the original descriptive information in the corrected reference factors and the attribute types of all candidate virtual objects. After analysis, it outputs the new object class and new information point class directly corresponding to the target third-party location identifier, providing clear identification results for subsequent correction of abnormal location binding relationships.

[0073] Following the example above, the target third-party location is identified as "Sensor_302". The original description information retrieved from the location binding relationship table is "Workshop 3 water pump outlet pressure sensor, and the attribute types of all selected candidate virtual objects are 'pipeline pressure' and 'fan pressure'". Correcting the description information involves directly combining the original description information with the attribute types of all candidate virtual objects to form text, for example: Workshop 3 water pump outlet pressure sensor, candidate virtual object attribute types: pipeline pressure, fan pressure.

[0074] S280. Based on the new object class, a new set of candidate virtual object instances is selected from the twin data of the digital twin platform, and the original description information is matched with the new set of candidate virtual object instances to obtain a new virtual object instance that matches the target third-party location.

[0075] Using the output new object class as the filtering condition, all virtual object instances belonging to that category are retrieved from the twin data of the digital twin platform to form a new set of candidate virtual object instances. Then, the original description information is compared with the metadata of each instance in the set to determine the similarity, and the instance with the highest score is selected as the new virtual object instance that matches the target third-party location.

[0076] S290. Based on the new virtual object instance, search for all new candidate virtual object attribute types contained in the new virtual object instance in the digital twin platform, and determine the new virtual object attribute type from the multiple new candidate virtual object attribute types based on the new information point class.

[0077] First, query all attribute types contained in the new virtual object instance in the digital twin platform to form a new set of candidate virtual object attribute types; then, based on the new information point class, select the corresponding attribute type from this set as the new virtual object attribute type.

[0078] S2100: Establish a new association between the target third-party location identifier, the new virtual object instance, the new virtual object attribute type, and the original description information, and modify the location binding relationship table based on the new association.

[0079] A new association is established between the target third-party location identifier, the new virtual object instance, the new virtual object attribute type, and the original description information. This new association replaces the original erroneous association record in the location binding relationship table, completing the automatic correction of the entire location binding anomaly. For example, the original erroneous association was "Sensor_302→Workshop 3 Water Pump A→Outlet Pressure", which is corrected to "Sensor_302→Workshop 3 Main Pipeline A→Pipeline Pressure", ensuring that subsequent data collection can accurately associate with the correct virtual object and attributes.

[0080] Furthermore, it may also include: if, based on the new object class, a new virtual object instance that meets the conditions cannot be obtained in the digital twin platform through similarity matching, an abnormal work order is generated that includes the target third-party location identifier, original description information, new object class, new information point class, and the reason for the matching failure, and the abnormal work order is output to the manual processing interface so as to manually correct the association relationship of the binding abnormal location in the binding relationship table.

[0081] This embodiment is a prerequisite for triggering abnormal work orders. After filtering candidate virtual object instances based on the new object class and completing the similarity matching between the original description information and the instances, the matching results are checked: if no instance with the required similarity is found, the automated matching is deemed to have failed, and the abnormal work order processing flow is initiated. An abnormal work order refers to a structured document that records key information about the bound abnormal location and the processing requirements. It is the core carrier for manual intervention and must contain sufficient information to support manual judgment. The manual processing interface refers to the functional module in the digital twin platform that allows staff to view and process abnormal work orders, supporting work order status tracking, information editing, and result submission. After viewing the work order through the manual processing interface, staff can make corrections based on their industry knowledge and the actual situation of the platform: if the platform lacks a corresponding virtual object instance, a new virtual object instance can be created in the digital twin platform first, then the association between the target third-party location identifier and the "new virtual object instance + new virtual object attribute type" can be manually established, and finally the correction result can be submitted through the interface. The platform automatically updates the location binding relationship table to complete the correction of the abnormal binding.

[0082] This embodiment implements a "automation first, manual backup" processing logic, which not only ensures efficient automated processing in most scenarios, but also covers special cases that automation cannot solve, ensuring that binding anomalies are fully corrected.

[0083] Example 3

[0084] Figure 3 This is a schematic diagram of an automatic correction device for point binding anomalies provided in Embodiment 3 of the present invention. Figure 3 As shown, the device includes:

[0085] The point data packet receiving module 310 is used to receive target point data packets from a third-party Internet of Things subsystem, wherein the target point data packets carry at least the target third-party point identifier and the target actual data value;

[0086] The point binding relationship acquisition module 320 is used to determine the target virtual object instance and the target virtual object attribute type that match the target third-party point identifier based on the point binding relationship table pre-stored in the digital twin platform; wherein, the point binding relationship table records the association relationship between the third-party point identifier, the virtual object instance name, the virtual object attribute type and the original description information;

[0087] The point binding anomaly determination module 330 is used to search for the target attribute value range corresponding to the target virtual object attribute type through the industry knowledge base. If the actual data value of the target does not fall into the target attribute value range, the target third-party point is marked as a binding anomaly point.

[0088] The virtual object attribute type filtering module 340 is used to filter all candidate attribute value ranges that the actual data value of the target can fall into in the industry knowledge base, and determine all candidate virtual object attribute types corresponding to the actual value of the target based on the attribute value ranges corresponding to each virtual object attribute type pre-stored in the industry knowledge base.

[0089] The point correction module 350 is used to combine the original description information and the attributes of each candidate virtual object to construct a correction reference factor corresponding to the binding abnormal point, input the correction reference factor into the pre-trained point recognition model to identify the original description information, and correct the abnormal binding relationship of the point based on the recognition result.

[0090] This invention eliminates the need for manual identification of location lists or manual mapping. It automates the identification and correction of location binding anomalies, significantly reducing binding processing time and lowering manual operation costs. By relying on standard attribute value ranges from an industry knowledge base to identify anomalies and combining this with a pre-trained model to accurately match correct binding relationships, it avoids misbinding issues caused by personnel fatigue or insufficient industry knowledge, improving the stability and accuracy of binding results. Furthermore, it detects binding anomalies during the data collection phase, rather than passively identifying them during data application, avoiding the accumulation of large amounts of historical junk data and directly improving the overall data quality of the digital twin platform.

[0091] Optionally, based on the above embodiments, it may further include: a point binding relationship table construction unit, used to receive a third-party point list provided by a third-party IoT subsystem before determining the target virtual object instance and target virtual object attribute type that match the target third-party point identifier based on the point binding relationship table pre-stored in the digital twin platform; wherein, the third-party point list includes the third-party point identifier and original description information corresponding to each third-party point;

[0092] The list of third-party locations is input into a location recognition model pre-trained based on an industry knowledge base. By reading the original description information corresponding to each third-party location, the object class and information point class of each third-party location are determined.

[0093] Based on the object class to which the current third-party location belongs, a set of candidate virtual object instances is selected from the twin data of the digital twin platform, and the original description information is matched with the set of candidate virtual object instances to obtain a virtual object instance that matches the current third-party location.

[0094] Based on the virtual object instance, search all candidate virtual object attribute types contained in the virtual object instance in the digital twin platform, and determine the virtual object attribute type from multiple candidate virtual object attribute types based on the information point class to which the current third-party location belongs;

[0095] Establish the association between the current third-party location identifier, virtual object instance, virtual object attribute type and original description information, and generate a location binding relationship table.

[0096] Optionally, based on the above embodiments, the point binding anomaly determination module 330 may include:

[0097] The attribute value range determination unit is used to retrieve the attribute value range stored in the industry knowledge base that is associated with the attribute type of the target virtual object, and determine the attribute value range as the target attribute value range;

[0098] The data value extraction unit is used to extract the actual data value of the target from the target location data packet and compare the actual data value of the target with the target attribute value range.

[0099] The interval value comparison unit is used to determine that the actual target data value does not fall within the target attribute value interval if the actual target data value is less than the minimum value of the target attribute value interval or greater than the maximum value of the target attribute value interval, and to mark the target third-party point corresponding to the actual target data value as a binding abnormal point.

[0100] Optionally, based on the above embodiments, the virtual object attribute type filtering module 340 may include:

[0101] The attribute range association list retrieval unit is used to retrieve the attribute range association list describing the correspondence between the attribute types and attribute value ranges of virtual objects from the industry knowledge base;

[0102] The candidate attribute value range filtering unit is used to traverse the attribute range association list and filter out all target attribute value ranges that the actual target data value can fall into, as candidate attribute value ranges.

[0103] The candidate virtual object attribute type matching unit is used to determine the attribute type of each candidate virtual object corresponding to each candidate attribute value range based on the attribute range association list, so as to obtain all candidate virtual object attribute types.

[0104] Optionally, based on the above embodiments, the position correction module 350 may include:

[0105] Based on the target third-party location identifier, query the corresponding original description information from the location binding relationship table;

[0106] The original description information is combined with the attribute types of all candidate virtual objects to generate corrected description information, which serves as a correction reference factor.

[0107] The correction reference factor is analyzed by the point identification model, and a new object class and a new information point class corresponding to the target third-party point identifier are output.

[0108] Optionally, based on the above embodiments, the position correction module 350 may further include:

[0109] Based on the new object class, a new set of candidate virtual object instances is selected from the twin data of the digital twin platform, and the original description information is matched with the new set of candidate virtual object instances to obtain a new virtual object instance that matches the target third-party location.

[0110] Based on the new virtual object instance, search all new candidate virtual object attribute types contained in the new virtual object instance in the digital twin platform, and determine the new virtual object attribute type from the multiple new candidate virtual object attribute types based on the new information point class.

[0111] Establish a new association between the target third-party location identifier, the new virtual object instance, the new virtual object attribute type, and the original description information, and revise the location binding relationship table based on the new association.

[0112] Optionally, based on the above embodiments, it may further include: a manual correction reporting unit, used to generate an abnormal work order containing the target third-party location identifier, original description information, new object class, new information point class and the reason for matching failure if, based on the new object class, a new virtual object instance that meets the conditions cannot be obtained in the digital twin platform through similarity matching, and output the abnormal work order to the manual processing interface so as to correct the association relationship of the binding abnormal location in the binding relationship table manually.

[0113] The automatic correction device for point binding anomalies provided in this embodiment of the invention can execute the automatic correction method for point binding anomalies provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0114] Example 4

[0115] Figure 4 A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0116] like Figure 4 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0117] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0118] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as an automatic correction method for bit-binding anomalies.

[0119] That is: receiving target location data packets from a third-party IoT subsystem, wherein the target location data packets carry at least the target third-party location identifier and the target actual data value;

[0120] Based on the location binding relationship table pre-stored in the digital twin platform, the target virtual object instance and target virtual object attribute type that match the target third-party location identifier are determined; wherein, the location binding relationship table records the association relationship between the third-party location identifier, the virtual object instance name, the virtual object attribute type, and the original description information;

[0121] The target attribute value range corresponding to the target virtual object attribute type is found by searching the industry knowledge base. If the actual data value of the target does not fall into the target attribute value range, the target third-party point is marked as a binding abnormal point.

[0122] In the industry knowledge base, filter all candidate attribute value ranges that the actual data value of the target can fall into. Based on the attribute value ranges corresponding to each virtual object attribute type pre-stored in the industry knowledge base, determine all candidate virtual object attribute types corresponding to the actual value of the target.

[0123] By combining the original description information and the attributes of each candidate virtual object, a correction reference factor corresponding to the binding abnormal point is constructed. The correction reference factor is input into the pre-trained point recognition model to identify the original description information, and the abnormal binding relationship of the point is corrected according to the recognition result.

[0124] In some embodiments, an automatic correction method for bit-binding anomalies may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the automatic correction method for bit-binding anomalies described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform an automatic correction method for bit-binding anomalies by any other suitable means (e.g., by means of firmware).

[0125] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0126] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0127] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0128] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0129] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0130] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0131] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0132] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. An automatic correction method for point binding anomalies, characterized in that, The method includes: Receive a target location data packet from a third-party IoT subsystem, wherein the target location data packet carries at least the target third-party location identifier and the target actual data value; Based on the location binding relationship table pre-stored in the digital twin platform, the target virtual object instance and target virtual object attribute type that match the target third-party location identifier are determined; wherein, the location binding relationship table records the association relationship between the third-party location identifier, the virtual object instance name, the virtual object attribute type, and the original description information; The target attribute value range corresponding to the target virtual object attribute type is found by searching the industry knowledge base. If the actual data value of the target does not fall into the target attribute value range, the target third-party point is marked as a binding abnormal point. Filter all candidate attribute value ranges that the actual data value of the target can fall into in the industry knowledge base, and determine all candidate virtual object attribute types corresponding to the actual data value of the target based on the attribute value ranges corresponding to each virtual object attribute type pre-stored in the industry knowledge base. Combining the original description information and the attribute types of each candidate virtual object, a correction reference factor corresponding to the binding abnormal point is constructed. The correction reference factor is input into the pre-trained point recognition model to identify the original description information, and the abnormal binding relationship of the point is corrected according to the recognition result. Combining the original description information and the attribute types of each candidate virtual object, a correction reference factor corresponding to the bound abnormal point is constructed. This correction reference factor is then input into a pre-trained point recognition model to identify the original description information, including: Based on the target third-party location identifier, query the corresponding original description information from the location binding relationship table; The original description information is combined with the attribute types of all candidate virtual objects to generate corrected description information, which serves as a correction reference factor. The correction reference factor is analyzed by the point identification model, and a new object class and a new information point class corresponding to the target third-party point identifier are output.

2. The method according to claim 1, characterized in that, Before determining the target virtual object instance and target virtual object attribute type that match the target third-party location identifier based on the pre-stored location binding relationship table on the digital twin platform, the following steps are also included: Receive a list of third-party locations provided by a third-party IoT subsystem; wherein the list of third-party locations includes the third-party location identifier and original description information corresponding to each third-party location; The list of third-party locations is input into a location recognition model pre-trained based on an industry knowledge base. By reading the original description information corresponding to each third-party location, the object class and information point class of each third-party location are determined. Based on the object class to which the current third-party location belongs, a set of candidate virtual object instances is selected from the twin data of the digital twin platform, and the original description information is matched with the set of candidate virtual object instances to obtain a virtual object instance that matches the current third-party location. Based on the virtual object instance, search all candidate virtual object attribute types contained in the virtual object instance in the digital twin platform, and determine the virtual object attribute type from multiple candidate virtual object attribute types based on the information point class to which the current third-party location belongs; Establish the association between the current third-party location identifier, virtual object instance, virtual object attribute type and original description information, and generate a location binding relationship table.

3. The method according to claim 1, characterized in that, The system searches the industry knowledge base for the target attribute value range corresponding to the target virtual object's attribute type. If the actual target data value does not fall within the target attribute value range, the target third-party point is marked as an abnormal binding point, including: Retrieve the attribute value range associated with the attribute type of the target virtual object from the industry knowledge base, and determine the attribute value range as the target attribute value range; Extract the actual data value of the target from the target location data packet, and compare the actual data value of the target with the target attribute value range; If the actual target data value is less than the minimum value of the target attribute value range or greater than the maximum value of the target attribute value range, it is determined that the actual target data value does not fall within the target attribute value range, and the target third-party point corresponding to the actual target data value is marked as a binding abnormal point.

4. The method according to claim 1, characterized in that, The industry knowledge base is used to filter all candidate attribute value ranges into which the target actual data value can fall. Based on the attribute value ranges corresponding to each virtual object attribute type pre-stored in the industry knowledge base, all candidate virtual object attribute types corresponding to the target actual data value are determined, including: Retrieve the attribute range association list from the industry knowledge base, which describes the correspondence between the attribute types and attribute value ranges of virtual objects; Traverse the attribute range association list and filter out all target attribute value ranges that the actual target data value can fall into, as candidate attribute value ranges; Based on the attribute range association list, determine the attribute type of each candidate virtual object corresponding to each candidate attribute value range, and obtain all candidate virtual object attribute types.

5. The method according to claim 1, characterized in that, Based on the identification results, the abnormal binding relationships of the points are corrected, including: Based on the new object class, a new set of candidate virtual object instances is selected from the twin data of the digital twin platform, and the original description information is matched with the new set of candidate virtual object instances to obtain a new virtual object instance that matches the target third-party location. Based on the new virtual object instance, search all new candidate virtual object attribute types contained in the new virtual object instance in the digital twin platform, and determine the new virtual object attribute type from multiple new candidate virtual object attribute types based on the new information point class. Establish a new association between the target third-party location identifier, the new virtual object instance, the new virtual object attribute type, and the original description information, and revise the location binding relationship table based on the new association.

6. The method according to claim 1, characterized in that, The method further includes: If, based on the new object class, a new virtual object instance that meets the conditions cannot be obtained in the digital twin platform through similarity matching, an abnormal work order is generated, which includes the target third-party location identifier, original description information, new object class, new information point class, and the reason for the matching failure. The abnormal work order is then output to the manual processing interface so that the association relationship of the abnormal location in the binding relationship table can be corrected manually.

7. An automatic correction device for point binding anomalies, characterized in that, The device includes: The point data packet receiving module is used to receive target point data packets from a third-party IoT subsystem, wherein the target point data packets carry at least the target third-party point identifier and the target actual data value; The point binding relationship acquisition module is used to determine the target virtual object instance and the target virtual object attribute type that match the target third-party point identifier based on the point binding relationship table pre-stored in the digital twin platform; wherein, the point binding relationship table records the association relationship between the third-party point identifier, the virtual object instance name, the virtual object attribute type, and the original description information; The point binding anomaly determination module is used to search for the target attribute value range corresponding to the target virtual object attribute type through the industry knowledge base. If the actual data value of the target does not fall into the target attribute value range, the target third-party point is marked as a binding anomaly point. The virtual object attribute type filtering module is used to filter all candidate attribute value ranges that the actual data value of the target can fall into in the industry knowledge base. Based on the attribute value ranges corresponding to each virtual object attribute type pre-stored in the industry knowledge base, it determines all candidate virtual object attribute types corresponding to the actual data value of the target. The point correction module is used to combine the original description information and the attribute types of each candidate virtual object to construct a correction reference factor corresponding to the binding abnormal point. The correction reference factor is input into the pre-trained point recognition model to identify the original description information, and the abnormal binding relationship of the point is corrected according to the recognition result. The location correction module includes: querying the corresponding original description information from the location binding relationship table based on the target third-party location identifier; The original description information is combined with the attribute types of all candidate virtual objects to generate corrected description information, which serves as a correction reference factor. The correction reference factor is analyzed by the point identification model, and a new object class and a new information point class corresponding to the target third-party point identifier are output.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores a computer program executable by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform an automatic correction method for point binding anomalies according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement an automatic correction method for point binding anomalies according to any one of claims 1-6.

Images