A data checking method for building digital twinning

Abstract

The application relates to the technical field of digital twinning, in particular to a data checking method for building digital twinning. A person triggers a data checking module of building digital twinning, the data checking module calls data from a building digital twinning database, simultaneously calls a rule set matched with the called data, checks the called data through the rule set, a computer statistically processes checking results, and calculates a correct rate. The data checking module checks batch data of building digital twinning, simultaneously calculates the correct rate to judge whether the data quality meets the building digital twinning standard. Based on the general understanding of each profession in the building, batch data generated by building digital twinning is checked, the problem that manual checking cannot be performed due to too much data is solved, and through the calculation of the correct rate, the problem that how to judge whether the twinning data meets the building digital twinning standard is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of digital twinning, in particular to a data checking method for building digital twinning. BACKGROUND

[0002] Building digital twinning is a technology based on computer technology for digitally describing and twinning objects such as spaces, equipment, and pipelines in buildings. It mainly takes spaces, equipment, pipelines, and systems as the core and twines the above several types of objects with information points and relationships. In such a system, due to the large number of objects and complex relationships between objects, the number of information points is extremely large, and it is difficult for humans to check and verify the data of digital twinning, and it is difficult to determine whether the data of digital twinning is accurate and meets the requirements. SUMMARY

[0003] (I) Technical problems to be solved

[0004] The present application solves the technical problem that too much data makes it difficult for humans to check, and computers cannot directly determine whether the twinning data meets the building digital twinning standard.

[0005] (II) Technical solutions

[0006] To solve the above technical problems, the present application provides a data checking method for building digital twinning, characterized in that it comprises the following steps: S1, triggering a human to start a data verification module of building digital twinning; S2, the data verification module calls data from a building digital twinning database and simultaneously calls a rule set matched with the called data; S3, the data verification module performs data verification on the called data through the rule set; S4, the verification result is output and displayed through a display; S5, the computer performs statistics on the verification result and calculates the accuracy rate; S6, the accuracy rate result is output and displayed through the display.

[0007] Further, the rule set is an inspection rule that cannot be understood by a computer converted into an inspection rule that can be understood by a computer through problem analysis.

[0008] Further, the rule set includes a condition judgment of "whether human secondary judgment is needed", and the rule instance with a value of "no" does not need human judgment, and the rule instance with a value of "yes" in "whether human secondary judgment is needed" needs to be fed back to human for judgment.

[0009] Further, the accuracy rate calculation only calculates the rule instance with a value of "no" in "whether human secondary judgment is needed", and does not calculate the rule instance with a value of "yes" in "whether human secondary judgment is needed".

[0010] Further, when the value in the "whether human secondary judgment is needed" is "no", the rule instance outputs the error checked by the verification result, and if the human cannot modify, the "must" give an explanation.

[0011] Further, when the value in the "whether human secondary judgment is needed" is "yes", the rule instance outputs the error checked by the verification result, and if the human cannot modify, the "try" give an explanation.

[0012] Further, the correctness rate calculation includes global correctness rate and local correctness rate, the global correctness rate = (total number of data verification instances - number of data verification instances of errors without filling in explanations) / total number of data verification instances, and the local correctness rate = (number of data verification instances of errors with filling in explanations) / total number of data verification instances, wherein the total number of data verification instances is the total number of times of checking data according to the configured rule class instance in each node of data.

[0013] Further, the data includes objects, relationships and information points.

[0014] (Three) beneficial effects

[0015] The above technical scheme of the present application has the following advantages: the data verification module checks the building digital twin data in batches, and calculates the correctness rate to determine whether the data quality meets the building digital twin standard. Based on the general understanding of each professional in the building, a large amount of data generated by the building digital twin is checked in batches, which solves the problem that manual checking is impossible due to excessive data volume, and through the calculation of the correctness rate, the problem of how to determine whether the twin data meets the building digital twin standard is solved. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 The building digital twin system data structure is provided.

[0017] Figure 2 The flowchart of the data checking method for the building digital twin is provided. DETAILED DESCRIPTION

[0018] The specific embodiments of the present application will be further described in detail below in combination with the drawings and examples. The following examples are used to illustrate the present application, but not to limit the scope of the present application.

[0019] Please refer to Figure 1 and Figure 2 The present application provides a data checking method for building digital twin, characterized in that it comprises the following steps:

[0020] S1, the human starts the data verification module of the building digital twin.

[0021] S2, the data verification module retrieves data from the building digital twin database and simultaneously calls the rule set that matches the retrieved data.

[0022] S3, the data verification module verifies the retrieved data using a set of rules.

[0023] S4, the verification result is displayed on the monitor.

[0024] S5, the computer performs statistical analysis on the verification results and calculates the accuracy rate.

[0025] S6, the accuracy results are displayed on the monitor.

[0026] The data verification module performs batch checks on building digital twin data and calculates the accuracy rate to determine whether the data quality meets building digital twin standards. Based on the general understanding of various disciplines within the building industry, this method performs batch checks on the large amounts of data generated by building digital twins, solving the problem of insufficient manual inspection due to excessive data volume. Furthermore, by calculating the accuracy rate, it addresses the question of how to determine whether the twin data conforms to building digital twin standards.

[0027] In some embodiments, the rule set is a set of inspection rules that are incomprehensible to computers, which are then transformed into computer-comprehensible inspection rules through problem parsing.

[0028] In some embodiments, the rule set includes a conditional judgment of "whether a second human judgment is required". Rule instances with a value of "no" do not require further human judgment, while rule instances with a value of "yes" in "whether a second human judgment is required" require feedback for human judgment. Secondary human judgment refers to situations that would not occur under normal engineering knowledge, but may occur if the owner violates general engineering regulations for other reasons. In other words, the computer can make the judgment, and human confirmation is only an aid to avoid misjudgment.

[0029] In some embodiments, the accuracy calculation only counts rule instances where the value of "whether a second human judgment is required" is "no", and does not count rule instances where the value of "whether a second human judgment is required" is "yes".

[0030] In some embodiments, for rule instances where the value of "whether a second human judgment is required" is "no", if the person triggering the error detected by the output verification result cannot modify it, then an explanation "must" be provided.

[0031] In some embodiments, for rule instances where the value of "whether a second human judgment is required" is "yes", the error detected by the output verification result will be explained "as much as possible" if the triggering person cannot modify it.

[0032] In some embodiments, the accuracy calculation includes global accuracy and perimeter accuracy. Global accuracy = (total number of data validation instances - number of data validation instances with errors and no explanation) / total number of data validation instances. Perimeter accuracy = (number of data validation instances with explanation) / total number of data validation instances, where the total number of data validation instances is the total number of times the data is checked according to the configured check rule class instances within each data node.

[0033] In some embodiments, the retrieved data includes objects, relationships, and information points. Objects: refer to equipment, spaces, or main functional areas, etc., that are manually modeled or generated through registration during the building digital twin process. Relationships: refer to the connections between various objects that are manually defined during the building digital twin process, including physical connections and functional associations. Information points: refer to multiple labeled values ​​used in the building digital twin process to describe the specification information of objects, such as information like "rated voltage 10kV".

[0034] I. Classification and Examples of Data Validation Rules

[0035] 1.1 Spatial Class Validation Rules

[0036]

[0037]

[0038] 1.2 Completeness Verification Rules for Object Information Points

[0039]

[0040]

[0041] 1.3 Object Relationship Completeness Verification Rules

[0042]

[0043] 1.4 Object Information Point Reasonableness Verification Rules

[0044]

[0045] 1.5 Object-Relationship Dependency Validation Rules

[0046]

[0047]

[0048] 1.6 Object Relationship Mutual Exclusion Validation Rules

[0049]

[0050] 1.7 Object dependency check rules

[0051]

[0052]

[0053] 1.8 Object mutual exclusion check rules

[0054]

[0055] 1.9 Relationship-equipment-information point check rules

[0056]

[0057] 1.10 Object relationship chain check rules

[0058]

[0059]

[0060] 1.11 Object relationship rationality check rules

[0061]

[0062] 1.12 Object-relationship-number check rules

[0063]

[0064]

[0065] II. Data check results and feedback

[0066] In the process of building each building digital twin, mandatory data checks need to be performed in fixed work nodes. The following describes the check results of such mandatory checks and the feedback of the check results:

[0067] The data check results of the mandatory data check nodes are fed back to the trigger person in the form of check result details. The trigger person should trigger different sets of data check rule instances at each delivery stage. After selecting the check stage, the trigger person should run all the check rule instances at this node every time the data check is triggered, and should check everything that should be checked. The trigger person should modify or explain the test results.

[0068] For all the above types of check rules, the value of "whether human secondary judgment is needed" in each type of check rule table is explained as follows.

[0069] 1、When the value of "whether human secondary judgment is needed" is "no", the trigger person must give an explanation if the error cannot be modified.

[0070] 2、When the value of "whether human secondary judgment is needed" is "yes", the trigger person should give an explanation if the error cannot be modified.

[0071] For all mandatory trigger data verification node check results, the feedback of the verification results should be collected, and whether there is a value in the "result processing" column of the verification result line where the value of "whether human secondary judgment is needed" is "no" and "unmodified, cannot be modified" should be determined.

[0072] Three, data verification accuracy calculation

[0073] When calculating the accuracy of each mandatory verification node feedback, the following rules should be followed.

[0074] 3.1 Calculation range

[0075] All accuracy calculations only calculate the rule instances where the value of "whether human secondary judgment is needed" is "no". The value of "whether human secondary judgment is needed" is "yes" is not calculated.

[0076] All accuracy calculations only calculate the global accuracy and the accuracy of the surrounding, and do not calculate the accuracy of a single rule instance.

[0077] 3.2 Calculation rules

[0078] The global accuracy calculation rule for each node feedback is as follows:

[0079] Global accuracy = (total number of data verification instances - number of data verification instances with errors that have not been explained) / total number of data verification instances

[0080] After each node completes the upload, the accuracy calculation rule is as follows:

[0081] Surrounding accuracy = (number of data verification instances with errors that have been explained) / total number of data verification instances

[0082] Total number of data verification instances:

[0083] The total number of data verification instances refers to the total number of times the data is checked according to the verification rule class instances configured in each node.

[0084] 3.3 Accuracy feedback

[0085] The accuracy is only fed back when the data check of each triggered mandatory node is passed, and the standard for each node is that the global accuracy is higher than 99%.

[0086] The above merely describes the preferred embodiments of the present application, and it should be noted that those skilled in the art can make several improvements and modifications without departing from the technical principles of the present application, and these improvements and modifications should also be considered as the protection scope of the present application.

Claims

1. A data checking method for building-oriented digital twinning, characterized in that, The method comprises the steps of: S1, triggering a person to start a data verification module of a building digital twin; S2, the data verification module calls data through a building digital twin database, and simultaneously calls a rule set matched with the called data; S3, the data verification module performs data verification on the called data through the rule set; S4, a verification result is output and displayed through a display; S5, a computer performs statistics on the verification result and calculates a correctness rate; S6, a correctness rate result is output and displayed through a display; The rule set comprises a condition judgment of "whether human secondary judgment is needed", wherein a rule instance with a value of "no" does not need human judgment, and a rule instance with a value of "yes" needs feedback for human judgment; All correctness rate calculations only calculate rule instances with a value of "no" in "whether human secondary judgment is needed", and do not calculate rule instances with a value of "yes" in "whether human secondary judgment is needed"; When a rule instance with a value of "no" in "whether human secondary judgment is needed", an output verification result checks an error, and if a trigger person cannot modify, an explanation must be given; When a rule instance with a value of "yes" in "whether human secondary judgment is needed", an output verification result checks an error, and if a trigger person cannot modify, an explanation is given as much as possible; All correctness rate calculations only calculate global correctness rate and surrounding correctness rate, and do not calculate single rule instance correctness rate; Global correctness rate calculation rules of feedback of each node are as follows: Global correctness rate=(total number of data verification instances-unfilled explanation error data verification instance number) / total number of data verification instances; After each node is completed, surrounding correctness rate calculation rules are as follows: Surrounding correctness rate=(filled explanation error data verification instance number) / total number of data verification instances; Wherein the total number of data verification instances is the total number of times of checking data according to the checking rule class instance configured in each node of the data; The correctness rate is only fed back when each trigger mandatory node performs data verification, and the standard passed by each node is that the global correctness rate is higher than 99%.

2. The data checking method for building-oriented digital twinning according to claim 1, wherein, The rule set is to convert checking rules that cannot be understood by a computer into checking rules that can be understood by the computer through problem analysis.

3. The data checking method for building digital twin according to any one of claims 1-2, characterized in that, The called data comprises objects, relationships and information points.

Images