Power transmission and transformation equipment manufacturing full-process data checking method and system, terminal and medium
By introducing data processing methods of field ontology models and semantic maps, the problems of data heterogeneity and abnormal identification in power transmission and transformation equipment manufacturing systems are solved, efficient data fusion and abnormal identification are achieved, and the accuracy and traceability of data verification are improved.
Patent Information
- Application Number
- CN202510764296.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-06-10
- Publication Date
- 2025-07-08
- Estimated Expiration
- 2045-06-10
AI Technical Summary
The data sources in the existing power transmission and transformation equipment manufacturing systems are heterogeneous and inconsistent, which leads to data being unable to be directly integrated and applied, and lacks intelligent algorithm support, resulting in misjudgment, misjudgment and information island problems, making it difficult to adapt to the dynamic manufacturing environment, and lacks a unified and trustworthy data archiving mechanism, making it difficult to trace out abnormal events.
Data processing methods based on the domain ontology model and semantic map are adopted, semantic mapping and normalization are performed through field semantic maps, data verification is performed in combination with principal component analysis and rule engines, and abnormal data is stored using blockchain to generate traceability structure and quality reports.
It realizes unified semantic representation and high-precision abnormality recognition of heterogeneous data, reduces the misjudgment rate and missed detection rate, improves data verification efficiency and traceability, and provides credible data storage and decision-making assistance.
Smart Images

Figure CN120277509A_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of data quality management, and particularly relates to a method, system, terminal and medium for checking data in the whole process of transmission and transformation equipment manufacturing. Background Art
[0002] As a typical discrete industrial process with high complexity and high reliability requirements, the data traceability, quality transparency and process controllability of the whole process of transmission and transformation equipment manufacturing are becoming the core technical requirements.
[0003] At present, the industry generally adopts a manufacturing mode of multi-system collaboration, including design systems, production execution systems (MES), quality management systems (QMS), equipment monitoring systems, etc. However, there are problems such as heterogeneous data sources, inconsistent semantics, lack of standards, and redundant records among these systems, resulting in the inability to directly integrate and apply data, especially forming an obvious bottleneck when performing quality checking tasks.
[0004] The existing data checking modes for transmission and transformation equipment manufacturing still mainly rely on manual verification and static rule checking. Although some enterprises have deployed some automatic verification tools, they usually only support field-level legality verification and cannot perceive key quality elements such as semantic associations between fields, process stage dependencies, and historical trend changes, which are prone to misjudgment and missed judgment, and even serious information island problems such as the same batch of data being repeatedly processed and misentered into the system. In addition, the existing systems are difficult to flexibly schedule the checking rules in a dynamic manufacturing environment and do not have the self-learning ability to adapt to complex working conditions changes.
[0005] The discrimination of abnormal data also mostly relies on empirical threshold methods or fixed templates, lacking intelligent algorithm support and being almost powerless for data matrices formed by high dimensions, multiple channels, and multiple process combinations. At the same time, there is a lack of a unified and reliable data archiving mechanism, resulting in the difficulty of tracing abnormal events and the difficulty of solving risk problems such as data fraud and version confusion from the source. Summary of the Invention
[0006] In view of the problems in the prior art, the present invention provides a method, system, terminal and medium for checking data in the whole process of transmission and transformation equipment manufacturing, and solves the problems of risks such as data fraud and version confusion caused by the fact that the discrimination of abnormal data in the prior art mostly relies on empirical threshold methods or fixed templates.
[0007] The technical solution adopted by the present invention is as follows: In a first aspect, the present application provides a method for checking data in the whole process of transmission and transformation equipment manufacturing, and the method includes the following steps: Step S1, obtaining the original business data generated in each link of transmission and transformation equipment manufacturing; Perform operations of unifying data types, initially eliminating abnormal fluctuations, and standardizing the collected data; Step S2: Transmit the preprocessed business data to the cloud for processing. Based on the industry process ontology model, construct a field semantic graph, and perform semantic mapping, field normalization, and attribute unification operations on heterogeneous fields through the field semantic graph to complete semantic fusion, obtaining standardized business data. Schedule the corresponding verification rule set corresponding to the business context from the rule warehouse according to the current manufacturing process status parameters; Step S3: Verify the obtained standardized business data through the corresponding verification rule set; Step S4: Perform abnormal marking processing on the data records with unqualified verification results, upload the field position, rule violation number, timestamp, and device ID carried by them as abnormal meta-information, and generate a prompt signal at the client interface; Step S5: Mark the data records that pass the verification with a "verification passed" label, write them into the blockchain data module, calculate the hash value for each piece of data to generate a digest, and store it on the chain to the distributed node to form a chained traceability structure; Step S6: Generate a quality verification report for this batch of manufacturing data according to the verification result vector and statistical parameters.
[0008] Preferably, in step S3, construct a structured matrix for the standardized business data:
[0009] where represents the value of the j-th field in the i-th business record, and the verification rule set is expressed as:
[0010] where represents the expression function of the rule, is the corresponding threshold, is the logical operation relationship; call the rule engine module to match each item of the data matrix element with the corresponding rule expression one by one to obtain the verification result vector:
[0011] where indicates that the i-th record satisfies all scheduling rules, indicates that at least one rule is not satisfied.
[0012] Preferably, the data preprocessing operation in step S1 includes the following sub-steps: Step S11: Perform field alignment and null value interpolation operations on various types of original data, and use the interpolation strategy based on the K-nearest neighbor method to process missing fields, where the missing field value is calculated as:
[0013] Among them, represents the j-th field value of the k-th similar record; Step S12, perform time series fluctuation smoothing based on the sliding window method, and the window length is , and the weighted moving average method is used for processing, and the output result is:
[0014] where is a preset weight factor, satisfying , and is used to weaken the influence of sampling errors.
[0015] Preferably, the semantic fusion in step S2 includes field normalization processing based on ontology mapping. The semantic ontology is constructed by domain experts and represents entity-attribute-relationship triples using the OWL language. Each field is mapped to the target ontology class through the SPARQL query interface and the following operations are performed: Identify field synonyms and abbreviation mappings; Normalize units and dimension conversions; Merge logically equivalent fields; The above operation results form a field-attribute mapping set ; The rule scheduling method constructs a rule activation graph based on the current business context , where the vertex set represents the rule set, and the edge set represents the dependency or mutual exclusion relationship between rules; By constructing a state variable vector , using the matching function:
[0016] Filter the current applicable rule set, and the activation path is represented by the rule trigger matrix for subsequent rule engine execution scheduling; Among them, is the activated rule set, is the i-th rule, is the rule trigger matrix.
[0017] Preferably, the rule checking engine in step S3 stores all rule expressions in an expression tree structure, where each rule expression is parsed into a boolean tree structure, with leaf nodes being field value or constant nodes and intermediate nodes being logical operators; during execution, a depth-first traversal strategy is adopted to evaluate each branch expression in sequence. If the calculation result of the root node is true, it is considered that the rule is hit; this mechanism supports expression-level dynamic compilation and registers an abstract syntax tree inside the rule engine for runtime invocation.
[0018] Preferably, step S3 further includes performing dimensionality reduction preprocessing on high-dimensional business data based on the principal component analysis method to discover potential abnormal samples. The specific steps are as follows: Calculate the covariance matrix of the original data matrix X ; Solve the principal component vector P and construct the principal component space ; Calculate the Mahalanobis distance:
[0019] If > , then it is considered that the sample is abnormal and transfer to step S4 for abnormal feedback processing; Among them, is the principal component vector of the i-th sample, is the principal component mean vector, is the abnormal determination threshold.
[0020] Preferably, the abnormal feedback module in step S4 adopts a structured logging mechanism to number each business data record with failed checking, locate the field position, identify the rule trigger ID and archive the original value, and generate a hash digest for its abnormal event:
[0021] Among them, ID is the business primary key, t is the event timestamp, f is the field name, and v is the original field value.
[0022] In a second aspect, the present application provides a full-process data checking system for the manufacture of power transmission and transformation equipment. The system includes: A data acquisition module, which is used to obtain original business data in real time from the raw material link, processing process link, and test and inspection link of the manufacture of power transmission and transformation equipment, and perform preprocessing on the obtained data, including operations such as field standardization, null value imputation, format normalization, and preliminary abnormal point elimination; A cloud processing module, which is used to receive the standardized data transmitted by the data acquisition module, and call the semantic fusion module to perform semantic mapping, attribute unification, and context normalization processing on data fields from different sources, and at the same time schedule the corresponding data checking rule set according to the manufacturing context; A data verification module, which is used to perform logical comparison and rule matching between the fused business data and the scheduled rule set, execute data quality verification at the field level and record level, and generate corresponding verification results; An exception feedback module, which is used to receive the data records that do not conform to the verification rules output by the data verification module, extract their field position information, trigger rule information, time information and source identifier, and submit the abnormal data to the alarm module and the client interface for prompting; A blockchain storage module, which is used to mark the data records passed the verification, generate their digest information, and then write the digest information and data records into the blockchain to achieve data traceability, secure storage and version control; A report generation module, which is used to generate a batch quality verification report according to the data verification results. The report includes indicators such as field compliance rate, abnormal type distribution, rule hit statistics, etc., and is presented in a structured form through the client display interface.
[0023] Thirdly, the present application provides a terminal, including: A memory, which is used to store the full-process data verification program for the manufacturing of power transmission and transformation equipment; A processor, which is used to execute the full-process data verification system for the manufacturing of power transmission and transformation equipment to implement the steps of the full-process data verification method for the manufacturing of power transmission and transformation equipment as described in the first aspect.
[0024] Fourthly, the present application provides a computer-readable storage medium. The storage medium stores computer instructions. When the computer reads the computer instructions in the storage medium, the computer executes the full-process data verification method for the manufacturing of power transmission and transformation equipment as described in the first aspect.
[0025] It can be seen from the above technical solutions that the advantages of the present invention are: (1). Introduce a field normalization mechanism based on the domain ontology model and semantic graph, which can automatically identify field synonymy, unit inconsistency and structural differences between different systems, ensure unified semantic representation of heterogeneous data before it is passed into the verification engine, and enable data from multiple systems such as design, process, test, and quality inspection to be fused and evaluated and collaboratively verified, avoiding the problems of "incompatible fields and unmatched data" existing in traditional systems.
[0026] (2). Introduce the principal component analysis and isolation forest algorithms before performing data quality verification, and identify abnormal samples by combining the covariance relationship and structural characteristics between fields, effectively improving the recognition accuracy of complex problems such as non-linear anomalies, isolated data points, and cross-field conflicts, and can assist the verification rules to filter systematic biases and random anomalies, significantly reducing the false positive rate and missed detection rate.
[0027] (3) By constructing a rule graph that includes rule nodes and their context dependencies, and introducing a rule activation mechanism based on business state vectors, the system can dynamically call and match subsets of rules for verification according to factors such as the current process, equipment status, and process section, avoiding the problems of "one-size-fits-all for all rules" or "static rules being divorced from reality" in traditional systems, and improving the rule adaptation degree and verification efficiency.
[0028] (4) Using a structured exception logging mechanism and blockchain distributed storage, generate summary records with meta-information such as timestamps, field positions, and source devices for all data that fails verification, and complete hash deposit and node signature verification on the chain to ensure the traceability and non-repudiation of exception events in subsequent quality issue analysis, auditing, and supervision scenarios.
[0029] (5) After completing the quality verification, automatically generate a data quality report covering multiple indicators such as field compliance rate, abnormal heat distribution, rule hit frequency, and abnormal clustering type, and visually display it in the form of charts, heat maps, and indicator cards through the client interface to assist managers in quickly locating the source of problems and optimizing the process flow, with decision-making assistance value. Description of the Drawings
[0030] To more clearly illustrate the technical solutions of the present invention, the drawings required for description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.
[0031] Figure 1 It is a schematic flowchart of the full-process data verification method for the manufacturing of power transmission and transformation equipment provided by the embodiments of the present invention. Detailed Embodiments
[0032] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some of the embodiments of the present invention, rather than all of them. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.
[0033] Please refer to Figure 1 As shown, the present invention provides a full-process data verification method for the manufacturing of power transmission and transformation equipment, including the following steps: Step S1: Obtain the original business data generated in each link of the manufacturing of power transmission and transformation equipment; Perform operations of unifying data types, initially removing abnormal fluctuations, and standardizing the collected data; Step S2: Transmit the pre - processed business data to the cloud for processing. Construct a field semantic graph based on the industrial process ontology model, and perform semantic mapping, field normalization, and attribute unification operations on heterogeneous fields to complete semantic fusion. Schedule the verification rule set corresponding to the business context from the rule repository according to the current manufacturing process status parameters; Step S3: Verify the obtained standardized business data through the corresponding verification rule set, including: Construct a structured matrix for the standardized business data:
[0034] where represents the value of the j - th field in the i - th business record, and the verification rule set is expressed as:
[0035] where represents the expression function of the rule, is the corresponding threshold, is the logical operation relationship; Call the rule engine module to match each item of all data matrix elements with the corresponding rule expression to obtain the verification result vector:
[0036] where indicates that the i - th record satisfies all scheduling rules, indicates that at least one rule is not satisfied; Step S4: Perform anomaly marking on the data records with unqualified verification results, upload the fields' positions, violated rule numbers, timestamps, and device IDs they carry as anomaly meta - information, and generate a prompt signal at the client interface; Step S5: Mark the data records with passed verification with a "passed verification" label and write them into the blockchain data module. Calculate the hash value for each piece of data to generate a digest, and store it on the chain in a distributed node to form a chained traceability structure; Step S6: Generate a quality verification report for this batch of manufacturing data according to the verification result vector and statistical parameters, and push the report to the client interface and support interface - call for downloading.
[0037] In some embodiments, the data pre - processing operations in Step S1 include the following sub - steps: Step S11: Perform field alignment and null value interpolation operations on various types of raw data, and use an interpolation strategy based on the K - nearest neighbor method to process missing fields, where the missing field value is calculated as:
[0038] where, Represents the j-th field value of the k-th similar record; Step S12: Smooth the time series fluctuations based on the sliding window method, and the window length is , and the weighted moving average method is used for processing. The output result is:
[0039] where is a preset weight factor that satisfies , and is used to weaken the influence of sampling errors.
[0040] In some embodiments, the semantic fusion in step S2 includes field normalization processing based on ontology mapping. The semantic ontology is constructed by domain experts and represents entity-attribute-relationship triples using the OWL language. Each field is mapped to the target ontology class through the SPARQL query interface and the following operations are performed: Identify field synonyms and abbreviation mappings; Normalize units and dimension conversions; Merge logically equivalent fields; The above operation results form a field-attribute mapping set ; The rule scheduling method constructs a rule activation graph based on the current business context , where the vertex set represents the rule set, and the edge set represents the dependency or mutual exclusion relationship between rules; By constructing a state variable vector , using the matching function:
[0041] Filter the current applicable rule set. The activation path is represented by a rule trigger matrix for subsequent rule engine execution scheduling; where, is the activated rule set, is the i-th rule, is the rule trigger matrix.
[0042] In some embodiments, the rule verification engine in step S3 stores all rule expressions in an expression tree structure. Each rule expression is parsed into a boolean tree structure, where the leaf nodes are field value or constant nodes, and the intermediate nodes are logical operators. During the execution process, the depth-first traversal strategy is used to evaluate each branch expression in turn. If the calculation result of the root node is true, the rule is considered to be hit. This mechanism supports expression-level dynamic compilation and registers an abstract syntax tree inside the rule engine for runtime invocation.
[0043] In some embodiments, the step S3 further includes preprocessing the high-dimensional service data for dimensionality reduction based on the principal component analysis method to discover potential abnormal samples. The specific steps are as follows: Calculate the covariance matrix of the original data matrix X ; Solve the principal component vector P and construct the principal component space ; Calculate the Mahalanobis distance:
[0044] If > , it is considered that the sample is abnormal, and the process proceeds to step S4 for abnormal feedback processing; Among them, is the principal component vector of the i-th sample, is the principal component mean vector, is the abnormal determination threshold.
[0045] In some embodiments, the abnormal feedback module in the step S4 adopts a structured log mechanism to number, locate the field position, identify the rule trigger ID, and archive the original value for each business data record with failed verification, and generate a hash digest for its abnormal event:
[0046] Among them, ID is the business primary key, t is the event timestamp, f is the field name, and v is the original field value.
[0047] In some embodiments, the data passing the verification in the step S5 is organized in a Merkle tree structure to form a root hash , and the root hash, together with the timestamp and batch identifier, is encapsulated into a new block to generate the following structure:
[0048] Among them, is the data structure of the t-th block, is the block generation timestamp, is the current batch identifier, is the node signature.
[0049] In some embodiments, the data items to be chained in the step S5 are compressed and encoded in CBOR format, and the hash values are calculated before and after encoding to verify the integrity. The hash before compression is:
[0050] Among them It is the i-th verified business record. The hash is used to verify whether structural damage occurs during the CBOR compression process. The compressed data is used for on-chain storage and a posteriori reconstruction.
[0051] In some embodiments, the abnormal feedback module in S4 is linked to the client interface to display the field-level abnormal heat map, and each heat value in the image rendering is represented by the function:
[0052] Calculated, among which Representation field The number of anomalies in the time window t, Indicates the maximum anomaly count per unit time of any field; the value range of the color gradient mapping function C is [0,1].
[0053] In some embodiments, the S3 step further includes identifying structural anomalies using an isolation forest algorithm and executing the following scoring function:
[0054] in is the average path length of sample x isolated in multiple random split trees, c(n) is the expected path normalization coefficient, and n is the total number of samples; if , then the sample x is judged to be structurally abnormal.
[0055] In some embodiments, the rule engine includes a reinforcement learning module, which uses a Q-learning algorithm to dynamically optimize the rule selection weights, and the update formula is:
[0056] in It represents the current valuation of taking action a in state s, r is the actual feedback result, α is the learning rate, γ is the discount factor, s′ is the next state, and a′ is the next action, which is used to optimize the activation order of rules.
[0057] The inter-field semantic completion process is modeled using a graph neural network (GNN), and the graph structure is defined as:
[0058] in is a collection of field nodes. Indicates field co-occurrence or dependency, For Field The embedded vector in this figure is subsequently used for missing value prediction and field verification through similarity measurement with the missing field.
[0059] In some embodiments, the structured verification report generated in step S6 includes abnormal type tags after clustering by the DBSCAN algorithm. The clustering core parameters are the density radius and the minimum number of points minPts, and the silhouette coefficient S is output to evaluate the clustering effectiveness:
[0060] where a(i) is the average distance from sample i to other points in the same cluster, and b(i) is the average distance from sample i to the nearest cluster.
[0061] In some embodiments, the present application provides a full-process data verification system for the manufacturing of power transmission and transformation equipment. The system includes: A data acquisition module, which is used to obtain original business data in real time from the raw material link, processing process link, and test and inspection link of the manufacturing of power transmission and transformation equipment, and preprocess the obtained data, including operations such as field standardization, null value imputation, format normalization, and preliminary abnormal point elimination; A cloud processing module, which is used to receive the standardized data transmitted by the data acquisition module, and call the semantic fusion module to perform semantic mapping, attribute unification, and context normalization processing on data fields from different sources, and at the same time schedule the corresponding data verification rule set according to the manufacturing context; A data verification module, which is used to perform logical comparison and rule matching between the fused business data and the scheduled rule set, execute data quality verification at the field level and record level, and generate corresponding verification results; An abnormal feedback module, which is used to receive the data records that do not conform to the verification rules output by the data verification module, extract their field position information, trigger rule information, time information, and source identification, and submit the abnormal data to the alarm module and the client interface for prompting; A blockchain storage module, which is used to mark the data records that pass the verification and generate their summary information, and then write the summary information and data records into the blockchain to achieve data traceability, secure storage, and version control; A report generation module, which is used to generate a batch quality verification report based on the data verification results. The report includes indicators such as field compliance rate, abnormal type distribution, and rule hit statistics, and is presented in a structured form through the client display interface.
[0062] In some embodiments, the present application provides a terminal, including: A memory, which is used to store the full-process data verification program for the manufacturing of power transmission and transformation equipment; A processor, which is used to execute the steps of the full-process data verification method for the manufacturing of power transmission and transformation equipment when implementing the full-process data verification system for the manufacturing of power transmission and transformation equipment.
[0063] In some embodiments, the present application provides a computer-readable storage medium storing computer instructions, and when a computer reads the computer instructions in the storage medium, the computer executes the full-process data verification method for manufacturing power transmission and transformation equipment.
[0064] The above content is only an example and illustration of the concept of the present invention. Those skilled in the art of the present technology can make various modifications or supplements to the described specific embodiments or use similar methods for substitution, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.
Claims
1. A full-process data verification method for the manufacturing of power transmission and transformation equipment, characterized in that, It includes the following steps: Step S1: Obtain the original business data generated in each link of power transmission and transformation equipment manufacturing; Perform operations such as unified data type, preliminary elimination of abnormal fluctuations, and standardization processing on the collected data; Step S2: Transmit the preprocessed business data to the cloud for processing, construct a field semantic graph based on the industry process ontology model, perform semantic mapping, field normalization, and attribute unification operations on heterogeneous fields through the field semantic graph to complete semantic fusion, obtain standardized business data, and dispatch the corresponding inspection rule set corresponding to the business context from the rule warehouse according to the current manufacturing process status parameters; Step S3: Verify the obtained standardized business data through the corresponding inspection rule set; Step S4: Perform abnormal marking processing on the data records with unqualified verification results, upload their carried field positions, violated rule numbers, timestamps, and device IDs as abnormal meta-information, and generate a prompt signal at the client interface; Step S5: Mark the "verification passed" label on the data records that pass the verification, write them into the blockchain data module, calculate the hash value for each piece of data to generate a digest, and store it on the chain to the distributed node to form a chain-like traceability structure; Step S6: Generate a quality inspection report for this batch of manufacturing data according to the verification result vector and statistical parameters.
2. The full-process data verification method for the manufacture of power transmission and transformation equipment according to claim 1, wherein, In step S3, construct a structured matrix for the standardized business data: where represents the value of the j-th field in the i-th service record, and the verification rule set is expressed as: Among them The expression function representing the rule Is the corresponding threshold Is the logical operation relationship; call the rule engine module to match each element of all data matrices with the corresponding rule expression item by item to obtain the verification result vector: wherein indicates that the i-th record satisfies all scheduling rules, indicating that at least one rule is not satisfied.
3. The full-process data verification method for the manufacture of power transmission and transformation equipment according to claim 2, characterized in that, The data preprocessing operations in step S1 include the following sub-steps: Step S11: Perform field alignment and null value interpolation operations on various types of original data, and use an interpolation strategy based on the K-nearest neighbor method to process missing fields, where the missing field value is calculated as follows: Among them, represents the j-th field value of the k-th similar record; Step S12: Perform time series fluctuation smoothing based on the sliding window method, with the window length being , and process it using the weighted moving average method. The output result is: wherein is a preset weight factor, satisfying , and is used to weaken the influence of sampling error.
4. The full-process data verification method for the manufacture of power transmission and transformation equipment according to claim 3, wherein The semantic fusion in the step S2 includes field normalization processing based on ontology mapping. The semantic ontology is constructed by domain experts and represents entity-attribute-relationship triples using the OWL language. Each field is mapped to the target ontology class through the SPARQL query interface and the following operations are performed: Identify field synonyms and abbreviation mappings; Normalize units and perform dimension conversion; Merge logically equivalent fields; The above operation results in a field-attribute mapping set ; The described rule scheduling method constructs a rule activation graph based on the current service context , where the vertex set represents a rule set, and the edge set represents the dependency or mutual exclusion relationship between rules; By constructing a state variable vector , using a matching function: Filter the currently applicable rule set and activate the path to the rule trigger matrix for subsequent execution scheduling by the rule engine; Among them, is the activation rule set, is the i-th rule, is the rule trigger matrix.
5. The full-process data verification method for the manufacturing of power transmission and transformation equipment according to claim 2, wherein The rule inspection engine in step S3 stores all rule expressions in an expression tree structure, where each rule expression is parsed into a boolean tree structure, with its leaf nodes being field value or constant nodes and the intermediate nodes being logical operators; During the execution process, adopt a depth-first traversal strategy to evaluate each branch expression in turn. If the calculation result of the root node is true, it is considered that the rule is hit; This mechanism supports expression-level dynamic compilation and registers an abstract syntax tree inside the rule engine for runtime calls.
6. The full-process data verification method for the manufacture of power transmission and transformation equipment according to claim 5, wherein Step S3 also includes performing dimensionality reduction preprocessing on high-dimensional business data based on the principal component analysis method to discover potential abnormal samples. The specific steps are as follows: Calculate the covariance matrix of the original data matrix X ; Solve the principal component vector P and construct the principal component space ; Calculate the Mahalanobis distance: If > , it is considered that the sample is abnormal, and the process proceeds to step S4 for abnormal feedback processing; Among them, is the principal component vector of the i-th sample, is the mean vector of the principal components, is the anomaly determination threshold.
7. The full-process data verification method for the manufacturing of power transmission and transformation equipment according to claim 1, characterized in that The abnormal feedback module in step S4 adopts a structured logging mechanism to number, locate the field position, identify the rule trigger ID, and archive the original value for each business data record with failed inspection, and generate a hash digest for its abnormal event; Among them, ID is the business primary key, t is the event timestamp, f is the field name, and v is the original field value.
8. A full-process data verification system for the manufacture of power transmission and transformation equipment, characterized in that, This system includes: A data acquisition module, which is used to obtain the original business data in real time from the raw material link, processing process link, and test and inspection link of power transmission and transformation equipment manufacturing, and perform preprocessing on the obtained data, including operations such as field standardization, null value imputation, format normalization, and preliminary abnormal point elimination; A cloud processing module, which is used to receive the standardized data transmitted by the data acquisition module, call the semantic fusion module to perform semantic mapping, attribute unification, and context normalization processing on data fields from different sources, and dispatch the corresponding data inspection rule set according to the manufacturing context; A data verification module, which is used to perform logical comparison and rule matching between the fused service data and the scheduled rule set, execute data quality verification at the field level and record level, and generate corresponding verification results; An exception feedback module, which is used to receive the data records that do not conform to the verification rules output by the data verification module, extract their field position information, trigger rule information, time information and source identifier, and submit the abnormal data to the alarm module and the client interface for prompting; A blockchain storage module, which is used to mark the data records passed the verification, generate their digest information, and then write the digest information and data records into the blockchain to achieve data traceability, secure evidence storage and version control; A report generation module, which is used to generate a batch quality verification report based on the data verification results. The report includes indicators such as field compliance rate, abnormal type distribution, rule hit statistics, etc., and is presented in a structured form through the client display interface.
9. A terminal, characterized in that, Including: A memory, which is used to store the full-process data verification program for the manufacturing of power transmission and transformation equipment; A processor, which is used to execute the steps of the full-process data verification method for the manufacturing of power transmission and transformation equipment as described in claim 1 when executing the full-process data verification system for the manufacturing of power transmission and transformation equipment.
10. A computer-readable storage medium, characterized in that, The storage medium stores computer instructions. When the computer reads the computer instructions in the storage medium, the computer executes the full-process data verification method for the manufacturing of power transmission and transformation equipment as described in claim 1.
Citation Information
Patent Citations
Distributed rule engine system, distributed rule engine construction method and rule processing method
CN104679790A
Method for realizing rule engine dynamics
CN111158691A
Multi-source department creation resource data fusion method
CN111708773A
Equipment abnormity detection method and device and electronic equipment
CN115987390A
Enterprise business data mining method and device, equipment and storage medium
CN118520404A
Cited By
IC carrier plate detection data processing method, system and equipment based on edge calculation
CN120448394A
Full-link data difference matching and quality dispute processing method
CN120782339A
Real-time data security supervision and early warning system based on data stream analysis
CN121037039A
Electric power material verification method and device based on dynamic rule tree
CN121094760A
Automatic quality inspection method, system and equipment based on low-efficiency land use data and medium
CN121413624A
