High-low voltage electrical cabinet continuous processing abnormal rapid response system

Abstract

This invention relates to the field of electrical equipment manufacturing technology and discloses a rapid response system for continuous processing anomalies in high and low voltage electrical cabinets. The system includes: a processing data acquisition module, which collects sensor signals, process parameters, and other elements to establish time-series processing data units; an edge computing preprocessing module, which retrieves preprocessing rule clauses based on the time-series data units and calculates a preprocessing complexity factor to determine data quality; an anomaly feature extraction module, which calculates a cross-time period feature correlation density index based on multiple time-series data units and constructs a multi-time period feature correlation map; and a response strategy generation module, which combines the map and data preprocessing status determination to generate a comprehensive anomaly response prompt. Furthermore, the system can also include an anomaly data storage module and an edge node collaboration module to realize anomaly data storage and cross-node collaborative analysis, enabling rapid response to processing anomalies and improving processing efficiency and accuracy.

Description

Technical Field

[0001] This invention relates to the field of electrical equipment manufacturing technology, specifically to a rapid response system for continuous processing of high and low voltage electrical cabinets. Background Technology

[0002] In the manufacturing process of high and low voltage electrical cabinets, continuous processing has become the mainstream trend. This processing mode improves production efficiency and product quality through automated equipment and assembly line operations. However, in actual continuous processing, due to factors such as the complexity of equipment operating conditions, the diversity of process parameters, and the uncertainty of the production environment, processing abnormalities often occur.

[0003] For example, sensors in processing equipment may experience signal abnormalities, leading to inaccurate data collection and affecting the judgment of the processing process; process parameters may deviate from preset benchmark values ​​during long-term operation, causing the processed electrical cabinet parts to fail to meet quality requirements; the equipment itself may also malfunction due to wear, aging, or other reasons, leading to production interruptions; in addition, changes in environmental parameters such as temperature and humidity in the production environment may also have an adverse impact on the processing process.

[0004] If processing anomalies are not detected and handled in a timely manner, it will not only lead to a large number of defective products and increase production costs, but also affect production schedules and reduce production efficiency. Traditional anomaly response methods often rely on manual inspection and experience-based judgment, which have problems such as slow response speed and low accuracy, and cannot meet the needs of continuous processing for rapid anomaly response.

[0005] With the development of industrial automation and information technology, although some enterprises have introduced data acquisition and monitoring systems, most of these systems can only collect and display data from a single device or a single process. They lack the ability to comprehensively analyze and correlate data from multiple time periods, multiple devices, and multiple process stages, and cannot promptly discover abnormal features and patterns hidden behind complex data.

[0006] Existing anomaly response systems are inadequate in data preprocessing, failing to effectively calibrate and assess data quality based on different processing techniques and equipment conditions, leading to inaccurate subsequent anomaly analysis. Furthermore, in terms of anomaly feature extraction and response strategy generation, they lack correlation analysis of data features across different time periods, making it impossible to construct a comprehensive feature correlation map, thus hindering the generation of reasonable anomaly response priorities and effective response strategies.

[0007] The storage and management of abnormal data are not systematic enough, making it impossible to effectively utilize and analyze historical abnormal data, and making it difficult to summarize the patterns and trends of abnormal occurrences to provide a reference for subsequent anomaly prevention and handling. Regarding edge node collaboration, the data sharing and collaborative processing capabilities between nodes are insufficient, making it impossible to achieve joint analysis of anomaly characteristics across nodes, thus affecting the overall efficiency and accuracy of anomaly response.

[0008] There is an urgent need for a system that can quickly respond to abnormal situations during the continuous processing of high and low voltage electrical cabinets. This system needs to have functions such as efficient data acquisition, preprocessing, abnormal feature extraction, response strategy generation, abnormal data storage, and edge node collaboration to improve the speed and accuracy of abnormal response and ensure the smooth progress of continuous processing. Summary of the Invention

[0009] The purpose of this invention is to provide a rapid response system for continuous processing of high and low voltage electrical cabinets to solve the problems mentioned in the background art.

[0010] To achieve the above objectives, the present invention provides a rapid response system for abnormalities in continuous processing of high and low voltage electrical cabinets, the system comprising:

[0011] The processing data acquisition module, based on the continuous processing equipment of high and low voltage electrical cabinets, collects sensor signals, process parameters, equipment status and environmental parameters, maps the elements to the preset data interface and establishes time-series association identifiers to establish time-series processing data units.

[0012] The edge computing preprocessing module, based on the time-series processing data unit, retrieves the preprocessing rule clauses that match the processing technology type and equipment of the edge nodes, calculates the preprocessing complexity factor according to the number and level depth of the matching clauses, and calibrates the processing data and clause benchmark values, time windows and parameter ranges item by item based on the preprocessing complexity factor, determines the data quality status, and generates a single data preprocessing status judgment.

[0013] The abnormal feature extraction module, based on multiple time-series processing data units, searches for shared equipment numbers, process steps, and parameter sequence entities across different processing time periods. It calculates the cross-time period feature correlation density index based on the number and category of shared entities. Based on the cross-time period feature correlation density index, it constructs a graph structure with processing time periods as nodes and shared relationships as edges, and establishes a multi-time period feature correlation graph.

[0014] The response strategy generation module calculates the abnormal response priority based on the multi-time period feature association map and the preprocessing status of individual data in the associated time period, and generates a comprehensive abnormal response prompt.

[0015] Preferably, the acquisition step of the time-series processing data unit is as follows: based on the continuous processing equipment for high and low voltage electrical cabinets, read the sensor signal content in the acquisition equipment item by item, perform signal type identification and matching, and perform parameter domain content splitting, extract sensor signals, process parameters, equipment status and environmental parameter elements one by one, and generate the original element set of the high and low voltage electrical cabinet processing equipment.

[0016] Based on the original set of elements of the high and low voltage electrical cabinet processing equipment, the mapping and matching verification of the interfaces are performed one by one, and the content of the successfully matched interfaces is converted into a time-series format and reconstructed to generate a time-series mapping interface set for the high and low voltage electrical cabinet processing equipment.

[0017] Based on the set of time-sequential mapping interfaces for high and low voltage electrical cabinet processing equipment, time-sequential association matching is performed between each time-sequential mapping interface, and a time-sequential association relationship identifier is established according to the matching results. The timestamps of the time-sequential mapping interfaces are constructed and the relationships are bound to generate time-sequential processing data units.

[0018] Preferably, the step of obtaining the preprocessing complexity factor is as follows: based on the processing technology type interface and equipment status interface in the time-series processing data unit, extract the original interface value and perform interface content standardization matching with the interface definition library, fill in the missing interfaces and unify the interface format and parameter expression method to obtain a standardized processing technology type interface and equipment status interface combination set.

[0019] Based on the standardized processing technology type interface and equipment status interface combination set, the preprocessing rule clauses that match the process type in the edge nodes are retrieved sequentially, and the clauses that match the equipment status interface and combination set are filtered. The hierarchical depth, number of logical judgments, number of clause references, coverage of clause activation time, overlap of applicable parameters and call frequency of each clause are extracted to generate a preprocessing rule clause matching structured information set.

[0020] The preprocessing complexity factor is calculated based on the preprocessing rule clauses matching the structured information set.

[0021] Preferably, the step of obtaining the single data preprocessing status determination is as follows: based on the preprocessing complexity factor, the processing technology type interface value and the clause benchmark value in the time-series processing data unit are called one by one, and the parameter threshold comparison, interface value range test and time window matching judgment between the interface value and the benchmark value are performed to form an intermediate result set of processing data and clause benchmark comparison.

[0022] Based on the intermediate result set of the comparison between the processing data and the clause benchmark, obtain the parameter range interface value in the time-series processing data unit, perform item-by-item mapping between the processing parameter interface and the applicable parameter range of the preprocessing clause, filter the clauses with successful parameter mapping, and generate a matching result set of processing data and parameter range;

[0023] Based on the matching result set of the processing data and parameter range, the data quality of each processing data item is determined, and a preprocessing status judgment for each data item is generated.

[0024] Preferably, the step of obtaining the cross-time period feature association density index is as follows: based on multiple time-series processing data units, extract the equipment number interface, process link interface and parameter sequence interface in each time-series processing data unit, perform entity classification matching and deduplication processing on all processing data according to interface type, and generate a processing shared entity set;

[0025] Based on the set of shared processing entities, the occurrence count of each entity in different processing periods and the category label of the entity are executed one by one. The repeated distribution information of each type of shared entity in each time-series processing data unit is recorded to obtain the set of shared entity distribution structure.

[0026] Based on the shared entity distribution structure set, the cross-time period feature correlation density index is calculated.

[0027] Preferably, the steps for obtaining the multi-time period feature association map are as follows: based on the cross-time period feature association density index, extract the equipment number interface, process link interface and parameter sequence interface from multiple time-series processing data units, perform entity normalization identification encoding and entity type labeling on each interface content, and generate a processing time period node set;

[0028] Calculate the sharing relationship strength value between processing time periods based on the set of processing time period nodes;

[0029] Based on the shared relationship strength value, a graph structure is constructed with processing time periods as nodes and shared relationship strength values ​​as edge weights. An entity edge connection mapping table between nodes is established to generate a multi-time period feature association graph.

[0030] Preferably, the step of obtaining the comprehensive anomaly response prompt is as follows: based on the multi-time period feature association map, extract the bidirectional paths of all processing time period nodes from the graph structure, and retrieve the single data preprocessing status judgment result, path length, path end node out-degree, path start node interface number and path intermediate node number corresponding to each path, and generate a multi-time period path attribute set.

[0031] Based on the set of multi-time period path attributes, calculate the abnormal response priority of the corresponding time period node;

[0032] Based on the aforementioned abnormal response priority, all processing time nodes are divided into response level ranges according to their abnormal response priority, and the response level color code and response warning node number are marked in each processing path to generate a comprehensive abnormal response prompt.

[0033] Preferably, the system further includes an abnormal data storage module, the working steps of which are: based on the time-series processed data unit and the single data preprocessing status judgment result, filtering out abnormal processed data whose data quality does not meet the requirements, marking the data timestamp and binding it with the device number, and generating an abnormal data identifier set;

[0034] Based on the set of abnormal data identifiers, the storage rule clauses matching the edge nodes are retrieved, abnormal data storage path allocation and storage capacity verification are performed, and an abnormal data storage address mapping table is generated.

[0035] Based on the abnormal data storage address mapping table, abnormal processing data is written to the corresponding storage address according to time sequence, an abnormal data index directory is established, and a historical abnormal data storage unit is generated.

[0036] Preferably, the system further includes an edge node collaboration module, the working steps of which are as follows: based on the time-series processing data units of multiple edge nodes, extract the equipment number interface, process link interface and parameter sequence interface of each node, perform interface content consistency verification and data format unification, and generate a cross-node collaboration data set;

[0037] Based on the cross-node collaboration data set, the collaboration rule clauses matching the edge nodes are retrieved, the data processing load and communication latency of each node are calculated, and the node collaboration priority ranking is generated.

[0038] Based on the node collaboration priority ranking, a data transmission channel and processing task allocation table are established between edge nodes, and cross-node anomaly feature joint analysis is performed to generate collaboration anomaly analysis results.

[0039] Preferably, the output steps of the abnormal response prompt are as follows: based on the comprehensive abnormal response prompt, extract the equipment number, process link and parameter sequence information of the abnormal time period node, perform information semantic conversion and visualization format adaptation, and generate a set of abnormal information visualization elements;

[0040] Based on the set of abnormal information visualization elements, the output rule clauses matching the edge nodes are retrieved, the display terminal type is identified and the display parameters are calibrated, and an abnormal information display template is generated.

[0041] Based on the aforementioned abnormal information display template, the abnormal information is displayed in layers according to risk level, and the time of occurrence, scope of impact, and response suggestions are marked to generate the final abnormal response prompt output.

[0042] Compared with the prior art, the beneficial effects of the present invention are:

[0043] The patented rapid response system for anomalies in continuous high and low voltage electrical cabinet processing demonstrates significant advantages in practical applications. Through a processing data acquisition module, the system comprehensively collects sensor signals, process parameters, equipment status, and environmental parameters from continuous high and low voltage electrical cabinet processing equipment. These elements are then mapped to preset data interfaces, establishing time-series correlation identifiers to form time-series processing data units. This comprehensive and orderly data acquisition method enables the system to obtain complete processing data, laying a solid foundation for subsequent anomaly analysis.

[0044] The edge computing preprocessing module, based on time-series processed data units, retrieves preprocessing rule clauses for matching processing technology types and equipment at edge nodes. By calculating the preprocessing complexity factor, it calibrates the processed data against the clause baseline values, time windows, and parameter ranges item by item, thereby determining the data quality status and generating a single data preprocessing status judgment. This process ensures the accuracy and reliability of the data, eliminates misjudgments caused by data quality issues, and provides high-quality data support for subsequent anomaly feature extraction.

[0045] The anomaly feature extraction module, based on multiple time-series processing data units, identifies shared equipment numbers, process steps, and parameter sequence entities across different processing time periods. It calculates the cross-time period feature correlation density index and constructs a multi-time period feature correlation map. This cross-time period feature correlation analysis can uncover hidden relationships and anomalies in the processing of different time periods, avoiding the limitations of single-time period analysis and improving the comprehensiveness and accuracy of anomaly detection.

[0046] The response strategy generation module calculates the anomaly response priority based on multi-time period feature correlation maps and the preprocessing status of individual data within the correlated time periods, generating comprehensive anomaly response prompts. This module can reasonably determine response priorities based on the severity and scope of the anomaly, providing clear guidance for staff in handling anomalies and improving the efficiency and relevance of anomaly responses.

[0047] The system also includes an abnormal data storage module. This module can filter out abnormal processed data that does not meet the quality requirements, timestamp and bind it to the device number to generate an abnormal data identifier set, then retrieve the storage rule clauses matching the edge nodes, allocate storage paths and verify capacity, write the abnormal data to the corresponding storage address according to time sequence, and establish an index directory. This abnormal data storage method of the system facilitates the querying and analysis of historical abnormal data, enabling the summarization of the patterns and trends of abnormal occurrences, and providing valuable reference for subsequent abnormal prevention and handling.

[0048] Furthermore, the edge node collaboration module, based on the time-series processing data units of multiple edge nodes, extracts the interface content of each node for consistency verification and format unification, generates a cross-node collaboration dataset, then retrieves collaboration rule clauses, calculates the data processing load and communication latency of each node, generates a node collaboration priority ranking, establishes data transmission channels and processing task allocation tables, and performs joint analysis of cross-node anomaly features. This edge node collaboration mechanism realizes data sharing and collaborative processing among nodes, improves the efficiency and accuracy of anomaly feature analysis, and enables a more comprehensive discovery and handling of anomalies in the processing process. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of the working principle of the continuous processing anomaly rapid response system for high and low voltage electrical cabinets described in this invention.

[0050] Figure 2 A flowchart for acquiring time-series processing data units;

[0051] Figure 3 A flowchart for preprocessing complexity factor calculation;

[0052] Figure 4 A flowchart generated for determining the preprocessing status of individual data items;

[0053] Figure 5 This is a flowchart for calculating the cross-time period feature correlation density index. Detailed Implementation

[0054] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 are within the scope of protection of the present invention.

[0055] Please see Figures 1-5This invention provides a rapid response system for anomalies in continuous processing of high and low voltage electrical cabinets. The system includes: a processing data acquisition module, an edge computing preprocessing module, an anomaly feature extraction module, and a response strategy generation module. Specific implementation details are as follows:

[0056] The processing data acquisition module, based on the continuous processing equipment for high and low voltage electrical cabinets, collects sensor signals, process parameters, equipment status, and environmental parameters. These elements are mapped to preset data interfaces, and time-series association identifiers are established to create time-series processing data units. Based on the continuous processing equipment, the module reads sensor signals from the acquisition equipment item by item, identifies and matches signal types, and decomposes parameter domain content. Sensor signals, process parameters, equipment status, and environmental parameters are extracted one by one to generate the original element set of the high and low voltage electrical cabinet processing equipment. Based on this original element set, the mapping and matching of interfaces are verified one by one. Successfully matched interfaces undergo time-series format conversion and interface reconstruction to generate a time-series mapped interface set. Based on this time-series mapped interface set, time-series association matching is performed between the time-series mapped interfaces. Based on the matching results, time-series association relationship identifiers are established, timestamps are constructed, and relationships are bound, generating time-series processing data units.

[0057] The edge computing preprocessing module is based on time-series processing data units. It retrieves preprocessing rule clauses that match the processing technology type and equipment of edge nodes. Based on the number and level depth of matching clauses, it calculates the preprocessing complexity factor. Based on the preprocessing complexity factor, it calibrates the benchmark values, time windows and parameter ranges of processing data and clauses item by item, determines the data quality status, and generates a single data preprocessing status judgment.

[0058] The anomaly feature extraction module is based on multiple time-series processing data units to find shared equipment numbers, process links and parameter sequence entities across different processing time periods. It calculates the cross-time period feature correlation density index based on the number and category of shared entities. Based on the cross-time period feature correlation density index, it constructs a graph structure with processing time periods as nodes and sharing relationships as edges to establish a multi-time period feature correlation graph.

[0059] The response strategy generation module calculates the priority of abnormal responses based on the feature correlation map of multiple time periods and the preprocessing status of individual data in the correlation time periods, and generates a comprehensive abnormal response prompt.

[0060] Example 1:

[0061] Acquiring time-series processing data units requires several specific steps. Based on the continuous processing equipment for high and low voltage electrical cabinets, the sensor signals from the acquisition equipment are read item by item. During the reading process, signal type identification and matching are performed simultaneously. This operation requires analyzing the characteristics of each sensor signal according to pre-defined signal type standards to determine its signal type, such as voltage, current, or temperature. Simultaneously, the parameter domain content is split, separating the various parameter information contained in the sensor signals to facilitate the subsequent extraction of sensor signals, process parameters, equipment status, and environmental parameters. Through this process, a raw element set for the high and low voltage electrical cabinet processing equipment is generated. This raw element set contains all the unprocessed raw data information collected from the processing equipment; this information exists in different element forms and has not yet formed an ordered structure.

[0062] Based on the generated set of original elements for high and low voltage electrical cabinet processing equipment, interface mapping and matching verification is performed on each element. This interface mapping and matching verification involves comparing the original elements with preset data interfaces to check if they meet the requirements for interface format, parameter type, etc. For successfully matched interfaces, time-series format conversion and interface reconstruction operations are required. Time-series format conversion ensures the data conforms to time-series format requirements for subsequent time-related analysis and processing; interface reconstruction adjusts and optimizes the interface structure to better meet the system's data processing needs. Through these operations, a time-series mapped interface set for high and low voltage electrical cabinet processing equipment is generated. The data in this time-series mapped interface set is now organized according to a specific time-series format and interface structure, exhibiting a clearer structure and more explicit time-series relationships compared to the original element set.

[0063] Based on the time-series mapping interface set for high and low voltage electrical cabinet processing equipment, time-series association matching is performed between each time-series mapping interface. This step requires analyzing the temporal order and interrelationships of the data from each interface to determine their time-series associations. A time-series association identifier is established based on the matching results. This identifier records the time-series association between the data from each interface, facilitating subsequent timestamp construction and relationship binding. After establishing the time-series association identifier, timestamp construction and relationship binding are performed on the time-series mapping interfaces. Timestamp construction involves adding a precise timestamp to each interface data point to record its generation time; relationship binding involves binding interface data with time-series associations according to their relationships, forming an organic whole. Through these operations, a time-series processing data unit is ultimately generated.

[0064] Throughout the acquisition of time-series processed data units, each step must strictly adhere to established procedures and standards. For example, during signal type identification and matching, accurate signal feature analysis algorithms are required to ensure that each sensor signal can be correctly identified as its type. During interface mapping matching and verification, the parameters of the original elements and the preset interfaces must be carefully checked, including parameter names, types, and value ranges, to ensure accurate matching. Elements that fail to match require appropriate processing, such as recording error messages or manual intervention, to guarantee data integrity and accuracy.

[0065] When performing time series format conversion, it is essential to select an appropriate conversion method based on different interface types and data characteristics to ensure that the converted data accurately reflects its time sequence and time intervals. During interface reconstruction, the system's data processing flow and subsequent analysis requirements need to be considered, and the interface structure should be reasonably adjusted to better support the system's various functions.

[0066] In the process of temporal correlation matching, multiple factors need to be considered comprehensively, such as the generation time of interface data and the logical relationship between data, in order to determine the correct temporal correlation between them. When establishing temporal correlation identifiers, it is necessary to ensure the accuracy and completeness of the identifiers so that timestamp construction and relationship binding can be performed accurately in the future.

[0067] Timestamp construction requires a high-precision clock source to ensure that the timestamps added to each interface data have sufficient accuracy to meet the system's time precision requirements. During the relationship binding process, the relevant interface data must be correctly bound according to the time sequence association identifier to ensure that the data in the generated time-series processed data units have accurate time sequence and logical relationships.

[0068] Example 2:

[0069] Obtaining the preprocessing complexity factor requires multiple steps. Based on the processing technology type interface and equipment status interface in the time-series processing data unit, the original interface values ​​are extracted. This extraction operation must strictly adhere to the interface definition specifications to ensure that the obtained original interface values ​​accurately reflect the actual state of the interface. After extraction, the original interface values ​​are matched against the interface definition library for standardized interface content. The interface definition library stores standard definitions for various interfaces, including interface formats and parameter expressions. During the matching process, the differences between the original interface values ​​and the standard definitions are checked one by one. For missing interfaces, they are completed according to the standard definitions; for interface formats and parameter expressions that do not conform to the standards, they are uniformly adjusted to obtain a standardized set of processing technology type interfaces and equipment status interfaces. The interface data in this set achieves a unified standard in format and expression, providing a standardized data foundation for subsequent processing.

[0070] Based on the standardized process type interface and equipment status interface combination set, preprocessing rule clauses matching the process type are sequentially retrieved from the edge nodes. The edge nodes store preprocessing rule clauses for different process types and equipment statuses. During retrieval, the corresponding rule clause is found within the edge nodes based on the process type information in the combination set. Simultaneously, clauses matching the equipment status interface with the combination set are filtered out, ensuring that the filtered clauses meet both the process type requirements and the equipment status interface. After filtering out relevant clauses, information such as the hierarchical depth, number of logical judgments, number of clause references, clause activation time coverage, overlap of applicable parameters, and call frequency are extracted to generate a structured information set of preprocessing rule clause matching. This information describes the characteristics of the matched clauses from multiple dimensions, providing comprehensive data support for subsequent calculation of the preprocessing complexity factor.

[0071] When extracting the hierarchical depth of each clause, it is necessary to clarify the hierarchical structure of the clause within the edge node rule system. The hierarchical depth reflects the complexity and importance of the clause. The number of logical judgments refers to the number of logical judgment conditions contained in the clause; the more logical judgments, the higher the complexity of the clause. The number of clause citations refers to the number of times the clause has been cited in previous processing; the more citations, the higher the importance of the clause. The clause activation time coverage refers to the time period within which the clause is active, which is important for assessing the clause's applicability and processing timeliness. The overlap of applicable parameters refers to the degree of overlap between the parameters applicable to the clause and the parameters in the combination set; the higher the overlap, the stronger the relevance of the clause to the current data. The call frequency refers to the number of times the clause is called per unit of time; the call frequency reflects the frequency of the clause's use.

[0072] Based on the structured information set matched with preprocessing rule clauses, a preprocessing complexity factor is calculated. During the calculation, various parameters in the structured information set need to be comprehensively considered. Through specific calculation methods and weight allocation, each parameter is transformed into a numerical value that reflects the preprocessing complexity. The weights of each parameter need to be determined based on their impact on preprocessing complexity. For example, hierarchy depth and the number of logical judgments may have a greater impact on preprocessing complexity, and therefore can be assigned higher weights; while the number of clause references and the frequency of invocation may have a relatively smaller impact on preprocessing complexity, and therefore can be assigned lower weights.

[0073] After determining the weights of each parameter, the parameters are standardized to transform parameters with different dimensions into comparable values. The standardization method can be chosen appropriately based on the characteristics of the parameters, such as min-max standardization or Z-score standardization. After standardization, each parameter is multiplied by its corresponding weight, and then all products are summed to obtain the preprocessing complexity factor.

[0074] Throughout the entire process of obtaining the preprocessing complexity factor, each step requires strict control over the accuracy of the data and the standardization of the processing. When extracting the original interface values, it is crucial to ensure the accuracy of the extraction process to avoid deviations in subsequent processing due to extraction errors. During the standardization matching with the interface definition library, each interface data point must be carefully verified to ensure that the completed and adjusted interface data conforms to the standard definition.

[0075] When retrieving and filtering preprocessing rule clauses, ensure the accuracy of the search criteria to avoid omitting important clauses. When extracting the parameters of each clause, accurately record the value of each parameter to ensure the data in the structured information set is authentic and reliable. When calculating the preprocessing complexity factor, reasonably set the weights of each parameter and select appropriate standardization and calculation methods to ensure that the calculation results accurately reflect the complexity of the preprocessing.

[0076] Example 3:

[0077] Obtaining the preprocessing status of a single data item requires multiple specific operations. Based on the preprocessing complexity factor, the processing technology type interface value and the clause benchmark value in the time-series processing data unit are called item by item. The parameter threshold comparison between the interface value and the benchmark value, the interface value range test, and the time window matching judgment are performed to form an intermediate result set of the comparison between the processing data and the clause benchmark.

[0078] During the parameter threshold comparison process, the interface value for each processing technology type needs to be compared with the corresponding clause baseline value. The difference between the two is calculated, and it is determined whether the difference is within the allowable threshold range. For example, if the clause baseline value of an interface is A, and the allowable threshold is ±B, then if the interface value is within the range [AB, A+B], the parameter threshold comparison is considered to pass; otherwise, it fails. Interface value range verification checks whether the interface value is within a specified value range. For example, if the value range of an interface is specified as [C, D], the verification passes if the interface value is within this range; otherwise, it fails. Time window matching judgment is for data with time attributes. It checks whether the timestamp of the data falls within a specified time window. If it falls within the time window, the match passes; otherwise, it fails. These three judgments form an intermediate result set, which records the pass / fail status of each interface value in each judgment.

[0079] Based on the intermediate result set comparing the processing data with the clause benchmark, the parameter range interface values ​​in the time-series processing data units are obtained. A step-by-step mapping is performed between the processing parameter interfaces and the applicable parameter ranges of the pre-processed clauses. Clauses with successful parameter mapping matches are filtered, generating a matching result set of processing data and parameter ranges. During the step-by-step mapping, for each processing parameter interface, its parameter range interface value is compared with the applicable parameter range specified in the pre-processed clause to determine if there is any overlap or inclusion relationship. If so, the clause is considered to have successfully matched the processing parameter interface; otherwise, the match fails. Through this mapping and filtering operation, a matching result set is obtained, which contains information on all clauses that successfully match the processing parameter interfaces.

[0080] Based on the matching result set of processed data and parameter ranges, the data quality of each piece of processed data is determined, and a preprocessing status judgment for each item is generated. When determining data quality, it is necessary to comprehensively consider the clause information in the matching result set and the judgment results in the intermediate result set. Data quality can be evaluated based on factors such as the number of successfully matched clauses, the importance of the clauses, and the pass rate of each judgment.

[0081] To more accurately assess data quality, the concept of a data quality assessment value is introduced, and its calculation formula is as follows:

[0082]

[0083] in, This represents the data quality assessment value, used to quantitatively reflect the data quality status of the processed data. This indicates the number of clauses that were successfully matched, i.e., the total number of clauses that participated in the evaluation; Indicates the first The weight of each successfully matched clause is determined based on factors such as the importance and scope of the clause. Different clauses have different weight values, and the larger the weight value, the greater the impact of the clause on data quality. Indicates the first The score corresponding to each successfully matched clause is calculated based on the pass / fail status of each judgment related to that clause in the intermediate result set. For example, if the parameter threshold comparison, interface value range check, and time window matching judgment for that clause all pass, then... The higher score will be awarded. If a partial pass or failure is not achieved, the corresponding score will be awarded based on the specific circumstances.

[0084] When calculating the data quality assessment value, it is first necessary to determine the weight of each successfully matched clause. Weights can be determined through methods such as expert evaluation and historical data statistics. For example, important process parameter clauses can be assigned higher weights, while general equipment status clauses can be assigned lower weights. Then, based on the intermediate result set, the judgment score corresponding to each clause is calculated. For example, if all three judgments corresponding to a certain clause pass, then... It can be set to 10 points; if two items pass and one item fails, then... It can be set to 7 points; if one item passes and two items fail, then... It can be set to 4 points; if all three items are failed, then It can be set to 1 point.

[0085] After confirming and Then, substitute the values ​​into the formula to calculate the data quality assessment value. .according to The size of the data can be used to classify data quality into different levels; for example, when... Data quality is considered excellent when the score is greater than or equal to 8; when Data quality is good when the score is between 6 and 8; When the score is between 4 and 6, the data quality is medium; when A score less than 4 indicates poor data quality. This method allows for an objective and accurate determination of the data quality of each processed data item, thereby generating a preprocessing status assessment for each individual data item.

[0086] When performing parameter threshold comparisons, interface value range checks, and time window matching judgments, it is crucial to accurately set the thresholds, value ranges, and time windows to ensure the rationality of the judgment criteria. When mapping the processing parameter interfaces to the applicable parameter ranges of the preprocessing clauses item by item, the parameter ranges must be carefully verified to avoid mapping errors. When determining the clause weights and judgment result scores, various factors must be fully considered to ensure that the weights and scores are set reasonably and scientifically.

[0087] Example 4:

[0088] Obtaining the cross-time period feature correlation density index requires multiple specific operations. Based on multiple time-series processing data units, the equipment number interface, process step interface, and parameter sequence interface are extracted from each time-series processing data unit. For example, in a continuous processing scenario of a high and low voltage electrical cabinet, there are three time-series processing data units, each corresponding to a different processing time period. In the time-series processing data unit of the first processing time period, the equipment number interface is displayed as "Equipment A-001", the process step interface is "cabinet welding", and the parameter sequence interface includes parameters such as welding current, voltage, and time. In the time-series processing data unit of the second processing time period, the equipment number interface is also "Equipment A-001", the process step interface is "cabinet assembly", and the parameter sequence interface includes parameters such as bolt torque and component dimensions. In the time-series processing data unit of the third processing time period, the equipment number interface is "Equipment B-002", the process step interface is "cabinet welding", and the parameter sequence interface includes parameters such as welding current, voltage, and time.

[0089] After extraction, all processing data is categorized and deduplicated according to interface type. For equipment number interfaces, "Equipment A-001" and "Equipment B-002" are categorized as different equipment entities; for process step interfaces, "cabinet welding" and "cabinet assembly" are categorized as different process step entities; for parameter sequence interfaces, welding current, voltage, time, bolt torque, component dimensions, etc., are categorized as different parameter entities. During deduplication, if different processing data units contain the same equipment number, process step, or parameter sequence, only one entity is retained to avoid duplicate calculations. Through this operation, a processing shared entity set is generated, which contains all equipment, process steps, and parameter entities that appear in multiple processing time periods.

[0090] Based on the set of shared processing entities, the occurrence count and category labeling of each entity in different processing time periods are performed. Taking the equipment number entity "Equipment A-001" as an example, it appears in the first and second processing time periods, so its occurrence count is 2. The process entity "Cabinet Welding" appears in the first and third processing time periods, with an occurrence count of 2. The parameter entity "Welding Current" appears in the first and third processing time periods, with an occurrence count of 2. Simultaneously, each entity is categorized, specifying whether it belongs to the equipment number, process stage, or parameter sequence category. The repetition distribution information of each type of shared entity in each time-series processing data unit is recorded. For example, equipment number "Equipment A-001" appears in processing time periods 1 and 2, process stage "Cabinet Welding" appears in processing time periods 1 and 3, and parameter "Welding Current" appears in processing time periods 1 and 3, thus obtaining the shared entity distribution structure set.

[0091] When counting occurrences, the time-series processing data units for each processing period must be carefully checked to ensure the accuracy of the statistical results. For example, when checking the equipment number interface, each character of the equipment number should be compared character by character to avoid statistical errors caused by subtle differences such as capitalization or spaces. When classifying, the classification should be strictly based on the interface type to ensure that the category label of each entity is accurate.

[0092] Based on the shared entity distribution structure set, the cross-time period feature association density index is calculated. The calculation process needs to comprehensively consider the number and category of shared entities, as well as their distribution across different processing time periods. For example, for two processing time periods, the greater the number of entities shared and the more diverse the entity categories, the stronger the feature association between these two processing time periods.

[0093] Suppose we want to calculate the cross-time-segment feature correlation index between processing time period 1 and processing time period 2. First, we count the number of shared equipment number entities, process step entities, and parameter sequence entities between these two processing time periods. In the example above, processing time periods 1 and 2 share 1 equipment number entity (equipment A-001), 0 process step entities, and 0 parameter sequence entities. Then, we count the number of shared entities between processing time periods 1 and 3: 0 equipment number entities, 1 process step entity (cabinet welding), and 1 parameter sequence entity (welding current).

[0094] Consider the impact of different entity categories on the degree of association. Sharing equipment number entities may mean using the same equipment for processing, resulting in a relatively high degree of association; sharing process step entities means performing the same process operations, with a slightly lower degree of association; sharing parameter sequence entities means using the same process parameters, resulting in a relatively low degree of association. Therefore, different weight coefficients can be assigned to different categories of entities, for example, a weight coefficient of 0.5 for equipment number entities, 0.3 for process step entities, and 0.2 for parameter sequence entities.

[0095] Based on the number of shared entities and their corresponding weight coefficients, the correlation strength value between each processing time period pair is calculated. For processing time period 1 and processing time period 2, the correlation strength value is: 1×0.5+0×0.3+0×0.2=0.5. For processing time period 1 and processing time period 3, the correlation strength value is: 0×0.5+1×0.3+1×0.2=0.5.

[0096] The impact of the frequency of shared entities appearing in different processing periods on the association strength can also be considered. For example, if an entity appears in both processing periods, and appears more frequently, it indicates a stronger association between the entity in those two periods. Therefore, the frequency of appearance can be used as a correction factor to adjust the association strength value. Assuming the correction factor for the frequency of appearance is the number of appearances divided by the maximum number of appearances, in the example above, the maximum number of appearances is 2. For the entity "Equipment A-001" (equipment number) appearing twice in processing periods 1 and 2, the correction factor is 2 ÷ 2 = 1. Therefore, the association strength value for processing periods 1 and 2 is adjusted to 0.5 × 1 = 0.5. Similarly, for the entity "Cabinet Welding" (process step) and the parameter entity "Welding Current" (parameter) appearing twice in processing periods 1 and 3, the correction factor is 2 ÷ 2 = 1. Therefore, the association strength value for processing periods 1 and 3 is adjusted to 0.5 × 1 = 0.5.

[0097] Through the above steps, the cross-time period feature correlation density index is calculated, which can quantitatively reflect the degree of feature correlation between different processing periods. In practical applications, the calculation method of the cross-time period feature correlation density index can accurately assess the degree of correlation between processing periods based on the sharing of equipment numbers, process links, and parameter sequence entities in the processing data, providing an important basis for the subsequent establishment of multi-time period feature correlation maps.

[0098] When extracting interface information, ensure that no important equipment number, process step, and parameter sequence interface are missed; when performing entity classification matching and deduplication, ensure accurate classification and thorough deduplication; when counting occurrences and labeling categories, be careful and meticulous to avoid errors; when calculating the correlation density index, set weight coefficients and correction factors reasonably to ensure that the index can accurately reflect the degree of feature correlation between processing periods.

[0099] Example 5:

[0100] Obtaining multi-time-period feature association maps requires multiple specific operations. Based on the cross-time-period feature association density index, the equipment number interface, process link interface, and parameter sequence interface are extracted from multiple time-series processing data units. For example, in a continuous processing scenario of a high and low voltage electrical cabinet, there are four time-series processing data units, corresponding to processing time periods 1 to 4. In the interface data of time period 1, the equipment number interface is "Equipment X-701", the process link interface is "plate cutting", and the parameter sequence interface includes parameters such as cutting speed and tool rotation speed; in time period 2, the equipment number interface is "Equipment X-701", the process link interface is "bending and forming", and the parameter sequence interface includes parameters such as bending angle and pressure value; in time period 3, the equipment number interface is "Equipment Y-802", the process link interface is "plate cutting", and the parameter sequence interface includes parameters such as cutting speed and tool rotation speed; in time period 4, the equipment number interface is "Equipment X-701", the process link interface is "welding assembly", and the parameter sequence interface includes parameters such as welding current and voltage.

[0101] After extracting the interfaces, each interface content is coded with entity normalization identifiers and labeled with entity type. Entity normalization identifier coding must follow a unified rule. For example, the equipment number interface "Equipment X-701" is coded as "EQ-X701", and "Equipment Y-802" is coded as "EQ-Y802"; the process step interface "Sheet metal cutting" is coded as "PR-CUT", "Bending" as "PR-BEND", and "Welding assembly" as "PR-WELD"; in the parameter sequence interface, "Cutting speed" is coded as "PA-CS", "Tool speed" as "PA-TR", and "Bending angle" as "PA-BA", etc. When labeling entity types, it is clearly defined that the equipment number interface corresponds to the "Equipment entity", the process step interface corresponds to the "Process entity", and the parameter sequence interface corresponds to the "Parameter entity", generating a set of processing time period nodes. In this set, each processing time period node contains a normalized encoded interface entity and type label. For example, the time period 1 node is represented as {EQ-X701 (equipment entity), PR-CUT (process entity), PA-CS (parameter entity), PA-TR (parameter entity)}.

[0102] Based on the set of nodes for each processing time period, the strength value of the sharing relationship between processing time periods is calculated. The calculation of the sharing relationship strength value is based on the number and type of shared entities between nodes. Taking time period 1 and time period 2 as examples, they share the equipment entity "EQ-X701" but do not share any process or parameter entities; time period 1 and time period 3 share the process entity "PR-CUT" and the parameter entities "PA-CS" and "PA-TR"; time period 1 and time period 4 share the equipment entity "EQ-X701". During the calculation, different weights are assigned to different types of entities: the weight of the equipment entity is set to 0.4, the weight of the process entity is set to 0.3, and the weight of the parameter entity is set to 0.2. The strength value of the sharing relationship between time period 1 and time period 2 is: 1 (number of equipment entities) × 0.4 = 0.4; the strength value of the sharing relationship between time period 1 and time period 3 is: 1 (number of process entities) × 0.3 + 2 (number of parameter entities) × 0.2 = 0.3 + 0.4 = 0.7; the strength value of the sharing relationship between time period 1 and time period 4 is: 1 × 0.4 = 0.4.

[0103] Based on the sharing relationship strength value, a graph structure is constructed with processing time periods as nodes and the sharing relationship strength value as the edge weight. Time periods 1 to 4 are used as nodes in the graph, and the edges between nodes are determined by the sharing relationship strength value. For example, there is an edge between time period 1 and time period 2 with a weight of 0.4; an edge between time period 1 and time period 3 with a weight of 0.7; and an edge between time period 1 and time period 4 with a weight of 0.4. Simultaneously, the sharing relationships between other time periods need to be checked. For example, time period 2 and time period 4 share the equipment entity "EQ-X701", whose sharing relationship strength value is 0.4. Therefore, there is an edge with a weight of 0.4 between time period 2 and time period 4.

[0104] After constructing the graph structure, a mapping table is created to represent the entity edge connections between nodes. This mapping table records the shared entity information corresponding to each edge. For example, the edge between time period 1 and time period 3 corresponds to the shared process entity "PR-CUT" and parameter entities "PA-CS" and "PA-TR"; the edge between time period 2 and time period 4 corresponds to the shared equipment entity "EQ-X701". The mapping table must be formatted correctly, including the starting node, ending node, edge weights, and a list of shared entities to clearly present the details of the relationships between nodes.

[0105] During the entity normalization identification coding process, it is necessary to ensure the consistency and uniqueness of coding rules to avoid different entities having the same code. For example, the code for equipment number "Equipment X-701" must uniquely correspond to this equipment and cannot be confused with the codes of other equipment numbers. Entity type labeling must strictly follow the interface type classification to ensure that the labeling of "Equipment Entity," "Process Entity," and "Parameter Entity" is accurate, providing a reliable foundation for subsequent calculation of the strength of shared relationships.

[0106] When calculating the strength of shared relationships, all shared entities must be comprehensively counted to avoid omissions. For example, the shared parameter entities "PA-CS" and "PA-TR" between time period 1 and time period 3 must all be included in the count, and their values ​​must be accurately multiplied and summed during weight calculation. For time period nodes that do not share any entities, such as time period 2 and time period 3, no edge connection is established between them.

[0107] When constructing the graph structure, the node representation must clearly reflect the processing time period information, and the edge connections must strictly adhere to the calculation results of the shared relation strength values. The visualization of the graph structure must conform to graph theory standards, with reasonable node placement and clearly labeled edge weights to facilitate subsequent analysis. The establishment of the entity edge connection mapping table must record the associated entities of each edge in detail, ensuring that the content of the mapping table is completely consistent with the graph structure, providing accurate data support for multi-time period feature association analysis.

[0108] Through the above steps, a multi-time period feature association map is generated. This map displays the degree of feature association between different processing time periods in an intuitive graph structure. Edges with higher shared relationship strength values ​​indicate a stronger feature association between corresponding time periods. In practical applications, the multi-time period feature association map can help the system quickly identify closely related time periods during processing, providing a visual basis for cross-time period analysis of abnormal features and helping to improve the efficiency and accuracy of anomaly response.

[0109] The system also includes an abnormal data storage module and an edge node collaboration module. The abnormal data storage module, based on the time-series processed data units and the preprocessing status judgment results of individual data, filters out abnormal processed data that does not meet the quality requirements, timestamps the data, binds it to the equipment number, and generates an abnormal data identifier set. Based on the abnormal data identifier set, it retrieves the storage rule clauses matching the edge nodes, allocates abnormal data storage paths and verifies storage capacity, and generates an abnormal data storage address mapping table. Based on the storage address mapping table, it writes the abnormal processed data to the corresponding storage addresses according to the time sequence, establishes an abnormal data index directory, and generates historical abnormal data storage units. The edge node collaboration module, based on the time-series processed data units of multiple edge nodes, extracts the equipment number interface, process link interface, and parameter sequence interface of each node, verifies the consistency of interface content and unifies the data format, and generates a cross-node collaborative data set. Based on the cross-node collaborative data set, it retrieves the collaboration rule clauses matching the edge nodes, calculates the data processing load and communication latency of each node, and generates a node collaboration priority ranking. Based on the priority ranking, it establishes a data transmission channel and processing task allocation table between edge nodes, performs joint analysis of cross-node abnormal features, and generates collaborative abnormal analysis results. The output steps for the comprehensive anomaly response prompt are as follows: Based on the comprehensive anomaly response prompt, extract the equipment number, process step, and parameter sequence information of the node during the anomaly period, perform semantic transformation and visualization format adaptation, and generate a set of anomaly information visualization elements; Based on the visualization element set, retrieve the output rule clauses that match the edge nodes, identify the display terminal type and calibrate the display parameters, and generate anomaly information display templates; Based on the display templates, display the anomaly information in layers according to risk level, mark the anomaly occurrence time, impact scope, and response suggestions, and generate the final anomaly response prompt output.

[0110] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0111] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A rapid response system for abnormalities in continuous processing of high and low voltage electrical cabinets, characterized in that, The system includes: The processing data acquisition module, based on the continuous processing equipment of high and low voltage electrical cabinets, collects sensor signals, process parameters, equipment status and environmental parameters, maps the elements to the preset data interface and establishes time-series association identifiers to establish time-series processing data units. The edge computing preprocessing module, based on the time-series processing data unit, retrieves the preprocessing rule clauses that match the processing technology type and equipment of the edge nodes, calculates the preprocessing complexity factor according to the number and level depth of the matching clauses, and calibrates the processing data and clause benchmark values, time windows and parameter ranges item by item based on the preprocessing complexity factor, determines the data quality status, and generates a single data preprocessing status judgment. The abnormal feature extraction module, based on multiple time-series processing data units, searches for shared equipment numbers, process steps, and parameter sequence entities across different processing time periods. It calculates the cross-time period feature correlation density index based on the number and category of shared entities. Based on the cross-time period feature correlation density index, it constructs a graph structure with processing time periods as nodes and shared relationships as edges, and establishes a multi-time period feature correlation graph. The response strategy generation module calculates the abnormal response priority based on the multi-time period feature association map and the preprocessing status of individual data in the associated time period, and generates a comprehensive abnormal response prompt. The steps for obtaining the comprehensive anomaly response prompt are as follows: Based on the multi-time period feature association map, extract the bidirectional paths of all processing time period nodes from the graph structure, and retrieve the single data preprocessing status judgment result, path length, path end node out-degree, path start node interface number and path intermediate node number corresponding to each path to generate a multi-time period path attribute set. Based on the set of multi-time period path attributes, calculate the abnormal response priority of the corresponding time period node; Based on the aforementioned abnormal response priority, all processing time nodes are divided into response level ranges according to their abnormal response priority, and the response level color code and response warning node number are marked in each processing path to generate a comprehensive abnormal response prompt. The system also includes an edge node collaboration module. The working steps of the edge node collaboration module are as follows: based on the time-series processing data units of multiple edge nodes, extract the equipment number interface, process link interface and parameter sequence interface of each node, perform interface content consistency verification and data format unification, and generate a cross-node collaboration data set. Based on the cross-node collaboration data set, the collaboration rule clauses matching the edge nodes are retrieved, the data processing load and communication latency of each node are calculated, and the node collaboration priority ranking is generated. Based on the node collaboration priority ranking, a data transmission channel and processing task allocation table between edge nodes are established, cross-node anomaly feature joint analysis is performed, and collaboration anomaly analysis results are generated. The steps for obtaining the cross-time period feature association density index are as follows: Based on multiple time-series processing data units, extract the equipment number interface, process link interface and parameter sequence interface in each time-series processing data unit, perform entity classification matching and deduplication processing on all processing data according to interface type, and generate a processing shared entity set; Based on the set of shared processing entities, the occurrence count of each entity in different processing periods and the category label of the entity are executed one by one. The repeated distribution information of each type of shared entity in each time-series processing data unit is recorded to obtain the set of shared entity distribution structure. Based on the shared entity distribution structure set, the cross-time period feature correlation density index is calculated.

2. The rapid response system for continuous processing anomalies in high and low voltage electrical cabinets according to claim 1, characterized in that, The steps for acquiring the time-series processing data unit are as follows: Based on the continuous processing equipment for high and low voltage electrical cabinets, read the sensor signal content in the acquisition equipment item by item, perform signal type identification and matching, and split the parameter domain content. Extract sensor signals, process parameters, equipment status and environmental parameter elements one by one to generate the original element set of the high and low voltage electrical cabinet processing equipment. Based on the original set of elements of the high and low voltage electrical cabinet processing equipment, the mapping and matching verification of the interfaces are performed one by one, and the content of the successfully matched interfaces is converted into a time-series format and reconstructed to generate a time-series mapping interface set for the high and low voltage electrical cabinet processing equipment. Based on the set of time-sequential mapping interfaces for high and low voltage electrical cabinet processing equipment, time-sequential association matching is performed between each time-sequential mapping interface, and a time-sequential association relationship identifier is established according to the matching results. The timestamps of the time-sequential mapping interfaces are constructed and the relationships are bound to generate time-sequential processing data units.

3. The rapid response system for continuous processing anomalies in high and low voltage electrical cabinets according to claim 1, characterized in that, The steps for obtaining the preprocessing complexity factor are as follows: Based on the processing technology type interface and equipment status interface in the time-series processing data unit, extract the original interface values ​​and perform interface content standardization matching with the interface definition library, fill in the missing interfaces and unify the interface format and parameter expression method to obtain a standardized processing technology type interface and equipment status interface combination set. Based on the standardized processing technology type interface and equipment status interface combination set, the preprocessing rule clauses that match the process type in the edge nodes are retrieved sequentially, and the clauses that match the equipment status interface and combination set are filtered. The hierarchical depth, number of logical judgments, number of clause references, coverage of clause activation time, overlap of applicable parameters and call frequency of each clause are extracted to generate a preprocessing rule clause matching structured information set. The preprocessing complexity factor is calculated based on the preprocessing rule clauses matching the structured information set.

4. The rapid response system for continuous processing anomalies in high and low voltage electrical cabinets according to claim 1, characterized in that, The steps for obtaining the preprocessing status determination of a single data item are as follows: based on the preprocessing complexity factor, the processing technology type interface value and the clause benchmark value in the time-series processing data unit are called item by item, and the parameter threshold comparison, interface value range test and time window matching judgment are performed between the interface value and the benchmark value to form an intermediate result set of the comparison between the processing data and the clause benchmark. Based on the intermediate result set of the comparison between the processing data and the clause benchmark, obtain the parameter range interface value in the time-series processing data unit, perform item-by-item mapping between the processing parameter interface and the applicable parameter range of the preprocessing clause, filter the clauses with successful parameter mapping, and generate a matching result set of processing data and parameter range; Based on the matching result set of the processing data and parameter range, the data quality of each processing data item is determined, and a preprocessing status judgment for each data item is generated.

5. The rapid response system for continuous processing anomalies in high and low voltage electrical cabinets according to claim 1, characterized in that, The steps for obtaining the multi-time period feature association map are as follows: Based on the cross-time period feature association density index, extract the equipment number interface, process link interface and parameter sequence interface from multiple time-series processing data units, perform entity normalization identification encoding and entity type labeling on each interface content, and generate a processing time period node set; Calculate the sharing relationship strength value between processing time periods based on the set of processing time period nodes; Based on the shared relationship strength value, a graph structure is constructed with processing time periods as nodes and shared relationship strength values ​​as edge weights. An entity edge connection mapping table between nodes is established to generate a multi-time period feature association graph.

6. The rapid response system for continuous processing anomalies in high and low voltage electrical cabinets according to claim 1, characterized in that, The system also includes an abnormal data storage module. The working steps of the abnormal data storage module are as follows: based on the time-series processed data unit and the single data preprocessing status judgment result, abnormal processed data that does not meet the data quality requirements are filtered out, data timestamps are marked and bound to the device number, and an abnormal data identifier set is generated. Based on the set of abnormal data identifiers, the storage rule clauses matching the edge nodes are retrieved, abnormal data storage path allocation and storage capacity verification are performed, and an abnormal data storage address mapping table is generated. Based on the abnormal data storage address mapping table, abnormal processing data is written to the corresponding storage address according to time sequence, an abnormal data index directory is established, and a historical abnormal data storage unit is generated.

7. A rapid response system for continuous processing anomalies in high and low voltage electrical cabinets according to claim 6, characterized in that, The output steps of the abnormal response prompt are as follows: Based on the comprehensive abnormal response prompt, extract the equipment number, process link and parameter sequence information of the abnormal time period node, perform information semantic conversion and visualization format adaptation, and generate a set of abnormal information visualization elements. Based on the set of abnormal information visualization elements, the output rule clauses matching the edge nodes are retrieved, the display terminal type is identified and the display parameters are calibrated, and an abnormal information display template is generated. Based on the aforementioned abnormal information display template, the abnormal information is displayed in layers according to risk level, and the time of occurrence, scope of impact, and response suggestions are marked to generate the final abnormal response prompt output.

Images