Power distribution equipment connector quality detection method and system based on big data

Abstract

The invention relates to the technical field of modern industrial manufacturing and equipment maintenance, and discloses a power distribution equipment connector quality detection method and system based on big data, historical operation data is analyzed to obtain preliminary fault mode classification, and a multi-dimensional analysis model is constructed in combination with real-time operation parameters and environmental factors, so that a power distribution equipment connector quality detection result is obtained. And extracting a core feature vector of an equipment operation state, predicting a potential fault, verifying the reliability of an early warning signal, dynamically tracking a high-risk state, analyzing the operation state in real time, and finally generating a fault early warning result and a maintenance scheduling instruction. Potential fault risks of the equipment can be effectively identified, early warning is performed in advance, maintenance measures are taken, the operation reliability and the maintenance efficiency of the equipment are improved, and the fault occurrence probability and the maintenance cost are reduced.

Description

Technical Field

[0001] This invention relates to the technical field of modern industrial manufacturing and equipment maintenance, and in particular to a method and system for quality inspection of power distribution equipment connectors based on big data. Background Technology

[0002] In modern industrial manufacturing and equipment maintenance, ensuring the stability and reliability of equipment operation is of paramount importance. This area directly relates to production efficiency and safety, and is one of the core pillars of industrial development. Currently, many solutions, when faced with complex equipment failure prediction, often rely on a single data source or simple analytical methods, making it difficult to comprehensively capture the underlying causes of failures. This results in insufficient prediction accuracy and often misses crucial early warning opportunities.

[0003] This field currently faces multiple challenges in its existing technologies. The most pressing is extracting valuable failure modes from massive amounts of historical data. Due to the complexity and diversity of data sources, relying solely on traditional methods makes it difficult to effectively integrate and uncover hidden patterns. Extending further to the dynamic monitoring of real-time operating parameters, equipment operating status is significantly affected by environmental factors. Without incorporating these external conditions into the analytical framework, it is difficult to construct a comprehensive feature system to reflect the true condition of the equipment. Simultaneously, the lack of accurate identification capabilities for potential failure modes currently only allows for reactive responses after failures occur, rather than proactive preventative measures, resulting in persistently high costs and risks associated with equipment maintenance.

[0004] Therefore, how to build a multi-dimensional analysis space by integrating historical data, real-time parameters, and environmental factors, and accurately identify potential failure modes through advanced means, has become a key issue in improving equipment maintenance efficiency and reliability. Summary of the Invention

[0005] This invention provides a method and system for quality inspection of power distribution equipment connectors based on big data, so as to improve equipment maintenance efficiency and reliability.

[0006] Firstly, to address the aforementioned technical problems, this invention provides a method for quality inspection of power distribution equipment connectors based on big data, comprising: Equipment operation records are obtained from stored historical data to obtain preliminary failure mode classification results and extract features to determine the association feature set between equipment operating status and failure mode; Environmental factors are acquired, and the environmental factors are fused with the associated feature set to obtain a comprehensive influence factor matrix. Dimensionality reduction is then performed to obtain the core feature vector of the equipment's operating status. Based on the core feature vector, potential fault modes are predicted. If the probability of anomalies in the prediction results exceeds a preset probability threshold, the core feature vector is marked as a high-risk state, and a warning signal for potential faults is obtained. The reliability of the warning signal is verified, it is determined whether the warning signal conforms to the characteristics of historical fault modes, the fault risk level is obtained, and the operating status of the equipment is continuously tracked. If the tracking result shows that the risk level continues to rise, an adjusted monitoring frequency strategy is generated, and a high-frequency monitoring period is determined. Data is collected and short-term fluctuation characteristics are extracted during the high-frequency monitoring period to determine whether there are signs of an impending fault, obtain the final fault warning result, generate scheduling instructions for equipment maintenance, and update the fault mode database according to the scheduling instructions.

[0007] Preferably, the step of obtaining equipment operation records from stored historical data to obtain preliminary fault mode classification results includes: The raw data of equipment operation records is obtained from the stored historical data. According to different equipment types and operating scenarios, the raw data is preprocessed using data cleaning tools to remove noise and redundant information, resulting in a cleaned equipment operation dataset. The cleaned equipment operation dataset is grouped and classified according to the equipment type and the operation scenario to determine the preliminary fault mode classification results.

[0008] Preferably, the step of acquiring environmental factors, and fusing the environmental factors with the associated feature set to obtain a comprehensive influence factor matrix, includes: External conditions are monitored in real time, and the environmental conditions are classified according to the equipment type and the operating scenario during monitoring to obtain classified environmental condition records. The classified environmental conditions are associated with the operating load and parameter fluctuations. During the mapping process, the influence of the external conditions is taken into account to determine a preliminary set of condition influences. If one of the indicators in the preliminary set of conditions exceeds a preset indicator threshold, the set of conditions is matched with a pre-established feature library to obtain the core influence elements related to the comprehensive influence factor matrix. The core influencing elements are fused with the associated feature set to obtain the final comprehensive influence factor matrix.

[0009] Preferably, the step of performing dimensionality reduction processing to obtain the core feature vector of the device's operating status includes: The external conditions and the operating load are classified to obtain a classified environmental impact dataset; The environmental impact dataset is associated with records of parameter fluctuations and data distribution. During the mapping process, the impact weights are combined to determine a preliminary set of dimensional features. If any of the indicators in the preliminary dimensional feature set exceeds the preset indicator threshold, the preliminary dimensional feature set is matched with the pre-established feature library to obtain the core dimensional features related to the state assessment. The core dimension features are integrated with the influence weights and the state assessment to obtain the core feature vector of the device's operating state.

[0010] Preferably, the step of predicting potential fault modes based on the core feature vector, and marking the core feature vector as a high-risk state if the anomaly probability in the prediction result exceeds a preset probability threshold, and obtaining a warning signal for potential faults, includes: The core feature vectors of the different equipment types are classified, and environmental impact records are combined during classification to obtain a preliminary feature set related to the fault mode. If the probability of anomalies in the preliminary feature set exceeds a preset probability threshold, the preliminary feature set is compared with a pre-established parameter fluctuation database to determine the specific feature combination that meets the high-risk state. The specific features are combined with the historical records of the state assessment and processed in conjunction with the weight information of the environmental impact to obtain a prediction result that matches the failure mode. The prediction results are used to mark specific characteristics of the high-risk state. When marking, the triggering conditions of the warning signal are combined to determine the final warning signal for the potential fault.

[0011] Preferably, verifying the reliability of the warning signal, determining whether the warning signal conforms to historical fault mode characteristics, and obtaining the fault risk level includes: The matching degree between the warning signal and the fault mode is compared. During the comparison, the influence weight of environmental variables is combined to determine whether the signal strength meets the preset strength threshold, and a preliminary reliability verification result is obtained. If the reliability verification result exceeds the preset reliability threshold, feature fragments related to the failure mode are extracted from the historical records. During extraction, the record integrity constraint is combined to obtain a set of reference modes that match the current signal. Based on the reference pattern set, the signal strength is correlated with the risk level. During the processing, the distribution characteristics of the pattern matching degree are combined to obtain the fault risk level classification.

[0012] Preferably, the continuous tracking of the equipment's operating status, and if the tracking result indicates a continuously rising risk level, generates an adjusted monitoring frequency strategy and determines high-frequency monitoring time periods, including: Based on the fault risk level, the equipment operating parameters are continuously tracked, and during the tracking, the difference between the real-time data and the environmental impact is combined to obtain the change trend related to the fluctuation of the parameters. If the trend shows that the risk increases beyond the preset threshold, the parameter fluctuations are matched with historical records. During the matching process, the logical constraints of the dynamic programming are combined to determine the key time periods in which the monitoring frequency needs to be adjusted. Based on the key time periods, an adjusted monitoring frequency strategy is generated. During the generation process, the correspondence between the fault risk level and the time period is combined to obtain a time arrangement suitable for high-frequency monitoring. The high-frequency monitoring tasks are assigned to the corresponding monitoring nodes through the aforementioned time arrangement. The final monitoring execution plan is determined by taking into account the real-time fluctuations of the operating parameters during the assignment.

[0013] Preferably, the step of collecting data and extracting short-term fluctuation characteristics during the high-frequency monitoring period, determining whether there are signs of an impending fault, and obtaining the final fault warning result includes: Based on the data collected during the high-frequency monitoring period, the data related to the operating status is cleaned and classified to obtain the original fluctuation characteristics related to the short-term fluctuations, and a set of fluctuation characteristics for subsequent processing is obtained. Based on the set of fluctuation features, the data within the high-frequency monitoring period is segmented to obtain short-term fluctuation changes related to the equipment status. If the magnitude of the change exceeds a preset magnitude threshold, it is determined that there is a need for anomaly detection, and a subset of fluctuation features is identified. Based on the aforementioned subset of fluctuation characteristics, the device status is analyzed in real time to obtain the distribution of anomalies related to fault signs, determine whether there are potential fault signs, and obtain a detailed record of the anomaly distribution. Based on the detailed records of the anomaly distribution, the anomalies are correlated and matched with the early warning results to determine whether there are any signs of an impending fault, and to determine the final fault early warning result.

[0014] Preferably, the step of generating scheduling instructions for equipment maintenance and updating the fault mode database according to the scheduling instructions includes: Based on the fault warning results, the equipment status monitoring data is classified and processed to extract information related to maintenance scheduling, generate corresponding control instructions, and obtain a set of scheduling instructions for subsequent processing. The set of scheduling instructions is sent to the target control unit, a receipt confirmation message is obtained, the operation record during the transmission process is recorded, and the integrity of the transmission process is determined. If the received confirmation information shows that the transmission was successful, the feedback information is sorted out, response data related to the device status is extracted, and it is determined whether the feedback information meets the preset threshold requirements to obtain the sorted feedback dataset. The feedback dataset is compared with historical operation records, and the relevant entries in the fault mode database are updated as reference data for subsequent optimization.

[0015] Secondly, the present invention provides a power distribution equipment connector quality inspection system based on big data, comprising: The first determination module is used to obtain equipment operation records from stored historical data, obtain preliminary fault mode classification results and extract features, and determine the association feature set between equipment operation status and fault mode; The first acquisition module is used to acquire environmental factors, fuse the environmental factors with the associated feature set to obtain a comprehensive influence factor matrix and perform dimensionality reduction processing to obtain the core feature vector of the equipment operating status; The second acquisition module is used to predict potential fault modes based on the core feature vector. If the abnormal probability in the prediction result exceeds a preset probability threshold, the core feature vector is marked as a high-risk state, and a warning signal for potential faults is obtained. The second determining module is used to verify the reliability of the warning signal, determine whether the warning signal conforms to the characteristics of historical fault modes, obtain the fault risk level, and continuously track the operating status of the equipment. If the tracking result is that the risk level continues to rise, an adjusted monitoring frequency strategy is generated to determine the high-frequency monitoring time period. The update module is used to collect data and extract short-term fluctuation characteristics during the high-frequency monitoring period, determine whether there are signs of an impending fault, obtain the final fault warning result, generate scheduling instructions for equipment maintenance, and update the fault mode database according to the scheduling instructions.

[0016] Compared to existing technologies, this invention discloses a method and system for quality inspection of power distribution equipment connectors based on big data. It analyzes historical operating data to obtain preliminary fault mode classification, and then combines real-time operating parameters and environmental factors to construct a multi-dimensional analysis model. This model extracts core feature vectors of the equipment's operating status, predicts potential faults, verifies the reliability of early warning signals, dynamically tracks high-risk states, and analyzes the operating status in real time. Finally, it generates fault warning results and maintenance scheduling instructions. This invention can effectively identify potential equipment fault risks, provide early warnings, and take maintenance measures, improving equipment operating reliability and maintenance efficiency, and reducing the probability of fault occurrence and maintenance costs. Attached Figure Description

[0017] Figure 1 This is a flowchart of a method for quality inspection of power distribution equipment connectors based on big data, provided in the first embodiment of the present invention. Figure 2 This is a schematic diagram of a power distribution equipment connector quality inspection system based on big data, provided in the second embodiment of the present invention. Detailed Implementation

[0018] 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.

[0019] Reference Figure 1 The first embodiment of the present invention provides a flowchart of a method for quality inspection of power distribution equipment connectors based on big data, including the following steps: S11. Obtain equipment operation records from stored historical data, obtain preliminary fault mode classification results and extract features to determine the association feature set between equipment operating status and fault mode; S12, obtain environmental factors, fuse the environmental factors with the associated feature set to obtain a comprehensive influence factor matrix and perform dimensionality reduction processing to obtain the core feature vector of the equipment operating status; S13, predict potential fault modes based on the core feature vector. If the abnormal probability in the prediction result exceeds a preset probability threshold, mark the core feature vector as a high-risk state and obtain a warning signal for potential faults. S14, verify the reliability of the warning signal, determine whether the warning signal conforms to the characteristics of historical fault modes, obtain the fault risk level, and continuously track the operating status of the equipment. If the tracking result is that the risk level continues to rise, generate an adjusted monitoring frequency strategy and determine the high-frequency monitoring time period. S15, during the high-frequency monitoring period, data is collected and short-term fluctuation characteristics are extracted to determine whether there are signs of an impending fault, the final fault warning result is obtained and a scheduling instruction for equipment maintenance is generated, and the fault mode database is updated according to the scheduling instruction.

[0020] In step S11, equipment operation records are obtained from the stored historical data to obtain preliminary fault mode classification results and extract features to determine the association feature set between equipment operating status and fault mode.

[0021] The process of retrieving equipment operation records from stored historical data to obtain preliminary fault mode classification results includes: The raw data of equipment operation records is obtained from the stored historical data. According to different equipment types and operating scenarios, the raw data is preprocessed using data cleaning tools to remove noise and redundant information, resulting in a cleaned equipment operation dataset. The cleaned equipment operation dataset is grouped and classified according to the equipment type and the operation scenario to determine the preliminary fault mode classification results. The process then includes: matching the preliminary fault mode classification results with a pre-established fault mode library; if the matching degree is lower than a preset matching threshold, regrouping is performed to obtain an adjusted fault mode classification set. Based on the adjusted fault mode classification set, a fault mode distribution chart corresponding to the different equipment types and the operating scenarios is generated.

[0022] For example, in processing equipment operation records, assuming the data pertains to industrial machinery such as injection molding machines and stamping presses, historical data includes parameters such as equipment operating time, temperature, and pressure. First, the application of data cleaning tools is crucial. The raw data may contain outliers caused by sensor malfunctions, such as a sudden temperature jump to 1000 degrees Celsius, significantly exceeding the normal range of 30 to 80 degrees Celsius. Cleaning tools can set thresholds to filter outliers and remove duplicate records. For example, multiple samples from the same time point are retained, only the first record is kept, resulting in a more accurate equipment operation dataset. This step effectively improves the reliability of subsequent analysis.

[0023] Specifically, when using the cleaned dataset for cluster analysis, it can be grouped based on features such as equipment type and operating scenario. Taking injection molding machines as an example, operating scenarios may be divided into two modes: high-temperature, high-pressure molding and low-temperature, low-pressure testing. The clustering tool will divide the data into two groups based on temperature and pressure parameters, initially identifying potential overload failure modes under high-temperature, high-pressure scenarios and abnormal startup modes under low-temperature, low-pressure scenarios. Such grouping helps to analyze the fault characteristics under different scenarios in a targeted manner, improving the accuracy of fault prediction.

[0024] For example, from multiple perspectives, the methods described above support each other in the stages of data cleaning, cluster analysis, matching and adjustment, and visualization, forming a complete technology chain. Cleaning ensures data quality, clustering and matching provide accurate classification, and visualization presents the results intuitively, collectively contributing to improved fault prediction and equipment management efficiency. Especially in industrial scenarios, reducing equipment downtime and lowering maintenance costs are direct and beneficial effects, demonstrating the practical value of the technology's application.

[0025] Information related to the device status is obtained from real-time operating parameters. For different device types and operating scenarios, the sensor data is processed using data acquisition tools to obtain preliminary parameter fluctuation records. Based on these preliminary records, time series processing tools are used to segment the records, extracting fluctuation features to determine combinations of fluctuation features related to the device status. If any feature value in a combination exceeds a preset threshold, a feature comparison tool is used to match the combination with a pre-established fault mode library to identify potential associated feature sets. Based on these associated feature sets, a data integration tool is used to fuse the sets with the fault mode classification results, incorporating background information on device type and operating scenario to obtain the final association feature results between the operating status and the fault mode.

[0026] For example, in industrial production, the acquisition and processing of real-time operating parameters for mechanical equipment such as injection molding machines allows for in-depth analysis of equipment status information from multiple perspectives. Firstly, regarding the application of data acquisition tools, assuming the equipment monitors parameters such as temperature, pressure, and vibration frequency in real time through sensors, the acquisition tool will organize this data into preliminary parameter fluctuation records by timestamp. For the high-temperature, high-pressure molding scenario of an injection molding machine, the acquisition tool will record the temperature change every minute. Assuming the normal range is 50 to 80 degrees Celsius, and the temperature frequently fluctuates to 90 degrees Celsius within a certain period, this record provides the basic data for subsequent analysis.

[0027] For example, in the segmentation process of time series processing tools, the preliminary records mentioned above can be segmented by hour or operating cycle to extract fluctuation characteristics. Suppose that in a one-hour data segment, temperature fluctuations exceeding 10 degrees Celsius occur five times, or the pressure value rises from 200 bar to 300 bar in a short period. The processing tool will mark these abnormal fluctuations as feature points, forming a combination of fluctuation characteristics related to equipment status. This process helps to focus on key anomalies, providing a basis for subsequent judgments.

[0028] For example, in the fusion stage of the associated feature set, data integration tools are used to combine the comparison results with the failure mode classification results, incorporating background information such as equipment type and operating scenario. Suppose that an injection molding machine operating under high temperature and high pressure conditions is classified as an overload risk due to abnormal temperature and pressure fluctuations. If the classification results also contain similar features, the fusion process ultimately confirms that this operating state is highly correlated with the overload failure mode. This fusion process integrates multi-dimensional information, ensuring the comprehensiveness of the judgment.

[0029] For example, from another perspective, the role of background information regarding the operating scenario in the fusion process can be understood as further calibration of the equipment's status. Suppose that in a low-temperature, low-pressure testing scenario, temperature fluctuations are small, but the vibration frequency is abnormal. By integrating the scenario information with the fusion tool, the abnormal vibration can be avoided from being misjudged as overload, and instead, it can be more likely to be a startup issue. This calibration effectively reduces misjudgments and improves the relevance of the analysis.

[0030] For example, in the overall process, the implementation methods for each technical topic are interconnected, forming a complete chain from data collection to feature extraction, comparison and matching, and then fusion analysis. For equipment like injection molding machines, the processing of real-time parameters and the correlation results of failure modes can provide maintenance teams with timely references and optimize equipment management strategies. This method has significant practical value in industrial scenarios.

[0031] In step S12, environmental factors are obtained, and the environmental factors are fused with the associated feature set to obtain a comprehensive influence factor matrix. Dimensionality reduction processing is then performed to obtain the core feature vector of the equipment's operating status.

[0032] The process of acquiring environmental factors involves fusing these factors with the associated feature set to obtain a comprehensive influence factor matrix, including: External conditions are monitored in real time, and the environmental conditions are classified according to the equipment type and the operating scenario during monitoring to obtain classified environmental condition records. The classified environmental conditions are associated with the operating load and parameter fluctuations. During the mapping process, the influence of the external conditions is taken into account to determine a preliminary set of condition influences. If one of the indicators in the preliminary set of conditions exceeds a preset indicator threshold, the set of conditions is matched with a pre-established feature library to obtain the core influence elements related to the comprehensive influence factor matrix. The core influencing elements are fused with the associated feature set to obtain the final comprehensive influence factor matrix.

[0033] For example, in the field of industrial equipment monitoring, when monitoring data related to ambient temperature, humidity levels, and vibration frequency in real time, data acquisition tools can be used to meticulously record external conditions. Taking injection molding machines as an example, assuming a high-temperature, high-pressure molding environment, ambient temperature may affect heat dissipation, humidity may cause electrical components to become damp, and vibration frequency may reflect external interference or instability of the equipment itself. Data acquisition tools will classify and process the collected data according to equipment type and scenario differences. For example, temperature data will be divided into high-temperature and normal-temperature environments, and humidity data will be distinguished into high-humidity and low-humidity, forming a categorized record of environmental conditions. This classification method helps to more accurately identify influencing factors in subsequent analysis.

[0034] For example, in the data integration phase, data integration tools are used to correlate and map the categorized environmental condition records with operating load and parameter fluctuation data. Suppose that the injection molding machine operates under high load for a certain period, the ambient temperature record shows 35 degrees Celsius, exceeding the normal range of 30 degrees Celsius, and the parameter fluctuation shows frequent pressure fluctuations. Combined with the influence of external disturbances, such as heat accumulation caused by the operation of other equipment in the workshop, it is preliminarily determined that the temperature increase has a direct impact on load and pressure fluctuations, forming a preliminary set of conditional influences. This process helps to identify the potential connections between environmental factors and equipment operating status.

[0035] For example, from another perspective, applying background information based on scene differences can further calibrate the analysis results. Suppose that in a low-temperature, low-humidity environment, the vibration frequency is abnormal, but the temperature and humidity are within the normal range. By incorporating scene information during fusion, the vibration anomaly can be attributed to external interference rather than internal equipment problems. This calibration method effectively improves the relevance of the analysis and reduces the possibility of misjudgment.

[0036] For example, in the overall process, from data collection to integration, comparison, and fusion, the implementation methods of each technical topic are closely linked. Taking injection molding machines as an example, the classification and recording of environmental conditions provides the foundation for subsequent mapping, while correlation mapping and feature comparison gradually focus on core influencing elements, ultimately merging to form a comprehensive matrix output. This complete chain can provide multi-dimensional reference for equipment management and optimize monitoring strategies.

[0037] The dimensionality reduction process, which obtains the core feature vector of the device's operating state, includes: The external conditions and the operating load are classified to obtain a classified environmental impact dataset; The environmental impact dataset is associated with records of parameter fluctuations and data distribution. During the mapping process, the impact weights are combined to determine a preliminary set of dimensional features. If any of the indicators in the preliminary dimensional feature set exceeds the preset indicator threshold, the preliminary dimensional feature set is matched with the pre-established feature library to obtain the core dimensional features related to the state assessment. The core dimension features are integrated with the influence weights and the state assessment to obtain the core feature vector of the device's operating state.

[0038] For example, in the field of industrial equipment monitoring, the operational status analysis of injection molding machines can be conducted by recording ambient temperature, humidity levels, and vibration frequencies in real time to deeply explore the impact of external conditions on the equipment. Regarding the use of data acquisition tools, assuming a high-temperature, high-pressure injection molding workshop, the tools would categorize temperature data into high-temperature ranges (e.g., 35 degrees Celsius) and normal-temperature ranges (e.g., 25 degrees Celsius) based on equipment type and scenario differences; humidity data would be differentiated into high humidity (e.g., 80%) and low humidity (e.g., 40%); and vibration frequencies would be classified into high-frequency (e.g., 50 Hz) and low-frequency (e.g., 20 Hz). This classification provides a clear data foundation for subsequent analysis, ensuring that the effects of different environmental conditions are accurately captured.

[0039] For example, if an indicator in the initial feature set exceeds a threshold, a feature comparison tool is used to match it with a pre-established feature library. Assuming the temperature threshold is set to 32 degrees Celsius, and the actual recorded temperature is 36 degrees Celsius, the feature library stores typical features of equipment overheating under high-temperature conditions. After comparison, temperature is determined to be a core feature related to condition assessment. This comparison process effectively filters out factors that significantly impact equipment operation, providing a clear basis for subsequent analysis.

[0040] For example, in the data fusion phase, data fusion tools are used to integrate core dimensional features with influencing weights and background information for status assessment. Assuming a high-temperature scenario, temperature is identified as the core feature with a weight of 60%. Combined with pressure fluctuation data and workshop background information, such as heat accumulation, the integrated data forms an operating status feature vector reflecting the correlation between temperature and pressure fluctuations. This integration method allows for multi-dimensional analysis of equipment status, ensuring results are more realistic.

[0041] For example, from another perspective, when applying background information based on scene differences, if the vibration frequency is abnormal (e.g., 60 Hz) in a low-temperature, low-humidity environment, but the temperature and humidity are normal, combining the background information during data fusion can attribute the vibration anomaly to external interference. This calibration improves the specificity of the analysis and reduces the possibility of misjudgment. Through the above multi-faceted analysis, from data acquisition to fusion, each step is closely linked, providing reliable references for equipment management and optimizing monitoring strategies.

[0042] In step S13, the potential fault mode is predicted based on the core feature vector. If the abnormal probability in the prediction result exceeds the preset probability threshold, the core feature vector is marked as a high-risk state, and a warning signal for the potential fault is obtained.

[0043] The step of predicting potential fault modes based on the core feature vector, and marking the core feature vector as a high-risk state if the anomaly probability in the prediction result exceeds a preset probability threshold, and obtaining a warning signal for potential faults, includes: The core feature vectors of the different equipment types are classified, and environmental impact records are combined during classification to obtain a preliminary feature set related to the fault mode. If the probability of anomalies in the preliminary feature set exceeds a preset probability threshold, the preliminary feature set is compared with a pre-established parameter fluctuation database to determine the specific feature combination that meets the high-risk state. The specific features are combined with the historical records of the state assessment and processed in conjunction with the weight information of the environmental impact to obtain a prediction result that matches the failure mode. The prediction results are used to mark specific characteristics of the high-risk state. When marking, the triggering conditions of the warning signal are combined to determine the final warning signal for the potential fault.

[0044] For example, in the field of industrial equipment monitoring, the analysis of the operating status of injection molding machines can explore how to identify potential faults through data comparison and feature mapping from multiple perspectives. Firstly, regarding the classification of different equipment types and the acquisition of a preliminary feature set based on environmental impact records, this can be achieved through hierarchical analysis of the environmental data during injection molding machine operation. Assuming a high-temperature workshop with an recorded temperature of 38 degrees Celsius, humidity of 75%, and vibration frequency of 45 Hz, these data are combined with the operating characteristics of equipment types such as high-speed and low-speed injection molding machines during classification. This allows for a preliminary assessment that high temperature may be the primary factor causing anomalies, forming a preliminary feature set. This process focuses on the interaction between environmental and equipment characteristics, laying the foundation for subsequent analysis.

[0045] For example, in the application of data integration tools, when associating specific feature combinations with historical status assessment records, environmental impact weighting information can be incorporated. Assuming temperature has a weight of 55% and vibration frequency a weight of 30%, the integration process reveals historical records of equipment overload shutdowns under similar high-temperature conditions. The prediction results show a high degree of match between the current state and the failure mode. This integration process, through weight allocation and historical correlation, enhances the targeting of predictions and provides a basis for fault early warning.

[0046] For example, when using signal generation tools to mark high-risk status characteristics, warning signal trigger conditions can be combined. Assuming a temperature exceeding 35 degrees Celsius for more than 2 hours is a trigger condition, and the current record shows 38 degrees Celsius for 3 hours, a warning signal is generated and marked as potential fault information. This marking method, by clearly defining the conditions, ensures that warning information is accurately transmitted to equipment managers, facilitating timely action. Through the above multi-faceted analysis, from classification and processing to signal generation, each step is closely linked, jointly supporting the determination of potential fault warning information. This method can effectively identify risk points in the operation of injection molding machines, providing accurate references for equipment maintenance and reducing the possibility of sudden failures.

[0047] In step S14, the reliability of the warning signal is verified, it is determined whether the warning signal conforms to the characteristics of historical fault modes, the fault risk level is obtained, and the operating status of the equipment is continuously tracked. If the tracking result shows that the risk level continues to rise, an adjusted monitoring frequency strategy is generated, and a high-frequency monitoring time period is determined.

[0048] The process of verifying the reliability of the warning signal, determining whether the warning signal conforms to historical fault mode characteristics, and obtaining the fault risk level includes: The matching degree between the warning signal and the fault mode is compared. During the comparison, the influence weight of environmental variables is combined to determine whether the signal strength meets the preset strength threshold, and a preliminary reliability verification result is obtained. If the reliability verification result exceeds the preset reliability threshold, feature fragments related to the failure mode are extracted from the historical records. During extraction, the record integrity constraint is combined to obtain a set of reference modes that match the current signal. Based on the reference pattern set, the signal strength is correlated with the risk level. During the processing, the distribution characteristics of the pattern matching degree are combined to obtain the fault risk level classification.

[0049] For example, in the field of industrial equipment monitoring, analyzing the operating status of injection molding machines can be explored from multiple perspectives, examining how data integration and comparison can identify potential risks. Firstly, aligning historical and real-time data along their timelines can be achieved using data integration tools. Assume real-time data originates from the injection molding machine's sensors, collecting temperature and vibration data every 5 minutes, while historical data is collected every 10 minutes from workshop logs. During integration, the tool aligns the two datasets by timestamp, unifying them into a single data point every 5 minutes, and interpolating to fill in any missing data, forming a comprehensive dataset. This process ensures data consistency, providing a reliable basis for subsequent comparisons.

[0050] For example, probabilistic calculation tools can be used to analyze the matching degree between warning signals and fault modes in a comprehensive dataset. Assuming that temperature has a weight of 60% and humidity has a weight of 40% among environmental variables, and the current warning signal shows a temperature of 39 degrees Celsius and humidity of 80%, comparing historical fault modes reveals that equipment under similar conditions has experienced overheating risks, resulting in a matching degree of 82%, exceeding the preset threshold of 75%. This analysis, combined with weight allocation, highlights the impact of key environmental factors, ensuring more accurate judgment of signal strength.

[0051] For example, to classify the risk level of a reference pattern set, a classification mapping tool can be used to associate signal strength with risk level. Assuming signal strength is divided into high, medium, and low levels, and the current signal matching degree is 82%, the distribution characteristics show that it is close to historical high-risk cases; the tool classifies it as a high-risk case. This process, through pattern distribution characteristics, ensures the detail of risk assessment and provides clear guidance for subsequent maintenance decisions. Through the above multi-faceted analysis, from data integration to risk classification, each link is closely connected, jointly supporting the accurate identification of potential injection molding machine faults. This method can effectively improve the accuracy of early warning and provide strong support for equipment management.

[0052] The continuous monitoring of the equipment's operating status, if the monitoring result indicates a continuously rising risk level, generates an adjusted monitoring frequency strategy and determines high-frequency monitoring time periods, including: Based on the fault risk level, the equipment operating parameters are continuously tracked, and during the tracking, the difference between the real-time data and the environmental impact is combined to obtain the change trend related to the fluctuation of the parameters. If the trend shows that the risk increases beyond the preset threshold, the parameter fluctuations are matched with historical records. During the matching process, the logical constraints of the dynamic programming are combined to determine the key time periods in which the monitoring frequency needs to be adjusted. Based on the key time periods, an adjusted monitoring frequency strategy is generated. During the generation process, the correspondence between the fault risk level and the time period is combined to obtain a time arrangement suitable for high-frequency monitoring. The high-frequency monitoring tasks are assigned to the corresponding monitoring nodes through the aforementioned time arrangement. The final monitoring execution plan is determined by taking into account the real-time fluctuations of the operating parameters during the assignment.

[0053] For example, in the field of industrial equipment monitoring, the tracking and risk management of injection molding machine operation status can be explored from multiple perspectives. This involves using data tools and strategy adjustments to identify signs of rising risks and optimize monitoring solutions. Firstly, for continuous tracking of equipment operating parameters, key parameters of the injection molding machine, such as pressure, speed, and temperature, can be acquired in real time using data acquisition tools. Assuming pressure data is collected every minute, while simultaneously recording changes in ambient humidity, the tool will compare this data with historical environmental conditions to analyze abnormal trends in pressure fluctuations when humidity increases. This approach helps to identify potential problems early.

[0054] For example, to match parameter fluctuations with historical records, data comparison tools can be used to analyze fluctuation characteristics. Suppose the current pressure value rises from the normal range of 50 units to 65 units within a short period, the tool will search for records of similar fluctuations leading to malfunctions over the past year, finding multiple cases where equipment malfunctioned when pressure exceeded 60 units. Through dynamic programming logic, the tool will identify the period of most significant pressure fluctuations—between 8 AM and 10 AM—as the key timeframe. This method ensures the rational allocation of monitoring resources.

[0055] For example, when generating an adjusted monitoring frequency strategy, a frequency scheduling tool can be used to allocate time slots based on risk levels. Assuming the current risk level is high, the tool will increase the monitoring frequency during key time periods from once per minute to once every 30 seconds, while maintaining a frequency of once every 5 minutes during non-key time periods. This strategy can more promptly capture abnormal changes during high-risk periods. Furthermore, by combining the correspondence between time periods and risk levels, the tool will also prioritize high-frequency monitoring during peak equipment operation periods to ensure maximum resource utilization.

[0056] For example, for the allocation of high-frequency monitoring tasks, a data distribution tool can be used to distribute tasks to specific monitoring nodes. Assuming there are five monitoring nodes in the factory, the tool will prioritize assigning tasks to nodes 1 and 2, which cover critical equipment areas, based on real-time fluctuation data, while ensuring that other nodes maintain basic monitoring. This allocation method can dynamically adapt to changes in equipment status, improving the targeting of data collection. Through the above multi-faceted analysis, from parameter tracking to task allocation, each link is closely connected, jointly supporting the optimization of injection molding machine risk management. This method can respond quickly when initial signs of rising risk appear, ensuring stable equipment operation.

[0057] In step S15, data is collected and short-term fluctuation characteristics are extracted during the high-frequency monitoring period to determine whether there are signs of an impending fault, obtain the final fault warning result, generate a scheduling instruction for equipment maintenance, and update the fault mode database according to the scheduling instruction.

[0058] The process of collecting data and extracting short-term fluctuation characteristics during the high-frequency monitoring period to determine whether there are signs of an impending fault, and obtaining the final fault warning result, includes: Based on the data collected during the high-frequency monitoring period, the data related to the operating status is cleaned and classified to obtain the original fluctuation characteristics related to the short-term fluctuations, and a set of fluctuation characteristics for subsequent processing is obtained. Based on the set of fluctuation features, the data within the high-frequency monitoring period is segmented to obtain short-term fluctuation changes related to the equipment status. If the magnitude of the change exceeds a preset magnitude threshold, it is determined that there is a need for anomaly detection, and a subset of fluctuation features is identified. Based on the aforementioned subset of fluctuation characteristics, the device status is analyzed in real time to obtain the distribution of anomalies related to fault signs, determine whether there are potential fault signs, and obtain a detailed record of the anomaly distribution. Based on the detailed records of the anomaly distribution, the anomalies are correlated and matched with the early warning results to determine whether there are any signs of an impending fault, and to determine the final fault early warning result.

[0059] For example, in the field of industrial equipment monitoring, the analysis of the operating status and risk warning of injection molding machines can be explored from multiple perspectives. How can data processing and feature analysis be used to identify anomalies and support early warning judgments? Firstly, for the cleaning and classification of data collected during high-frequency monitoring periods, data preprocessing tools can be used to process the operating data of the injection molding machine. Assuming that pressure and temperature data are collected every 30 seconds during the key monitoring period from 8:00 AM to 10:00 AM, the tool will first remove obviously abnormal noise data, such as records where pressure values ​​suddenly jump to unreasonable ranges. Then, the data will be classified into pressure and temperature groups according to time sequence, forming a structured dataset that facilitates subsequent analysis.

[0060] For example, segmenting the set of fluctuation characteristics can be achieved using feature extraction tools combined with sliding window techniques for data analysis. Assuming a 5-minute window unit, the tool will extract the amplitude and frequency characteristics of pressure data segment by segment, observing whether there are significant changes in the short term. If the pressure value rapidly rises from 50 units to 62 units within a window, exceeding a preset threshold of 10 units, the tool will mark that data segment as an abnormal demand interval, generating a subset of fluctuation characteristics that requires further analysis. This segmentation method helps to focus on short-term fluctuation changes.

[0061] For example, in anomaly detection tools, real-time analysis of subsets of fluctuation characteristics can identify signs of malfunction. Suppose the tool detects that the distribution of abnormal pressure fluctuations is concentrated around 9 AM within a certain time period, and the fluctuation values ​​repeatedly exceed 60 units. Combining this with historical data showing similar distributions leading to equipment overload, the tool will determine that there are potential signs of malfunction and generate a detailed record of the anomaly distribution. This real-time analysis provides data support for subsequent early warning systems.

[0062] The process of generating scheduling instructions for equipment maintenance and updating the fault mode database based on the scheduling includes: Based on the fault warning results, the equipment status monitoring data is classified and processed to extract information related to maintenance scheduling, generate corresponding control instructions, and obtain a set of scheduling instructions for subsequent processing. The set of scheduling instructions is sent to the target control unit, a receipt confirmation message is obtained, the operation record during the transmission process is recorded, and the integrity of the transmission process is determined. If the received confirmation information shows that the transmission was successful, the feedback information is sorted out, response data related to the device status is extracted, and it is determined whether the feedback information meets the preset threshold requirements to obtain the sorted feedback dataset. The feedback dataset is compared with historical operation records, and the relevant entries in the fault mode database are updated as reference data for subsequent optimization.

[0063] For example, in the business area of ​​equipment condition monitoring, the processing of fault early warning results can be analyzed and implemented in detail from multiple perspectives. First, regarding the application of instruction generation tools, the principle is to extract information related to maintenance scheduling by classifying and processing monitoring data.

[0064] For example, in an industrial equipment monitoring scenario, suppose a device triggers a high-temperature warning during a high-frequency operating period. The instruction generation tool will categorize the temperature data, operating time, and other information into the "urgent maintenance required" category and generate control instructions to reduce load or shut down the device, forming a set of scheduling instructions. This classification process can quickly respond to potential risks and ensure equipment safety.

[0065] Specifically, in terms of data transmission interfaces, the set of scheduling instructions is transmitted over a network to the target control unit, such as a control terminal located in a factory workshop. During transmission, the system records the sending time, receiving time, and possible delays for each instruction. Assuming a transmission time of 0.5 seconds and a successful reception confirmation message, the integrity of the transmission can be verified. This recording method helps to trace the root cause of subsequent problems and improves system reliability.

[0066] For example, in the application of information integration tools, the organization of feedback information is a crucial step. Suppose that after the control unit executes a command, it returns equipment status data, such as the temperature dropping to the normal range of 35 degrees Celsius and the operating vibration frequency meeting a preset threshold. The information integration tool will extract and categorize this data into a "normal response" feedback dataset. This organization method can intuitively reflect the command execution effect, providing a basis for subsequent decision-making.

[0067] Specifically, the use of database management tools primarily involves comparing feedback datasets with historical operational records to update the fault mode database. For example, if historical records show that a similar overheating issue previously caused equipment downtime, and timely intervention prevented the fault this time, the system will update the database with the corresponding fault mode response strategy, recording this successful intervention case. This update mechanism continuously optimizes the fault prediction model, providing more accurate reference data for the future.

[0068] For example, looking at the logical progression from the core solution to the extended solution, the core lies in forming a closed-loop management system through instruction generation, transmission, feedback processing, and database updates. Extended solutions can include adding multi-dimensional data analysis, such as incorporating factors like environmental humidity and equipment aging, to further refine the instruction generation logic. This multi-faceted support approach ensures full-process coverage from early warning to maintenance, improves the intelligence level of equipment management, reduces the probability of failures, and guarantees production continuity.

[0069] In summary, this invention discloses a big data-based method for quality inspection of power distribution equipment connectors. It analyzes historical operating data to obtain preliminary fault mode classification, then combines real-time operating parameters and environmental factors to construct a multi-dimensional analysis model. This model extracts core feature vectors of the equipment's operating status, predicts potential faults, verifies the reliability of early warning signals, dynamically tracks high-risk states, and analyzes the operating status in real time. Finally, it generates fault warning results and maintenance scheduling instructions. This invention can effectively identify potential equipment fault risks, provide early warnings, and take maintenance measures, improving equipment operating reliability and maintenance efficiency, and reducing the probability of fault occurrence and maintenance costs.

[0070] Reference Figure 2 The second embodiment of the present invention provides a big data-based power distribution equipment connector quality inspection system, comprising: The first determining module 201 is used to obtain equipment operation records from stored historical data, obtain preliminary fault mode classification results and extract features, and determine the association feature set between equipment operation status and fault mode; The first acquisition module 202 is used to acquire environmental factors, fuse the environmental factors with the associated feature set to obtain a comprehensive influence factor matrix and perform dimensionality reduction processing to acquire the core feature vector of the equipment operating status. The second acquisition module 203 is used to predict potential fault modes based on the core feature vector. If the abnormal probability in the prediction result exceeds a preset probability threshold, the core feature vector is marked as a high-risk state, and a warning signal for potential faults is obtained. The second determining module 204 is used to verify the reliability of the warning signal, determine whether the warning signal conforms to the characteristics of historical fault modes, obtain the fault risk level, and continuously track the operating status of the equipment. If the tracking result is that the risk level continues to rise, an adjusted monitoring frequency strategy is generated to determine the high-frequency monitoring time period. The update module 205 is used to collect data and extract short-term fluctuation characteristics during the high-frequency monitoring period, determine whether there are signs of an impending fault, obtain the final fault warning result, generate a scheduling instruction for equipment maintenance, and update the fault mode database according to the scheduling instruction.

[0071] It should be noted that the big data-based power distribution equipment connector quality inspection system provided in this embodiment of the invention is used to execute all the process steps of the big data-based power distribution equipment connector quality inspection method in the above embodiment. The working principles and beneficial effects of the two are one-to-one, so they will not be described again.

[0072] This invention also provides an electronic device. The electronic device includes a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a big data-based power distribution equipment connector quality inspection program. When the processor executes the computer program, it implements the steps in the various big data-based power distribution equipment connector quality inspection method embodiments described above, for example... Figure 1 The step S11 shown. Alternatively, when the processor executes the computer program, it implements the functions of each module / unit in the above-described device embodiments, such as the first determining module.

[0073] For example, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the electronic device.

[0074] The electronic device may be a desktop computer, laptop, handheld computer, or smart tablet, etc. The electronic device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. It may include more or fewer components than described above, or combine certain components, or different components. For example, the electronic device may also include input / output devices, network access devices, buses, etc.

[0075] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the electronic device, connecting all parts of the electronic device via various interfaces and lines.

[0076] The memory can be used to store the computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, phonebook, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0077] Wherein, if the modules / units integrated in the electronic device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the methods of the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. It should be noted that the content included in the computer-readable medium can be appropriately added or removed according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electrical carrier signals and telecommunication signals.

[0078] It should be noted that the device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can be specifically implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0079] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A method for quality inspection of power distribution equipment connectors based on big data, characterized in that, include: Equipment operation records are obtained from stored historical data to obtain preliminary failure mode classification results and extract features to determine the association feature set between equipment operating status and failure mode; Environmental factors are acquired, and the environmental factors are fused with the associated feature set to obtain a comprehensive influence factor matrix. Dimensionality reduction is then performed to obtain the core feature vector of the equipment's operating status. Based on the core feature vector, potential fault modes are predicted. If the probability of anomalies in the prediction results exceeds a preset probability threshold, the core feature vector is marked as a high-risk state, and a warning signal for potential faults is obtained. The reliability of the warning signal is verified, it is determined whether the warning signal conforms to the characteristics of historical fault modes, the fault risk level is obtained, and the operating status of the equipment is continuously tracked. If the tracking result shows that the risk level continues to rise, an adjusted monitoring frequency strategy is generated, and a high-frequency monitoring period is determined. Data is collected and short-term fluctuation characteristics are extracted during the high-frequency monitoring period to determine whether there are signs of an impending fault, obtain the final fault warning result, generate scheduling instructions for equipment maintenance, and update the fault mode database according to the scheduling instructions.

2. The method for quality inspection of power distribution equipment connectors based on big data according to claim 1, characterized in that, The process of retrieving equipment operation records from stored historical data to obtain preliminary fault mode classification results includes: The raw data of equipment operation records is obtained from the stored historical data. According to different equipment types and operating scenarios, the raw data is preprocessed using data cleaning tools to remove noise and redundant information, resulting in a cleaned equipment operation dataset. The cleaned equipment operation dataset is grouped and classified according to the equipment type and the operation scenario to determine the preliminary fault mode classification results.

3. The method for quality inspection of power distribution equipment connectors based on big data according to claim 2, characterized in that, The process of acquiring environmental factors involves fusing these factors with the associated feature set to obtain a comprehensive influence factor matrix, including: External conditions are monitored in real time, and the environmental conditions are classified according to the equipment type and the operating scenario during monitoring to obtain classified environmental condition records. The classified environmental conditions are associated with the operating load and parameter fluctuations. During the mapping process, the influence of the external conditions is taken into account to determine a preliminary set of condition influences. If one of the indicators in the preliminary set of conditions exceeds the preset indicator threshold, the set of conditions is matched with a pre-established feature library to obtain the core influence elements related to the comprehensive influence factor matrix. The core influencing elements are fused with the associated feature set to obtain the final comprehensive influence factor matrix.

4. The method for quality inspection of power distribution equipment connectors based on big data according to claim 1, characterized in that, The dimensionality reduction process, which obtains the core feature vector of the device's operating state, includes: External conditions and operational loads are classified to obtain a classified environmental impact dataset. The environmental impact dataset is associated with records of parameter fluctuations and data distribution. During the mapping process, the impact weights are combined to determine a preliminary set of dimensional features. If one of the indicators in the preliminary set of dimensional features exceeds a preset indicator threshold, the preliminary set of dimensional features is matched with a pre-established feature library to obtain core dimensional features related to state assessment. The core dimension features are integrated with the influence weights and the state assessment to obtain the core feature vector of the device's operating state.

5. The method for quality inspection of power distribution equipment connectors based on big data according to claim 2, characterized in that, The step of predicting potential fault modes based on the core feature vector, and marking the core feature vector as a high-risk state if the anomaly probability in the prediction result exceeds a preset probability threshold, and obtaining a warning signal for potential faults, includes: The core feature vectors of the different equipment types are classified, and environmental impact records are combined during classification to obtain a preliminary feature set related to the fault mode. If the probability of anomalies in the preliminary feature set exceeds a preset probability threshold, the preliminary feature set is compared with a pre-established parameter fluctuation database to determine the specific feature combination that meets the high-risk state. The specific feature combination is associated with the historical records of state assessment, and the weight information of the environmental impact is combined during the processing to obtain a prediction result that matches the failure mode. The prediction results are used to mark specific characteristics of the high-risk state. When marking, the triggering conditions of the warning signal are combined to determine the final warning signal for the potential fault.

6. The method for quality inspection of power distribution equipment connectors based on big data according to claim 1, characterized in that, The process of verifying the reliability of the early warning signal, determining whether the early warning signal conforms to historical fault mode characteristics, and obtaining the fault risk level includes: The matching degree between the warning signal and the fault mode is compared. During the comparison, the influence weight of environmental variables is combined to determine whether the signal strength meets the preset strength threshold, and a preliminary reliability verification result is obtained. If the reliability verification result exceeds the preset reliability threshold, feature fragments related to the failure mode are extracted from the historical records. During extraction, the record integrity constraint is combined to obtain a set of reference modes that match the current signal. Based on the reference pattern set, the signal strength is correlated with the risk level. During the processing, the distribution characteristics of the pattern matching degree are combined to obtain the fault risk level classification.

7. The method for quality inspection of power distribution equipment connectors based on big data according to claim 1, characterized in that, The continuous monitoring of the equipment's operating status, if the monitoring result indicates a continuously rising risk level, generates an adjusted monitoring frequency strategy and determines high-frequency monitoring time periods, including: Based on the fault risk level, the equipment operating parameters are continuously tracked. During the tracking, the difference between real-time data and environmental impact is combined to obtain the change trend related to parameter fluctuations. If the trend shows that the risk increases beyond the preset threshold, the parameter fluctuations are matched with historical records. During the matching process, the logical constraints of dynamic programming are combined to determine the key time periods in which the monitoring frequency needs to be adjusted. Based on the key time periods, an adjusted monitoring frequency strategy is generated. During the generation process, the correspondence between the fault risk level and the time period is combined to obtain a time arrangement suitable for high-frequency monitoring. The high-frequency monitoring tasks are assigned to the corresponding monitoring nodes through the aforementioned time arrangement. The final monitoring execution plan is determined by taking into account the real-time fluctuations of the operating parameters during the assignment.

8. The method for quality inspection of power distribution equipment connectors based on big data according to claim 1, characterized in that, The process of collecting data and extracting short-term fluctuation characteristics during the high-frequency monitoring period to determine whether there are signs of an impending fault, and obtaining the final fault warning result, includes: Based on the data collected during the high-frequency monitoring period, the data related to the operating status are cleaned and classified to obtain the original fluctuation characteristics related to the short-term fluctuations, and a set of fluctuation characteristics for subsequent processing is obtained. Based on the set of fluctuation features, the data within the high-frequency monitoring period is segmented to obtain short-term fluctuation changes related to the equipment status. If the magnitude of the change exceeds a preset magnitude threshold, it is determined that there is a need for anomaly detection, and a subset of fluctuation features is identified. Based on the aforementioned subset of fluctuation characteristics, the device status is analyzed in real time to obtain the distribution of anomalies related to fault signs, determine whether there are potential fault signs, and obtain a detailed record of the anomaly distribution. Based on the detailed records of the anomaly distribution, the anomalies are correlated and matched with the early warning results to determine whether there are any signs of an impending fault, and to determine the final fault early warning result.

9. The method for quality inspection of power distribution equipment connectors based on big data according to claim 1, characterized in that, The process of generating scheduling instructions for equipment maintenance and updating the fault mode database based on the scheduling includes: Based on the fault warning results, the equipment status monitoring data is classified and processed to extract information related to maintenance scheduling, generate corresponding control instructions, and obtain a set of scheduling instructions for subsequent processing. The set of scheduling instructions is sent to the target control unit, a receipt confirmation message is obtained, the operation record during the transmission process is recorded, and the integrity of the transmission process is determined. If the received confirmation information shows that the transmission was successful, the feedback information is sorted out, response data related to the device status is extracted, and it is determined whether the feedback information meets the preset threshold requirements to obtain the sorted feedback dataset. The feedback dataset is compared with historical operation records, and the relevant entries in the fault mode database are updated as reference data for subsequent optimization.

10. A quality inspection system for power distribution equipment connectors based on big data, characterized in that, include: The first determination module is used to obtain equipment operation records from stored historical data, obtain preliminary fault mode classification results and extract features, and determine the association feature set between equipment operation status and fault mode; The first acquisition module is used to acquire environmental factors, fuse the environmental factors with the associated feature set to obtain a comprehensive influence factor matrix and perform dimensionality reduction processing to obtain the core feature vector of the equipment operating status; The second acquisition module is used to predict potential fault modes based on the core feature vector. If the abnormal probability in the prediction result exceeds a preset probability threshold, the core feature vector is marked as a high-risk state, and a warning signal for potential faults is obtained. The second determining module is used to verify the reliability of the warning signal, determine whether the warning signal conforms to the characteristics of historical fault modes, obtain the fault risk level, and continuously track the operating status of the equipment. If the tracking result is that the risk level continues to rise, an adjusted monitoring frequency strategy is generated to determine the high-frequency monitoring time period. The update module is used to collect data and extract short-term fluctuation characteristics during the high-frequency monitoring period, determine whether there are signs of an impending fault, obtain the final fault warning result, generate scheduling instructions for equipment maintenance, and update the fault mode database according to the scheduling instructions.

Images