MES-based power distribution cabinet quality inspection analysis and process regulation and control method and system

Abstract

The invention provides an MES-based power distribution cabinet quality inspection analysis and process regulation and control method and system, and the method comprises the steps: converting a quality inspection standard into an executable rule, generating a standardized quality inspection template, evaluating the quality of a single project and a batch, monitoring the mean value and fluctuation of a production process based on a statistical process control theory, and calculating a process capability index. A dynamic threshold value formula is combined to adjust a control limit in real time, process stability is evaluated, equipment data is standardized, production batches are associated, data consistency is verified and ensured, graded early warning is performed according to the deviation degree, a machine learning model is combined to realize predictive maintenance, process parameters are dynamically corrected, historical data are analyzed through machine learning, and the process quality is improved. According to the method, high-order feature quantification influence factors are extracted, abnormal parameter combination is detected in combination with an isolated forest, a process strategy library is updated, closed-loop optimization is formed, full-process automation from data acquisition, real-time analysis and parameter adaptive adjustment to predictive maintenance is realized, and quality inspection efficiency and process stability are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of quality inspection analysis and process control of power distribution cabinets, and in particular to a method and system for quality inspection analysis and process control of power distribution cabinets based on MES. Background Art

[0002] In the traditional power distribution cabinet manufacturing process, quality control mainly relies on manual inspection and static process parameter setting, which has limitations. It relies on manual filling out of paper forms to record data, which is prone to missed detection, misjudgment or data delay, and cannot monitor the production process in real time. Although quality fluctuations can be monitored through X-bar charts and R charts, the threshold settings are fixed and cannot dynamically adapt to equipment aging, environmental changes or maintenance cycle differences, resulting in high false alarm rates or delayed fault identification. The optimization of process parameters depends on the experience of engineers and lacks a data-driven adaptive adjustment mechanism. Quality inspection data, equipment operation data, environmental data, etc. are stored in a scattered manner, and a unified analysis framework has not been formed, making it difficult to explore deep-level quality-related factors.

[0003] Existing MES systems focus more on production scheduling and basic data recording, lack the ability to conduct in-depth correlation analysis of quality inspection data, equipment parameters, and environmental data, and mostly use fixed thresholds and preset rules. They are unable to dynamically adjust control boundaries based on the real-time operating status of the equipment. Even if machine learning is used to predict quality defects, there is still a lack of a closed-loop mechanism to convert prediction results into executable parameter adjustments. The physical limits and process feasibility constraints of parameter adjustments are not considered, and analysis that relies on historical data is mostly post-mortem. There is a lack of real-time identification of abnormal process parameter combinations and verification of preset strategies, resulting in a long implementation cycle for optimization strategies and uncontrollable results. Summary of the invention

[0004] The purpose of the present invention is to provide a method and system for quality inspection analysis and process control of power distribution cabinets based on MES.

[0005] The problem to be solved by the present invention is: to solve the problems of inconsistent implementation of quality inspection standards, delayed adjustment of process parameters, insufficient monitoring of quality fluctuations, weak equipment failure prediction ability and reliance on manual experience for process optimization in the production process of distribution cabinets. By integrating the quality inspection rule library, real-time data analysis, dynamic threshold warning, adaptive parameter adjustment and process optimization driven by machine learning through the MES system, full-process automated monitoring, predictive maintenance, intelligent parameter adjustment and data-driven process strategy iteration are realized, thereby improving quality stability, reducing failure risks and optimizing production efficiency.

[0006] The method of quality inspection analysis and process control of power distribution cabinet based on MES adopts the following technical solutions: S1: Collect all the quality inspection items, specifications, and tolerance information in the quality inspection standards and quality control plan, quantify the collected information, set up corresponding inspection forms and record templates in the MES system, and build a quality inspection rule base; S2: Based on statistical process control theory, create X-bar charts and R charts for quality control points including plate thickness and welding strength, calculate the process capability index Cp, and set the Cp threshold to evaluate the capability of the production process; S3: Fill in the production data measurement values ​​in the MES system and make qualitative descriptions. For production equipment that directly outputs digital signals, the production data is transmitted to the MES system and the data verification function is integrated; S4: Real-time analysis of the collected production data, combined with the equipment operation data, once a threshold is detected, the MES system will immediately trigger the early warning mechanism, build a predictive maintenance model, and identify equipment that may fail in advance; S5: When the situation in S4 exceeds the threshold setting range, the MES system automatically adjusts the corresponding parameters based on the preset adjustment strategy; S6: Use machine learning algorithms to analyze historical quality inspection results data, identify factors that affect product quality, and preset process control strategies.

[0007] Furthermore, the statistical information is quantified in S1, and corresponding inspection forms and record templates are set in the MES system to build a quality inspection rule base, including: Define measurement indicators and allowable error ranges for each quality inspection item, create a standardized template library containing various quality inspection items, and use the quality score formula to judge the eligibility of any quality inspection item. ,in is the quality score of any quality inspection item, is the actual measured value of any quality inspection item, is the lower specification limit of any quality inspection item, is the upper specification limit of any quality inspection item; The rule engine is integrated into the MES system to make a qualification judgment on each individual quality inspection item and a comprehensive evaluation of all quality inspection items in each batch. The comprehensive evaluation formula is: ,in is the comprehensive quality score, is the weight of the i-th quality inspection item, is the quality score of the i-th quality inspection item, The number of all quality inspection items.

[0008] Furthermore, in S2, an X-bar chart and an R chart are created, the process capability index Cp is calculated, and the capability of the production process is evaluated, including: S21: Draw n samples from the production line and calculate their mean and standard deviation s, the X-bar chart monitors the changing trend of the sample mean, and the R chart monitors the changing trend of the sample range. The sample range refers to the difference between the maximum and minimum values ​​in the sample; S22: The calculation formula of Cp process capability index is ,in is the lower specification limit of any quality inspection item, is the upper specification limit of any quality inspection item, The standard deviation of sample data for any quality inspection item is set according to the quality inspection standard and quality control plan. If the Cp process capability index is less than the threshold, it indicates that there is volatility in the production process. S23: Update X-bar chart and R chart regularly to evaluate the changing trend of production process.

[0009] Furthermore, the S3 fills in the production data measurement values ​​in the MES system and makes a qualitative description, integrating the data verification function, including: S31: Preset a standardized production data form template in the MES system, including all production data standards and production data items in the quality control plan, and define a unified qualitative description for each production data item; S32: The MES system has a built-in data parsing module that automatically converts equipment data into a unified format and stores it. It uses regular expressions to extract fields including equipment ID, measurement value, and timestamp, stores the data in the MES database, and associates it with the corresponding production batch; S33: Perform data consistency verification. ,in To verify the results, is the actual production data measurement value, is the lower specification limit of any quality inspection item, is the upper specification limit of any quality inspection item, It is the deviation between the actual production data measurement value and the standard value.

[0010] Furthermore, in S4, the collected production data is analyzed in real time, combined with the equipment operation data, to build a predictive maintenance model to identify equipment that may fail in advance, including: S41: Integrate the production data and equipment operation data collected in the MES system in real time and establish a dynamic threshold formula. ,in is the dynamic threshold of any production process, is the mean of historical data of any production process, is the standard deviation of historical data of any production process, is the coefficient, is the attenuation coefficient, is the current device running time, The last maintenance time is displayed, and the X-bar chart and R chart are used to track the mean and fluctuation in real time; S42: The warning level is divided according to the degree of deviation. If the data exceeds the dynamic threshold by three historical data standard deviations, the machine needs to be stopped immediately for inspection. If the data exceeds the dynamic threshold by two historical data standard deviations but does not exceed the dynamic threshold by three historical data standard deviations, the process parameters should be adjusted. If the data does not exceed the dynamic threshold by two historical data standard deviations, the data trend should be paid attention to. S43: Compare the equipment operation data with the quality score of any quality inspection item Integrate to form a multi-dimensional feature vector, use machine learning algorithms to train prediction models, and predict the probability of equipment failure , ,in is the feature vector, is temperature, vibration frequency, quality score, running time, is the weight coefficient, is the bias term.

[0011] Furthermore, the MES system in S5 automatically adjusts corresponding parameters based on a preset adjustment strategy, including: When the dynamic threshold index of any production process exceeds the preset range, the MES system automatically adjusts the equipment process parameters and sends the adjusted parameters to the equipment control system. The MES system calculates the subsequent collected new batch data to determine whether the corresponding parameter adjustment is still required. If it still fails to meet the standard after three consecutive adjustments, a manual intervention request is triggered and the current batch of products is frozen; ,in is the adjustment amount of the i-th device parameter, is the basic adjustment coefficient, is the quality score of the quality inspection item of the i-th equipment, For parameters right The sensitivity of is obtained based on historical data regression. is the nonlinear attenuation coefficient; ,in is the adjustment amount of the i-th device parameter, is the basic adjustment coefficient, is the quality score of the quality inspection item of the i-th equipment, For parameters right The sensitivity of is obtained based on historical data regression. is the nonlinear attenuation coefficient; Constraints satisfied , , and ,in and For parameters The physical limit of is the nonlinear constraint coefficient, is the process feasibility gain coefficient, is the Sigmoid slope coefficient.

[0012] Furthermore, the factors affecting product quality are found in S6, and the process control strategy is preset, including: Integrate historical quality inspection data, equipment operating parameters, environmental data, and process parameter data in the MES system, use automatic feature extraction algorithms to extract high-order features from raw data, quantify feature importance based on the gradient boosting tree method, divide training sets and test sets, optimize hyperparameters through cross-validation, use accuracy and F1 scores to measure model performance, identify parameters that have the greatest impact on quality inspection results based on feature importance scores, use time series analysis to verify whether parameter changes significantly affect quality inspection results, and identify abnormal process parameter combinations based on the isolation forest method; Based on the identified abnormal process parameter combinations, preset process control strategies are implemented, the effects of the strategies are simulated and tested, and the strategies that are executed are compared with those that are not executed in batches on the production line. The models and strategy libraries are updated based on the latest data in the MES.

[0013] Furthermore, a system for quality inspection, analysis and process control of a power distribution cabinet based on MES is used to implement any of the above-mentioned methods for quality inspection, analysis and process control of a power distribution cabinet based on MES. The system for quality inspection, analysis and process control of a power distribution cabinet based on MES includes: a quality inspection rule base management module, a statistical process control module, a data integration and verification module, a real-time analysis and early warning module, an automatic parameter adjustment module, and a process optimization and strategy preset module: Quality inspection rule base management module: converts quality inspection standards into executable rules, including measurement indicators, tolerance ranges and weights, generates standardized quality inspection forms and record templates, and evaluates the eligibility of individual quality inspection items and the overall quality of the batch; Statistical process control module: monitors sample mean and range based on X-bar chart and R chart, identifies mean shift and fluctuation in production process, calculates Cp index to evaluate process capability, establishes dynamic threshold formula based on equipment operation time and maintenance history, and adjusts quality control threshold in real time; Data integration and verification module: preset standardized data forms, use regular expressions to extract equipment data and associate it with production batches for consistency verification; Real-time analysis and early warning module: Dynamic threshold early warning is graded according to the degree of deviation, and the probability of equipment failure is predicted through machine learning, which automatically freezes unqualified batches and triggers the recall process; Automatic parameter adjustment module: Establish an adaptive adjustment strategy, calculate the adjustment amount based on the quality score and parameter sensitivity of the quality inspection items of the i-th equipment, and impose conditional constraints based on physical limitations and process feasibility; Process optimization and strategy preset module: extract high-order features from historical data, quantify feature importance based on the gradient boosting tree method, detect abnormal parameter combinations using isolation forests, compare strategy effects in batches on the production line, and regularly update the model and strategy library based on the latest data.

[0014] The beneficial effects of the present invention are: converting the quality inspection standards into executable quantitative rules, realizing the automated qualification judgment of the quality inspection items and the comprehensive scoring of batches, avoiding the errors of manual subjective judgment, ensuring the consistency of quality inspection results of different batches and different operators, and significantly improving the quality inspection efficiency and traceability; Trigger early warnings based on the degree of deviation, monitor production process fluctuations in real time, dynamically adjust thresholds, identify insufficient process capabilities in advance, and reduce scrap rates caused by fluctuations. Build a predictive maintenance model based on equipment operation data to identify equipment failure risks in advance, which effectively reduces downtime caused by sudden failures and ensures production continuity. Adaptive process parameter optimization improves production efficiency, dynamically calculates adjustment amounts, and combines physical limit constraints and process feasibility indicators to ensure that parameter adjustments are both accurate and safe, respond to fluctuations more quickly, and reduce resource waste caused by over-adjustment; Through high-order feature extraction and gradient boosting tree algorithm, the impact of parameters on quality is quantified. Combined with time series analysis and isolation forest detection, abnormal process combinations are generated to generate preset strategies. Combined with the latest data, the model and strategy library are regularly updated to avoid strategy rigidity and ensure that process parameters are always in the optimal state. Integrate MES, production equipment, quality inspection data, environmental data, etc. to achieve full process automation from data collection to process optimization, reduce manual intervention and reduce human errors. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is a flow chart of the quality inspection analysis and process control method of the power distribution cabinet based on MES; Figure 2 This is the system module diagram for quality inspection analysis and process control of distribution cabinets based on MES. DETAILED DESCRIPTION

[0016] The present invention is further clearly and completely described below, but the protection scope of the present invention is not limited thereto.

[0017] The method of quality inspection analysis and process control of power distribution cabinet based on MES adopts the following technical solutions: S1: Collect all the quality inspection items, specifications, and tolerance information in the quality inspection standards and quality control plan, quantify the collected information, set up corresponding inspection forms and record templates in the MES system, and build a quality inspection rule base; S2: Based on statistical process control theory, create X-bar charts and R charts for quality control points including plate thickness and welding strength, calculate the process capability index Cp, and set the Cp threshold to evaluate the capability of the production process; S3: Fill in the production data measurement values ​​in the MES system and make qualitative descriptions. For production equipment that directly outputs digital signals, the production data is transmitted to the MES system and the data verification function is integrated; S4: Real-time analysis of the collected production data, combined with the equipment operation data, once a threshold is detected, the MES system will immediately trigger the early warning mechanism, build a predictive maintenance model, and identify equipment that may fail in advance; S5: When the situation in S4 exceeds the threshold setting range, the MES system automatically adjusts the corresponding parameters based on the preset adjustment strategy; S6: Use machine learning algorithms to analyze historical quality inspection results data, identify factors that affect product quality, and preset process control strategies.

[0018] refer to Figure 1 The figure shows a flow chart of the quality inspection analysis and process control method of power distribution cabinets based on MES.

[0019] Furthermore, the statistical information is quantified in S1, and corresponding inspection forms and record templates are set in the MES system to build a quality inspection rule base, including: Define measurement indicators and allowable error ranges for each quality inspection item, create a standardized template library containing various quality inspection items, and use the quality score formula to judge the eligibility of any quality inspection item. ,in is the quality score of any quality inspection item, is the actual measured value of any quality inspection item, is the lower specification limit of any quality inspection item, is the upper specification limit of any quality inspection item; The rule engine is integrated into the MES system to make a qualification judgment on each individual quality inspection item and a comprehensive evaluation of all quality inspection items in each batch. The comprehensive evaluation formula is: ,in is the comprehensive quality score, is the weight of the i-th quality inspection item, is the quality score of the i-th quality inspection item, The number of all quality inspection items.

[0020] Furthermore, in S2, an X-bar chart and an R chart are created, the process capability index Cp is calculated, and the capability of the production process is evaluated, including: S21: Draw n samples from the production line and calculate their mean and standard deviation s, the X-bar chart monitors the changing trend of the sample mean, and the R chart monitors the changing trend of the sample range. The sample range refers to the difference between the maximum and minimum values ​​in the sample; S22: The calculation formula of Cp process capability index is ,in is the lower specification limit of any quality inspection item, is the upper specification limit of any quality inspection item, The standard deviation of sample data for any quality inspection item is set according to the quality inspection standard and quality control plan. If the Cp process capability index is less than the threshold, it indicates that there is volatility in the production process. S23: Update X-bar chart and R chart regularly to evaluate the changing trend of production process.

[0021] Furthermore, the S3 fills in the production data measurement values ​​in the MES system and makes a qualitative description, integrating the data verification function, including: S31: Preset a standardized production data form template in the MES system, including all production data standards and production data items in the quality control plan, and define a unified qualitative description for each production data item; S32: The MES system has a built-in data parsing module that automatically converts equipment data into a unified format and stores it. It uses regular expressions to extract fields including equipment ID, measurement value, and timestamp, stores the data in the MES database, and associates it with the corresponding production batch; S33: Perform data consistency verification. ,in To verify the results, is the actual production data measurement value, is the lower specification limit of any quality inspection item, is the upper specification limit of any quality inspection item, It is the deviation between the actual production data measurement value and the standard value.

[0022] Furthermore, in S4, the collected production data is analyzed in real time, combined with the equipment operation data, to build a predictive maintenance model to identify equipment that may fail in advance, including: S41: Integrate the production data and equipment operation data collected in the MES system in real time and establish a dynamic threshold formula. ,in is the dynamic threshold of any production process, is the mean of historical data of any production process, is the standard deviation of historical data of any production process, is the coefficient, is the attenuation coefficient, is the current device running time, The last maintenance time is displayed, and the X-bar chart and R chart are used to track the mean and fluctuation in real time; S42: The warning level is divided according to the degree of deviation. If the data exceeds the dynamic threshold by three historical data standard deviations, the machine needs to be stopped immediately for inspection. If the data exceeds the dynamic threshold by two historical data standard deviations but does not exceed the dynamic threshold by three historical data standard deviations, the process parameters should be adjusted. If the data does not exceed the dynamic threshold by two historical data standard deviations, the data trend should be paid attention to. S43: Compare the equipment operation data with the quality score of any quality inspection item Integrate to form a multi-dimensional feature vector, use machine learning algorithms to train prediction models, and predict the probability of equipment failure , ,in is the feature vector, is temperature, vibration frequency, quality score, running time, is the weight coefficient, is the bias term.

[0023] Furthermore, the MES system in S5 automatically adjusts corresponding parameters based on a preset adjustment strategy, including: When the dynamic threshold index of any production process exceeds the preset range, the MES system automatically adjusts the equipment process parameters and sends the adjusted parameters to the equipment control system. The MES system calculates the subsequent collected new batch data to determine whether the corresponding parameter adjustment is still required. If it still fails to meet the standard after three consecutive adjustments, a manual intervention request is triggered and the current batch of products is frozen; ,in is the adjustment amount of the i-th device parameter, is the basic adjustment coefficient, is the quality score of the quality inspection item of the i-th equipment, For parameters right The sensitivity of is obtained based on historical data regression. is the nonlinear attenuation coefficient; Constraints satisfied , , and ,in and For parameters The physical limit of is the nonlinear constraint coefficient, is the process feasibility gain coefficient, is the Sigmoid slope coefficient.

[0024] Furthermore, the factors affecting product quality are found in S6, and the process control strategy is preset, including: Integrate historical quality inspection data, equipment operating parameters, environmental data, and process parameter data in the MES system, use automatic feature extraction algorithms to extract high-order features from raw data, quantify feature importance based on the gradient boosting tree method, divide training sets and test sets, optimize hyperparameters through cross-validation, use accuracy and F1 scores to measure model performance, identify parameters that have the greatest impact on quality inspection results based on feature importance scores, use time series analysis to verify whether parameter changes significantly affect quality inspection results, and identify abnormal process parameter combinations based on the isolation forest method; Based on the identified abnormal process parameter combinations, preset process control strategies are implemented, the effects of the strategies are simulated and tested, and the strategies that are executed are compared with those that are not executed in batches on the production line. The models and strategy libraries are updated based on the latest data in the MES.

[0025] Furthermore, a system for quality inspection, analysis and process control of a power distribution cabinet based on MES is used to implement any of the above-mentioned methods for quality inspection, analysis and process control of a power distribution cabinet based on MES. The system for quality inspection, analysis and process control of a power distribution cabinet based on MES includes: a quality inspection rule base management module, a statistical process control module, a data integration and verification module, a real-time analysis and early warning module, an automatic parameter adjustment module, and a process optimization and strategy preset module: Quality inspection rule base management module: converts quality inspection standards into executable rules, including measurement indicators, tolerance ranges and weights, generates standardized quality inspection forms and record templates, and evaluates the eligibility of individual quality inspection items and the overall quality of the batch; Statistical process control module: monitors sample mean and range based on X-bar chart and R chart, identifies mean shift and fluctuation in production process, calculates Cp index to evaluate process capability, establishes dynamic threshold formula based on equipment operation time and maintenance history, and adjusts quality control threshold in real time; Data integration and verification module: preset standardized data forms, use regular expressions to extract equipment data and associate it with production batches for consistency verification; Real-time analysis and early warning module: Dynamic threshold early warning is graded according to the degree of deviation, and the probability of equipment failure is predicted through machine learning, which automatically freezes unqualified batches and triggers the recall process; Automatic parameter adjustment module: Establish an adaptive adjustment strategy, calculate the adjustment amount based on the quality score and parameter sensitivity of the quality inspection items of the i-th equipment, and impose conditional constraints based on physical limitations and process feasibility; Process optimization and strategy preset module: extract high-order features from historical data, quantify feature importance based on the gradient boosting tree method, detect abnormal parameter combinations using isolation forests, compare strategy effects in batches on the production line, and regularly update the model and strategy library based on the latest data.

[0026] refer to Figure 2 The figure shows the system module diagram of quality inspection analysis and process control of power distribution cabinets based on MES.

[0027] The present invention provides a method and system for quality inspection analysis and process control of power distribution cabinets based on MES, which converts quality inspection standards into executable rules, generates standardized quality inspection templates, evaluates the quality of single items and batches, monitors the mean and fluctuation of the production process based on statistical process control theory, calculates the process capability index, adjusts the control limit in real time in combination with the dynamic threshold formula, evaluates process stability, standardizes equipment data and associates production batches, verifies and ensures data consistency, issues graded warnings according to the degree of deviation, and realizes predictive maintenance in combination with a machine learning model, dynamically corrects process parameters, analyzes historical data through machine learning, extracts high-order features to quantify influencing factors, detects abnormal parameter combinations in combination with isolation forests, updates the process strategy library, forms a closed-loop optimization, realizes full-process automation from data collection, real-time analysis, parameter adaptive adjustment to predictive maintenance, and improves quality inspection efficiency and process stability.

Claims

1. The method of quality inspection analysis and process control of power distribution cabinet based on MES is characterized by: include: S1: Collect all the quality inspection items, specifications, and tolerance information in the quality inspection standards and quality control plan, quantify the collected information, set up corresponding inspection forms and record templates in the MES system, and build a quality inspection rule base; S2: Based on statistical process control theory, create X-bar charts and R charts for quality control points including plate thickness and welding strength, calculate the process capability index Cp, and set the Cp threshold to evaluate the capability of the production process; S3: Fill in the production data measurement values ​​in the MES system and make qualitative descriptions. For production equipment that directly outputs digital signals, the production data is transmitted to the MES system and the data verification function is integrated; S4: Real-time analysis of the collected production data, combined with the equipment operation data, once a threshold is detected, the MES system will immediately trigger the early warning mechanism, build a predictive maintenance model, and identify equipment that may fail in advance; S5: When the situation in S4 exceeds the threshold setting range, the MES system automatically adjusts the corresponding parameters based on the preset adjustment strategy; S6: Use machine learning algorithms to analyze historical quality inspection results data, identify factors that affect product quality, and preset process control strategies.

2. The method for quality inspection, analysis and process control of power distribution cabinets based on MES as claimed in claim 1, characterized in that: In S1, the statistical information is quantified, and corresponding inspection forms and record templates are set in the MES system to build a quality inspection rule base, including: Define measurement indicators and allowable error ranges for each quality inspection item, create a standardized template library containing various quality inspection items, and use the quality score formula to judge the eligibility of any quality inspection item. ,in is the quality score of any quality inspection item, is the actual measured value of any quality inspection item, is the lower specification limit of any quality inspection item, is the upper specification limit of any quality inspection item; The rule engine is integrated into the MES system to make a qualification judgment on each individual quality inspection item and a comprehensive evaluation of all quality inspection items in each batch. The comprehensive evaluation formula is: ,in is the comprehensive quality score, is the weight of the i-th quality inspection item, is the quality score of the i-th quality inspection item, The number of all quality inspection items.

3. The method for quality inspection, analysis and process control of power distribution cabinets based on MES as claimed in claim 1, characterized in that: In S2, X-bar chart and R chart are created, process capability index Cp is calculated, and the capability of the production process is evaluated, including: S21: Draw n samples from the production line and calculate their mean and standard deviation s, the X-bar chart monitors the changing trend of the sample mean, and the R chart monitors the changing trend of the sample range. The sample range refers to the difference between the maximum and minimum values ​​in the sample; S22: The calculation formula of Cp process capability index is ,in is the lower specification limit of any quality inspection item, is the upper specification limit of any quality inspection item, The standard deviation of sample data for any quality inspection item is set according to the quality inspection standard and quality control plan. If the Cp process capability index is less than the threshold, it indicates that there is volatility in the production process. S23: Update X-bar chart and R chart regularly to evaluate the changing trend of production process.

4. The method for quality inspection, analysis and process control of power distribution cabinets based on MES as claimed in claim 1, characterized in that: The S3 fills in the production data measurement values ​​in the MES system and makes a qualitative description, integrating data verification functions, including: S31: Preset a standardized production data form template in the MES system, including all production data standards and production data items in the quality control plan, and define a unified qualitative description for each production data item; S32: The MES system has a built-in data parsing module that automatically converts equipment data into a unified format and stores it. It uses regular expressions to extract fields including equipment ID, measurement value, and timestamp, stores the data in the MES database, and associates it with the corresponding production batch; S33: Perform data consistency verification. ,in To verify the results, is the actual production data measurement value, is the lower specification limit of any quality inspection item, is the upper specification limit of any quality inspection item, It is the deviation between the actual production data measurement value and the standard value.

5. The method for quality inspection, analysis and process control of power distribution cabinets based on MES as claimed in claim 1, characterized in that: The S4 performs real-time analysis on the collected production data, combines it with the equipment operation data, builds a predictive maintenance model, and identifies equipment that may fail in advance, including: S41: Integrate the production data and equipment operation data collected in the MES system in real time and establish a dynamic threshold formula. ,in is the dynamic threshold of any production process, is the mean of historical data of any production process, is the standard deviation of historical data of any production process, is the coefficient, is the attenuation coefficient, is the current device running time, The last maintenance time is displayed, and the X-bar chart and R chart are used to track the mean and fluctuation in real time; S42: The warning level is divided according to the degree of deviation. If the data exceeds the dynamic threshold by three historical data standard deviations, the machine needs to be stopped immediately for inspection. If the data exceeds the dynamic threshold by two historical data standard deviations but does not exceed the dynamic threshold by three historical data standard deviations, the process parameters should be adjusted. If the data does not exceed the dynamic threshold by two historical data standard deviations, the data trend should be paid attention to. S43: Compare the equipment operation data with the quality score of any quality inspection item Integrate to form a multi-dimensional feature vector, use machine learning algorithms to train prediction models, and predict the probability of equipment failure , ,in is the feature vector, is temperature, vibration frequency, quality score, running time, is the weight coefficient, is the bias term.

6. The method for quality inspection, analysis and process control of power distribution cabinets based on MES as claimed in claim 1, characterized in that: The MES system in S5 automatically adjusts corresponding parameters based on the preset adjustment strategy, including: When the dynamic threshold index of any production process exceeds the preset range, the MES system automatically adjusts the equipment process parameters and sends the adjusted parameters to the equipment control system. The MES system calculates the subsequent collected new batch data to determine whether the corresponding parameter adjustment is still required. If it still fails to meet the standard after three consecutive adjustments, a manual intervention request is triggered and the current batch of products is frozen; ,in is the adjustment amount of the i-th device parameter, is the basic adjustment coefficient, is the quality score of the quality inspection item of the i-th equipment, For parameters right The sensitivity of is obtained based on historical data regression. is the nonlinear attenuation coefficient; Constraints satisfied , , and ,in and For parameters The physical limit of is the nonlinear constraint coefficient, is the process feasibility gain coefficient, is the Sigmoid slope coefficient.

7. The method for quality inspection, analysis and process control of power distribution cabinets based on MES as claimed in claim 1, characterized in that: In S6, factors affecting product quality are found and process control strategies are preset, including: Integrate historical quality inspection data, equipment operating parameters, environmental data, and process parameter data in the MES system, use automatic feature extraction algorithms to extract high-order features from raw data, quantify feature importance based on the gradient boosting tree method, divide training sets and test sets, optimize hyperparameters through cross-validation, use accuracy and F1 scores to measure model performance, identify parameters that have the greatest impact on quality inspection results based on feature importance scores, use time series analysis to verify whether parameter changes significantly affect quality inspection results, and identify abnormal process parameter combinations based on the isolation forest method; Based on the identified abnormal process parameter combinations, preset process control strategies are implemented, the effects of the strategies are simulated and tested, and the strategies that are executed are compared with those that are not executed in batches on the production line. The models and strategy libraries are updated based on the latest data in the MES.

8. The system of quality inspection, analysis and process control of power distribution cabinet based on MES is characterized by: The method for implementing the quality inspection analysis and process control of the power distribution cabinet based on MES as described in any one of claims 1 to 7, the system for quality inspection analysis and process control of the power distribution cabinet based on MES comprises: a quality inspection rule base management module, a statistical process control module, a data integration and verification module, a real-time analysis and early warning module, an automatic parameter adjustment module, and a process optimization and strategy preset module: Quality inspection rule base management module: converts quality inspection standards into executable rules, including measurement indicators, tolerance ranges and weights, generates standardized quality inspection forms and record templates, and evaluates the eligibility of individual quality inspection items and the overall quality of the batch; Statistical process control module: monitors sample mean and range based on X-bar chart and R chart, identifies mean shift and fluctuation in production process, calculates Cp index to evaluate process capability, establishes dynamic threshold formula based on equipment operation time and maintenance history, and adjusts quality control threshold in real time; Data integration and verification module: preset standardized data forms, use regular expressions to extract equipment data and associate it with production batches for consistency verification; Real-time analysis and early warning module: Dynamic threshold early warning is graded according to the degree of deviation, and the probability of equipment failure is predicted through machine learning, which automatically freezes unqualified batches and triggers the recall process; Automatic parameter adjustment module: Establish an adaptive adjustment strategy, calculate the adjustment amount based on the quality score and parameter sensitivity of the quality inspection items of the i-th equipment, and impose conditional constraints based on physical limitations and process feasibility; Process optimization and strategy preset module: extract high-order features from historical data, quantify feature importance based on the gradient boosting tree method, detect abnormal parameter combinations using isolation forests, compare strategy effects in batches on the production line, and regularly update the model and strategy library based on the latest data.

Images