SPC control method, system and device based on dynamic threshold optimization and medium

Abstract

The invention provides an SPC control method, system and device based on dynamic threshold optimization and a medium, and can solve the problems caused by a traditional SPC static threshold, and the method comprises the steps: collecting and preprocessing an original data stream containing key quality characteristics or key process parameters from various data sources on a production line in real time, obtaining a structured valid data sequence; intercepting current sliding window data from the structured valid data sequence by adopting a sliding time window mechanism, performing abnormal value processing, and dynamically calculating a control limit of the SPC control chart based on the processed sliding window data to obtain a first control limit; when the process parameter step change of the production line is monitored, updating the control limit of the SPC control chart to obtain a second control limit; and comparing the latest effective data output in the structured effective data sequence with the first control limit or the second control limit in real time, and triggering an alarm when judging that the production process is abnormal according to a comparison result.

Description

Technical Field

[0001] The present application relates to the technical field of statistical process control, and in particular to an SPC control method, system, device and medium based on dynamic threshold optimization. Background Art

[0002] Traditional Statistical Process Control (SPC) systems typically use fixed control limits (UCL / LCL) calculated based on historical data under stable process conditions. For example, the limits of a control chart are set based on the principle of three standard deviations of the mean (X-bar ±3°).

[0003] However, as modern manufacturing develops towards intelligence, flexibility, and high precision, the dynamics and complexity of production processes are increasing. The traditional SPC method based on static thresholds has gradually exposed its inherent limitations in practical applications: (1) Insufficient adaptability of static thresholds: Static thresholds are difficult to respond to dynamic changes such as equipment aging and process fine-tuning, resulting in high false alarms, low process capability index and missed alarms.

[0004] (2) Data processing and response lag: Relying on periodic offline analysis and manual adjustments, it cannot match the millisecond-level process fluctuation requirements.

[0005] (3) Limitations of data outlier processing: Outliers that are not effectively identified will widen the control limits and mask the true variation.

[0006] These limitations restrict SPC from realizing its full potential in modern intelligent manufacturing environments, making it difficult to meet companies' higher demands for lean production and excellent quality control. Therefore, there is an urgent need for a new SPC control solution that can dynamically optimize control thresholds and intelligently adapt to changes in the production process. Summary of the Invention

[0007] The present application provides an SPC control method, system, device, and medium based on dynamic threshold optimization to solve the problems existing in related technologies. The technical solution is as follows: In a first aspect, an embodiment of the present application provides an SPC control method based on dynamic threshold optimization, comprising: Collect raw data streams containing key quality characteristics or key process parameters from various data sources on the production line in real time; Preprocessing the original data stream to obtain a structured valid data sequence; Adopting a sliding time window mechanism, intercepting the valid data in the most recent period of time in the structured valid data sequence to obtain the current sliding window data; Performing outlier processing on the sliding window data, and dynamically calculating control limits of an SPC control chart based on the processed sliding window data to obtain first control limits; When a step change in process parameters of the production line is detected, the control limits of the SPC control chart are updated based on a preset control line adjustment mechanism to obtain a second control limit; The latest valid data output from the structured valid data sequence is compared with the first control limit or the second control limit in real time, and an alarm is triggered when it is determined that an abnormality occurs in the production process based on the comparison result.

[0008] In one embodiment, preprocessing the original data stream to obtain a structured valid data sequence includes: Cleaning the original data stream to obtain a valid data stream; Performing data type conversion on the valid data stream to obtain a valid data sequence; The valid data sequences are aligned according to timestamps to obtain a structured valid data sequence.

[0009] In one embodiment, performing outlier processing on the sliding window data, and dynamically calculating the control limits of the SPC control chart based on the processed sliding window data, to obtain the first control limits includes: Performing outlier detection on the sliding window data, filtering out abnormal data in the sliding window data, and obtaining the processed sliding window data; Performing a goodness of fit test on the processed sliding window data, and adaptively selecting the most suitable target statistical distribution model according to the test result; Based on the processed sliding window data and the target statistical distribution model, the control limits of the SPC control chart are dynamically calculated to obtain the first control limits.

[0010] In one embodiment, performing outlier detection on the sliding window data, filtering out abnormal data in the sliding window data, and obtaining processed sliding window data includes: The Tukey method is used to calculate the first quartile Q1, the third quartile Q3 and the interquartile range IQR of the sliding window data; The outlier limit is set using the criterion of IQR ≥ K, where the upper limit of the outlier limit is Q3 + K*IQR and the lower limit is Q1-K*IQR, and K>1.5; Based on the outlier limit, outlier data in the sliding window data is identified and filtered out to obtain the processed sliding window data.

[0011] In one embodiment, when a step change in a process parameter of the production line is detected, the control limits of the SPC control chart are updated based on a preset control line adjustment mechanism to obtain the second control limits, including: When a step change in process parameters of the production line is detected, resetting or clearing the sliding window data; Determining the sliding window data as a data change point; Adopting a sliding time window mechanism, intercepting valid data located after the data change point in the structured valid data sequence to obtain new sliding window data; Outlier processing is performed on the new sliding window data, and the control limits of the SPC control chart are recalculated based on the processed new sliding window data to obtain the second control limits.

[0012] In one embodiment, the method further comprises: When receiving production change instruction information from an external system, updating the control limits of the SPC control chart based on the control line adjustment mechanism to obtain third control limits; The latest valid data output from the structured valid data sequence is compared with the third control limit in real time, and an alarm is triggered when it is determined based on the comparison result that an abnormality occurs in the production process.

[0013] In one embodiment, the method further comprises: Dynamically update and draw SPC control charts and process capability dashboards on the graphical human-computer interaction interface; All relevant data of SPC control are persistently stored in the database.

[0014] In a second aspect, an embodiment of the present application further provides an SPC control system based on dynamic threshold optimization, comprising: Data acquisition module, used to collect raw data streams containing key quality characteristics or key process parameters from various data sources on the production line in real time; A data preprocessing module, configured to preprocess the original data stream to obtain a structured valid data sequence; A dynamic threshold calculation engine module is configured to use a sliding time window mechanism to intercept valid data within a recent period of time in the structured valid data sequence to obtain current sliding window data; perform outlier processing on the sliding window data, and dynamically calculate the control limits of the SPC control chart based on the processed sliding window data to obtain first control limits; an intelligent collaborative control logic module, configured to update the control limits of the SPC control chart based on a preset control line adjustment mechanism to obtain second control limits when a step change in process parameters of the production line is detected; The process status monitoring and alarm module is used to compare the latest valid data output from the structured valid data sequence with the first control limit or the second control limit in real time, and trigger an alarm when it is determined that an abnormality occurs in the production process based on the comparison result.

[0015] In one embodiment, when the data preprocessing module is used to preprocess the original data stream to obtain a structured valid data sequence, it is specifically used to: Cleaning the original data stream to obtain a valid data stream; Performing data type conversion on the valid data stream to obtain a valid data sequence; The valid data sequences are aligned according to timestamps to obtain a structured valid data sequence.

[0016] In one embodiment, the dynamic threshold calculation engine module is used to perform outlier processing on the sliding window data, and dynamically calculate the control limits of the SPC control chart based on the processed sliding window data to obtain the first control limit, specifically for: Performing outlier detection on the sliding window data, filtering out abnormal data in the sliding window data, and obtaining the processed sliding window data; Performing a goodness of fit test on the processed sliding window data, and adaptively selecting the most suitable target statistical distribution model according to the test result; Based on the processed sliding window data and the target statistical distribution model, the control limits of the SPC control chart are dynamically calculated to obtain the first control limits.

[0017] In one embodiment, when the dynamic threshold calculation engine module is used to perform outlier detection on the sliding window data, filter out abnormal data in the sliding window data, and obtain processed sliding window data, it is specifically used to: The Tukey method is used to calculate the first quartile Q1, the third quartile Q3 and the interquartile range IQR of the sliding window data; The outlier limit is set using the criterion of IQR ≥ K, where the upper limit of the outlier limit is Q3 + K*IQR and the lower limit is Q1-K*IQR, and K>1.5; Based on the outlier limit, outlier data in the sliding window data is identified and filtered out to obtain the processed sliding window data.

[0018] In one embodiment, the intelligent collaborative control logic module is configured to update the control limits of the SPC control chart based on a preset control line adjustment mechanism when detecting a step change in process parameters of the production line to obtain a second control limit, specifically for: When a step change in process parameters of the production line is detected, resetting or clearing the sliding window data; Determining the sliding window data as a data change point; Adopting a sliding time window mechanism, intercepting valid data located after the data change point in the structured valid data sequence to obtain new sliding window data; Outlier processing is performed on the new sliding window data, and the control limits of the SPC control chart are recalculated based on the processed new sliding window data to obtain the second control limits.

[0019] In one embodiment, the intelligent collaborative control logic module is further configured to update the control limits of the SPC control chart based on the control line adjustment mechanism to obtain third control limits upon receiving production change instruction information from an external system; The process status monitoring and alarm module is further used to compare the latest valid data output from the structured valid data sequence with the third control limit in real time, and trigger an alarm when it is determined that an abnormality occurs in the production process based on the comparison result.

[0020] In one embodiment, the system further comprises: Visualization and human-computer interaction module, used to dynamically update and draw SPC control charts and process capability dashboards on a graphical human-computer interaction interface; The data storage and traceability module is used to store all relevant data of SPC control persistently in the database.

[0021] In a third aspect, an embodiment of the present application further provides an electronic device comprising: a memory and a processor, wherein the memory stores instructions, and the instructions are loaded and executed by the processor to implement a method in any one of the above-mentioned embodiments, wherein the memory and the processor communicate with each other through an internal connection path.

[0022] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored. When the computer program runs on a computer, the method in any one of the above-mentioned embodiments is implemented.

[0023] The advantages or beneficial effects of the above technical solution include at least: (a) Significantly reduce the false alarm rate and improve monitoring accuracy: This application dynamically calculates and updates the control limits of the SPC control chart based on sliding window data, which can better accommodate normal fluctuations in the process and avoid erroneous judgments caused by overly strict or loose fixed thresholds or interference from abnormal data. It can significantly reduce the false alarm rate in the production process from the common 8%-12% in traditional SPC to ≤2%. This greatly reduces unnecessary production interruptions and resource waste, and enhances operators' trust and reliance on the SPC control system.

[0024] (b) Effectively Improve the Process Capability Index (CPK): This application utilizes a sliding time window mechanism combined with process parameter step change monitoring to dynamically calculate and update the control limits of SPC control charts. This allows for more precise and adaptive control limits to the actual process state, helping to truly reflect the inherent variation and excursion of the process, thereby guiding more effective process improvement activities. When process stability is effectively monitored and improved, the Process Capability Index (CPK) will naturally improve.

[0025] (c) Enhanced adaptability to dynamic production environments: This application utilizes a sliding time window mechanism to enable the control limits of SPC control charts to smoothly follow the natural evolution of the process (such as slow drift caused by equipment wear). Furthermore, combined with a mechanism for monitoring and responding to step changes in process parameters (for example, automatically resetting the sliding window when the variance change rate is ≥15%), the system can quickly adapt to sudden changes in process conditions caused by equipment adjustments, raw material batch changes, production mode switching, and other factors, ensuring the effectiveness and timeliness of control limits in various dynamic scenarios.

[0026] (d) Significantly Improved Response Efficiency and Automation: This application achieves end-to-end automation from data acquisition, preprocessing, control limit calculation, process state determination, and alarming. In particular, process step changes can be automatically identified and rapidly adjusted without manual intervention. This automated control response time can be reduced from hours with traditional methods to seconds (e.g., ≤1 second), significantly improving the efficiency of problem detection and resolution and reducing potential quality losses.

[0027] (e) Optimizing data processing procedures and improving the scientificity and robustness of system decisions: This application effectively filters abnormal data values, making the basis of statistical analysis more solid and reliable, thereby making subsequent control decisions (such as determining whether the process is out of control and whether process parameters need to be adjusted) more scientific and accurate, and enhancing the robustness of the entire SPC control system.

[0028] In summary, this application can solve the problems caused by traditional SPC static thresholds, can adapt to the dynamic characteristics of the production process in real time, intelligently adjust control limits, and effectively filter out abnormal data interference, and can achieve more accurate, efficient, and intelligent process quality control, thereby improving overall production quality and efficiency.

[0029] The above summary is for illustrative purposes only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments and features described above, further aspects, embodiments and features of the present application will be readily apparent by reference to the accompanying drawings and the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] In the accompanying drawings, unless otherwise specified, the same reference numerals throughout the multiple drawings represent the same or similar components or elements. These drawings are not necessarily drawn to scale. It should be understood that these drawings only depict some embodiments disclosed in this application and should not be construed as limiting the scope of this application.

[0031] Figure 1 An example process method of an SPC control method based on dynamic threshold optimization provided in an embodiment of the present application; Figure 2 Another example process method of an SPC control method based on dynamic threshold optimization provided in an embodiment of the present application; Figure 3 Another example process method of an SPC control method based on dynamic threshold optimization provided in an embodiment of the present application; Figure 4 A structural block diagram of an SPC control system based on dynamic threshold optimization provided in an embodiment of the present application; Figure 5 This is a structural block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0032] Hereinafter, only certain exemplary embodiments are briefly described. As will be appreciated by those skilled in the art, the described embodiments may be modified in various ways without departing from the spirit or scope of the present application. Therefore, the drawings and description are to be regarded as illustrative in nature and not restrictive.

[0033] Statistical Process Control (SPC) is a classic quality management tool. Its core idea is to prevent defects and continuously improve quality by real-time monitoring and analysis of key quality characteristics (KQCss) or key process parameters (KPPs) in the production process, distinguishing between random fluctuations (common causes) and abnormal fluctuations (special causes) in the process.

[0034] To better understand the above limitations of the traditional static threshold-based SCP method, these limitations are further explained below.

[0035] 1. Static thresholds are not adaptable enough The biggest problem with traditional SPC systems is that once the control limits are set, they remain unchanged for a long period of time. This static nature makes it difficult to adapt to the dynamically changing production environment. Specifically, 1) Sluggish response to changes in the production environment: During the production process, equipment gradually ages due to natural wear and tear, raw material batches may have inherent variations, process parameters may be fine-tuned to meet production needs, and environmental conditions (such as temperature and humidity) may fluctuate. These factors can cause slow drifts or step changes in the process mean or variation. Consequently, fixed control limits cannot capture the dynamic changes in equipment status, raw material characteristics, process parameters, and environmental conditions during the production process, rendering control charts ineffective in name only.

[0036] 2) High False Alarm Rate: When process characteristics experience normal, acceptable minor deviations (still within specifications) or the data itself exhibits glitches, rigid control limits can be overly sensitive, triggering frequent alarms. Traditional SPC systems can experience false alarm rates as high as 8%-12%. High false alarm rates not only consume significant engineering resources to troubleshoot unrealistic anomalies, disrupting normal production, but more seriously, they can gradually erode operators' trust in the SPC system, leading to a "cry wolf" effect where signals of genuine process anomalies are drowned out or ignored. SPC control charts inherently carry the risk of false alarms, and static thresholds amplify this risk in a dynamic environment.

[0037] 3) Low Process Capability Index (CPK) and Assessment Distortion: Process capability indices (such as CPK and PPK) are key indicators of a process's ability to meet specifications. Static and potentially inappropriate control limits, if they fail to accurately reflect the true state of the process, can lead to inaccurate assessments of process capability (either too high or too low). Furthermore, irrational alarms can lead to incorrect "improvement" measures, hindering actual CPK improvement. In static threshold systems, the process capability index (CPK) often only maintains a level of 1.0-1.2, indicating a high potential for defective product, significantly below the industry-wide target of CPK ≥1.33 or even 1.67.

[0038] 4) Risk of Missing True Signals: In contrast to high false alarms, when a process deteriorates slowly and continuously, or experiences small but statistically significant abnormal fluctuations, fixed control limits that are too loose or no longer suitable for the current process state may not be sensitive enough to issue alarms in a timely manner, resulting in quality risks not being discovered and addressed early, which may lead to subsequent batches of defective products.

[0039] (2) Data processing and response lag Traditional SPC systems rely on offline analysis and manual adjustment of control limits for data processing and response. This results in response delays of up to hours to millisecond-level process fluctuations, making them unable to meet real-time control requirements. This is manifested in the following ways: 1) Offline Analysis and Periodic Adjustment: Many traditional SPC implementations rely on periodic (e.g., weekly or monthly) offline statistical analysis of data to reassess and adjust control limits. This approach is unable to adapt to the millisecond or even microsecond variations in process parameters found in modern production. Existing technologies rely on offline statistical analysis and are unable to respond in real time to millisecond-level process fluctuations. This results in delays in identifying quality issues often reaching hours, which is unacceptable for production lines with fast cycles and high precision requirements, such as electronics manufacturing and precision machining.

[0040] 2) Manual intervention is inefficient and highly subjective: Even when control limits need adjustment, traditional methods often rely on human judgment and manual adjustments. This is not only time-consuming and labor-intensive, but the effectiveness of these adjustments is highly dependent on the engineer's experience and subjective judgment, lacking consistency and optimality. The frequency of these adjustments is far from sufficient to match the actual rate of change in the process.

[0041] (3) Limitations of Data Outlier Processing Data collected during the production process is inevitably subject to various interferences, resulting in outliers, such as transient sensor failures, recording errors, or extreme disturbances that cannot be reproduced in the short term. Traditional SPC methods, when calculating control limits, fail to effectively identify and address these outliers. These outliers directly factor into the calculation of the mean and standard deviation, potentially widening the control limits unreasonably. This reduces the monitoring sensitivity of the control chart and obscures the true process variation. In other words, unaddressed outliers distort the calculation of control limits and reduce monitoring sensitivity.

[0042] In summary, existing SPC systems using static thresholds face significant bottlenecks in addressing the dynamic nature of production processes, improving alarm accuracy, enhancing response speed, optimizing the process capability index (CPK), and robustly handling data anomalies. These shortcomings limit SPC's full potential in modern intelligent manufacturing environments and make it difficult to meet companies' demands for lean production and superior quality control. Robust handling of data anomalies specifically refers to the system's robust anti-interference capabilities, the ability to accurately identify and appropriately handle outliers in the data, ensuring stable and reliable calculation results and judgments. The system also automatically adjusts the outlier identification criteria and handling methods based on data characteristics, thereby avoiding false positives or negatives caused by outliers.

[0043] In addition, the limitations of traditional static threshold alarms are becoming increasingly prominent in digital system operations and maintenance, with false alarms and missed alarms occurring frequently, which also confirms similar challenges faced in the SPC field.

[0044] In order to address the above-mentioned limitations of the traditional SCP method based on static thresholds, the embodiments of the present application provide a new SPC control solution that can dynamically optimize control thresholds and intelligently adapt to changes in the production process.

[0045] The technical solutions provided in the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0046] Figure 1 FIG. 1 is a flow chart showing an SPC control method based on dynamic threshold optimization according to an embodiment of the present application. Figure 1 As shown, the method may include the following steps: S100, collects raw data streams containing key quality characteristics or key process parameters from various data sources on the production line in real time.

[0047] In one implementation, the data collection frequency can be pre-configured based on process requirements, up to milliseconds. Based on this frequency, raw data streams containing key quality characteristics (KQCss) or key process parameters (KPPs) are collected in real time from various data sources on the production line.

[0048] As an example, various data sources may include, but are not limited to, programmable logic controllers (PLCs), supervisory control and data acquisition (SCADA) systems, manufacturing execution systems (MES), sensors, and detection equipment.

[0049] As an example, it can support communication connections with various data sources through multiple industrial communication protocols (such as OPC UA, Modbus TCP / IP, MQTT, etc.) and database interfaces, ensuring seamless integration with existing production information systems.

[0050] In an embodiment of the present application, by executing step S100, the original data stream including at least one key quality characteristic or at least one key process parameter from the production process can be captured in real time, thereby improving the accuracy of subsequent real-time monitoring and analysis.

[0051] S110: Preprocess the original data stream to obtain a structured valid data sequence.

[0052] In one embodiment, the implementation process of step S110 may include the following steps: S111. Clean the original data stream to obtain a valid data stream.

[0053] In specific implementation, the original data stream can be cleaned as follows: removing obvious noise and processing missing values ​​(such as interpolating or eliminating missing values). In this way, the interfering data in the original data stream can be removed to obtain a valid data stream.

[0054] S112. Perform data type conversion on the valid data stream to obtain a valid data sequence.

[0055] During specific implementation, the following data type conversion can be performed on the valid data stream: First, the valid data stream is converted into a unified data type, and then the data is normalized to obtain a valid data sequence, which can facilitate subsequent real-time monitoring and analysis.

[0056] Among them, data normalization (Scaling), also known as data scale normalization, is to map the value range of a certain attribute of the feature (a certain dimension of the feature vector) to a specific range, so as to eliminate the influence of different size ranges of numerical attributes on the fairness of the results of distance-based classification methods.

[0057] S113. Align the valid data sequence according to the timestamp to obtain a structured valid data sequence.

[0058] In specific implementation, by aligning valid data sequences according to timestamps, consistency of timestamps of multi-source data can be ensured.

[0059] In another embodiment, after the valid data sequences are aligned according to the timestamps to obtain the structured valid data sequences, the structured valid data sequences may be further processed as needed, such as normalized or standardized.

[0060] In the embodiment of the present application, by executing step S110, a high-quality structured valid data sequence can be output.

[0061] S120: Using a sliding time window mechanism, intercept valid data within a recent period of time in the structured valid data sequence to obtain current sliding window data.

[0062] In one embodiment, the sliding window data may be obtained by intercepting valid data within a recent period of time in the structured valid data sequence according to a preset sliding window and a preset sliding step size.

[0063] As an example, the preset sliding window may slide according to a first-in-first-out principle.

[0064] As an example, the preset sliding window may include N data points, and the value of N may range from 50 to 200. It can be understood that the size of the preset window may be determined by the N data points. The value of N may be adjusted according to the stability and sensitivity of the specific process.

[0065] As an example, the preset sliding step size may be set to 1, which can be understood as that each time a new data point is received, the preset sliding window slides forward one position.

[0066] In an embodiment of the present application, by executing step S120, the current sliding window data can be used as the data basis for calculating the control limits of the current SPC control chart, that is, the sliding window data can be understood as the data subset currently used for analysis.

[0067] S130 , performing outlier processing on the sliding window data, and dynamically calculating the control limits of the SPC control chart based on the processed sliding window data to obtain first control limits.

[0068] In one embodiment, the implementation process of step S130 may include the following steps: S131 . Perform outlier detection on the sliding window data, filter out the abnormal data in the sliding window data, and obtain processed sliding window data.

[0069] In a specific implementation, the Tukey method may be used to calculate the first quartile Q1, the third quartile Q3, and the interquartile range (IQR) of the sliding window data, where IQR=Q3-Q1.

[0070] In a specific implementation, the outlier limit can be set by applying the criterion of IQR≥K, wherein the upper limit of the outlier limit is Q3+K*IQR, the lower limit is Q1-K*IQR, and K>1.5.

[0071] As an example, the embodiment of the present application may set the K value to 3 to correspond to a stricter outlier screening standard, while the traditional Tukey method usually sets the K value to 1.5 to identify mild outliers.

[0072] Of course, during specific implementation, other values ​​can be configured according to actual needs, and the embodiments of this application do not specifically limit this.

[0073] In a specific implementation, outlier data in the sliding window data can be identified and filtered out based on the outlier limit to obtain processed sliding window data. For example, data in the sliding window data that exceeds the outlier limit can be considered as outlier data that needs to be processed, and these outlier data can be selected to be eliminated or marked with a lower weight. The eliminated or marked outlier data does not participate in the subsequent control limit calculation of the SPC control chart.

[0074] It should be understood that in the implementation of the present application, the abnormal data in the sliding window data is processed only temporarily, such as temporarily eliminated or temporarily marked, so that these abnormal data do not participate in the subsequent control limit calculation of the SPC control chart.

[0075] That is, in the embodiment of the present application, based on the data set of the current sliding window, the Tukey method is applied to identify and filter abnormal data values ​​within the sliding window through IQR.

[0076] In an embodiment of the present application, by executing step S131, extreme data points caused by short-term disturbances or measurement errors can be effectively eliminated, ensuring that the data used to calculate the control limits of the SPC control chart are purer and more representative, preventing the control limits of the SPC control chart from being distorted by outliers, and helping to improve the stability and robustness of the control limits of the SPC control chart.

[0077] S132: Perform a goodness of fit test on the processed sliding window data, and adaptively select the most suitable target statistical distribution model based on the test result.

[0078] In a specific implementation, multiple statistical distribution models can be pre-set. This allows for adaptation of multiple statistical distribution models, improving the accuracy of analysis of different data types. This allows for automatic selection or user configuration of the distribution model that best describes the data characteristics of the current sliding window data. These multiple statistical distribution models may include, but are not limited to, normal distribution models, exponential distribution models, and Poisson distribution models.

[0079] During specific implementation, a goodness of fit test can be performed on the processed sliding window data, and the most suitable target statistical distribution model can be adaptively selected based on the test results.

[0080] As an example, a goodness-of-fit test, such as a Kolmogorov-Smirnov test or a Chi-squared test, can be performed on the processed sliding window data. Then, based on the test results (such as P-value, AIC / BIC criteria, etc.), the most suitable statistical distribution model is selected or confirmed for the current sliding window data. For example, the normal distribution model is used by default, but if the test results indicate that the exponential distribution model is better, the exponential distribution model is switched to, that is, the exponential distribution model is selected as the currently most suitable target statistical distribution model.

[0081] In actual production, many industrial process parameters do not strictly follow a normal distribution. For example, some lifespan data may follow an exponential distribution, and count values ​​may follow a Poisson distribution. Therefore, by executing step S132, the present embodiment can adapt the correct statistical distribution model to the current sliding window data, enabling more accurate estimation of process parameters and calculation of control limits, thereby improving the effectiveness of SPC analysis.

[0082] S133. Based on the processed sliding window data and the target statistical distribution model, dynamically calculate the control limits of the SPC control chart to obtain first control limits.

[0083] In specific implementation, the control limits of the SPC control chart can be dynamically calculated based on the processed sliding window data and the characteristics of the target statistical model (such as mean, standard deviation or parameter estimation of a specific distribution, etc.) to obtain the first control limits, that is, the first control limits include the dynamic center line (CL_dynamic), dynamic upper control limit (UCL_dynamic) and dynamic lower control limit (LCL_dynamic) at the current moment.

[0084] For example, taking the target statistical distribution model as a normal distribution model, CL_dynamic, UCL_dynamic, and LCL_dynamic can be calculated by the following formulas (1) to (3) respectively.

[0085]

[0086]

[0087]

[0088] Here, u_dynamic refers to the dynamic mean, o_dynamic refers to the dynamic standard deviation, and N is usually set to 3 (corresponding to a confidence interval of approximately 99.73%), but can also be adjusted according to process risk assessment and management needs.

[0089] Among them, for other statistical distribution models, the control limit calculation formula of the corresponding distribution can be used, and the embodiments of the present application do not limit this.

[0090] In the embodiment of the present application, by executing step S130, the control limits of the SPC control chart can be adjusted in real time according to the latest process data through the sliding time window mechanism to adapt to the natural fluctuations and slow drifts of the process, thereby ensuring the real-time calculation of the control limits of the SPC control chart, so that it can accurately reflect the actual status of the production process in the near future, rather than relying on long-term historical data.

[0091] S140: When a step change in process parameters of the production line is detected, the control limits of the SPC control chart are updated based on a preset control line adjustment mechanism to obtain second control limits.

[0092] In one embodiment, the implementation process of step S140 may include the following steps: S141. When a step change in process parameters of the production line is detected, the sliding window data is reset or cleared.

[0093] In a specific implementation, the statistical characteristic data of the sliding window data can be analyzed. Based on the analysis results, if the rate of change of the statistical characteristic data exceeds a preset threshold or changes significantly within a short period of time, it is determined that a step change in the process parameters of the production line has occurred. The statistical characteristic data can include variance, mean shift, etc.

[0094] For example, the statistical characteristic data of the sliding window data can be monitored (e.g., in real time or periodically) for changes within a short period of time (e.g., within a number of consecutive data points, such as 3-5 data points). For example, by calculating the variance of the sliding window data and comparing the currently calculated variance with the baseline variance of the previous stable period (which can be dynamically determined). When it is detected that the variance has changed significantly within the short period of time, and its rate of change exceeds a preset threshold (e.g., variance change rate ≥ 15%, this threshold can be configured based on process sensitivity), or when a significant and sustained shift in the mean is detected, it is determined that a step change in the process parameters may have occurred (e.g., a major equipment adjustment, replacement of a major raw material, switching of production modes, etc.), and a step change in the process parameters of the production line is determined to have occurred, and the sliding window data can be (immediately) reset or cleared.

[0095] In an embodiment of the present application, by executing step S141, when a step change in the production line process parameters is monitored, the historical data accumulated in the current sliding window (i.e., the current sliding window data) can be immediately reset or cleared to achieve sliding window reset and rapid update of control limits, thereby avoiding the contamination of the control limit calculation of the new state by the data of the old state.

[0096] S142: Determine the sliding window data as a data change point.

[0097] By executing step S142 , the sliding window can be refilled with new valid data collected after the data change point.

[0098] S143. Using a sliding time window mechanism, intercept valid data located after the data change point in the structured valid data sequence to obtain new sliding window data.

[0099] In specific implementation, the implementation process of step S143 is similar to the implementation process of the above-mentioned step S120, and will not be repeated here.

[0100] It should be noted that in step S143, once the new sliding window data reaches a certain amount (e.g., half or all of the minimum window size), step S144 can be immediately executed to quickly calculate dynamic control limits that adapt to the new process state. During this transition period, strategies such as temporarily wider control limits or suspending alarms can be adopted to avoid unnecessary disturbances.

[0101] S144. Perform outlier processing on the new sliding window data, and recalculate the control limits of the SPC control chart based on the processed new sliding window data to obtain second control limits.

[0102] In specific implementation, the implementation process of step S144 is the same as or similar to the implementation process of the above-mentioned step S130, and will not be repeated here.

[0103] In an embodiment of the present application, by executing step S140, it can be ensured that the control limits of the SPC control chart can quickly adapt to the actual state mutation of the process, avoiding monitoring lags or misjudgments caused by using control limits containing old process state data. This is especially important when major changes, planned or unplanned, occur in the process.

[0104] That is, the embodiment of the present application innovatively introduces process parameter step change detection logic (such as based on variance change rate ≥ 15%). Once a sudden change in the process is detected, it can automatically reset the data window and quickly update the control limit, which can greatly shorten the system response delay.

[0105] S150. Compare the latest valid data output from the structured valid data sequence with the first control limit or the second control limit in real time, and trigger an alarm when it is determined that an abnormality occurs in the production process based on the comparison result.

[0106] It should be understood that in step S150, when it is monitored that no process parameter step change has occurred on the production line, the latest valid data output from the structured valid data sequence is compared with the first control limit in real time, and an alarm is triggered when it is determined based on the comparison result that an abnormality has occurred in the production process. Alternatively, when it is monitored that a process parameter step change has occurred on the production line, the latest valid data output from the structured valid data sequence is compared with the second control limit in real time, and an alarm is triggered when it is determined based on the comparison result that an abnormality has occurred in the production process.

[0107] In one embodiment, SPC exception judgment rules can be pre-configured to comprehensively determine whether the process is currently in a statistically controlled state, a warning state, or an out-of-control state. These SPC exception judgment rules may include, but are not limited to, basic Nelson rules and more complex pattern recognition rules. During implementation, these SPC exception judgment rules should be adaptively adjusted under dynamic thresholds.

[0108] In one embodiment, when the comparison result is based on the SPC abnormality judgment rule and it is determined that the production process is out of control (OOC) or a warning signal requiring attention appears, it is determined that the production process is abnormal and an alarm is immediately triggered.

[0109] In one embodiment, in step S150, various alarm forms may be used for alarm, such as sound and light prompts on the system interface, and sending alarm information (such as email, text message, or enterprise instant messaging tool message, etc.) to designated personnel (such as operators, engineers, or managers). The alarm information may record in detail the time, parameters, data points, current control limits of the SPC control chart, and triggering rules of the alarm.

[0110] In an applicable scenario provided in the embodiment of the present application, combined with Figure 1 and Figure 2 As shown, the SPC control method based on dynamic threshold optimization provided in the embodiment of the present application may further include the following steps: S160 : When production change instruction information is received from an external system, the control limits of the SPC control chart are updated based on the control line adjustment mechanism to obtain third control limits.

[0111] In one embodiment, the external system may be a system related to the production process, such as a Manufacturing Execution System (MES).

[0112] In one embodiment, the production instruction information may be order change instruction information, product switching instruction information, process path adjustment instruction information, etc. In the actual production process, when an external system generates a production change event, it may trigger the external system to initiate production change instruction information.

[0113] In one embodiment, the implementation process of step S160 may include the following steps: S161. When receiving production change instruction information from an external system, reset or clear the sliding window data.

[0114] S162: Determine the sliding window data as a data change point.

[0115] S163. Adopting a sliding time window mechanism, intercepting valid data located after the data change point in the structured valid data sequence to obtain new sliding window data.

[0116] S164. Perform outlier processing on the new sliding window data, and recalculate the control limits of the SPC control chart based on the processed new sliding window data to obtain third control limits.

[0117] In specific implementation, the implementation process of step S161 to step S164 is the same as or similar to the implementation process of the above-mentioned step S141 to step S144, and will not be repeated here.

[0118] In an embodiment of the present application, by executing step S160, linkage with an external system can be achieved, and the production change event of the external system can be used as a basis for triggering a sliding window reset or adjusting a dynamic threshold calculation strategy.

[0119] That is, the embodiments of the present application have the potential to interact with external systems such as MES, and can lay the foundation for achieving higher-level intelligent manufacturing collaborative control.

[0120] S170. Compare the latest valid data output from the structured valid data sequence with the third control limit in real time, and trigger an alarm when it is determined based on the comparison result that the production process is out of control.

[0121] In specific implementation, the implementation process of step S170 is the same as or similar to the implementation process of the above-mentioned step S150, and will not be repeated here.

[0122] That is, in the embodiment of the present application, by executing step S150 and step S170, each latest valid data output in the structured valid data sequence can be compared with the corresponding control limit according to the corresponding situation to determine whether there is an abnormality in the production process.

[0123] In another applicable scenario provided in the embodiment of the present application, combined with Figure 1-Figure 3 As shown, the SPC control method based on dynamic threshold optimization provided in the embodiment of the present application may further include the following steps: S180, dynamically update and draw SPC control charts and process capability dashboards on the graphical human-computer interaction interface.

[0124] In one embodiment, the latest process data points, dynamically changing control limits, process status (normal / warning / out of control), alarm information, etc. can be plotted and updated in real time on relevant SPC control charts (such as Xbar-R charts, I-MR charts, etc.) and process capability dashboards.

[0125] For example, data points, dynamically updated center lines, and control limits (UCL / LCL) can be plotted in real time on an SPC control chart. Furthermore, real-time CPK, PPK, and other indicators and their trends can be dynamically displayed on a process capability dashboard.

[0126] In one embodiment, the graphical human-computer interaction interface can provide functions such as alarm list and historical alarm query, parameter configuration interface, etc.

[0127] In an embodiment of the present application, by executing step S180, authorized users can be allowed to view historical data trends, query SPC control charts for a specific time period, confirm and handle alarm events (and also record handling measures and results), and adjust system operating parameters (such as sliding window size, k value for outlier filtering, variance change rate threshold for step change detection, etc.).

[0128] S190. All relevant data of SPC control are persistently stored in the database.

[0129] In one embodiment, all relevant data may include, but are not limited to: original collected data (such as original data stream), pre-processed data (such as structured valid data sequence), previously calculated dynamic control limits (such as first control limit, second control limit and third control limit) and their corresponding calculated sliding window data, alarm event record details (time, type, parameters, value, processing measures, person in charge, etc.), user operations and system events (such as system operation logs, user configuration change records), etc.

[0130] In one embodiment, the database may be a time series database or a relational database, which is not limited in this embodiment of the present application.

[0131] In one embodiment, all relevant data of SPC control can be persistently stored in a database in real time or in batches for query and tracing.

[0132] In the embodiment of the present application, by executing step S190, complete data support can be provided for post-quality analysis, problem tracing, process improvement effect evaluation, and audit requirements.

[0133] In practical applications, the SPC control method based on dynamic threshold optimization provided in the embodiments of the present application can be executed periodically and iteratively. For example, it can start from step S100 or step S110 (if the preprocessing is continuous streaming) and continuously and periodically iterate steps S110 to S150, S110 to S170, or S110 to S190 as new production data continues to arrive, thereby achieving uninterrupted dynamic monitoring and control of the production process. The frequency of iteration depends on the frequency of data collection and the system processing capacity.

[0134] From the above description, it can be seen that the SPC control method based on dynamic threshold optimization provided in the embodiment of the present application can achieve the following beneficial effects: (a) Significantly Reduce False Alarm Rates and Improve Monitoring Accuracy: By dynamically calculating and updating the control limits of the SPC control chart based on sliding window data, the embodiments of the present application can better accommodate normal fluctuations in the process and avoid erroneous judgments caused by overly strict or loose fixed thresholds or interference from abnormal data. The false alarm rate in the production process can be significantly reduced from the 8%-12% common in traditional SPC to ≤2%. This greatly reduces unnecessary production interruptions and waste of resources, and enhances operators' trust and reliance on the SPC control system.

[0135] At the same time, the Tukey method is used to filter outliers in the sliding window data to ensure the purity of the data used to calculate the control limits, which can further reduce the false alarm rate and improve monitoring accuracy.

[0136] (b) Effectively improve the process capability index (CPK): The embodiment of the present application uses a sliding time window mechanism combined with process parameter step change monitoring to dynamically calculate and update the control limits of the SPC control chart. This can more accurately and more adapt to the control limits of the actual state of the process, and help to truly reflect the inherent variation and offset of the process, thereby guiding more effective process improvement activities. When process stability is effectively monitored and improved, the process capability index (CPK) will naturally be improved. In actual production applications, the embodiment of the present application is expected to increase the process capability index CPK from the general static threshold level of 1.0-1.2 to above 1.5. A CPK of 1.5 usually means that the process is in good condition, the defective rate is extremely low, and product quality is significantly guaranteed.

[0137] (c) Enhanced adaptability to dynamic production environments: This embodiment of the present application utilizes a sliding time window mechanism, enabling the control limits of SPC control charts to smoothly follow the natural evolution of the process (e.g., slow drift caused by equipment wear). Furthermore, combined with a mechanism for monitoring and responding to step changes in process parameters (e.g., automatically resetting the sliding window when the variance change rate is ≥15%), this allows for rapid adaptation to sudden changes in process conditions caused by equipment adjustments, raw material batch changes, production mode switching, and other factors, ensuring the effectiveness and timeliness of control limits in various dynamic scenarios.

[0138] (d) Significantly Improved Response Efficiency and Automation: The present embodiments achieve end-to-end automation from data acquisition, preprocessing, control limit calculation, process state determination, and alarming. In particular, process step changes can be automatically identified and rapidly adjusted without manual intervention. This automated control response time can be reduced from hours with traditional methods to seconds (e.g., ≤1 second), significantly improving the efficiency of problem detection and resolution and reducing potential quality losses.

[0139] (e) Optimizing the data processing process and improving the scientificity and robustness of system decisions: The embodiments of the present application effectively filter abnormal data values, making the basis of statistical analysis more solid and reliable, thereby making subsequent control decisions (such as determining whether the process is out of control, whether process parameters need to be adjusted, etc.) more scientific and accurate, and enhancing the robustness of the entire SPC control system.

[0140] At the same time, the embodiment of the present application can further make the basis of statistical analysis more solid and reliable through adaptive selection of data distribution models.

[0141] (f) Providing comprehensive data-driven decision support and a foundation for continuous improvement: The present embodiments not only provide real-time monitoring but also comprehensively record raw data, processing progress, control limit changes, alarm events, and processing results. This historical data, combined with powerful visualization and analysis tools, provides process engineers with in-depth insights into process behavior, supporting data-driven decision-making and providing a solid data foundation and validation platform for continuous quality improvement and process optimization (e.g., parameter optimization and reduction of sources of variation).

[0142] In summary, the SPC control method based on dynamic threshold optimization provided in the embodiments of the present application can solve the problems caused by traditional SPC static thresholds, can adapt to the dynamic characteristics of the production process in real time, intelligently adjust the control limits, and effectively filter out abnormal data interference, and can achieve more accurate, efficient, and intelligent process quality control, thereby improving overall production quality and efficiency.

[0143] Figure 4 FIG. 1 shows a structural block diagram of an SPC control system based on dynamic threshold optimization according to an embodiment of the present application. Figure 4 As shown, the system may include: The data acquisition module 210 is used to collect raw data streams containing key quality characteristics or key process parameters from various data sources on the production line in real time; The data preprocessing module 220 is used to preprocess the original data stream to obtain a structured valid data sequence; The dynamic threshold calculation engine module 230 is configured to use a sliding time window mechanism to intercept valid data within a recent period of time in the structured valid data sequence to obtain current sliding window data; perform outlier processing on the sliding window data, and dynamically calculate the control limits of the SPC control chart based on the processed sliding window data to obtain first control limits; The intelligent collaborative control logic module 240 is used to update the control limits of the SPC control chart based on a preset control line adjustment mechanism to obtain second control limits when a step change in process parameters of the production line is detected; The process status monitoring and alarm module 250 is used to compare the latest valid data output from the structured valid data sequence with the first control limit or the second control limit in real time, and trigger an alarm when it is determined that an abnormality occurs in the production process based on the comparison result.

[0144] In one embodiment, when the data preprocessing module 220 is used to preprocess the original data stream to obtain a structured valid data sequence, it is specifically used to: Clean the original data stream to obtain a valid data stream; Perform data type conversion on the valid data stream to obtain a valid data sequence; The valid data sequences are aligned according to the timestamps to obtain a structured valid data sequence.

[0145] In one embodiment, the dynamic threshold calculation engine module 230 is used to perform outlier processing on the sliding window data and dynamically calculate the control limits of the SPC control chart based on the processed sliding window data to obtain the first control limits, specifically for: Perform outlier detection on the sliding window data, filter out the abnormal data in the sliding window data, and obtain the processed sliding window data; Perform a goodness of fit test on the processed sliding window data and adaptively select the most suitable target statistical distribution model based on the test results; Based on the processed sliding window data and the target statistical distribution model, the control limits of the SPC control chart are dynamically calculated to obtain the first control limits.

[0146] In one embodiment, the dynamic threshold calculation engine module 230 is used to perform outlier detection on the sliding window data, filter out the abnormal data in the sliding window data, and obtain the processed sliding window data, specifically to: Tukey's method was used to calculate the first quartile Q1, third quartile Q3 and interquartile range IQR of the sliding window data; The outlier limit is set using the criterion of IQR ≥ K, where the upper limit of the outlier limit is Q3 + K*IQR and the lower limit is Q1-K*IQR, and K>1.5; Based on the outlier limit, the outlier data in the sliding window data is identified and filtered out to obtain the processed sliding window data.

[0147] In one embodiment, the intelligent collaborative control logic module 240 is configured to update the control limits of the SPC control chart based on a preset control line adjustment mechanism when a step change in a process parameter of the production line is detected. When the second control limits are obtained, the intelligent collaborative control logic module 240 is configured to: When a step change in process parameters of the production line is detected, the sliding window data is reset or cleared; Determine the sliding window data as the data change point; Adopting the sliding time window mechanism, the valid data after the data change point in the structured valid data sequence is intercepted to obtain the new sliding window data; The new sliding window data is processed for outliers, and the control limits of the SPC control chart are recalculated based on the processed new sliding window data to obtain the second control limits.

[0148] In one embodiment, the intelligent collaborative control logic module 240 is further configured to update the control limits of the SPC control chart based on the control line adjustment mechanism upon receiving production change instruction information from an external system to obtain third control limits; The process status monitoring and alarm module 250 is further used to compare the latest valid data output from the structured valid data sequence with the third control limit in real time, and trigger an alarm when it is determined that an abnormality occurs in the production process based on the comparison result.

[0149] In one embodiment, the system further comprises: A visualization and human-computer interaction module 260 is used to dynamically update and draw SPC control charts and process capability dashboards on a graphical human-computer interaction interface; The data storage and tracing module 270 is used to persistently store all relevant data of SPC control in a database.

[0150] In practical applications, these modules can exchange data and pass commands through efficient internal message queues or APIs. Integration with external systems (such as MES and SCADA) can rely on standard industrial network protocols and interfaces. Furthermore, the overall design of the SPC control system based on dynamic threshold optimization in the embodiments of this application should ensure real-time data flow and low latency in system response.

[0151] The functions of each module in the SPC control system based on dynamic threshold optimization in the embodiment of the present application can be found in the corresponding description in the above method, and will not be repeated here.

[0152] Figure 5 FIG. 1 shows a structural block diagram of an electronic device according to an embodiment of the present application. Figure 5 As shown, the electronic device includes: a memory 310 and a processor 320. The memory 310 stores instructions, which are loaded and executed by the processor 320 to implement the SPC control method based on dynamic threshold optimization in the above embodiment. The number of the memory 310 and the processor 320 can be one or more.

[0153] The electronic device also includes: The communication interface 330 is used to communicate with external devices and perform data exchange transmission.

[0154] If the memory 310, processor 320, and communication interface 330 are implemented independently, the memory 310, processor 320, and communication interface 330 can be connected to each other via a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one thick line is used in the diagram, but this does not mean that there is only one bus or one type of bus.

[0155] Optionally, in a specific implementation, if the memory 310, the processor 320 and the communication interface 330 are integrated on a chip, the memory 310, the processor 320 and the communication interface 330 can communicate with each other through an internal interface.

[0156] An embodiment of the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program runs on a computer, the method provided in the embodiment of the present application is implemented.

[0157] An embodiment of the present application also provides a chip, which includes a processor for calling and executing instructions stored in a memory, so that a communication device equipped with the chip executes the method provided in the embodiment of the present application.

[0158] An embodiment of the present application also provides a chip, including: an input interface, an output interface, a processor and a memory. The input interface, the output interface, the processor and the memory are connected through an internal connection path. The processor is used to execute the code in the memory. When the code is executed, the processor is used to execute the method provided in the embodiment of the application.

[0159] It should be understood that the processor described above may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor. It is worth noting that the processor may be a processor that supports the Advanced RISC Machines (ARM) architecture.

[0160] Furthermore, optionally, the above-mentioned memory may include read-only memory and random access memory, and may also include non-volatile random access memory. The memory may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Among them, the non-volatile memory may include read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM) or flash memory. The volatile memory may include random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available. For example, static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM) and direct rambus RAM (DR RAM).

[0161] In the above embodiments, all or part of the embodiments may be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments may be implemented in the form of a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions according to the present application are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another.

[0162] In the description of this specification, the reference terms "one embodiment," "some embodiments," "example," "specific example," or "some examples" mean that the specific features, structures, materials, or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. Moreover, the specific features, structures, materials, or characteristics described may be combined in any appropriate manner in any one or more embodiments or examples. In addition, those skilled in the art may combine and combine different embodiments or examples described in this specification, as well as features of different embodiments or examples, unless they are mutually inconsistent.

[0163] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. Throughout the description of this application, "plurality" means two or more, unless otherwise specifically defined.

[0164] Any process or method description in a flow chart or otherwise described herein can be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a specific logical function or process. The scope of the preferred embodiments of the present application includes additional implementations in which the functions may be performed in a different order than shown or discussed, including in a substantially simultaneous manner or in a reverse order depending on the functions involved.

[0165] The logic and / or steps represented in the flowchart or otherwise described herein may be considered, for example, as a sequenced list of executable instructions for implementing the logical functions, and may be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device).

[0166] It should be understood that various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. All or part of the steps of the above embodiment method can be completed by instructing the relevant hardware through a program, which can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0167] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing module, or each unit may exist physically separately, or two or more units may be integrated into a single module. The aforementioned integrated modules may be implemented in the form of hardware or in the form of software functional modules. If the aforementioned integrated modules are implemented in the form of software functional modules and sold or used as independent products, they may also be stored in a computer-readable storage medium. The storage medium may be a read-only memory, a magnetic disk, or an optical disk, etc.

[0168] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any person skilled in the art can easily conceive of various modifications or substitutions within the technical scope disclosed in this application, and such modifications or substitutions should be included within the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. An SPC control method based on dynamic threshold optimization, characterized in that: include: Collect raw data streams containing key quality characteristics or key process parameters from various data sources on the production line in real time; Preprocessing the original data stream to obtain a structured valid data sequence; Adopting a sliding time window mechanism, intercepting the valid data in the most recent period of time in the structured valid data sequence to obtain the current sliding window data; Performing outlier processing on the sliding window data, and dynamically calculating control limits of an SPC control chart based on the processed sliding window data to obtain first control limits; When a step change in process parameters of the production line is detected, the control limits of the SPC control chart are updated based on a preset control line adjustment mechanism to obtain a second control limit; The latest valid data output from the structured valid data sequence is compared with the first control limit or the second control limit in real time, and an alarm is triggered when it is determined that an abnormality occurs in the production process based on the comparison result.

2. The method according to claim 1, characterized in that Preprocessing the original data stream to obtain a structured valid data sequence includes: Cleaning the original data stream to obtain a valid data stream; Performing data type conversion on the valid data stream to obtain a valid data sequence; The valid data sequences are aligned according to timestamps to obtain a structured valid data sequence.

3. The method according to claim 1, characterized in that Performing outlier processing on the sliding window data, and dynamically calculating the control limits of the SPC control chart based on the processed sliding window data, to obtain the first control limits includes: Performing outlier detection on the sliding window data, filtering out abnormal data in the sliding window data, and obtaining the processed sliding window data; Performing a goodness of fit test on the processed sliding window data, and adaptively selecting the most suitable target statistical distribution model according to the test result; Based on the processed sliding window data and the target statistical distribution model, the control limits of the SPC control chart are dynamically calculated to obtain the first control limits.

4. The method according to claim 3, characterized in that Performing outlier detection on the sliding window data, filtering out abnormal data in the sliding window data, and obtaining processed sliding window data includes: The Tukey method is used to calculate the first quartile Q1, the third quartile Q3 and the interquartile range IQR of the sliding window data; The outlier limit is set using the criterion of IQR ≥ K, where the upper limit of the outlier limit is Q3 + K*IQR and the lower limit is Q1-K*IQR, and K>1.5; Based on the outlier limit, outlier data in the sliding window data is identified and filtered out to obtain the processed sliding window data.

5. The method according to claim 1, wherein When a step change in process parameters of the production line is detected, the control limits of the SPC control chart are updated based on a preset control line adjustment mechanism, and the second control limits include: When a step change in process parameters of the production line is detected, resetting or clearing the sliding window data; Determining the sliding window data as a data change point; Adopting a sliding time window mechanism, intercepting valid data located after the data change point in the structured valid data sequence to obtain new sliding window data; Outlier processing is performed on the new sliding window data, and the control limits of the SPC control chart are recalculated based on the processed new sliding window data to obtain the second control limits.

6. The method according to any one of claims 1 to 5, characterized in that The method further comprises: When receiving production change instruction information from an external system, updating the control limits of the SPC control chart based on the control line adjustment mechanism to obtain third control limits; The latest valid data output from the structured valid data sequence is compared with the third control limit in real time, and an alarm is triggered when it is determined based on the comparison result that an abnormality occurs in the production process.

7. The method according to any one of claims 1 to 5, characterized in that The method further comprises: Dynamically update and draw SPC control charts and process capability dashboards on the graphical human-computer interaction interface; All relevant data of SPC control are persistently stored in the database.

8. An SPC control system based on dynamic threshold optimization, characterized in that: include: Data acquisition module, used to collect raw data streams containing key quality characteristics or key process parameters from various data sources on the production line in real time; A data preprocessing module, configured to preprocess the original data stream to obtain a structured valid data sequence; A dynamic threshold calculation engine module is configured to use a sliding time window mechanism to intercept valid data within a recent period of time in the structured valid data sequence to obtain current sliding window data; perform outlier processing on the sliding window data, and dynamically calculate the control limits of the SPC control chart based on the processed sliding window data to obtain first control limits; an intelligent collaborative control logic module, configured to update the control limits of the SPC control chart based on a preset control line adjustment mechanism to obtain second control limits when a step change in process parameters of the production line is detected; The process status monitoring and alarm module is used to compare the latest valid data output from the structured valid data sequence with the first control limit or the second control limit in real time, and trigger an alarm when it is determined that an abnormality occurs in the production process based on the comparison result.

9. An electronic device, characterized in that: include: A memory and a processor, wherein the memory stores instructions, and the instructions are loaded and executed by the processor to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is run on a computer, the method according to any one of claims 1 to 7 is implemented.

Images