A data processing method and device based on a data acquisition and monitoring control system

Abstract

The application discloses a data processing method and device based on a data acquisition and monitoring control system, comprising the following steps: receiving configuration parameters sent by a user, and obtaining corresponding process quality characteristic data from a plurality of preset device bit numbers according to the configuration parameters; for each process quality characteristic data, performing preprocessing on the process quality characteristic data according to an adaptive and interpolation algorithm to obtain standardized sub-group data; performing adaptive processing on the standardized sub-group data by using an adaptive algorithm to obtain target sub-group data; judging whether an industrial device of the device bit number corresponding to the target sub-group data exists process abnormality according to a pre-configured discrimination rule; and if the process abnormality exists, generating alarm information based on the target sub-group data. Thus, by deeply fusing SCADA data acquisition and SPC analysis, performing standardized preprocessing and adaptive processing on process quality characteristic data by using an adaptive and interpolation algorithm, real-time connection of data and analysis is realized.

Description

Technical Field

[0001] This application relates to the field of industrial automation and control technology, and in particular to a data processing method and apparatus based on a data acquisition and monitoring control system. Background Technology

[0002] Supervisory Control and Data Acquisition (SCADA) is a crucial technology widely used for monitoring and controlling industrial processes. It utilizes field sensors and programmable logic controllers (PLCs) to acquire industrial data, monitor status, and remotely control processes. Statistical Process Control (SPC), based on mathematical statistics principles, analyzes quality characteristic data, calculates process capability indices, and generates control charts. It identifies abnormal fluctuations based on anomaly detection rules, providing early warnings for the process. In industrial settings, SCADA and SPC respectively play the roles of monitoring and quality analysis.

[0003] Currently, mainstream SCADA software primarily focuses on data acquisition, equipment monitoring, and process visualization, generally lacking built-in process statistical analysis capabilities. This means that when users need to perform process capability analysis or create control charts, they must first export data from the SCADA system and then import it into third-party SPC software via a dump method. Subsequently, they configure parameters such as subgroup size and control limit type in professional statistical software, and finally use that software to create control charts and manually analyze the results.

[0004] Because mainstream SCADA software lacks built-in process statistical analysis capabilities, SCADA data acquisition and SPC analysis are completely separated. This makes it impossible to achieve real-time triggering analysis, dynamic updating of control charts, and integrated display within SCADA, resulting in a serious lag in the detection of process anomalies, as well as complex operation and fragmented interfaces. Summary of the Invention

[0005] In view of the shortcomings of the prior art, this application provides a data processing method and apparatus based on a data acquisition and monitoring control system to solve the problem of complete separation between SCADA data acquisition and SPC analysis.

[0006] To achieve the above objectives, this application provides the following technical solution:

[0007] The first aspect of this application provides a data processing method based on a data acquisition and monitoring control system, including:

[0008] Receive configuration parameters sent by the user, and obtain corresponding process quality characteristic data from multiple preset device tag numbers according to the configuration parameters;

[0009] For each of the process quality feature data, the process quality feature data is preprocessed according to an adaptive and interpolation algorithm to obtain standardized subgroup data;

[0010] The standardized subgroup data is adaptively processed using an adaptive algorithm to obtain the target subgroup data;

[0011] Based on the pre-configured anomaly detection rules, determine whether there is a process anomaly in the industrial equipment corresponding to the device tag number of the target subgroup data;

[0012] If an industrial device with the device tag number corresponding to the target subgroup data has a process abnormality, an alarm message is generated based on the target subgroup data.

[0013] Optionally, in the above-described data processing method based on a data acquisition and monitoring control system, the step of obtaining corresponding process quality characteristic data from a preset plurality of device tag numbers according to the configuration parameters includes:

[0014] Extract the sampling frequency, subgroup value and sampling interval corresponding to multiple preset device tag numbers from the configuration parameters;

[0015] For each device tag number, real-time data is obtained from the device tag number according to the sampling frequency corresponding to the device tag number;

[0016] The real-time data is grouped according to the subgroup values ​​and the sampling interval to obtain feature data;

[0017] Each set of feature data is preprocessed to obtain process quality feature data.

[0018] Optionally, in the above-described data processing method based on data acquisition and monitoring control system, the step of preprocessing each process quality feature data according to an adaptive and interpolation algorithm to obtain standardized subgroup data includes:

[0019] For each device tag number, the sampling interval corresponding to the device tag number is extracted from the configuration parameters;

[0020] Using the system clock as a reference, the timestamp of the process quality characteristic data corresponding to the device tag number is converted to obtain the target process quality characteristic data;

[0021] Based on the sampling interval, the device tag number is periodically identified to obtain the actual sampling period;

[0022] Determine whether the actual sampling period is less than the sampling interval;

[0023] If the actual sampling period is less than the sampling interval, the target process quality characteristic data are aggregated using a sliding window algorithm to obtain subgroup data.

[0024] If the actual sampling period is greater than the sampling interval, the target process quality feature data is filled using the nearest neighbor forward filling algorithm to obtain subgroup data.

[0025] Detect whether there are any missing data in the subgroup data;

[0026] If there are missing data in the subgroup data, a triple imputation strategy is used to imputate the subgroup data to obtain standardized subgroup data.

[0027] If there are no missing data in the subgroup data, then the subgroup data is used as standardized subgroup data.

[0028] Optionally, the data processing method based on the data acquisition and monitoring control system described above further includes:

[0029] The standardized subgroup data is labeled using a box plot method, and the labeled standardized subgroup data is then fed back to the user.

[0030] Optionally, in the above-described data processing method based on data acquisition and monitoring control system, the step of adaptively processing the standardized subgroup data using an adaptive algorithm to obtain the target subgroup data includes:

[0031] Calculate the mean and process standard deviation of the standardized subgroup data;

[0032] Based on the mean and the process standard deviation, it is determined whether the current process of the target industrial equipment is in a stable state; wherein, the target industrial equipment refers to the industrial equipment with the equipment tag number corresponding to the standardized subgroup data;

[0033] If the current process of the target industrial equipment is in a stable state, the control limits and process capabilities of the target industrial equipment are calculated asymptotically based on the mean and the process standard deviation.

[0034] If the current process of the target industrial equipment is not in a stable state, then obtain the historical false alarm rate and historical false alarm rate of the target industrial equipment;

[0035] Based on the historical false alarm rate and historical false alarm rate, the control limits of the target industrial equipment are updated, and the process capability of the target industrial equipment is calculated based on the mean and the process standard deviation.

[0036] The control limits, the process capability, and the standardized subgroup data are used as the target subgroup data.

[0037] Optionally, the data processing method based on the data acquisition and monitoring control system described above further includes:

[0038] Extract the control limits and process capabilities from the target subgroup data;

[0039] Based on the control limits, process capabilities, and alarm information, a control chart is drawn and displayed on a visual interface.

[0040] Optionally, the data processing method based on the data acquisition and monitoring control system described above further includes:

[0041] When the update instruction for the control limit is received, the control limit is smoothed using an exponentially weighted moving average algorithm to obtain the target control limit.

[0042] A second aspect of this application provides a data processing device based on a data acquisition and monitoring control system, comprising:

[0043] The data acquisition unit is used to receive configuration parameters sent by the user and obtain corresponding process quality characteristic data from a plurality of preset device tag numbers according to the configuration parameters;

[0044] The preprocessing unit is used to preprocess each process quality feature data according to an adaptive and interpolation algorithm to obtain standardized subgroup data.

[0045] An adaptive processing unit is used to adaptively process the standardized subgroup data using an adaptive algorithm to obtain target subgroup data;

[0046] An anomaly detection unit is used to determine whether there is a process anomaly in the industrial equipment corresponding to the device tag number of the target subgroup data, based on pre-configured anomaly detection rules.

[0047] The generation unit is used to generate alarm information based on the target subgroup data if an industrial device with the device tag number corresponding to the target subgroup data has a process abnormality.

[0048] Optionally, in the data processing device based on the data acquisition and monitoring control system described above, the data acquisition unit includes:

[0049] The first extraction unit is used to extract the sampling frequency, subgroup value and sampling interval corresponding to multiple preset device tag numbers from the configuration parameters;

[0050] The first acquisition unit is used to acquire corresponding real-time data from the device tag number according to the sampling frequency corresponding to the device tag number, respectively.

[0051] A grouping unit is used to group the real-time data according to the subgroup value and the sampling interval to obtain feature data;

[0052] The processing unit is used to preprocess each set of feature data to obtain process quality feature data.

[0053] Optionally, in the data processing device based on the data acquisition and monitoring control system described above, the preprocessing unit includes:

[0054] The second extraction unit is used to extract the sampling interval corresponding to each device tag number from the configuration parameters for each device tag number.

[0055] The conversion unit is used to convert the timestamp of the process quality characteristic data corresponding to the device tag number based on the system clock to obtain the target process quality characteristic data.

[0056] The identification unit is used to periodically identify the device tag number according to the sampling interval to obtain the actual sampling period;

[0057] The judgment unit is used to determine whether the actual sampling period is less than the sampling interval;

[0058] An aggregation unit is used to aggregate the target process quality characteristic data using a sliding window algorithm to obtain subgroup data if the actual sampling period is less than the sampling interval.

[0059] A filling unit is used to fill the target process quality feature data using the nearest neighbor forward filling algorithm if the actual sampling period is greater than the sampling interval, so as to obtain subgroup data.

[0060] The detection unit is used to detect whether there are any missing data in the subgroup data;

[0061] An interpolation processing unit is used to interpolate the subgroup data using a triple interpolation strategy if there are missing data in the subgroup data, so as to obtain standardized subgroup data.

[0062] The first unit is used to treat the subgroup data as standardized subgroup data if there are no missing data in the subgroup data.

[0063] Optionally, the data processing device based on the data acquisition and monitoring control system described above further includes:

[0064] The labeling unit is used to label the standardized subgroup data using the box plot method and then feed the labeled standardized subgroup data back to the user.

[0065] Optionally, in the data processing device based on the data acquisition and monitoring control system described above, the adaptive processing unit includes:

[0066] The first calculation unit is used to calculate the mean and process standard deviation of the standardized subgroup data;

[0067] A state determination unit is used to determine whether the current process of the target industrial equipment is in a stable state based on the mean and the process standard deviation; wherein, the target industrial equipment refers to the industrial equipment with the equipment tag number corresponding to the standardized subgroup data;

[0068] The second calculation unit is used to calculate the control limits and process capabilities of the target industrial equipment using an asymptotic method based on the mean and the process standard deviation if the current process of the target industrial equipment is in a stable state.

[0069] The second acquisition unit is used to acquire the historical false alarm rate and historical false alarm rate of the target industrial equipment if the current process of the target industrial equipment is not in a stable state.

[0070] An update unit is used to update the control limits of the target industrial equipment based on the historical false alarm rate and the historical false alarm rate, and to calculate the process capability of the target industrial equipment based on the mean and the process standard deviation.

[0071] The second unit is used to use the control limits, the process capabilities, and the standardized subgroup data as target subgroup data.

[0072] Optionally, the data processing device based on the data acquisition and monitoring control system described above further includes:

[0073] The second extraction unit is used to extract the control limits and the process capabilities from the target subgroup data;

[0074] The display unit is used to draw and display a control chart on a visual display interface based on the control limits, the process capabilities, and the alarm information.

[0075] Optionally, the data processing device based on the data acquisition and monitoring control system described above further includes:

[0076] The smoothing processing unit is used to perform smooth transition processing on the control limit using an exponentially weighted moving average algorithm when it receives the update instruction of the control limit, so as to obtain the target control limit.

[0077] This application provides a data processing method based on a data acquisition and monitoring control system. It receives configuration parameters sent by the user and retrieves corresponding process quality characteristic data from multiple preset device tag numbers based on these parameters. Then, for each process quality characteristic data, it preprocesses the data using adaptive and interpolation algorithms to obtain standardized subgroup data. Next, it adaptively processes the standardized subgroup data using an adaptive algorithm to obtain target subgroup data. Finally, based on pre-configured anomaly detection rules, it determines whether the industrial equipment corresponding to the device tag number of the target subgroup data has a process anomaly. If the industrial equipment corresponding to the device tag number of the target subgroup data has a process anomaly, an alarm message is generated based on the target subgroup data. Thus, by deeply integrating SCADA data acquisition and SPC analysis, and using adaptive and interpolation algorithms for standardized preprocessing and adaptive processing of process quality characteristic data, it achieves real-time data and analysis integration. It automatically identifies process anomalies and generates alarms based on anomaly detection rules, effectively solving the drawbacks of their separation and improving the timeliness and accuracy of anomaly detection. Attached Figure Description

[0078] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of this application. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0079] Figure 1 A flowchart illustrating a data processing method based on a data acquisition and monitoring control system provided in this application embodiment;

[0080] Figure 2 A flowchart illustrating a method for preprocessing process quality characteristic data, provided as another embodiment of this application;

[0081] Figure 3 A flowchart illustrating a method for acquiring standardized subgroup data according to another embodiment of this application;

[0082] Figure 4 A flowchart illustrating a method for acquiring target subgroup data according to another embodiment of this application;

[0083] Figure 5 A flowchart illustrating a method for drawing and displaying control charts, provided in another embodiment of this application;

[0084] Figure 6 A system network topology diagram provided in another embodiment of this application;

[0085] Figure 7 This is a schematic diagram of the structure of a data processing device based on a data acquisition and monitoring control system, which is provided in another embodiment of this application. Detailed Implementation

[0086] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0087] In this application, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0088] In one optional embodiment, this application provides a data processing method based on a data acquisition and monitoring control system, the core of which lies in the deep integration of the Supervisory Control and Data Acquisition (SCADA) system with statistical process control (SPC) functions. This method adopts a front-end / back-end separated architecture, deploying the SPC analysis engine as a core service within the SCADA system.

[0089] Specifically, the front-end and back-end separated architecture consists of four parts: The first part is the core of the SPC analysis engine built on the JAVA framework, which realizes statistical calculation and process capability analysis. The second part is the embedded data acquisition and processing module, which directly obtains process quality characteristic data from the SCADA real-time database. The third part is the web-based configuration management module, which provides a browser-accessible process information configuration interface. The fourth part is the visualization module, which embeds control charts and analysis results into the SCADA native human-machine interface (HMI).

[0090] The SPC analysis engine, developed using the Java framework, serves as the core component for statistical calculations, integrating three main functions: control chart calculation, process capability analysis, and outlier detection. The control chart calculation module supports various chart types, including P Chart (Number of Nonconforming Items), NP Chart (Number of Nonconforming Items), C Chart (Number of Defects), U Chart (Number of Defects per Unit), and quantitative control charts (Xbar-R Chart, Xbar-S Chart, I-MR Chart). Based on the Poisson or binomial distribution assumptions, this module automatically calculates the center line (CL), upper control limit (UCL), and lower control limit (LCL). The U Chart employs an independent control limit strategy, calculating independent UCL and LCL for each subgroup based on its actual sample size. The process capability analysis module calculates key indicators such as Cp, Cpk, Pp, Ppk, Ca (accuracy index), DPMO, and process sigma level, and automatically assesses the process capability level (A+ to D) based on the Cpk value. The anomaly detection module incorporates commonly used anomaly detection rules such as Western Electric Rules and Nelson Rules, including rules such as "one point exceeds the control limit", "seven consecutive points are on the same side of the center line", and "six consecutive points show an increasing or decreasing trend", to detect out-of-control states in real time.

[0091] Based on the aforementioned front-end and back-end separation architecture, this application provides a data processing method based on a data acquisition and monitoring control system, such as... Figure 1 As shown, the specific steps include:

[0092] S101. Receive the configuration parameters sent by the user, and obtain the corresponding process quality characteristic data from multiple preset device tag numbers according to the configuration parameters.

[0093] Specifically, configuration parameters refer to a set of structured control variables pre-defined by users (such as process engineers or system operators) through a web-based configuration management module. These parameters include at least a list of equipment tag numbers, sampling frequency, subgroup values, sampling intervals, control chart type, anomaly detection rule combinations, specification upper and lower limits, and target values. In essence, configuration parameters define the scope, granularity, grouping rules, and analysis rules for data acquisition.

[0094] The device tag number is a unique identifier for each independently controllable device or measuring point in an industrial field, such as the tag corresponding to a temperature sensor, pressure transmitter, or flow meter.

[0095] Process quality characteristic data refers to the original measurement sequence that directly reflects the quality status of an industrial process, such as the temperature value, pressure value, number of defects, or number of non-standard items collected for a certain equipment tag over a period of time.

[0096] It should be noted that process quality characteristic data can be deeply integrated with the original data acquisition engine of the SCADA system through the embedded data acquisition module, and directly obtain process quality characteristic data from the SCADA real-time database without the need for additional data export or dumping operations.

[0097] Optionally, in another embodiment of this application, a specific implementation of step S101, which involves obtaining corresponding process quality characteristic data from a plurality of preset device tag numbers according to configuration parameters, is as follows: Figure 2 As shown, the specific steps include:

[0098] S201. Extract the sampling frequency, subgroup value and sampling interval corresponding to multiple preset device tag numbers from the configuration parameters.

[0099] Specifically, the sampling frequency represents the number of times data is collected from the device tag number per unit time. The subgroup value refers to the number of sample data points required to constitute a statistical subgroup; for example, each subgroup contains 5 consecutively collected measurements. The sampling interval refers to the time difference between two adjacent sampling operations. For example, the system extracts these three parameters corresponding to each device tag number by parsing the key-value pair fields in the configuration parameters.

[0100] S202. For each device tag number, obtain the corresponding real-time data from the device tag number according to the sampling frequency corresponding to the device tag number.

[0101] As can be understood, real-time data refers to the latest measurement values ​​read from a specified device tag and written to the real-time database by the SCADA system under the drive of the data acquisition engine, according to the sampling frequency configured by the user on the web-based configuration management module. For each device tag, the system generates a data read request based on the sampling frequency specified in its configuration parameters, and retrieves the raw reading sequence within the current moment or the most recent continuous time window from the real-time data buffer of that tag. This step utilizes the SCADA system's inherent data acquisition channel, eliminating the need for additional acquisition programs.

[0102] The web-based configuration management module adopts a front-end / back-end separation architecture, providing an independent web configuration interface. Users can complete the entire SPC analysis configuration process through a browser. This module is deeply integrated with the SCADA system's device tag number system, supporting the direct binding of configuration parameters to SCADA device tag numbers. Configuration content includes: monitoring variable configuration (selecting the device tag number to be monitored and the data acquisition point), process specification configuration (setting the upper limit (USL), lower limit (LSL), and target value (Target), subgroup parameter configuration (setting the subgroup size, number of subgroups, and sampling interval), control chart parameter configuration (selecting the control chart type and setting the control limit coefficients, defaulting to 3σ), and outlier rule configuration (selecting the combination of outlier rules to be enabled). All configuration information is stored in the SCADA system configuration library in structured data form, and the SPC analysis engine automatically reads and applies these configurations at runtime. The configuration interface uses a visual form and drag-and-drop operation, eliminating the need for users to write any code or perform complex data mapping. Even operators without a statistical background can quickly complete the deployment of SPC analysis tasks.

[0103] S203. Based on the subgroup values ​​and sampling intervals, the real-time data is grouped to obtain feature data.

[0104] Specifically, grouping real-time data involves dividing continuously collected real-time data into multiple independent or partially overlapping data blocks according to a fixed number of samples (i.e., subgroup values) and a fixed time span (i.e., sampling interval). Each data block is a feature data unit. For example, if the subgroup value is set to 5 and the sampling interval is 1 minute, then every 5 consecutive measurements are divided into a group, and the starting time interval of each group is 1 minute. The grouping operation ensures that the data structure on which subsequent statistical calculations are based has a uniform organizational format.

[0105] S204. Preprocess each set of feature data to obtain process quality feature data.

[0106] Specifically, preprocessing includes at least data cleaning, outlier filtering, and dimensional normalization. Data cleaning removes invalid data points that are significantly beyond the equipment's measurement range or caused by communication errors. Outlier filtering can use methods based on fixed thresholds or statistical rules (such as the 3σ principle) to eliminate outliers. Dimensional normalization converts all feature data into a unified physical unit (e.g., all converted to degrees Celsius or megapascals). After the above preprocessing, each set of feature data is transformed into process quality feature data that can be directly used for statistical calculations.

[0107] It should be noted that the embedded data acquisition module also automatically summarizes the total number of samples, the number of qualified products, the number of unqualified products, and the number of defects for each subgroup. For measurement data (such as dimensions, weight, and temperature), the module automatically calculates the mean, range, or standard deviation for each subgroup. The entire process is fully automated, requiring no manual intervention, ensuring a seamless transition from data acquisition to statistical analysis.

[0108] S102. For each process quality feature data, preprocess the process quality feature data according to the adaptive and interpolation algorithm to obtain standardized subgroup data.

[0109] Specifically, adaptive and interpolation algorithms are a set of methods that can automatically adjust processing strategies and compensate for missing data based on the actual data acquisition rhythm. This can solve data loss problems caused by inconsistent data acquisition frequencies for different device tag numbers in SCADA systems, difficulty in aligning timestamps, data packet loss, and network latency. The algorithm first identifies the deviation between the actual sampling period and the theoretical sampling interval for each device tag number, and finally selects either aggregation or imputation operations. Standardized subgroup data refers to a set of subgroup data that has undergone time axis alignment, missing data compensation, and anomaly marking. Its time index is unified, data points are complete, and the number of samples in each subgroup is equal to the preset subgroup value.

[0110] Optionally, in another embodiment of this application, one specific implementation of step S102 is as follows: Figure 3 As shown, the specific steps include:

[0111] S301. For each device tag number, extract the sampling interval corresponding to the device tag number from the configuration parameters.

[0112] Understandably, the system iterates through all process quality characteristic data obtained in step S101, and for each data point corresponding to a device tag number, reads the preset sampling interval for that tag number from the previously stored configuration parameters. This sampling interval serves as the benchmark value for subsequent periodic comparisons.

[0113] S302. Using the system clock as a reference, convert the timestamp of the process quality characteristic data corresponding to the device tag number to obtain the target process quality characteristic data.

[0114] Specifically, the system clock refers to the unified hardware clock or Network Time Protocol (NTP) synchronization clock of the server hosting the SCADA system. Since the raw data from different device tags may contain timestamps recorded by their respective local clocks (e.g., the PLC clock and the SCADA server clock may drift), these timestamps need to be uniformly converted to millisecond-precision timestamps based on the system clock, and a unified time index sequence needs to be established. The conversion process includes time offset compensation and resolution alignment. The data obtained after conversion is called the target process quality characteristic data, and its timeline has global consistency and comparability.

[0115] S303. Based on the sampling interval, periodically identify the device tag number to obtain the actual sampling period.

[0116] Understandably, periodicity identification refers to the statistical analysis of the timestamp sequence of target process quality characteristic data, such as calculating the mode or median of the differences between adjacent timestamps, to determine the actual sampling period of the equipment tag during actual operation. The actual sampling period may differ from the theoretical sampling interval due to communication delays, equipment load, or network packet loss.

[0117] S304. Determine whether the actual sampling period is less than the sampling interval.

[0118] Specifically, in order to distinguish between the two scenarios of excessively dense and sparse data collection and to perform corresponding data processing, it is necessary to compare the actual sampling period with the preset sampling interval in the configuration parameters. If the actual sampling period is less than the sampling interval, then step S305 is executed. If the actual sampling period is greater than the sampling interval, then step S306 is executed.

[0119] S305. Use the sliding window algorithm to aggregate the target process quality characteristic data to obtain subgroup data.

[0120] Specifically, when the actual sampling period is less than the sampling interval, the sliding window algorithm calculates statistical characteristics (such as arithmetic mean, median, or maximum value) for each data point in the target process quality characteristic data falling within each window, thereby downsampling high-frequency data into equally spaced low-frequency data. The time interval of the subgroup data obtained after aggregation processing is consistent with the preset sampling interval.

[0121] S306. Use the nearest neighbor forward filling algorithm to fill the target process quality feature data to obtain subgroup data.

[0122] Understandably, when the actual sampling period is longer than the sampling interval, the nearest neighbor forward fill algorithm is needed to fill the target process quality feature data. Specifically, the nearest neighbor forward fill algorithm means that for each missing time point on the equally spaced target time axis, it searches for the nearest non-missing measurement value along the decreasing time direction (i.e., forward) and copies that measurement value to the current missing position. After filling, the original sparse data is transformed into an equally spaced complete data sequence. This algorithm assumes that the process characteristics do not change significantly within a short time window; therefore, the previous valid value can approximately represent the value at the current missing time.

[0123] S307. Check if there are any missing data in the subgroup data.

[0124] It should be noted that after completing the above aggregation or filling operations, there may still be consecutive gaps in the subgroup data due to communication interruptions, equipment failures, or other reasons (exceeding the reasonable range of nearest neighbor forward filling). This step checks whether each expected time point in the subgroup data has a valid numerical marker to determine whether there are any unfilled gaps. Therefore, if there are gaps in the subgroup data, step S308 is executed. If there are no gaps in the subgroup data, step S309 is executed.

[0125] S308. Using a triple interpolation strategy, the subgroup data is interpolated to obtain standardized subgroup data.

[0126] Specifically, when missing data exists in a subgroup, a triple imputation strategy is used to impute the data. This triple imputation strategy is a combined missing value estimation method. The first imputation is based on time-series prediction of historical data from the same period, for example, using an exponential smoothing model to predict the current missing value based on data trends over several past periods. The second imputation is based on multivariate regression of related reference digits, that is, using measurements of other related reference digits at the same timestamp to estimate the missing value of the current reference digit through a regression model. The third imputation serves as a fallback, using linear interpolation between adjacent time points (i.e., taking the weighted average of two consecutive valid values). Finally, the system weights the triple imputation results according to preset weights (e.g., 0.5, 0.3, 0.2) to obtain the final imputed value, and simultaneously records the data quality label of this value as "imputed data".

[0127] S309. Treat the subgroup data as standardized subgroup data.

[0128] Understandably, when the subgroup data is complete (i.e., all expected time points have valid original data or there are no gaps after nearest neighbor forward filling), triple imputation is no longer triggered, and the current subgroup data is directly output as the standardized subgroup data.

[0129] Optionally, after interpolation is completed, the interpolated or filled data can be marked so that the user can be aware of it. Therefore, in another embodiment of this application, after performing step S309, the method further includes:

[0130] The standardized subgroup data is labeled using the box plot method, and the labeled standardized subgroup data is fed back to the user.

[0131] Specifically, an improved box plot method is used to automatically identify standardized subgroup data and mark outliers that exceed the normal range by 3 times the IQR. Feedback is then sent to the user, who decides whether to remove, correct, or retain the data. This achieves automated preprocessing of SCADA heterogeneous data, reducing data preparation time from hours to seconds.

[0132] S103. Adaptive processing of the standardized subgroup data is performed using an adaptive algorithm to obtain the target subgroup data.

[0133] Specifically, an adaptive algorithm refers to an algorithm that can automatically select the control limit calculation path based on whether the current process is in a statistical steady state. This algorithm evaluates the process mean and standard deviation in real time through a sliding window and dynamically adjusts the control limit coefficients based on historical false alarm and false negative rates. The target subgroup data refers to a composite data structure that includes standardized subgroup data and their corresponding control limits (centerline CL, upper control limit UCL, lower control limit LCL) and process capability indicators (such as Cp, Cpk).

[0134] Optionally, in another embodiment of this application, one specific implementation of step S103 is as follows: Figure 4 As shown, the specific steps include:

[0135] S401. Calculate the mean and process standard deviation of the standardized subgroup data.

[0136] Specifically, it is the arithmetic mean of all data points within the current standardized subgroup. Process standard deviation is a statistic used to estimate the inherent variation in a production process; in econometric control charts, it is typically expressed as the moving range method or the average of the sample standard deviations divided by a relevant constant.

[0137] S402. Based on the mean and process standard deviation, determine whether the current process of the target industrial equipment is in a stable state.

[0138] Here, the target industrial equipment refers to the industrial equipment whose equipment tag number corresponds to the standardized subgroup data.

[0139] Understandably, a stable state is defined as a process affected only by random factors, with no attributable special cause variation. Specific criteria include: the current subgroup mean does not exceed the temporary control limit, no outlier rules have been triggered in several consecutive subgroups in the past, and the process capability index Cpk is not lower than 1.33 (i.e., the process has sufficient capability margin). If all the above conditions are met, it is determined to be a stable state, and step S403 is executed; otherwise, it is an unstable state, and step S404 is executed.

[0140] S403. Based on the mean and process standard deviation, use asymptotic calculation to determine the control limits and process capabilities of the target industrial equipment.

[0141] Specifically, when the current process of the target industrial equipment is in a steady state, asymptotic calculation refers to expressing the control limit formula based on the normal distribution assumption, with the control limit calculation logic being UCL / LCL = _t±k· _t, where k=3 is the default value. When the process is in a steady state, the control limit calculation logic... (mean estimation) and The (standard deviation estimate) will be continuously updated, and the center line and upper and lower limits of the control chart will move horizontally with the process truth, progressively updating the control limits as the process progresses to reflect the latest steady state of the process. USL and LSL are the user-configured upper and lower limits of the specifications, respectively. Process capability is assessed based on the mean and process standard deviation, and then step S406 is executed.

[0142] S404. Obtain the historical false alarm rate and historical false alarm rate of the target industrial equipment.

[0143] Understandably, when the current process of the target industrial equipment is not in a stable state, that is, when the process undergoes attributable special cause variation (such as tool change or batch switching), the control limit calculation baseline is automatically reset. At this time, a process capability feedback mechanism is introduced to dynamically adjust the control limit coefficient k based on the historical false alarm rate and false alarm rate. Therefore, it is necessary to obtain the historical false alarm rate and historical false alarm rate of the target industrial equipment. The false alarm rate is calculated based on the collected standardized subgroup data and alarm data. The total number of points in the most recent five subgroups is used as the window interval (the subgroup size is configured as 25, so the total number of points is 125). The false alarm rate calculation logic is MR_Window = (the number of points that occurred during the "true out of control" period but did not trigger an alarm within the window) / (the total number of points marked as "true out of control" within the window). The false alarm rate calculation logic is FAR_Window = (the number of points that occurred during the "steady-state label" period and triggered an alarm within the window) / (the total number of points marked as "steady-state" within the window).

[0144] S405. Based on the historical false alarm rate and historical false alarm rate, update the control limits of the target industrial equipment, and calculate the process capability of the target industrial equipment based on the mean and process standard deviation.

[0145] Specifically, when the false alarm rate is >5%, the k value is automatically increased to tighten the control limit; when the false negative rate is >1%, the k value is automatically decreased to relax the control limit. The adjustment range of the k value is limited to [2.5, 3.5] to ensure statistical validity. The calculation of process capability still uses the ratio of specification limit to process standard deviation and is not affected by the adjustment of control limit.

[0146] S406. Use control limits, process capability, and standardized subgroup data as target subgroup data.

[0147] It is understandable that, regardless of whether the process is in a stable or unstable state, the calculated control limits (centerline, upper control limit, lower control limit), process capability indicators (Cp, Cpk, etc.), and the original standardized subgroup data will eventually be packaged and combined into a single data structure, namely the target subgroup data.

[0148] S104. Based on the pre-configured anomaly detection rules, determine whether there is a process anomaly in the industrial equipment corresponding to the device tag number of the target subgroup data.

[0149] It should be noted that the anomaly detection rules refer to a set of statistical detection criteria pre-selected by the user. Each anomaly detection rule triggers an event with a confidence score (0-100 points), taking into account factors such as deviation magnitude (the ratio of the distance a data point exceeds the control limit to 3σ), duration (the number of subgroups where the abnormal pattern persists), and historical similarity (the degree of pattern matching with historical real abnormal events). Alarms with a confidence score below the threshold are automatically downgraded to the "alert" level and do not trigger SCADA alarms. Combined with equipment status information in the SCADA system (such as whether the equipment is in manual mode, during maintenance, or in a process changeover phase), invalid alarms in non-production states are automatically suppressed. User responses to alarms (confirmation / ignore / false alarm) are recorded, and an alarm feedback dataset is constructed. The parameters of the anomaly detection rules are periodically optimized through incremental learning: for alarm events marked as "false alarms" by the user, the trigger sensitivity of the corresponding rule combination is reduced; for alarm events marked as "confirmation" by the user, the trigger sensitivity of the corresponding rule combination is increased. The learning results are stored in the form of a rule weight matrix. When the number of alarms exceeds the threshold within a unit of time, the alarm aggregation mechanism is automatically triggered, merging alarms with the same equipment and the same rule type into a single summary alarm. Therefore, based on these anomaly detection rules, it is determined whether there is a process anomaly in the industrial equipment corresponding to the device tag number of the target subgroup data. When an anomaly detection rule is triggered, it is considered that the process is abnormal, and step S105 is executed. When no anomaly detection rule is triggered, it is determined that the industrial equipment is in a normal state.

[0150] S105. Generate alarm information based on the target subgroup data.

[0151] Specifically, when an industrial device corresponding to the device tag number of the target subgroup data experiences a process anomaly, the SPC analysis engine immediately performs calculations and anomaly detection rules after each data update. Once the anomaly condition is triggered, real-time alarm information is immediately generated through the SCADA system's alarm center. The alarm information includes: the timestamp of the anomaly location, the subgroup number, the name of the violated anomaly detection rule, the current value, and control limits, among other detailed information. The alarm information is simultaneously pushed to the SCADA alarm bar and client devices (such as WeChat, email, and SMS) to ensure relevant personnel are informed of the process anomaly immediately. Regarding resource synchronization, this application supports the synchronous sharing of configuration information and control chart templates across different SCADA clients. Configurations completed on one client can be automatically synchronized to other clients without requiring repeated configuration.

[0152] Optionally, in actual industrial field monitoring, simply using alarm information for anomaly notification is insufficient to intuitively reflect the dynamic changing trends of process parameters. Operators often need to use graphical tools to quickly identify whether subgroup data exceeds control limits or exhibits a specific arrangement pattern. To embed statistical calculation results into the SCADA system's monitoring interface in a visual manner, thereby improving the efficiency of anomaly identification and the intuitiveness of process management, another embodiment of this application also provides a control chart display method for graphically presenting control limits, process capabilities, and alarm information in target subgroup data. Optionally, as... Figure 5 As shown, another embodiment of this application provides a method for drawing and displaying control charts, specifically including the following steps:

[0153] S501. Extract control limits and process capabilities from the target subgroup data.

[0154] The target subgroup data refers to the composite data structure obtained after adaptive processing in step S103. This data structure encapsulates standardized subgroup data, control limits (including centerline CL, upper control limit UCL, and lower control limit LCL), and process capability indicators (at least Cp and Cpk, and optionally Pp, Ppk, Ca, DPMO, and process sigma level). In this step, the system parses the memory layout or data structure fields of the target subgroup data, reads the values ​​of the control limit field and the process capability field, and temporarily stores them in the internal variables of the plotting module. It is easy to understand that control limits are used to determine the positions of the upper and lower warning boundaries in the control chart, and process capability is used to quantify the degree to which the process meets specifications. Both need to be extracted separately from the target subgroup data so that they can be used for plotting coordinate positioning and capability level identification in subsequent steps.

[0155] S502. Based on the control limits, process capabilities, and alarm information, draw and display the control chart on the visualization interface.

[0156] It should be noted that the visualization module directly embeds control charts, statistical indicators, and alarm information into the SCADA native HMI or its accompanying visualization interface. Control charts are dynamically drawn using CanvasG technology, supporting real-time updates. The charts automatically update with each new data point, eliminating the need for manual page refresh by the user. The control charts clearly mark the center line (CL), upper control limit (UCL), and lower control limit (LCL). Data points exceeding the control limits are highlighted in red. Points violating exception rules are marked with special symbols (such as triangles or circles) and the specific rule name is displayed in a floating window. Process capability analysis results are displayed in dashboard format, including Cp / Cpk values, process sigma level, DPMO, and capability level (A+ to D). Values ​​outside the normal range are indicated by color (green for normal, yellow for warning, and red for abnormal). All analysis results support historical playback, allowing users to view process trends and control chart changes over any time period. Operators can perform process monitoring and quality analysis directly within the SCADA interface, eliminating the need to switch between different software.

[0157] Optionally, in the foregoing embodiments, when the process state changes (e.g., tool change, batch change) or the control limit coefficient is dynamically adjusted based on the false alarm rate and missed alarm rate, the newly calculated control limit may deviate significantly from the currently used control limit. Directly replacing the old control limit with the new one can easily lead to batch exceeding or falling back in the data points on the control chart relative to the new control limit, resulting in a "quake oscillation" problem where a large number of consecutive alarms or alarms suddenly disappear within a short period, affecting the operator's accurate judgment of the process state. To achieve a smooth transition between the old and new boundaries during the control limit update process and avoid false alarms or missed alarms caused by sudden changes in the control limit, another embodiment of this application provides a control limit smooth transition processing method for gradually adjusting the control limit during the update process. Optionally, another embodiment of this application provides a control limit smooth transition processing method, specifically including the following steps:

[0158] When an update instruction for the control limits is received, the control limits are smoothly transitioned using an exponentially weighted moving average algorithm to obtain the target control limits.

[0159] It should be noted that when the process is in a steady state, it indicates that the process has reached its optimal state under the current control limits. When a control limit update instruction is received, the k value needs to be reduced to pursue higher process requirements. Conversely, if the process frequently triggers anomaly rules and generates alarms, it indicates that the current control limits are not applicable to the process at this time, and the k value needs to be increased to gradually achieve process optimization. Smoothing the k value transition can prevent the occurrence of spurious data. For example, at time T6, due to data updates, the control limits suddenly narrow from [95, 105] to [96, 104]. The data 104.5 at time T7 is normal according to the previous control limits, but "out of control" according to the new control limits. This alarm is caused by spurious data alarm oscillations due to sudden changes in control limits. Therefore, an exponentially weighted moving average (EWMA) is needed for a smooth transition. The update instruction refers to the control limit recalculation signal triggered manually by the user (e.g., clicking the "Update Control Limits" button in the configuration interface) or automatically generated by the system (e.g., after the dynamic control limit adjustment mechanism determines that a change in the k value is needed, or after a process variation due to attributable special causes).

[0160] In an optional embodiment, this application provides a data processing method based on a data acquisition and monitoring control system, relying on a data mining and process analysis system for industrial SCADA software. The network deployment architecture of this system is as follows: Figure 6 As shown.

[0161] This application provides a data processing method based on a data acquisition and monitoring control system. It receives configuration parameters sent by the user and retrieves corresponding process quality characteristic data from multiple preset device tag numbers based on these parameters. Then, for each process quality characteristic data, it preprocesses the data using adaptive and interpolation algorithms to obtain standardized subgroup data. Next, it adaptively processes the standardized subgroup data using an adaptive algorithm to obtain target subgroup data. Finally, based on pre-configured anomaly detection rules, it determines whether the industrial equipment corresponding to the device tag number of the target subgroup data has a process anomaly. If the industrial equipment corresponding to the device tag number of the target subgroup data has a process anomaly, an alarm message is generated based on the target subgroup data. Thus, by deeply integrating SCADA data acquisition and SPC analysis, and using adaptive and interpolation algorithms for standardized preprocessing and adaptive processing of process quality characteristic data, it achieves real-time data and analysis integration. It automatically identifies process anomalies and generates alarms based on anomaly detection rules, effectively solving the drawbacks of their separation and improving the timeliness and accuracy of anomaly detection.

[0162] Another embodiment of this application provides a data processing device based on a data acquisition and monitoring control system, such as... Figure 7 As shown, it includes the following units:

[0163] The data acquisition unit 701 is used to receive configuration parameters sent by the user and obtain corresponding process quality characteristic data from multiple preset device tag numbers according to the configuration parameters.

[0164] The preprocessing unit 702 is used to preprocess the process quality feature data for each process quality feature data according to the adaptive and interpolation algorithm to obtain standardized subgroup data.

[0165] The adaptive processing unit 703 is used to adaptively process the standardized subgroup data using an adaptive algorithm to obtain the target subgroup data.

[0166] The anomaly detection unit 704 is used to determine whether there is a process anomaly in the industrial equipment corresponding to the device tag number of the target subgroup data, based on the pre-configured anomaly detection rules.

[0167] The generation unit 705 is used to generate alarm information based on the target subgroup data if there is a process abnormality in the industrial equipment with the device tag number corresponding to the target subgroup data.

[0168] It should be noted that the specific working process of the above modules in the embodiments of this application can be referred to steps S101 to S105 in the above method embodiments, and will not be repeated here.

[0169] Optionally, in another embodiment of this application, a data processing device based on a data acquisition and monitoring control system includes a data acquisition unit 701, comprising:

[0170] The first extraction unit is used to extract the sampling frequency, subgroup value and sampling interval corresponding to multiple preset device tag numbers from the configuration parameters.

[0171] The first acquisition unit is used to acquire the corresponding real-time data from the device tag number according to the sampling frequency corresponding to the device tag number.

[0172] The grouping unit is used to group real-time data according to subgroup values ​​and sampling intervals to obtain feature data.

[0173] The processing unit is used to preprocess each set of feature data to obtain process quality feature data.

[0174] Optionally, in another embodiment of this application, a data processing device based on a data acquisition and monitoring control system includes a preprocessing unit 702, comprising:

[0175] The second extraction unit is used to extract the sampling interval corresponding to each device tag number from the configuration parameters for each device tag number.

[0176] The conversion unit is used to convert the timestamp of the process quality characteristic data corresponding to the device tag number based on the system clock to obtain the target process quality characteristic data.

[0177] The identification unit is used to periodically identify the device tag number according to the sampling interval to obtain the actual sampling period.

[0178] The judgment unit is used to determine whether the actual sampling period is less than the sampling interval.

[0179] The aggregation unit is used to aggregate the target process quality feature data using a sliding window algorithm if the actual sampling period is less than the sampling interval, thereby obtaining subgroup data.

[0180] The filling unit is used to fill the target process quality feature data with the nearest neighbor forward filling algorithm if the actual sampling period is greater than the sampling interval, so as to obtain subgroup data.

[0181] The detection unit is used to detect whether there are any missing data in the subgroup data.

[0182] The imputation processing unit is used to impute the subgroup data using a triple imputation strategy if there are missing data in the subgroup data, so as to obtain standardized subgroup data.

[0183] The first unit is used to treat the subgroup data as standardized subgroup data if there are no missing data in the subgroup data.

[0184] Optionally, another embodiment of this application provides a data processing device based on a data acquisition and monitoring control system, which further includes:

[0185] The labeling unit is used to label the standardized subgroup data using the box plot method and then feed the labeled standardized subgroup data back to the user.

[0186] Optionally, in another embodiment of this application, a data processing device based on a data acquisition and monitoring control system includes an adaptive processing unit 703, comprising:

[0187] The first calculation unit is used to calculate the mean and process standard deviation of the standardized subgroup data.

[0188] The state determination unit is used to determine whether the current process of the target industrial equipment is in a stable state based on the mean and process standard deviation. Here, the target industrial equipment refers to the industrial equipment with the corresponding equipment tag number in the standardized subgroup data.

[0189] The second calculation unit is used to calculate the control limits and process capabilities of the target industrial equipment asymptotically based on the mean and process standard deviation if the current process of the target industrial equipment is in a stable state.

[0190] The second acquisition unit is used to acquire the historical false alarm rate and historical false alarm rate of the target industrial equipment if the current process of the target industrial equipment is not in a stable state.

[0191] The update unit is used to update the control limits of the target industrial equipment based on the historical false alarm rate and the historical false alarm rate, and to calculate the process capability of the target industrial equipment based on the mean and process standard deviation.

[0192] The second unit is used to take control limits, process capability, and standardized subgroup data as target subgroup data.

[0193] Optionally, another embodiment of this application provides a data processing device based on a data acquisition and monitoring control system, which further includes:

[0194] The second extraction unit is used to extract control limits and process capabilities from the target subgroup data.

[0195] The display unit is used to draw and display control charts on a visual display interface based on control limits, process capabilities, and alarm information.

[0196] Optionally, another embodiment of this application provides a data processing device based on a data acquisition and monitoring control system, which further includes:

[0197] The smoothing unit is used to perform smooth transition processing on the control limits using an exponentially weighted moving average algorithm when it receives an update instruction for the control limits, so as to obtain the target control limits.

[0198] It should be noted that the specific working process of each module provided in the above embodiments of this application can be referred to the corresponding steps in the above method embodiments, and will not be repeated here.

[0199] It should also be noted that the data processing device based on the data acquisition and monitoring control system provided in this application has the technical effects of any of the above embodiments, and will not be described in detail here.

[0200] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0201] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A data processing method based on a data acquisition and monitoring control system, characterized in that, include: Receive configuration parameters sent by the user, and obtain corresponding process quality characteristic data from multiple preset device tag numbers according to the configuration parameters; For each of the process quality feature data, the process quality feature data is preprocessed according to an adaptive and interpolation algorithm to obtain standardized subgroup data; The standardized subgroup data is adaptively processed using an adaptive algorithm to obtain the target subgroup data; Based on the pre-configured anomaly detection rules, determine whether there is a process anomaly in the industrial equipment corresponding to the device tag number of the target subgroup data; If an industrial device with the device tag number corresponding to the target subgroup data has a process abnormality, an alarm message is generated based on the target subgroup data.

2. The method according to claim 1, characterized in that, The step of obtaining corresponding process quality characteristic data from a preset set of device tag numbers according to the configuration parameters includes: Extract the sampling frequency, subgroup value and sampling interval corresponding to multiple preset device tag numbers from the configuration parameters; For each device tag number, real-time data is obtained from the device tag number according to the sampling frequency corresponding to the device tag number; The real-time data is grouped according to the subgroup values ​​and the sampling interval to obtain feature data; Each set of feature data is preprocessed to obtain process quality feature data.

3. The method according to claim 1, characterized in that, For each of the process quality feature data, preprocessing is performed on the process quality feature data according to an adaptive and interpolation algorithm to obtain standardized subgroup data, including: For each device tag number, the sampling interval corresponding to the device tag number is extracted from the configuration parameters; Using the system clock as a reference, the timestamp of the process quality characteristic data corresponding to the device tag number is converted to obtain the target process quality characteristic data; Based on the sampling interval, the device tag number is periodically identified to obtain the actual sampling period; Determine whether the actual sampling period is less than the sampling interval; If the actual sampling period is less than the sampling interval, the target process quality characteristic data are aggregated using a sliding window algorithm to obtain subgroup data. If the actual sampling period is greater than the sampling interval, the target process quality feature data is filled using the nearest neighbor forward filling algorithm to obtain subgroup data. Detect whether there are any missing data in the subgroup data; If there are missing data in the subgroup data, a triple imputation strategy is used to imputate the subgroup data to obtain standardized subgroup data. If there are no missing data in the subgroup data, then the subgroup data is used as standardized subgroup data.

4. The method according to claim 3, characterized in that, Also includes: The standardized subgroup data is labeled using a box plot method, and the labeled standardized subgroup data is then fed back to the user.

5. The method according to claim 1, characterized in that, The step of adaptively processing the standardized subgroup data using an adaptive algorithm to obtain the target subgroup data includes: Calculate the mean and process standard deviation of the standardized subgroup data; Based on the mean and the process standard deviation, it is determined whether the current process of the target industrial equipment is in a stable state; wherein, the target industrial equipment refers to the industrial equipment with the equipment tag number corresponding to the standardized subgroup data; If the current process of the target industrial equipment is in a stable state, the control limits and process capabilities of the target industrial equipment are calculated asymptotically based on the mean and the process standard deviation. If the current process of the target industrial equipment is not in a stable state, then obtain the historical false alarm rate and historical false alarm rate of the target industrial equipment; Based on the historical false alarm rate and historical false alarm rate, the control limits of the target industrial equipment are updated, and the process capability of the target industrial equipment is calculated based on the mean and the process standard deviation. The control limits, the process capability, and the standardized subgroup data are used as the target subgroup data.

6. The method according to claim 5, characterized in that, Also includes: Extract the control limits and process capabilities from the target subgroup data; Based on the control limits, process capabilities, and alarm information, a control chart is drawn and displayed on a visual interface.

7. The method according to claim 5, characterized in that, Also includes: When the update instruction for the control limit is received, the control limit is smoothed using an exponentially weighted moving average algorithm to obtain the target control limit.

8. A data processing device based on a data acquisition and monitoring control system, characterized in that, include: The data acquisition unit is used to receive configuration parameters sent by the user and obtain corresponding process quality characteristic data from a plurality of preset device tag numbers according to the configuration parameters; The preprocessing unit is used to preprocess each process quality feature data according to an adaptive and interpolation algorithm to obtain standardized subgroup data. An adaptive processing unit is used to adaptively process the standardized subgroup data using an adaptive algorithm to obtain target subgroup data; An anomaly detection unit is used to determine whether there is a process anomaly in the industrial equipment corresponding to the device tag number of the target subgroup data, based on pre-configured anomaly detection rules. The generation unit is used to generate alarm information based on the target subgroup data if an industrial device with the device tag number corresponding to the target subgroup data has a process abnormality.

9. The apparatus according to claim 8, characterized in that, The data acquisition unit includes: The first extraction unit is used to extract the sampling frequency, subgroup value and sampling interval corresponding to multiple preset device tag numbers from the configuration parameters; The first acquisition unit is used to acquire corresponding real-time data from the device tag number according to the sampling frequency corresponding to the device tag number, respectively. A grouping unit is used to group the real-time data according to the subgroup value and the sampling interval to obtain feature data; The processing unit is used to preprocess each set of feature data to obtain process quality feature data.

10. The apparatus according to claim 8, characterized in that, The preprocessing unit includes: The second extraction unit is used to extract the sampling interval corresponding to each device tag number from the configuration parameters for each device tag number. The conversion unit is used to convert the timestamp of the process quality characteristic data corresponding to the device tag number based on the system clock to obtain the target process quality characteristic data. The identification unit is used to periodically identify the device tag number according to the sampling interval to obtain the actual sampling period; The judgment unit is used to determine whether the actual sampling period is less than the sampling interval; An aggregation unit is used to aggregate the target process quality characteristic data using a sliding window algorithm to obtain subgroup data if the actual sampling period is less than the sampling interval. A filling unit is used to fill the target process quality feature data using the nearest neighbor forward filling algorithm if the actual sampling period is greater than the sampling interval, so as to obtain subgroup data. The detection unit is used to detect whether there are any missing data in the subgroup data; An interpolation processing unit is used to interpolate the subgroup data using a triple interpolation strategy if there are missing data in the subgroup data, so as to obtain standardized subgroup data. The first unit is used to treat the subgroup data as standardized subgroup data if there are no missing data in the subgroup data.

Images