Metering control chart construction performance evaluation method and system based on Bayesian algorithm

Abstract

The invention relates to the technical field of data processing, and particularly discloses a metering control chart construction performance evaluation method and system based on a Bayesian algorithm, and the method comprises the steps: obtaining an updating period of a constructed metering control chart through the Bayesian algorithm, and analyzing the preprocessing operation compliance of a platform to which the metering control chart belongs, according to the process, the compliance of platform preprocessing operation can be improved, then the visual delay degree value of the metering control chart is analyzed, and an accurate operation data support is provided for subsequently judging whether visual delay adjustment is executed or not; according to the method, the platform to which the metering control chart belongs executes the visual delay adjustment of the metering control chart, a reasonable data basis is provided, and finally, whether data importance screening needs to be carried out is judged by evaluating the load operation compliance value of the platform to which the metering control chart belongs, so that the normal visualization capability of the data of the metering control chart is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for constructing a performance evaluation method of a metrological control chart based on a Bayesian algorithm. Background Art

[0002] Data visualization is an important part of the performance evaluation of metrological control charts. As a statistical process control tool, control charts are widely used in production and quality management to monitor the stability and consistency of processes. The existence of data visualization delays indicates that there is some abnormality in the process, which may lead to product quality problems or reduced production efficiency. Therefore, the detection of data visualization and the evaluation of its impact on the overall control chart are important steps to ensure the quality of the production process.

[0003] For example, the invention patent with announcement number CN108717356B announces a method and device for holographic visualization of metrology, which constructs the Canvas class library ZRender with source data obtained from provincial metrology automation system and / or regional dispatching system and / or production management system and / or geographic information system and / or video monitoring system and / or meteorological environment system, and uses the DataV class data visualization tool for visualization.

[0004] For example, the invention patent with publication number CN111694883A discloses a visualization system for historical geographic information data, including a historical geographic information acquisition module for acquiring and sending historical geographic information data; a historical geographic data conversion module for receiving historical geographic information data and converting the historical geographic information data into modern scientific statistical data through historical geographic measurement methods; a scientific statistical data processing module for analyzing and processing scientific statistical data and generating analysis reports; and a data visualization display module for visually displaying data in the form of charts based on the analysis report.

[0005] Combined with the above technical solutions, it is found that currently when visualizing the data in the graph, the data is usually displayed directly in the graph, but this process will result in low data visualization capabilities due to the unreasonable operation of the platform, and subsequent visualization will gradually experience delays or data offsets, making the distribution of data points in the graph more uneven, which is not conducive to subsequent data visualization processing. Summary of the invention

[0006] In view of the deficiencies in the prior art, the present invention provides a performance evaluation method and system for constructing a metrological control chart based on a Bayesian algorithm, which can effectively solve the problems involved in the above-mentioned background technology.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions: In the first aspect of the present invention, a performance evaluation method for constructing a measurement control chart based on the Bayesian algorithm is provided, including: setting and updating the time frequency of the constructed measurement control chart through the Bayesian algorithm to obtain the update period of the measurement control chart, and collecting the preprocessing operation information of the platform to which the measurement control chart belongs during the first sub-update cycle to determine whether to perform preprocessing again; obtaining the visualization delay information of the measurement control chart during the second sub-update cycle to determine whether to perform visualization delay adjustment; collecting the load operation information of the platform to which the measurement control chart belongs during the third sub-update cycle, and determining whether data screening is required.

[0008] As a further method, the process of determining whether to perform preprocessing again is as follows: Analyze the preprocessing operation information of the platform to which the measurement control chart belongs during the first sub-update cycle, obtain the preprocessing operation compliance degree of the platform to which the measurement control chart belongs, and compare it with the preset preprocessing operation compliance threshold in the evaluation database. If the preprocessing operation compliance degree of the platform to which the measurement control chart belongs is higher than or equal to the preprocessing operation compliance threshold, there is no need to preprocess the data again. If the preprocessing operation compliance degree of the platform to which the measurement control chart belongs is lower than the preprocessing operation compliance threshold, send an abnormal preprocessing operation instruction of the platform to which the measurement control chart belongs to the platform to which the measurement control chart belongs, and the platform to which the measurement control chart belongs performs an inspection of the preprocessing operation to complete the preprocessing of the data.

[0009] As a further method, the process of determining whether to perform visualization delay adjustment is as follows: Process the visualization delay information of the measurement control chart during the second sub-update cycle to obtain the visualization delay degree value of the measurement control chart, and compare it with the preset visualization delay degree threshold in the evaluation database. If the visualization delay degree value of the measurement control chart is lower than the visualization delay degree threshold, there is no need to perform visualization delay adjustment. If the visualization delay degree value of the measurement control chart is higher than or equal to the visualization delay degree threshold, perform visualization delay adjustment.

[0010] As a further method, the process of determining whether data screening is required is as follows: According to the load operation information of the platform to which the measurement control chart belongs during the third sub-update cycle, obtain the load operation compliance value of the platform to which the measurement control chart belongs, and compare it with the preset load operation compliance threshold in the evaluation database; if the load operation compliance value of the platform to which the measurement control chart belongs is higher than or equal to the load operation compliance threshold, there is no need to perform data screening; if the load operation compliance value of the platform to which the measurement control chart belongs is lower than the load operation compliance threshold, data screening is required, and data visualization is performed again after completing the data screening.

[0011] The second aspect of the present invention provides a system for evaluating the performance of constructing a measurement control chart based on the Bayesian algorithm, including: a preprocessing operation determination module, which is used to update the time frequency of the constructed measurement control chart through the Bayesian algorithm, obtain the update period of the measurement control chart, and collect the preprocessing operation information of the platform to which the measurement control chart belongs during the first sub-period of the update, and determine whether to perform preprocessing again; a visualization delay determination module, which is used to obtain the visualization delay information of the measurement control chart during the second sub-period of the update and determine whether to perform visualization delay adjustment; a data screening determination module, which is used to collect the load operation information of the platform to which the measurement control chart belongs during the third sub-period of the update and determine whether data screening is required.

[0012] Compared with the prior art, the embodiments of the present invention have at least the following advantages or beneficial effects: (1) By providing a method and system for evaluating the performance of constructing a measurement control chart based on the Bayesian algorithm, the present invention first updates the time frequency of the constructed measurement control chart through the Bayesian algorithm, obtains the update period of the measurement control chart, and collects the preprocessing operation information of the platform to which the measurement control chart belongs during the first sub-period of the update, and analyzes the preprocessing operation compliance of the platform to which the measurement control chart belongs to determine whether to perform preprocessing again. This process can improve the accuracy of data preprocessing anomalies. Then, it obtains the visualization delay information of the measurement control chart during the second sub-period of the update and analyzes the visualization delay degree value of the measurement control chart, providing accurate operation data support for subsequent determination of whether to perform visualization delay adjustment, making the platform to which the measurement control chart belongs more reasonable in performing visualization delay adjustment of the measurement control chart based on data. Finally, by collecting the load operation information of the platform to which the measurement control chart belongs during the third sub-period of the update, it evaluates the load operation compliance value of the platform to which the measurement control chart belongs and determines whether data screening is required to improve the normal visualization ability of the data of the measurement control chart.

[0013] (2) By obtaining the update period of the measurement control chart, collecting and analyzing the data preprocessing operation duration, data anomaly processing duration, sampling frequency, and data throughput of the platform to which the measurement control chart belongs during the first sub-period of the update, the quantified preprocessing operation compliance of the platform to which the measurement control chart belongs is obtained and compared with the preprocessing operation compliance threshold. By the comparison result, it is determined whether to perform preprocessing operation anomaly inspection, which can prevent the subsequent actual data visualization process from being affected by data anomalies in the measurement control chart itself and obtain more accurate and data-supported normal data of the measurement control chart.

[0014] (3) The present invention analyzes the proportion of unshown data, the average data delay duration, and the network response duration within the updated second sub-cycle of the measurement control chart, and combines the preprocessing operation compliance of the platform to which the measurement control chart belongs to obtain the visualization delay degree value of the measurement control chart. By comparing it with the visualization delay degree threshold, it determines whether there is a visualization delay phenomenon in the data in the measurement control chart according to the comparison result, thereby improving the problem of unreasonable data distribution caused by visualization delay in the measurement control chart.

[0015] (4) The present invention comprehensively analyzes the CPU utilization rate, memory usage, load average, preprocessing operation compliance of the platform to which the measurement control chart belongs, and the visualization delay degree value of the measurement control chart within the updated third sub-cycle to obtain the load operation compliance value of the platform to which the measurement control chart belongs. By comparing it with the load operation compliance threshold, it determines whether it is necessary to perform importance screening on the data in the measurement control chart according to the comparison result, and by performing importance screening on the data, it processes the complex and changeable dynamic measurement control chart and increases the visualization distribution continuity and uniformity of the data points in the measurement control chart. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The present invention is further described with reference to the accompanying drawings. However, the embodiments in the drawings do not constitute any limitation to the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to the following drawings.

[0017] Figure 1 It is a schematic flowchart of the method steps of the present invention; Figure 2 It is a schematic diagram of the connection of the system modules of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0019] Refer to Figure 1 As shown, the first aspect of the present invention provides a method for evaluating the construction performance of a measurement control chart based on the Bayesian algorithm, including: setting and updating the time frequency of the constructed measurement control chart through the Bayesian algorithm, obtaining the update period of the measurement control chart, and collecting the preprocessing operation information of the platform to which the measurement control chart belongs within the updated first sub-cycle to determine whether to perform preprocessing again.

[0020] It should be noted that the above-mentioned time frequency of the constructed measurement control chart is set and updated through the Bayesian algorithm. The specific process is as follows: First, based on historical data, the prior distributions of relevant parameters (such as mean, variance, etc.) of the measurement control chart are set. At the same time, during the production process of this product, new observed data are collected. The newly collected observed data are combined with the prior distribution, and the posterior distribution is calculated through Bayes' theorem. The feasibility of the current time frequency is evaluated using the posterior distribution, so as to determine whether the time frequency needs to be updated. For example, a certain factory produces a certain electronic component, and the goal is to keep the quantity produced per hour within a certain range to ensure production efficiency and quality. The initial goal is that the average value of the quantity produced per hour is μ = 100, and the standard deviation σ = 10. We hope to monitor the quantity produced per hour through a measurement control chart (such as an X-bar chart) and dynamically adjust the monitoring time frequency according to the actual production data. Therefore, the specific adjustment process is as follows: Based on the historical data of the measurement control chart, assume that the mean μ of the quantity produced per hour follows a normal distribution, and the initial prior is: , where, (historical mean), posterior variance (reflecting the initial uncertainty about the mean). At the same time, assume that the standard deviation σ = 10. If 102 are produced in the first hour, 98 in the second hour, and 101 in the third hour, then for the first hour ; for the second hour ; for the third hour ; where, for the second hour , which is expressed as the operation result of the first hour; n represents the number of times of observing products during this production period. From the above expressions, it can be known that n is 1 for all. If the set posterior variance threshold is 15, then according to the above expressions, it can be known that the production process is becoming more and more stable, and the monitoring frequency needs to be reduced. That is, after the third hour, the monitoring frequency needs to be reduced from once per hour to once every two hours. Therefore, data may not be updated immediately at the 4th hour, but the next update will be at the 6th hour.

[0021] It should be noted that through the above-mentioned Bayesian update, historical prior information can be combined with real-time production data, and according to the changes in the posterior distribution, the monitoring frequency of the control chart can be dynamically adjusted to adapt to the stability changes of the production process, improve the monitoring efficiency, and save monitoring resources at the same time.

[0022] Among them, the update period is obtained through the platform to which the measurement control chart belongs, and the platform to which the measurement control chart belongs can be a quality management system, a manufacturing execution system, an industrial Internet platform, a statistical process control software, or a data analysis and visualization tool; these platforms usually have real-time data collection, analysis, visualization, and reporting functions. By visually detecting and adjusting the latency of the measurement control chart, the enterprise to which the platform belongs can better respond to market changes and customer needs, while optimizing the production process and product quality.

[0023] Specifically, the determination of whether to perform preprocessing again is as follows: Analyze the preprocessing operation information of the platform to which the measurement control chart belongs during the first sub-update period, obtain the preprocessing operation compliance of the platform to which the measurement control chart belongs, and compare it with the preset preprocessing operation compliance threshold in the evaluation database. If the preprocessing operation compliance of the platform to which the measurement control chart belongs is higher than or equal to the preprocessing operation compliance threshold, there is no need to preprocess the data again. If the preprocessing operation compliance of the platform to which the measurement control chart belongs is lower than the preprocessing operation compliance threshold, send an instruction of abnormal preprocessing operation of the platform to which the measurement control chart belongs to the platform to which the measurement control chart belongs, and the platform to which the measurement control chart belongs performs an inspection of the preprocessing operation to complete the preprocessing of the data.

[0024] Furthermore, the specific analysis process of the preprocessing operation compliance of the platform to which the measurement control chart belongs is as follows: The preprocessing operation information of the platform to which the measurement control chart belongs during the first sub-update period specifically includes the data preprocessing operation duration of the platform to which the measurement control chart belongs, the sampling frequency of the platform to which the measurement control chart belongs, and the data throughput of the platform to which the measurement control chart belongs.

[0025] Among them, the first sub-update period represents the time period for detecting the preprocessing operation ability of the platform to which the measurement control chart belongs, and it is directly obtained through the platform to which the measurement control chart belongs; the preprocessing operation information can be obtained through the platform to which the measurement control chart belongs; the data preprocessing operation duration represents the time for the platform to preprocess the uploaded data, where the data preprocessing includes the process of data cleaning, that is, handling missing values, handling outliers, removing duplicate data, and correcting incorrect data.

[0026] Normalize and weight the data preprocessing operation duration and data throughput respectively with the corresponding preprocessing operation weights preset in the evaluation database. At the same time, perform a difference process by combining the sampling frequency with the preset sampling reference frequency in the evaluation database to obtain the preprocessing operation compliance of the platform to which the measurement control chart belongs. The preprocessing operation compliance of the platform to which the measurement control chart belongs represents the compliance quantification value of the influence degree of the data preprocessing operation duration, sampling frequency, and data throughput on the preprocessing operation of the platform to which the measurement control chart belongs, and is used to quantify the preprocessing operation compliance degree. The specific representation method is as follows: ; Among them, FB is the data preprocessing operation duration of the platform to which the measurement control chart belongs, is the preprocessing operation weight corresponding to the data preprocessing operation duration preset in the evaluation database, FC is the sampling frequency of the platform to which the measurement control chart belongs, is the preset sampling reference frequency in the evaluation database, is the preprocessing operation weight corresponding to the sampling frequency preset in the evaluation database, FP is the data throughput of the platform to which the measurement control chart belongs, is the preprocessing operation weight corresponding to the data throughput preset in the evaluation database.

[0027] It should be noted that the above preprocessing operation weights corresponding to the data preprocessing operation duration, sampling frequency, and data throughput are respectively used to quantify the influence degrees of the data preprocessing operation duration, sampling frequency, and data throughput on the preprocessing operation compliance degree; the evaluation database stores the corresponding relationships between the data preprocessing operation duration, sampling frequency, and data throughput and their corresponding preprocessing operation weights. For example, when the data preprocessing operation duration, sampling frequency, and data throughput are input into the evaluation database, the evaluation database can match the preprocessing operation weights corresponding to the data preprocessing operation duration, sampling frequency, and data throughput, and their value ranges are all between 0 and 1.

[0028] To measure the compliance of the preprocessing operation of the platform to which the measurement control chart belongs, in actual operation, if the data preprocessing can maintain a short duration, the overall operation efficiency of the platform will increase and the data quality will be guaranteed; however, if the data preprocessing is delayed, it may affect the overall preprocessing compliance. Therefore, if the platform can maintain a short preprocessing duration, it usually means that the platform can process data efficiently, thus supporting a higher data throughput; similarly, an appropriate sampling frequency can enable the platform to make full use of the high throughput advantage without causing data congestion, while an unreasonable sampling frequency (too high or too low) may lead to a mismatch in data processing load and affect the overall operation compliance of the platform; therefore, through the comprehensive analysis of the above parameters, the preprocessing operation degree of the platform will be more efficient and have higher data quality.

[0029] Specifically, the platform to which the measurement control chart belongs performs an inspection of the preprocessing operation. The specific execution process is as follows: Obtain the time period between the updated first sub-cycle and the updated second sub-cycle and record it as the preprocessing detection period, which can be directly obtained through the platform to which it belongs.

[0030] When the preprocessing operation compliance of the platform to which the measurement control chart belongs is lower than the preprocessing operation compliance threshold, the data processing module of the platform to which the measurement control chart belongs synchronously sends an instruction for abnormal preprocessing operation of the platform to which the measurement control chart belongs. The data processing module refers to the module that inspects the data preprocessing process of the platform, that is, it can be calculated through the corresponding expression of the above preprocessing operation compliance, and this module has the function of data comparison. After detecting an abnormality, it will directly issue a preset instruction to the platform to which the measurement control chart belongs.

[0031] Finally, the platform to which the measurement control chart belongs conducts a preprocessing inspection on the data within the preprocessing detection period through a preset preprocessing operation inspection module. After the inspection is completed, the platform to which the measurement control chart belongs directly visualizes the data.

[0032] It should be explained that the platform to which the measurement control chart belongs conducts a preprocessing inspection on the data within the preprocessing detection period through a preset preprocessing operation inspection module. The function of the preprocessing operation inspection module is to check whether there are loopholes in the data conversion of different data sources, whether data dimensionality reduction processing is required, and whether data enhancement processing is required. Therefore, through the function of the preprocessing operation inspection module, the platform specifically preprocesses the abnormal data again and makes the preprocessing operation compliance of the platform to which the measurement control chart belongs higher than the corresponding threshold.

[0033] Specifically, the determination of whether to perform visualization delay adjustment, the specific determination process is as follows: Process the visualization delay information of the measurement control chart within the updated second sub-cycle, obtain the visualization delay degree value of the measurement control chart, and compare it with the preset visualization delay degree threshold in the evaluation database. If the visualization delay degree value of the measurement control chart is lower than the visualization delay degree threshold, there is no need to perform visualization delay adjustment. If the visualization delay degree value of the measurement control chart is higher than or equal to the visualization delay degree threshold, perform visualization delay adjustment.

[0034] Further, the specific analysis process of the visualization delay degree value of the measurement control chart is as follows: The visualization delay information of the measurement control chart within the updated second sub-cycle specifically includes the proportion of un-displayed data of the measurement control chart, the average data delay duration of the measurement control chart, and the network response duration of the platform to which the measurement control chart belongs.

[0035] The above-mentioned updated second sub-cycle represents the time period for performing delay analysis on the visualization ability of the platform, which can be obtained through the platform to which the measurement control chart belongs; and the visualization delay information can also be obtained through the platform to which the measurement control chart belongs; where the proportion of un-displayed data represents the ratio of the amount of data points in the measurement control chart that are not displayed within the specified time to the total amount of data; the average data delay duration represents the time period from the start of visualization of the data in the measurement control chart to the completion of visualization.

[0036] Normalize and weight the proportion of un-displayed data, the average data delay duration, the network response duration, and the preprocessing operation compliance of the platform to which the measurement control chart belongs respectively with the preset delay impact weights in the evaluation database to obtain the visualization delay degree value of the measurement control chart. The visualization delay degree value of the measurement control chart represents the quantitative value of the impact of the proportion of un-displayed data, the average data delay duration, the network response duration, and the preprocessing operation compliance of the platform to which the measurement control chart belongs on the visualization delay of the measurement control chart, and is used to quantify the visualization delay degree. The specific expression method is as follows: ; Among them, YS is the proportion of un-displayed data of the measurement control chart, is the delay impact weight corresponding to the proportion of un-displayed data preset in the evaluation database, YJ is the average data delay duration of the measurement control chart, is the delay impact weight corresponding to the average data delay duration preset in the evaluation database, YC is the network response duration of the platform to which the measurement control chart belongs, is the delay impact weight corresponding to the network response duration preset in the evaluation database, is the preprocessing operation compliance of the platform to which the measurement control chart belongs, is the delay impact weight corresponding to the preprocessing operation compliance preset in the evaluation database.

[0037] It should be elaborated that the delay impact weights corresponding to the above-mentioned unshown data ratio, the delay impact weights corresponding to the average data delay duration, the delay impact weights corresponding to the network response duration, and the delay impact weights corresponding to the preprocessing operation compliance are respectively used to represent the influence degrees of the unshown data ratio, the average data delay duration, the network response duration, and the preprocessing operation compliance on the visualization delay degree value; the corresponding relationships between the unshown data ratio, the average data delay duration, the network response duration, and the preprocessing operation compliance stored in the evaluation database and their corresponding delay impact weights are evaluated. For example, when the unshown data ratio, the average data delay duration, the network response duration, and the preprocessing operation compliance are input into the evaluation database, the evaluation database can match the delay impact weights corresponding to the unshown data ratio, the delay impact weights corresponding to the average data delay duration, the delay impact weights corresponding to the network response duration, and the delay impact weights corresponding to the preprocessing operation compliance, and their value ranges are all between 0 and 1.

[0038] For the accuracy value of the outlier operation of the platform to which the measurement control chart belongs, an increase in the average data delay often causes some data to fail to be displayed due to timeout or expiration, thereby pushing up the unshown data ratio; at the same time, a high missing rate may also imply delays in the data collection or transmission process, and the two form a feedback loop, exacerbating the visualization delay; among them, the network response duration is a key link in the data transmission link. Network delay directly lengthens the average data delay, and a poor network response duration may affect the timeliness of the preprocessing module to receive data, making it difficult for the preprocessing operation to meet the standardized requirements and reducing the preprocessing compliance; therefore, if the preprocessing operation compliance is low, it may cause data to be lost or misclassified during the processing, thereby increasing the unshown data ratio; and the increase in the unshown data ratio may be a signal of low efficiency or data loss problems in the preprocessing process, further affecting the overall visualization timeliness.

[0039] Specifically, the execution of the visualization delay adjustment is carried out as follows: The difference between the visualization delay degree value of the measurement control chart and the visualization delay degree threshold is calculated to obtain the visualization delay deviation value of the measurement control chart. The platform to which the measurement control chart belongs adjusts the visualization delay ratio of the preprocessing operation information through the visualization delay deviation value of the measurement control chart to complete the visualization delay adjustment of the measurement control chart.

[0040] It should be noted that the above visualization delay adjustment is specifically to remind relevant personnel to perform corresponding operations. For example, if the visualization delay deviation value of the measurement control chart is H, the operation is to replace the algorithm with a long processing time in the data processing flow with a more efficient version, upgrade the network hardware or optimize the transmission protocol, and increase the CPU resources allocated to the visualization module to achieve the predetermined visualization delay ratio. Through the above operations, H is reduced to H×75%, and the visualization delay degree value of the measurement control chart is made lower than the visualization delay degree threshold.

[0041] Specifically, the process of determining whether data screening is required is as follows: Based on the load operation information of the platform to which the measurement control chart belongs during the third sub-update cycle, obtain the load operation compliance value of the platform to which the measurement control chart belongs, and compare it with the preset load operation compliance threshold in the evaluation database; if the load operation compliance value of the platform to which the measurement control chart belongs is higher than or equal to the load operation compliance threshold, data screening is not required; if the load operation compliance value of the platform to which the measurement control chart belongs is lower than the load operation compliance threshold, data screening is required, and after completing the data screening, data visualization is performed again.

[0042] It should be noted that the above data screening is to perform a difference process on the load operation compliance threshold and the load operation compliance value of the platform to which the measurement control chart belongs to obtain the load operation deviation value, and match it with the corresponding data screening amount in each load operation deviation value interval preset in the evaluation database, thereby obtaining the data screening amount of the platform to which the measurement control chart belongs. Finally, the data is screened through the data screening amount. The specific screening process is to obtain the data point distribution weight in the evaluation database. This weight is set by the preset personnel for data screening according to the data occurrence frequency in historical data. Based on the data screening amount and the data point distribution weight, weighted sampling is performed through the weighted sampling algorithm to obtain the sampling result, that is, the screened data, and it is transmitted to the visualization module (such as the measurement control chart) to ensure that the displayed data meets the requirements of real-time and accuracy; other data is saved to the storage module of the platform.

[0043] Furthermore, the specific analysis process of the load operation compliance value of the platform to which the measurement control chart belongs is as follows: The load operation information of the platform to which the measurement control chart belongs during the third sub-update cycle specifically includes the CPU utilization rate of the platform to which the measurement control chart belongs, the memory usage of the platform to which the measurement control chart belongs, and the load average of the platform to which the measurement control chart belongs.

[0044] It should be noted that the above-mentioned updated third sub-cycle represents the time period for detecting the load operation of the platform to which the measurement control chart belongs, which can be obtained through the platform to which the measurement control chart belongs; and the load operation information can also be obtained through the platform to which the measurement control chart belongs; among them, the load average represents the average number of processes in the active state (i.e., running or waiting to run) in the platform operation queue within a certain period of time, and this indicator reflects the overall workload and resource usage of the platform.

[0045] The CPU utilization rate, memory usage, load average, preprocessing operation compliance of the platform to which the measurement control chart belongs, and the visualization delay degree value of the measurement control chart are respectively subjected to normalization processing and weighting processing with the corresponding load operation compliance weights preset in the evaluation database, and the load operation compliance value of the platform to which the measurement control chart belongs is coupled. The load operation compliance value of the platform to which the measurement control chart belongs represents the compliance quantization value of the influence degree of the CPU utilization rate, memory usage, load average, preprocessing operation compliance of the platform to which the measurement control chart belongs, and the visualization delay degree value on the load operation of the platform to which the measurement control chart belongs, and is used to quantify the load operation compliance degree. The specific expression method is as follows: ; Among them, ZC is the CPU utilization rate of the platform to which the measurement control chart belongs, is the load operation compliance weight of the CPU utilization rate preset in the evaluation database, ZD is the memory usage of the platform to which the measurement control chart belongs, is the load operation compliance weight of the memory usage preset in the evaluation database, ZG is the load average of the platform to which the measurement control chart belongs, is the load operation compliance weight of the load average preset in the evaluation database, is the preprocessing operation compliance of the platform to which the measurement control chart belongs, is the load operation compliance weight of the preprocessing operation compliance preset in the evaluation database, is the visualization delay degree value of the measurement control chart, is the load operation compliance weight of the visualization delay degree value preset in the evaluation database.

[0046] It should be noted that the load operation compliance weights of the above CPU utilization rate, memory usage, load average, preprocessing operation compliance degree, and visualization delay degree value are respectively used to quantify the influence degrees of the CPU utilization rate, memory usage, load average, preprocessing operation compliance degree, and visualization delay degree value on the load operation compliance value; evaluate the corresponding relationship between the CPU utilization rate, memory usage, load average, preprocessing operation compliance degree, and visualization delay degree value stored in the evaluation database and their corresponding load operation compliance weights. For example, input the CPU utilization rate, memory usage, load average, preprocessing operation compliance degree, and visualization delay degree value into the evaluation database, and the evaluation database can match the load operation compliance weights of the CPU utilization rate, memory usage, load average, preprocessing operation compliance degree, and visualization delay degree value. The value ranges are all between 0 and 1.

[0047] For the operation compliance value of the measurement control chart, a high CPU utilization rate often tends to push up the load average because when the CPU is in a high-load state for a long time, the situation of task queuing and waiting for scheduling will increase, thus reducing the load operation ability of the platform; the memory usage directly affects the caching and intermediate result storage in the data preprocessing process. Sufficient and highly efficient memory usage can ensure the smooth progress of the preprocessing process, thereby improving the compliance degree. However, if the memory occupancy is too high, it may cause delays or abnormalities in the preprocessing process, indirectly affecting the overall load operation state of the system; at the same time, when the system load is low, the preprocessing tasks can obtain the required computing resources faster, thereby improving the compliance degree, and the decrease in the preprocessing operation compliance degree (such as due to data errors or delays) may also be reflected in the instantaneous fluctuations of the system load, forming a feedback effect; therefore, a highly compliant preprocessing process can ensure the timely and accurate transmission of data to the visualization module, reducing the overall delay. On the contrary, a high visualization delay may also indicate problems in the preprocessing link (such as data accumulation, untimely exception handling), thus affecting the load operation compliance value; therefore, through the collaborative analysis of the above parameters, a more comprehensive platform load capacity can be obtained, providing more accurate and comprehensive data support for subsequent data screening.

[0048] Refer to Figure 2 As shown in the figure, the second aspect of the present invention provides a system for evaluating the construction performance of a measurement control chart based on the Bayesian algorithm, including: a preprocessing operation determination module, a visualization delay determination module, a data screening determination module, and an evaluation database.

[0049] The preprocessing operation determination module is connected to the visualization delay determination module, and the visualization delay determination module is connected to the data screening determination module. The preprocessing operation determination module, the visualization delay determination module, and the data screening determination module are all connected to the evaluation database.

[0050] The preprocessing operation determination module is used to update the time frequency of the constructed measurement control chart through the Bayesian algorithm, obtain the update period of the measurement control chart, and collect the preprocessing operation information of the platform to which the measurement control chart belongs during the first sub-update period, and determine whether to perform preprocessing again.

[0051] The visualization delay determination module is used to obtain the visualization delay information of the measurement control chart during the second sub-update period and determine whether to perform visualization delay adjustment.

[0052] The data screening determination module is used to collect the load operation information of the platform to which the measurement control chart belongs during the third sub-update period and determine whether data screening is required.

[0053] The evaluation database is used to store the weight factors corresponding to the above preprocessing operation information, the weight factors corresponding to the visualization delay information, the weight factors corresponding to the load operation information, and multiple thresholds, etc.

[0054] The above content is only an example and explanation of the structure of the present invention. Those skilled in the art of this technology make various modifications or supplements or use similar methods to replace the specific embodiments described, as long as they do not deviate from the structure of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A performance evaluation method for constructing a measurement control chart based on a Bayesian algorithm, characterized in that: include: The time frequency setting of the constructed metrology control chart is updated through the Bayesian algorithm to obtain the update cycle of the metrology control chart, and the preprocessing operation information of the platform to which the metrology control chart belongs in the first sub-cycle of the update is collected to determine whether to perform preprocessing again; Obtaining visualization delay information of the measurement control chart during the second sub-period of updating, and determining whether to perform visualization delay adjustment; Collect the load operation information of the platform to which the metering control chart belongs during the update of the third sub-period, and determine whether data screening is required.

2. The method for constructing a performance evaluation method for a measurement control chart based on a Bayesian algorithm according to claim 1, characterized in that: The specific determination process of whether to perform preprocessing again is as follows: Analyze the preprocessing operation information of the platform to which the metrology control chart belongs during the first sub-period of updating, obtain the preprocessing operation compliance of the platform to which the metrology control chart belongs, and compare it with the preprocessing operation compliance threshold preset in the evaluation database; if the preprocessing operation compliance of the platform to which the metrology control chart belongs is higher than or equal to the preprocessing operation compliance threshold, there is no need to preprocess the data again; if the preprocessing operation compliance of the platform to which the metrology control chart belongs is lower than the preprocessing operation compliance threshold, send a preprocessing operation exception instruction of the platform to which the metrology control chart belongs to the metrology control chart to the platform to which the metrology control chart belongs, and the platform to which the metrology control chart belongs performs a preprocessing operation check to complete the data preprocessing.

3. The method for constructing a performance evaluation method for a measurement control chart based on a Bayesian algorithm according to claim 2, characterized in that: The platform to which the metrology control chart belongs performs a pre-processing operation check, and the specific execution process is as follows: Obtaining a time period between updating the first sub-cycle and updating the second sub-cycle, and recording it as a preprocessing detection period; When the preprocessing operation compliance of the platform to which the metrology control chart belongs is lower than the preprocessing operation compliance threshold, the data processing module of the platform to which the metrology control chart belongs synchronously sends the preprocessing operation abnormal instruction of the platform to which the metrology control chart belongs to the metrology control chart to the platform to which the metrology control chart belongs. Finally, the platform to which the metrology control chart belongs performs preprocessing inspection on the data within the preprocessing detection period through the preset preprocessing operation inspection module. After the inspection is completed, the platform to which the metrology control chart belongs directly visualizes the data.

4. The method for constructing a performance evaluation method for a measurement control chart based on a Bayesian algorithm according to claim 2, characterized in that: The pre-processing operation compliance of the platform to which the measurement control chart belongs is analyzed in detail as follows: The preprocessing operation information of the platform to which the metering control chart belongs during the updating of the first sub-period, specifically including the data preprocessing operation duration of the platform to which the metering control chart belongs, the sampling frequency of the platform to which the metering control chart belongs, and the data throughput of the platform to which the metering control chart belongs; The data preprocessing running time and data throughput are weighted with the corresponding preprocessing running weights preset in the evaluation database, and the sampling frequency is combined with the sampling reference frequency preset in the evaluation database for difference processing to obtain the preprocessing running compliance of the platform to which the metrology control chart belongs. The preprocessing running compliance of the platform to which the metrology control chart belongs represents the influence of the data preprocessing running time, sampling frequency and data throughput on the preprocessing running of the platform to which the metrology control chart belongs. The compliance quantitative value is used to quantify the compliance degree of the preprocessing operation.

5. The method for constructing a performance evaluation method for a measurement control chart based on a Bayesian algorithm according to claim 1, characterized in that: The specific process of determining whether to perform visualization delay adjustment is as follows: Process the visualization delay information of the metrology control chart during the second sub-period of updating, and obtain the visualization delay degree value of the metrology control chart, and compare it with the visualization delay degree threshold preset in the evaluation database. If the visualization delay degree value of the metrology control chart is lower than the visualization delay degree threshold, there is no need to perform visualization delay adjustment. If the visualization delay degree value of the metrology control chart is higher than or equal to the visualization delay degree threshold, perform visualization delay adjustment.

6. The method for constructing a performance evaluation method for a measurement control chart based on a Bayesian algorithm according to claim 5 is characterized in that: The specific execution process of the visualization delay adjustment is as follows: The visualization delay degree value of the metering control chart is differenced with the visualization delay degree threshold to obtain the visualization delay deviation value of the metering control chart. The platform to which the metering control chart belongs adjusts the visualization delay ratio of the pre-processed operation information according to the visualization delay deviation value of the metering control chart to complete the visualization delay adjustment of the metering control chart.

7. The method for constructing a performance evaluation method for a measurement control chart based on a Bayesian algorithm according to claim 5, characterized in that: The visualization delay degree value of the measurement control chart, the specific analysis process is as follows: The visualization delay information of the metering control chart in the second sub-period of updating specifically includes the proportion of undisplayed data of the metering control chart, the average data delay duration of the metering control chart, and the network response duration of the platform to which the metering control chart belongs; The proportion of undisplayed data, the average data delay time, the network response time, and the preprocessing operation compliance of the platform to which the metering control chart belongs are weighted respectively with the delay impact weights preset in the evaluation database to obtain the visualization delay degree value of the metering control chart. The visualization delay degree value of the metering control chart represents a quantitative numerical value of the influence of the proportion of undisplayed data, the average data delay time, the network response time, and the preprocessing operation compliance of the platform to which the metering control chart belongs on the visualization delay of the metering control chart, and is used to quantify the visualization delay degree.

8. The method for constructing a performance evaluation method for a measurement control chart based on a Bayesian algorithm according to claim 1, characterized in that: The specific process of determining whether data screening is required is as follows: According to the load operation information of the platform to which the metering control diagram belongs during the third sub-period of updating, the load operation compliance value of the platform to which the metering control diagram belongs is obtained, and compared with the load operation compliance threshold value preset in the evaluation database; If the load operation compliance value of the platform to which the metering control chart belongs is higher than or equal to the load operation compliance threshold, no data screening is required; If the load operation compliance value of the platform to which the metering control chart belongs is lower than the load operation compliance threshold, data filtering is required, and data visualization is performed again after the data filtering is completed.

9. The method for constructing a performance evaluation method for a measurement control chart based on a Bayesian algorithm according to claim 8, characterized in that: The load operation compliance value of the platform to which the metering control diagram belongs, the specific analysis process is as follows: The load operation information of the platform to which the metering control graph belongs during the third sub-period of updating specifically includes the CPU utilization rate of the platform to which the metering control graph belongs, the memory usage of the platform to which the metering control graph belongs, and the load average of the platform to which the metering control graph belongs; The CPU utilization, memory usage, load mean, preprocessing operation compliance of the platform to which the metering control chart belongs, and the visualization delay degree value of the metering control chart are weighted with the corresponding load operation compliance weights preset in the evaluation database, and coupled to obtain the load operation compliance value of the platform to which the metering control chart belongs. The load operation compliance value of the platform to which the metering control chart belongs represents a compliance quantitative value of the degree of influence of the CPU utilization, memory usage, load mean, preprocessing operation compliance of the platform to which the metering control chart belongs, and the visualization delay degree value of the metering control chart on the load operation of the platform to which the metering control chart belongs, and is used to quantify the load operation compliance degree.

10. A system for constructing a performance evaluation method using a measurement control chart based on a Bayesian algorithm as claimed in any one of claims 1 to 9, characterized in that: include: The preprocessing operation determination module is used to set and update the time frequency of the constructed metrology control chart through the Bayesian algorithm, obtain the update cycle of the metrology control chart, and collect the preprocessing operation information of the platform to which the metrology control chart belongs within the first sub-cycle of the update to determine whether to perform preprocessing again; A visualization delay determination module is used to obtain visualization delay information of the metering control chart during the second sub-period of updating, and determine whether to perform visualization delay adjustment; The data screening and determination module is used to collect the load operation information of the platform to which the metering control diagram belongs during the third sub-period of updating, and to determine whether data screening is required.

Images