Production monitoring system of aluminum alloy storage tank

By designing an aluminum alloy storage tank production monitoring system, real-time monitoring and deep correction of production data, a health deviation evaluation index is generated, and the problem of low accuracy of monitoring and early warning of aluminum alloy storage tank production is solved, and more accurate early warning and production process control is achieved.

CN119087892APending Publication Date: 2024-12-06NANTONG CHENGDA METAL EQUIP MFG CO LTD

Patent Information

Application Number
CN202411564210.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-05
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

The accuracy of monitoring and early warning of aluminum alloy storage tank production is low, resulting in the inability to conduct early warnings in a timely manner.

Method used

An aluminum alloy storage tank production monitoring system was designed, including production line monitoring module, monitoring precision module, production line control traceability module, control feedback prediction module, health deviation evaluation module and production line monitoring and early warning module, to monitor the production process in real time The key parameters in the system are in-depth correction compensation and data processing, and a health deviation evaluation index is generated, and an early warning is performed based on the embedded warning table.

Benefits of technology

It improves the accuracy of production monitoring and early warning, can timely identify and warn the health status of the production line, and ensures the stability and quality of the production process.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119087892A_ABST
    Figure CN119087892A_ABST
Patent Text Reader

Abstract

The invention discloses a production monitoring system for an aluminum alloy storage tank, and relates to the technical field of monitoring control, and the production monitoring system comprises the steps: carrying out the real-time monitoring of a production line, and obtaining J aluminum alloy storage tank production monitoring data streams; depth correction compensation is carried out based on the J aluminum alloy storage tank production monitoring data streams, and a jth production line monitoring block corresponding to a jth aluminum alloy storage tank production line is obtained; a jth production line production control block corresponding to a jth aluminum alloy storage tank production line is called; health feedback prediction is carried out based on the production control block of the j-th production line, and a health feedback block of the j-th production line is obtained; and performing health deviation evaluation on the j-th production line monitoring block, generating a j-th production line health deviation evaluation index, and performing early warning on the j-th aluminum alloy storage tank production line based on an embedded production line health deviation grading early warning table. The technical problem that the production monitoring and early warning precision of the aluminum alloy storage tank is low in the prior art is solved, and the technical effect of improving the production monitoring and early warning precision is achieved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of monitoring and control, and in particular to a production monitoring system for an aluminum alloy storage tank. Background Art

[0002] With the rapid development of industrial production and transportation, aluminum alloy storage tanks have been widely used in many fields. However, with the continuous expansion of production scale and the complexity of process flow, the production monitoring and management of aluminum alloy storage tanks are also facing increasing challenges. Traditional production monitoring methods often have low monitoring and early warning accuracy, resulting in the problem of failure to provide early warning in a timely manner. Summary of the invention

[0003] The present application provides a production monitoring system for aluminum alloy storage tanks, which is used to solve the technical problem of low accuracy of production monitoring and early warning for aluminum alloy storage tanks in the prior art.

[0004] In view of the above problems, the present application provides a production monitoring system for aluminum alloy storage tanks.

[0005] The present application provides a production monitoring system for aluminum alloy storage tanks, the system comprising: A production line monitoring module, which is used to perform real-time monitoring of J aluminum alloy storage tank production lines, and obtain J aluminum alloy storage tank production monitoring data streams, where J is a positive integer greater than 1, and each aluminum alloy storage tank production line corresponds to an aluminum alloy storage tank production node; a monitoring refinement module, which is used to perform deep correction compensation based on the J aluminum alloy storage tank production monitoring data streams, and obtain the j-th production line monitoring block corresponding to the j-th aluminum alloy storage tank production line, where j is a positive integer, 1≤j≤J; a production line control tracing module, which is used to retrieve the j-th aluminum alloy storage tank production line corresponding to the production line monitoring block; The j-th production line production control block; a control feedback prediction module, the control feedback prediction module is used to perform health feedback prediction based on the j-th production line production control block to obtain the j-th production line health feedback block; a health deviation evaluation module, the health deviation evaluation module is used to perform health deviation evaluation on the j-th production line monitoring block based on the j-th production line health feedback block to generate the j-th production line health deviation evaluation index; a production line monitoring and early warning module, the production line monitoring and early warning module is based on the embedded production line health deviation graded early warning table, combined with the j-th production line health deviation evaluation index to warn the j-th aluminum alloy storage tank production line.

[0006] The technical solution provided in this application has at least the following technical effects or advantages: The present application performs real-time monitoring on J aluminum alloy storage tank production lines to obtain J aluminum alloy storage tank production monitoring data streams, wherein J is a positive integer greater than 1, and each aluminum alloy storage tank production line corresponds to an aluminum alloy storage tank production node; performs deep correction compensation based on the J aluminum alloy storage tank production monitoring data streams to obtain the j-th production line monitoring block corresponding to the j-th aluminum alloy storage tank production line, wherein j is a positive integer, 1≤j≤J; retrieves the j-th production line production control block corresponding to the j-th aluminum alloy storage tank production line; performs health feedback prediction based on the j-th production line production control block to obtain the j-th production line health feedback block; performs health deviation evaluation on the j-th production line monitoring block based on the j-th production line health feedback block to generate the j-th production line health deviation evaluation index; based on the embedded production line health deviation graded warning table, the j-th aluminum alloy storage tank production line is warned in combination with the j-th production line health deviation evaluation index. The present invention solves the technical problem of low accuracy of production monitoring and early warning of aluminum alloy storage tanks in the prior art, and achieves the technical effect of improving the accuracy of production monitoring and early warning by real-time monitoring of key parameters in the production process, performing deep correction compensation and data processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0008] Figure 1 A schematic diagram of the structure of a production monitoring system for an aluminum alloy storage tank provided in an embodiment of the present application; Figure 2 A schematic flow chart of a production monitoring method for aluminum alloy storage tanks provided in an embodiment of the present application.

[0009] Explanation of the accompanying drawings: production line monitoring module 11, monitoring refinement module 12, production line control tracing module 13, control feedback prediction module 14, health deviation evaluation module 15, production line monitoring and early warning module 16. DETAILED DESCRIPTION

[0010] The present application provides a production monitoring system for aluminum alloy storage tanks, aiming to solve the technical problem of low accuracy of production monitoring and early warning of aluminum alloy storage tanks in the prior art. By real-time monitoring of key parameters in the production process, deep correction compensation and data processing are performed to achieve the technical effect of improving the accuracy of production monitoring and early warning.

[0011] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0012] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices. Embodiment 1

[0013] like Figure 1 As shown, the present application provides a production monitoring system for aluminum alloy storage tanks, which is used to perform the following Figure 2 A production monitoring method for an aluminum alloy storage tank is shown, the system comprising: The production line monitoring module 11 is used to perform real-time monitoring on J aluminum alloy storage tank production lines to obtain J aluminum alloy storage tank production monitoring data streams, wherein J is a positive integer greater than 1, and each aluminum alloy storage tank production line corresponds to an aluminum alloy storage tank production node.

[0014] In an embodiment of the present application, high-precision sensors are installed on each aluminum alloy storage tank production line, and the sensors include temperature sensors, pressure sensors, liquid level sensors, etc., which are used to collect in real time the key parameters of the aluminum alloy storage tank production nodes corresponding to each aluminum alloy storage tank production line during the production process, such as temperature, pressure, flow rate, liquid level, etc.

[0015] The production line monitoring module performs real-time monitoring through high-precision sensors installed on each aluminum alloy storage tank production line to obtain J aluminum alloy storage tank production monitoring data streams.

[0016] The monitoring refinement module 12 is used to perform deep correction compensation based on the J aluminum alloy storage tank production monitoring data streams to obtain the j-th production line monitoring block corresponding to the j-th aluminum alloy storage tank production line, wherein j is a positive integer, 1≤j≤J.

[0017] In an embodiment of the present application, when performing depth correction and compensation based on J aluminum alloy storage tank production monitoring data streams, first extract the j-th aluminum alloy storage tank production monitoring data stream from the J aluminum alloy storage tank production monitoring data streams. Standardize the extracted data to generate the j-th production line monitoring data block. Next, perform denoising on the standardized j-th production line monitoring data block to generate the j-th production line monitoring denoised data block. Retrieve the j-th production line monitoring equipment status data set corresponding to the j-th production line monitoring denoised data block, and perform monitoring error correction on the j-th production line monitoring denoised data block according to the j-th production line monitoring equipment status data set, wherein the monitoring error correction includes the steps of anomaly detection, accuracy impact evaluation, and anomaly compensation.

[0018] Through the above process, the j-th production line monitoring block corresponding to the j-th aluminum alloy storage tank production line is obtained.

[0019] Furthermore, in the system provided by the embodiment of the application, the monitoring refinement module is used to perform depth correction compensation based on the J aluminum alloy storage tank production monitoring data streams to obtain the j-th production line monitoring block corresponding to the j-th aluminum alloy storage tank production line, including: According to the J aluminum alloy storage tank production monitoring data streams, the j-th aluminum alloy storage tank production monitoring data stream is extracted; according to the j-th aluminum alloy storage tank production monitoring data stream, standardization processing is performed to obtain the j-th production line monitoring data block; according to the moving average filter embedded in the monitoring refinement module, the j-th production line monitoring data block is denoised to obtain the j-th production line monitoring denoised data block; the j-th production line monitoring equipment status data set corresponding to the j-th production line monitoring denoised data block is retrieved; according to the j-th production line monitoring equipment status data set, monitoring error correction is performed on the j-th production line monitoring denoised data block to generate the j-th production line monitoring block.

[0020] In an embodiment of the present application, the j-th aluminum alloy storage tank production monitoring data stream is first extracted from J aluminum alloy storage tank production monitoring data streams, and the j-th aluminum alloy storage tank production monitoring data stream is standardized. Through data scaling, centralization and other operations, the dimensional differences between different monitoring devices or different monitoring parameters are eliminated, so that the data has a unified measurement standard, and the j-th production line monitoring data block is obtained.

[0021] Next, the moving average filter embedded in the monitoring refinement module is used to denoise the standardized j-th production line monitoring data block. The moving average filter creates a continuous average value of the input signal through moving average filtering to smooth the data sequence, eliminate random noise and interference signals in the data, extract more accurate and stable monitoring signals, and generate the j-th production line monitoring denoised data block. Then, the j-th production line monitoring equipment status data set corresponding to the j-th production line monitoring denoised data block is retrieved. The equipment status data set contains the equipment's operating status information, such as working status, fault records, and maintenance records, which are stored in the historical monitoring database.

[0022] Finally, based on the j-th production line monitoring equipment status data set, the monitoring error correction is performed. When performing monitoring error correction, anomaly detection is first performed to identify abnormal conditions in the monitoring data, detect abnormal data points, and generate the j-th monitoring abnormal condition data. Then, an accuracy impact evaluation is performed to evaluate the impact of abnormal conditions on the accuracy of the monitoring data, and the j-th monitoring accuracy impact coefficient is obtained. If the j-th monitoring accuracy impact coefficient is greater than or equal to the preset monitoring accuracy impact threshold, the embedded monitoring anomaly compensation model is activated to generate the j-th monitoring anomaly compensation data. Finally, based on the j-th monitoring anomaly compensation data, compensation correction is performed on the j-th production line monitoring denoised data block.

[0023] Furthermore, in the system provided by the embodiment of the application, the monitoring error correction of the j-th production line monitoring denoised data block is performed according to the j-th production line monitoring equipment status data set to generate the j-th production line monitoring block, including: Perform anomaly detection based on the j-th production line monitoring equipment status data set to obtain the j-th monitoring abnormal status data; perform monitoring accuracy impact evaluation based on the j-th monitoring abnormal status data to obtain the j-th monitoring accuracy impact coefficient; determine whether the j-th monitoring accuracy impact coefficient is greater than / equal to the monitoring accuracy impact threshold; if the j-th monitoring accuracy impact coefficient is greater than / equal to the monitoring accuracy impact threshold, activate the monitoring abnormality compensation model embedded in the monitoring refinement module; based on the j-th monitoring abnormal status data and according to the monitoring abnormality compensation model, obtain the j-th monitoring abnormality compensation data; perform compensation and correction on the j-th production line monitoring denoised data block according to the j-th monitoring abnormality compensation data to obtain the j-th production line monitoring block.

[0024] In the embodiment of the present application, firstly, anomaly detection is performed using the isolation forest algorithm based on the j-th production line monitoring equipment status data set. The isolation forest isolates data points by constructing multiple random trees. Abnormal points are usually easier to isolate, so the average path length of abnormal points is shorter. By calculating the average path length of each data point, the abnormal state in the monitoring data is identified, and the j-th monitoring abnormal state data is generated.

[0025] Next, the difference between the monitoring data and the equipment status data is evaluated by the root mean square error according to the jth monitoring abnormal state data, and the jth monitoring accuracy influence coefficient is calculated. The obtained jth monitoring accuracy influence coefficient is compared with the preset monitoring accuracy influence threshold. When the jth monitoring accuracy influence coefficient is greater than or equal to the preset monitoring accuracy influence threshold, it indicates that the abnormal state has a significant impact on the monitoring data and needs to be compensated. At this time, the monitoring abnormality compensation model embedded in the monitoring refinement module is activated. The monitoring abnormality compensation model is pre-trained. During training, historical monitoring data, equipment status data and abnormal event records over a long period of time are obtained from the historical database. Next, key features such as temperature fluctuations and pressure changes are extracted from the historical monitoring data, and operating status features such as equipment health status and fault frequency are extracted from the equipment status data. Time series models such as ARIMA are used for training. The extracted features are input into the model to learn the data patterns under normal and abnormal conditions. Through abnormal event records, the model is trained to identify and compensate for abnormal state data. The monitoring abnormality compensation model trained by the above data and steps is obtained.

[0026] The jth monitoring abnormal state data is input into the monitoring abnormality compensation model to obtain the jth monitoring abnormality compensation data. Finally, according to the jth monitoring abnormality compensation data, the jth production line monitoring denoised data block is compensated and corrected. During the compensation correction process, the weighted average method is used to combine the compensation data with the denoised data, where the weights of the compensation data and the denoised data are pre-set, and the jth production line monitoring block is obtained by calculation.

[0027] The production line control tracing module 13 is used to retrieve the j-th production line production control block corresponding to the j-th aluminum alloy storage tank production line.

[0028] In the embodiment of the present application, the production control data of each production line is stored in an industrial database, such as SQL Server. The production control data includes production plans, process parameters, equipment operation records, quality inspection data, etc., which are indexed and managed by key fields such as timestamps and production batches.

[0029] When the production control data of a production line needs to be traced, the production line control traceability module retrieves the corresponding production control data from the database through SQL query. Specifically, according to the j-th production line and time period, the SQL query statement is used to retrieve the relevant data in the database. In the same way, the production line control traceability module uses the SQL query statement to retrieve the j-th production line production control block corresponding to the j-th aluminum alloy storage tank production line.

[0030] The control feedback prediction module 14 is used to perform health feedback prediction based on the j-th production line production control block to obtain the j-th production line health feedback block.

[0031] Furthermore, in the system provided by the embodiment of the application, the control feedback prediction module is used to perform health feedback prediction based on the j-th production line production control block to obtain the j-th production line health feedback block, including: According to the control feedback prediction module, the production line control health monitoring sample set corresponding to the j-th aluminum alloy storage tank production line is retrieved; according to the j-th production line production control block, feature association selection is performed on the production line control health monitoring sample set to obtain the production line health monitoring association source; according to the production line health monitoring association source, a multi-factor production line health monitoring association diagram is constructed; according to the multi-factor production line health monitoring association diagram, a central trend evaluation is performed to obtain the multi-factor production line health monitoring characteristic value; according to the multi-factor production line health monitoring characteristic value, the j-th production line health feedback block is generated.

[0032] In an embodiment of the present application, when performing health feedback prediction, the production line control health monitoring sample set corresponding to the j-th aluminum alloy storage tank production line is first retrieved from the industrial database according to the control feedback prediction module. The sample set includes historical production data, process parameters, equipment operation records and quality inspection results. When performing feature association selection based on the retrieved production line control health monitoring sample set, the key feature parameters of the production control block, such as temperature, pressure, flow, speed, etc., are first extracted. Then, correlation analysis methods, such as the Pearson correlation coefficient and mutual information, are used to calculate the correlation between each feature and the health status. According to the correlation analysis results, the correlation coefficient that reaches the preset threshold is screened out as the key feature that has a significant impact on the health status, and the screened out key features are used as production line health monitoring association sources.

[0033] Next, when constructing a multi-factor production line health monitoring association graph based on the production line health monitoring association source, a graph structure representation method is used, each feature parameter is used as a node in the graph, and the strength of the association between features is used as the weight of the edge. Specifically, each key feature is used as a node, and a correlation analysis method such as the Pearson correlation coefficient is used to calculate the strength of the association between features. Edges are created based on the strength of the association, and the weight of the edge represents the strength of the association. This process completes the construction of the multi-factor production line health monitoring association graph.

[0034] After the correlation diagram is constructed, the central tendency evaluation is performed, and the principal component analysis method is used to extract the main eigenvalues, reduce the dimension and capture the central tendency of the data. PCA standardizes the data, calculates the covariance matrix, calculates the eigenvalues ​​and eigenvectors, selects the eigenvector that explains the maximum variance, and generates the eigenvalues ​​of multi-factor production line health monitoring. Explaining the maximum variance means selecting those eigenvectors that can explain the most variation in the data, that is, selecting the eigenvector with the largest eigenvalue.

[0035] Finally, cluster analysis, such as K-means, is used to aggregate and predict the eigenvalues ​​based on the extracted multi-factor production line health monitoring eigenvalues. Specifically, K cluster centers are initialized, the distance from each data point to the cluster center is calculated, and the data points are assigned to the nearest cluster center. The average value of each cluster is calculated as the new cluster center. Repeat the above steps until the cluster center no longer changes, and generate the j-th production line health feedback block, which includes the current health status assessment and future health status prediction.

[0036] Furthermore, in the system provided by the embodiment of the application, feature correlation selection is performed on the production line control health monitoring sample set according to the j-th production line production control block to obtain the production line health monitoring correlation source, including: According to the production line control health monitoring sample set, a first production line control health monitoring sample is extracted, wherein the first production line control health monitoring sample includes a first production line production control block sample and a first production line health monitoring block sample corresponding to the j-th aluminum alloy storage tank production line; a feature association evaluation is performed on the first production line production control block sample according to the j-th production line production control block to obtain a first control block feature association coefficient; it is determined whether the first control block feature association coefficient satisfies the control block feature association constraint; if the first control block feature association coefficient satisfies the control block feature association constraint, the first production line health monitoring block sample is added to the production line health monitoring association source; based on the control block feature association constraint and the j-th production line production control block, feature association selection is continued for the production line control health monitoring sample set to generate the production line health monitoring association source.

[0037] In the embodiment of the present application, the first production line control health monitoring sample is first extracted from the production line control health monitoring sample set. The sample includes the first production line production control block sample and the first production line health monitoring block sample corresponding to the j-th aluminum alloy storage tank production line. The first production line production control block sample contains process parameters such as temperature, pressure, flow, etc., and the first production line health monitoring block sample contains health status information, such as equipment health score, fault record, etc.

[0038] Next, based on the j-th production line production control block, the feature correlation evaluation is performed on the first production line production control block sample, and the correlation analysis method, such as the Pearson correlation coefficient, is used to calculate the correlation between each feature and the health status. Specifically, the correlation coefficient between each feature and the corresponding health monitoring data is calculated, and the feature correlation coefficient of the first control block is obtained by calculating the correlation coefficient. Then, it is determined whether the feature correlation coefficient of the first control block satisfies the preset control block feature correlation constraint. The correlation constraint is a preset threshold value used to filter out features that are significantly correlated with the health status. If the feature correlation coefficient is greater than or equal to the threshold, it means that the feature has a significant correlation with the health status. If the feature correlation coefficient of the first control block satisfies the control block feature correlation constraint, the first production line health monitoring block sample is added to the production line health monitoring association source.

[0039] Based on the control block feature association constraints and the j-th production line production control block, continue to perform feature association selection on the production line control health monitoring sample set. Repeat the above process, perform feature association evaluation on each production control block sample in the sample set, filter out the features that meet the association constraints, and add the corresponding health monitoring block samples to the production line health monitoring association source.

[0040] The health deviation evaluation module 15 is used to perform a health deviation evaluation on the j-th production line monitoring block based on the j-th production line health feedback block to generate a health deviation evaluation index for the j-th production line.

[0041] Furthermore, in the system provided by the embodiment of the application, the health deviation evaluation module is used to perform a health deviation evaluation on the j-th production line monitoring block based on the j-th production line health feedback block to generate a health deviation evaluation index for the j-th production line, including: According to the j-th production line health feedback block, a multi-feature deviation comparison is performed on the j-th production line monitoring block to obtain a plurality of characteristic health deviation vectors; a plurality of health deviation risk assessment domains are constructed, and the plurality of health deviation risk assessment domains are embedded into the health deviation assessment module; the plurality of characteristic health deviation vectors are input into the plurality of health deviation risk assessment domains to obtain a plurality of health deviation risk coefficients; a deviation depth evaluation is performed according to the plurality of characteristic health deviation vectors to obtain a plurality of health deviation depth coefficients; and the health deviation evaluation index of the j-th production line is calculated according to the plurality of health deviation risk coefficients and the plurality of health deviation depth coefficients.

[0042] In an embodiment of the present application, when performing a multi-feature deviation comparison on the j-th production line monitoring block according to the j-th production line health feedback block, relevant data is obtained from the j-th production line health feedback block and the j-th production line monitoring block. The j-th production line health feedback block includes health status assessment and prediction results, and the j-th production line monitoring block includes actual monitoring data and operating parameters. The acquired data is preprocessed, including data cleaning, deduplication and standardization. According to the predicted value in the j-th production line health feedback block and the actual value in the j-th production line monitoring block, a multi-feature deviation comparison is performed, and the compared features include key parameters such as temperature, pressure, flow, speed, etc. When performing a feature deviation comparison, the deviation value is obtained by subtracting the predicted value from the actual value and then removing the predicted value. By calculating the deviation values ​​of multiple features, the deviation value of each feature is used as a vector to generate multiple feature health deviation vectors.

[0043] When constructing multiple health deviation risk assessment domains, first extract and identify the attributes of the characteristic health deviation vector, collect sample data based on these attributes for distribution feature identification, and construct a health deviation risk assessment coordinate system. Then use statistical methods and machine learning models, such as SVM and random forest, for training to generate health deviation risk assessment domains. Embed the constructed multiple health deviation risk assessment domains into the health deviation assessment module, and each health deviation risk assessment domain is used to assess the health deviation risk of a specific feature. Input multiple characteristic health deviation vectors into multiple health deviation risk assessment domains to obtain multiple health deviation risk coefficients, each of which reflects the degree of influence of characteristic deviation on health status. For example, input the temperature health deviation vector into the temperature health deviation risk assessment domain to obtain the temperature health deviation risk coefficient.

[0044] Next, the deviation depth is evaluated based on multiple characteristic health deviation vectors. Clustering algorithms such as K-means and DBSCAN are used to perform cluster analysis on the characteristic health deviation vectors, determine the risk level of each characteristic deviation, and calculate the deviation depth coefficient. Specifically, the characteristic health deviation vectors are divided into K clusters using the K-means clustering algorithm, and each cluster represents a risk level. The center point of each cluster and the distance from the sample point to the center point are calculated, where the center point is the mean of all samples in each cluster, and the distance from each sample point to the center point of its cluster is its health deviation depth coefficient, generating multiple health deviation depth coefficients.

[0045] Finally, multiple health deviation risk coefficients and multiple health deviation depth coefficients are substituted into the health deviation evaluation analytical function to calculate the health deviation evaluation index of the jth production line.

[0046] Furthermore, in the system provided in the application embodiment, multiple health deviation risk assessment domains are constructed, including: According to the multiple characteristic health deviation vectors, a first characteristic health deviation vector is extracted; according to the first characteristic health deviation vector, attribute identification is performed to obtain a first characteristic health deviation attribute; according to the first characteristic health deviation attribute, health deviation risk assessment samples are collected for the j-th aluminum alloy storage tank production line to obtain a first characteristic health deviation vector sample set and a first health deviation risk coefficient sample set; distribution feature identification is performed according to the first characteristic health deviation vector sample set and the first health deviation risk coefficient sample set to obtain first sample distribution feature information; according to the first characteristic health deviation attribute and the first sample distribution feature information, a first health deviation risk assessment coordinate system is constructed; based on the first characteristic health deviation vector sample set and the first health deviation risk coefficient sample set, according to the first health deviation risk assessment coordinate system, a first health deviation risk assessment domain is constructed, and the first health deviation risk assessment domain is added to the multiple health deviation risk assessment domains.

[0047] In an embodiment of the present application, first, a characteristic health deviation vector is randomly extracted from multiple characteristic health deviation vectors as a first characteristic health deviation vector, for example, a temperature health deviation vector. Next, the first characteristic health deviation vector is attributed and its related characteristic attributes, such as numerical range, change trend, etc., are determined. For example, the attributes of the temperature health deviation vector include maximum value, minimum value, average value, variance, etc. Based on the identified characteristic health deviation attributes, health deviation risk assessment samples are collected for the j-th aluminum alloy storage tank production line. The specific operation is to extract relevant historical data from the database to form a first characteristic health deviation vector sample set and a first health deviation risk coefficient sample set.

[0048] According to the first characteristic health deviation vector sample set and the first health deviation risk coefficient sample set, distribution feature identification is performed. The mean, variance, skewness and kurtosis of the sample data are calculated, and the distribution characteristics of the data are identified using statistical analysis methods, such as whether the data conforms to the normal distribution. Based on the identified distribution feature information, a health deviation risk assessment coordinate system is constructed. Determine the dimension and measurement standard of the coordinate system, for example, the horizontal axis represents the health deviation value and the vertical axis represents the health risk coefficient. For the temperature health deviation vector, the horizontal axis represents the temperature deviation value and the vertical axis represents the corresponding health risk coefficient. In the health deviation risk assessment coordinate system, a health deviation risk assessment domain is constructed. Use a data fitting method, such as a linear regression model, to fit the relationship between the health deviation value and the risk coefficient. After the fitting is completed, a temperature health deviation risk assessment domain is formed. After the temperature health deviation risk assessment domain is constructed, it is added to multiple health deviation risk assessment domains. Then other features, such as pressure, flow, speed, etc., are processed in the same way to construct the corresponding health deviation risk assessment domain.

[0049] Furthermore, in the system provided in the embodiment of the application, the health deviation evaluation index of the j-th production line is calculated according to the multiple health deviation risk coefficients and the multiple health deviation depth coefficients, including: The multiple health deviation risk coefficients and the multiple health deviation depth coefficients are input into the health deviation evaluation analytical function embedded in the health deviation evaluation module to obtain the health deviation evaluation index of the j-th production line. The health deviation evaluation analytical function is: ; Among them, phy j Characterizes the production line health deviation evaluation index corresponding to the j-th aluminum alloy storage tank production line, hrx jr Characterizes the rth health deviation risk coefficient corresponding to the jth aluminum alloy storage tank production line, r is a positive integer, 1≤r≤R, R is the total number of health deviation risk coefficients corresponding to the jth aluminum alloy storage tank production line, and the total number of health deviation risk coefficients is equal to the total number of health deviation depth coefficients corresponding to the jth aluminum alloy storage tank production line, hdx jr It represents the rth health deviation depth coefficient corresponding to the jth aluminum alloy storage tank production line, α and β represent the analytical weight conditions of health deviation evaluation, 0<α<1, 0<β<1, and the sum of α and β is 1.

[0050] In the embodiment of the present application, in order to calculate the health deviation evaluation index of the j-th production line, the health deviation evaluation index of the j-th production line is calculated by inputting multiple health deviation risk coefficients and multiple health deviation depth coefficients into the health deviation evaluation analytical function embedded in the health deviation evaluation module. In the health deviation evaluation analytical function, α and β represent the health deviation evaluation analytical weight conditions, and their values ​​are pre-set by technical experts.

[0051] The production line monitoring and early warning module 16 issues an early warning to the j-th aluminum alloy storage tank production line based on the embedded production line health deviation graded early warning table and the j-th production line health deviation evaluation index.

[0052] In an embodiment of the present application, a production line health deviation graded warning table is embedded in the production line monitoring and early warning module, which defines different warning levels according to different health deviation evaluation index ranges. For example, a health deviation evaluation index between 0.0 and 0.3 is normal, between 0.3 and 0.6 is a slight deviation, between 0.6 and 0.8 is a significant deviation, and between 0.8 and 1.0 is a serious deviation. According to the health deviation evaluation index of the j-th production line, it is compared with each warning level in the warning table to determine the corresponding warning level. According to the warning level, the production line monitoring and early warning module generates corresponding warning information, and the warning information includes a description of the current health status of the production line.

[0053] Finally, the production line monitoring and early warning module will notify relevant personnel of the generated early warning information through real-time alarms, emails, etc. to complete the early warning.

[0054] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects: The present application performs real-time monitoring on J aluminum alloy storage tank production lines to obtain J aluminum alloy storage tank production monitoring data streams, wherein J is a positive integer greater than 1, and each aluminum alloy storage tank production line corresponds to an aluminum alloy storage tank production node; performs deep correction compensation based on the J aluminum alloy storage tank production monitoring data streams to obtain the j-th production line monitoring block corresponding to the j-th aluminum alloy storage tank production line, wherein j is a positive integer, 1≤j≤J; retrieves the j-th production line production control block corresponding to the j-th aluminum alloy storage tank production line; performs health feedback prediction based on the j-th production line production control block to obtain the j-th production line health feedback block; performs health deviation evaluation on the j-th production line monitoring block based on the j-th production line health feedback block to generate the j-th production line health deviation evaluation index; based on the embedded production line health deviation graded warning table, the j-th aluminum alloy storage tank production line is warned in combination with the j-th production line health deviation evaluation index. The present invention solves the technical problem of low accuracy of production monitoring and early warning of aluminum alloy storage tanks in the prior art, and achieves the technical effect of improving the accuracy of production monitoring and early warning by real-time monitoring of key parameters in the production process, performing deep correction compensation and data processing.

[0055] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0056] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.

[0057] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A production monitoring system for aluminum alloy storage tanks, characterized in that: The system comprises: A production line monitoring module, which is used to monitor J aluminum alloy storage tank production lines in real time and obtain J aluminum alloy storage tank production monitoring data streams, where J is a positive integer greater than 1, and each aluminum alloy storage tank production line corresponds to an aluminum alloy storage tank production node; A monitoring refinement module, wherein the monitoring refinement module is used to perform depth correction compensation based on the J aluminum alloy storage tank production monitoring data streams to obtain the j-th production line monitoring block corresponding to the j-th aluminum alloy storage tank production line, wherein j is a positive integer, 1≤j≤J; A production line control tracing module, the production line control tracing module is used to retrieve the j-th production line production control block corresponding to the j-th aluminum alloy storage tank production line; A control feedback prediction module, the control feedback prediction module is used to perform health feedback prediction based on the j-th production line production control block to obtain the j-th production line health feedback block; A health deviation evaluation module, wherein the health deviation evaluation module is used to perform a health deviation evaluation on the j-th production line monitoring block based on the j-th production line health feedback block to generate a health deviation evaluation index for the j-th production line; A production line monitoring and early warning module, which issues an early warning to the j-th aluminum alloy storage tank production line based on an embedded production line health deviation graded early warning table and in combination with the j-th production line health deviation evaluation index.

2. A production monitoring system for aluminum alloy storage tanks as claimed in claim 1, characterized in that: The monitoring refinement module is used to perform depth correction compensation based on the J aluminum alloy storage tank production monitoring data streams to obtain the j-th production line monitoring block corresponding to the j-th aluminum alloy storage tank production line, including: Extracting the j-th aluminum alloy storage tank production monitoring data stream according to the J aluminum alloy storage tank production monitoring data streams; Performing standardization processing on the j-th aluminum alloy storage tank production monitoring data stream to obtain the j-th production line monitoring data block; Performing denoising processing on the j-th production line monitoring data block according to the moving average filter embedded in the monitoring refinement module to obtain the j-th production line monitoring denoised data block; Retrieving the j-th production line monitoring equipment status data set corresponding to the j-th production line monitoring denoised data block; The j-th production line monitoring denoised data block is subjected to monitoring error correction according to the j-th production line monitoring equipment status data set to generate the j-th production line monitoring block.

3. A production monitoring system for aluminum alloy storage tanks as claimed in claim 2, characterized in that: Performing monitoring error correction on the j-th production line monitoring denoised data block according to the j-th production line monitoring equipment status data set to generate the j-th production line monitoring block, including: Performing anomaly detection according to the j-th production line monitoring equipment status data set to obtain j-th monitoring abnormal status data; Performing a monitoring accuracy impact evaluation according to the j-th monitoring abnormal state data to obtain a j-th monitoring accuracy impact coefficient; Determine whether the j-th monitoring accuracy influence coefficient is greater than / equal to a monitoring accuracy influence threshold; If the j-th monitoring accuracy influence coefficient is greater than / equal to the monitoring accuracy influence threshold, activating the monitoring anomaly compensation model embedded in the monitoring precision module; Based on the j-th monitoring abnormal state data, and according to the monitoring abnormality compensation model, obtaining the j-th monitoring abnormality compensation data; The j-th production line monitoring denoised data block is compensated and corrected according to the j-th monitoring abnormality compensation data to obtain the j-th production line monitoring block.

4. A production monitoring system for aluminum alloy storage tanks as claimed in claim 1, characterized in that: The control feedback prediction module is used to perform health feedback prediction based on the j-th production line production control block to obtain the j-th production line health feedback block, including: According to the control feedback prediction module, the production line control health monitoring sample set corresponding to the j-th aluminum alloy storage tank production line is retrieved; Performing feature correlation selection on the production line control health monitoring sample set according to the j-th production line production control block to obtain a production line health monitoring correlation source; Constructing a multi-factor production line health monitoring association graph according to the production line health monitoring association source; Performing a central tendency evaluation based on the multi-factor production line health monitoring association diagram to obtain a multi-factor production line health monitoring characteristic value; The j-th production line health feedback block is generated according to the multi-factor production line health monitoring characteristic value.

5. A production monitoring system for aluminum alloy storage tanks as claimed in claim 4, characterized in that: According to the j-th production line production control block, feature correlation selection is performed on the production line control health monitoring sample set to obtain a production line health monitoring correlation source, including: Extracting a first production line control health monitoring sample according to the production line control health monitoring sample set, wherein the first production line control health monitoring sample includes a first production line production control block sample and a first production line health monitoring block sample corresponding to the j-th aluminum alloy storage tank production line; Performing feature correlation evaluation on the first production line production control block samples according to the jth production line production control block to obtain a first control block feature correlation coefficient; Determining whether the first control block feature association coefficient satisfies a control block feature association constraint; If the first control block feature association coefficient satisfies the control block feature association constraint, adding the first production line health monitoring block sample to the production line health monitoring association source; Based on the control block feature association constraint and the j-th production line production control block, feature association selection is continued for the production line control health monitoring sample set to generate the production line health monitoring association source.

6. A production monitoring system for aluminum alloy storage tanks as claimed in claim 1, characterized in that: The health deviation evaluation module is used to perform health deviation evaluation on the j-th production line monitoring block based on the j-th production line health feedback block to generate a health deviation evaluation index for the j-th production line, including: Performing a multi-feature deviation comparison on the j-th production line monitoring block according to the j-th production line health feedback block to obtain a plurality of feature health deviation vectors; constructing a plurality of health deviation risk assessment domains, and embedding the plurality of health deviation risk assessment domains into the health deviation assessment module; Inputting the plurality of characteristic health deviation vectors into the plurality of health deviation risk assessment domains to obtain a plurality of health deviation risk coefficients; Perform deviation depth evaluation according to the plurality of characteristic health deviation vectors to obtain a plurality of health deviation depth coefficients; The j-th production line health deviation evaluation index is calculated based on the multiple health deviation risk coefficients and the multiple health deviation depth coefficients.

7. A production monitoring system for aluminum alloy storage tanks as claimed in claim 6, characterized in that: Construct multiple health deviation risk assessment domains, including: Extracting a first characteristic health deviation vector according to the plurality of characteristic health deviation vectors; Performing attribute recognition according to the first characteristic health deviation vector to obtain a first characteristic health deviation attribute; According to the first characteristic health deviation attribute, health deviation risk assessment samples are collected for the j-th aluminum alloy storage tank production line to obtain a first characteristic health deviation vector sample set and a first health deviation risk coefficient sample set; Performing distribution feature identification based on the first characteristic health deviation vector sample set and the first health deviation risk coefficient sample set to obtain first sample distribution feature information; Constructing a first health deviation risk assessment coordinate system according to the first characteristic health deviation attribute and the first sample distribution characteristic information; Based on the first characteristic health deviation vector sample set and the first health deviation risk coefficient sample set, a first health deviation risk assessment domain is constructed according to the first health deviation risk assessment coordinate system, and the first health deviation risk assessment domain is added to the multiple health deviation risk assessment domains.

8. A production monitoring system for aluminum alloy storage tanks as claimed in claim 6, characterized in that: Calculating the j-th production line health deviation evaluation index according to the multiple health deviation risk coefficients and the multiple health deviation depth coefficients, including: The multiple health deviation risk coefficients and the multiple health deviation depth coefficients are input into the health deviation evaluation analytical function embedded in the health deviation evaluation module to obtain the health deviation evaluation index of the j-th production line. The health deviation evaluation analytical function is: ; Among them, phy j Characterizes the production line health deviation evaluation index corresponding to the j-th aluminum alloy storage tank production line, hrx jr Characterizes the rth health deviation risk coefficient corresponding to the jth aluminum alloy storage tank production line, r is a positive integer, 1≤r≤R, R is the total number of health deviation risk coefficients corresponding to the jth aluminum alloy storage tank production line, and the total number of health deviation risk coefficients is equal to the total number of health deviation depth coefficients corresponding to the jth aluminum alloy storage tank production line, hdx jr It represents the rth health deviation depth coefficient corresponding to the jth aluminum alloy storage tank production line, α and β represent the analytical weight conditions of health deviation evaluation, 0<α<1, 0<β<1, and the sum of α and β is 1.

Citation Information

Patent Citations

  • Temperature measurement method and device, electronic equipment and storage medium

    CN113008404A

  • Production monitoring and early warning system based on industrial internet

    CN116757546A

  • Early fault early warning and identification method and device for electromechanical equipment equipped with SCADA (supervisory control and data acquisition) system

    CN117992859A

  • Big data processing display unit control method and system

    CN118312658A

  • Industrial control host abnormal behavior identification method based on multi-modal data fusion

    CN118378196A

Cited By

  • Abnormality early warning method and device for liquid valve die-casting production line and medium

    CN120894900A