Industrial Doctor Enterprise Comprehensive Service Platform

Abstract

The present invention discloses an industrial doctor enterprise comprehensive service platform, comprising a data acquisition module, a data preprocessing and storage module, an anomaly diagnosis module, an early warning transmission module, and a user visualization and touch-based management interface. The present invention relates to the technical field of management service platforms. The industrial doctor enterprise comprehensive service platform establishes an anomaly diagnosis model and introduces data analysis to determine whether there are anomalies in the data of various sensors or production line equipment on the production line. Based on the anomalies, the platform performs tracing and locating operations on specific problems, generates diagnostic results for data transmission, and realizes real-time monitoring of each process while completing rapid fault diagnosis operations, reducing downtime and further improving production efficiency. The comprehensive service platform also uses unified command management and automated anomaly diagnosis to achieve traceability and location operations for faulty equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of management service platforms, and specifically to an industrial doctor enterprise comprehensive service platform. Background Art

[0002] Currently, monitoring the operating status of production line equipment typically relies solely on manual visual inspection or the equipment's built-in display screens. This manual online management places high demands on on-site personnel, requiring them to accurately and promptly monitor the situation. This is particularly true during night shifts, where oversight is often lacking, resulting in significant losses to production and safety. This also hinders rational allocation of personnel and prevents real-time online monitoring of the operation of equipment and process status across the company's production line.

[0003] The reference patent name is: An online operation and maintenance management system for enterprise production lines (patent publication number: CN116562500A, patent publication date: 2023-08-08), including: a data acquisition unit for acquiring enterprise production line data; a data transmission module for transmitting enterprise production line data to a data analysis module in real time through wireless communication technology; a data analysis module for analyzing enterprise production line data in real time to obtain the operating status of the enterprise production line, and the operating status includes: normal state and abnormal state; an abnormality handling module for processing the enterprise production line data corresponding to the abnormal state based on the abnormal state, and restoring the normal operating state of the enterprise production line. It can monitor the operation of each equipment and process status of the enterprise production line in real time online. If an abnormal condition occurs in the enterprise production line, it can handle the abnormal condition in time, and at the same time, it can grasp the abnormal position of the entire production line at the first time, and accurately judge the specific location of the abnormal operation condition.

[0004] Based on the description in the above documents, in the existing industrial production line processes, the industrial production line equipment or sensors often have abnormalities. In the current monitoring and management operations, although it is known that there are abnormalities in the operation process, the specific location of the abnormality is unknown, and the specific abnormal equipment cannot be diagnosed, so it is difficult to achieve real-time monitoring and rapid diagnosis operations, resulting in the problem that equipment failures cannot be handled in a timely manner. For this reason, the present invention provides an industrial doctor enterprise comprehensive service platform. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the present invention provides an industrial doctor enterprise comprehensive service platform, which solves the problem that in the existing industrial production line process, the industrial production line equipment or sensors often have abnormalities. In the current monitoring and management operations, although it is known that there are abnormalities in the operation process, the specific location of the abnormality is unknown, and the specific abnormal equipment cannot be diagnosed, making it difficult to achieve real-time monitoring and rapid diagnosis operations, resulting in the problem that equipment failures cannot be handled in a timely manner.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: Industrial Doctor Enterprise Comprehensive Service Platform, including:

[0007] The data acquisition module uses multiple sensors and combines IoT transmission to realize real-time collection of operating status data of industrial production line equipment;

[0008] The data preprocessing and storage module receives and stores the collected data in the database, and simultaneously implements data classification and screening operations to form the required data set;

[0009] The abnormality diagnosis module establishes an abnormality diagnosis model and introduces data analysis to determine whether there are abnormalities in the data of each sensor or production line equipment on the production line. It then traces and locates the specific problem based on the abnormality and generates diagnostic results for data transmission.

[0010] The early warning transmission module sends abnormal instructions to the nearest device terminal held by personnel and marks the abnormal location with a flashing red light;

[0011] The user visualization and touch management interface provides users with intuitive data display, transmits instructions according to on-site conditions, and performs integrated management and control operations.

[0012] Preferably, the operations for implementing data classification and screening in the data preprocessing and storage module are:

[0013] A1. Frame the collected video data to form the required image data;

[0014] A2. Set the required data template, extract the corresponding image category name from the data template, and classify the image data name according to the corresponding image category name;

[0015] A3. Classify the image data corresponding to the sensors in the industrial production line process sequence to form an initial image dataset labeled J.

[0016] Preferably, the framing operation of the video data collected in A1 is as follows:

[0017] a11. Take a segment of video data and set the processing frame period according to the production line's operating rate. Then, segment the video data from the beginning based on the frame period. Images separated by corresponding frame periods form sequential image data of the video data.

[0018] a12. Compare the sequential image data. If the adjacent image data in the sequential image data are the same, remove the preceding image data and retain the subsequent image data. If the adjacent image data in the sequential image data are different, retain both the preceding and subsequent image data.

[0019] Preferably, the abnormality diagnosis model is established by:

[0020] B1. Based on actual equipment operation data, a simulation model similar to the actual industrial production line equipment is introduced and constructed;

[0021] B2. Extract historical data of industrial production line equipment from the database based on product image data at the end of each process, and aggregate them into a result image dataset labeled K;

[0022] B3. Introduce the result image dataset K into the simulation model for data combination to establish the required abnormality diagnosis model.

[0023] Preferably, the abnormality diagnosis module analyzes whether there is abnormality in the data of each sensor or production line equipment on the production line:

[0024] C1. Introduce the classified initial image dataset J into the abnormality diagnosis model, and divide the image data corresponding to the sensor in the initial image dataset J into image data groups corresponding to each different process in the industrial production line, labeled as j. m (j1,j2,…,j m ), first realize the image data group j m The traversal operation;

[0025] C2, through the image data set j m (j1,j2,…,j m ) performs a feature comparison operation with the result image dataset K in the abnormality diagnosis model;

[0026] C3. If there is a difference in the feature data of the corresponding image data, the currently detected image data is abnormal and a mark extraction is performed. Conversely, if the feature data of the corresponding image data remains consistent, the currently detected image data is not abnormal.

[0027] Preferably, the C1 pairs of image data sets j m The traversal operation is:

[0028] c11, realize the image data group j m The number of inspection operations is performed, and the image data group j is divided according to the actual number of sensors. m The number of sensors is compared with the number of actual sensors. If the two numbers are consistent, the image data set j m The number of is normal, if the two numbers are inconsistent, then the image data set j m Some image data is missing, and some sensor devices have recognition problems;

[0029] c12. Realize image data set j mThe product features are checked in the image data, and the image data is identified in turn. If there are product features that do not exist in the image data, the device that generated the image data is directly traced back to determine that there is a problem with the production line equipment here.

[0030] Preferably, the specific steps of the feature comparison operation in C2 are:

[0031] c21. Select image data group j m (j1,j2,…,j m ) in which the last image data in a single process is used as the image to be compared;

[0032] c22, then performing pixel conversion on the selected image data to form the same size as the image data corresponding to the result image dataset K;

[0033] Grayscale the selected image data and the result image data to achieve segmentation of product features and background features. Reference points P1 and P2 are set for the product features of the selected image data, and reference points Q1 and Q2 are set at the same positions corresponding to the product features of the result image data.

[0034] According to the direction of the background features, the corner points of the two image data are aligned to achieve the overlay operation of the image data, and the positions of the corresponding reference points on the two image data are compared;

[0035] c23. Extract and mark the parts where the feature data are different in the comparison.

[0036] Preferably, the operation of comparing the positions of corresponding reference points on the two image data in c22 is:

[0037] D1. Establish an XY coordinate axis based on the lower left corner of the image data as the starting point. Establish the X axis from the starting point toward the bottom edge of the image data. Establish the Y axis from the starting point toward the side edge of the image data. The X axis and Y axis intersect at right angles at the starting point.

[0038] D2. Based on the coordinate axis, the coordinates of reference points P1 and P2 in the image data are (x1, y1) and (x2, y2), respectively, while the coordinates of reference points Q1 and Q2 are (x3, y3) and (x4, y4), respectively.

[0039] D3. Calculate the deviation between reference point P1 and reference point P2 and reference point Q2. If the value is equal to 0, the product feature is normal, and the corresponding equipment and sensor are also normal. Otherwise, if the value is not equal to 0, the product feature position is abnormal, and the corresponding equipment is abnormal.

[0040] Preferably, the formula for calculating the deviation value in D3 is:

[0041] F=∑[|x1-x3|+|y1-y3|+|x2-x4|+|y2-y4|];

[0042] F represents the calculated deviation value, |x1-x3| is the absolute value of the deviation between reference point P1 and reference point Q1 on the X-axis, |y1-y3| is the absolute value of the deviation between reference point P1 and reference point Q1 on the Y-axis, |x2-x4| is the absolute value of the deviation between reference point P2 and reference point Q2 on the X-axis, and |y2-y4| is the absolute value of the deviation between reference point P2 and reference point Q2 on the Y-axis;

[0043] If F=0, the corresponding device and sensor are normal; if F>0, the corresponding device is abnormal.

[0044] Preferably, the abnormality diagnosis module performs the following steps to trace and locate a specific problem:

[0045] E1. After discovering an abnormality, it is diagnosed that the fault is caused by production line equipment or sensor;

[0046] E2. Based on the sorting and positioning of the image data, trace the location of the corresponding device or sensor where the abnormal image data appears, and perform maintenance and repair operations by personnel;

[0047] The early warning transmission module will use the corresponding device or sensor location found as the center point and send out a signal. The signal will be sent first to the person holding the device terminal closest to the center point, informing the person to provide timely processing instructions.

[0048] The present invention provides an integrated service platform for industrial doctors. Compared with the existing technology, it has the following advantages:

[0049] (1) The Industrial Doctor Enterprise Comprehensive Service Platform establishes an abnormality diagnosis model and introduces data analysis to determine whether there are abnormalities in the data of each sensor or production line equipment on the production line. It then traces and locates the specific problem based on the abnormality, generates diagnostic results for data transmission, and realizes real-time monitoring of each process while completing rapid fault diagnosis operations, reducing downtime and further improving production efficiency. It also uses the comprehensive service platform to complete unified instruction management and automated abnormality diagnosis, and realizes traceability and location operations for faulty equipment.

[0050] (2) The industrial doctor enterprise comprehensive service platform introduces the classified initial image data set into the abnormality diagnosis model, and divides the image data corresponding to the sensor in the initial image data set to form an image data group corresponding to each different process in the industrial production line, and realizes the traversal operation of the image data group, so as to know in advance the number of image data groups corresponding to the number of processes, thereby identifying the abnormality of the sensor, and at the same time checking whether there are product features in the image data to determine the problem of the production line equipment, avoiding subsequent losses caused by not knowing the problem in time after it occurs, and avoiding the problem of long-term downtime for maintenance.

[0051] (3) The industrial doctor enterprise comprehensive service platform realizes the segmentation operation of product features and background features by gray-scaling the selected image data and the result image data, sets reference points for the product features of the selected image data and the result image data, and aligns the corner points of the two image data according to the direction of the background features to realize the overlay operation of the image data, and compares the positions of the corresponding reference points on the two image data. In this way, the abnormality of the equipment can be diagnosed based on the abnormality of the product features, and the nearest person holding the equipment terminal can be informed by tracing the source, completing the processing operation quickly and effectively, and improving the efficiency of service management. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] Figure 1 This is a principle block diagram of the integrated service platform of the present invention;

[0053] Figure 2 This is an operational flow chart of the data preprocessing and storage module of the present invention;

[0054] Figure 3 An operational flow chart for establishing the abnormality diagnosis model of the present invention;

[0055] Figure 4 This is a flowchart of the operation of the abnormality diagnosis module of the present invention to determine abnormalities. DETAILED DESCRIPTION

[0056] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0057] See also Figure 1-Figure 4 , the present invention provides two technical solutions:

[0058] Example 1: Industrial Doctor Enterprise Comprehensive Service Platform, including:

[0059] The data acquisition module uses multiple sensors and combines IoT transmission to realize real-time collection of operating status data of industrial production line equipment;

[0060] The sensors are used to collect the operating data of the industrial production line and control it through video. The video data parameters collected by the sensors are kept consistent, and multiple sensors are set up directly above the transmission routes of each process.

[0061] The data preprocessing and storage module receives and stores the collected data in the database, and simultaneously implements data classification and screening operations to form the required data set;

[0062] The database can not only store newly collected data, but also compress and store historical data to facilitate subsequent comparison and data traceability;

[0063] The abnormality diagnosis module establishes an abnormality diagnosis model and introduces data analysis to determine whether there are abnormalities in the data of each sensor or production line equipment on the production line. It then traces and locates the specific problem based on the abnormality and generates diagnostic results for data transmission.

[0064] The early warning transmission module sends abnormal instructions to the nearest device terminal held by personnel and marks the abnormal location with a flashing red light;

[0065] The user visualization and touch-screen management interface provides users with intuitive data display in the form of charts and text, and transmits instructions according to on-site conditions for integrated management and control operations.

[0066] Among them, by establishing an abnormal diagnosis model and introducing data analysis to determine whether there are abnormalities in the data of each sensor or production line equipment on the production line, and tracing and locating specific problems based on the abnormalities, generating diagnostic results for data transmission, and realizing real-time monitoring of each process, while completing rapid diagnosis operations of faults, reducing downtime, and further improving production efficiency, and using the integrated service platform to complete unified instruction management and automated abnormal diagnosis, realizing traceability and positioning operations of faulty equipment.

[0067] In the embodiment of the present invention, the operations for implementing data classification and screening in the data preprocessing and storage module are:

[0068] A1. Frame the collected video data to form the required image data;

[0069] A2. Set the required data template, extract the corresponding image category name from the data template, and classify the image data name according to the corresponding image category name;

[0070] A3. Classify the image data corresponding to the sensors in the industrial production line process sequence to form an initial image dataset labeled J.

[0071] In the embodiment of the present invention, the frame processing operation of the video data collected in A1 is as follows:

[0072] a11. Take a segment of video data and set the processing frame period according to the production line's operating rate. Then, segment the video data from the beginning based on the frame period. Images separated by corresponding frame periods form sequential image data of the video data.

[0073] a12. Compare the sequential image data. If the adjacent image data in the sequential image data are the same, remove the preceding image data and retain the subsequent image data. If the adjacent image data in the sequential image data are different, retain both the preceding and subsequent image data.

[0074] In the embodiment of the present invention, the operation of establishing the abnormality diagnosis model is as follows:

[0075] B1. Based on actual equipment operation data, a simulation model similar to the actual industrial production line equipment is introduced and constructed;

[0076] B2. Extract historical data of industrial production line equipment from the database based on product image data at the end of each process, and aggregate them into a result image dataset labeled K;

[0077] B3. Introduce the result image dataset K into the simulation model for data combination to establish the required abnormality diagnosis model.

[0078] In the embodiment of the present invention, the abnormality diagnosis module analyzes the data of each sensor or production line equipment on the production line to see whether there is an abnormality. The steps are as follows:

[0079] C1. Introduce the classified initial image dataset J into the abnormality diagnosis model, and divide the image data corresponding to the sensor in the initial image dataset J into image data groups corresponding to each different process in the industrial production line, labeled as j. m (j1,j2,…,j m ), first implement the image data set j m The traversal operation;

[0080] C2, through the image data set j m (j1,j2,…,j m ) performs a feature comparison operation with the result image dataset K in the abnormality diagnosis model;

[0081] C3. If there is a difference in the feature data of the corresponding image data, the currently detected image data is abnormal and a mark extraction is performed. Conversely, if the feature data of the corresponding image data remains consistent, the currently detected image data is not abnormal.

[0082] In the embodiment of the present invention, C1 is used to compare the image data set j m The traversal operation is:

[0083] c11, realize the image data group j m The number of inspection operations is performed, and the image data group j is divided according to the actual number of sensors. m The number of sensors is compared with the number of actual sensors. If the two numbers are consistent, the image data set j m The number of is normal, if the two numbers are inconsistent, then the image data set j m Some image data is missing, and some sensor devices have recognition problems;

[0084] c12. Realize image data set j m The product features are checked in the image data, and the image data is identified in turn. If there are product features that do not exist in the image data, the device that generated the image data is directly traced back to determine that there is a problem with the production line equipment here.

[0085] Among them, by introducing the classified initial image data set into the abnormality diagnosis model, and dividing the image data corresponding to the sensor in the initial image data set, an image data group corresponding to each different process in the industrial production line is formed, and a traversal operation of the image data group is realized, so as to know in advance the number of image data groups in the corresponding number of processes, thereby identifying sensor abnormalities, and at the same time checking whether there are product features in the image data to determine problems with production line equipment, avoiding subsequent losses caused by not knowing the problem in time after it occurs, and avoiding the problem of long-term downtime for maintenance.

[0086] In the embodiment of the present invention, the specific steps of the feature comparison operation in C2 are:

[0087] c21. Select image data group j m (j1,j2,…,j m ) in which the last image data in a single process is used as the image to be compared;

[0088] c22, then performing pixel conversion on the selected image data to form the same size as the image data corresponding to the result image dataset K;

[0089] Grayscale the selected image data and the result image data to achieve segmentation of product features and background features. Reference points P1 and P2 are set for the product features of the selected image data, and reference points Q1 and Q2 are set at the same positions corresponding to the product features of the result image data.

[0090] According to the direction of the background features, the corner points of the two image data are aligned to achieve the overlay operation of the image data, and the positions of the corresponding reference points on the two image data are compared;

[0091] c23. Extract and mark the parts where the feature data are different in the comparison.

[0092] In the embodiment of the present invention, the operation of comparing the positions of corresponding reference points on the two image data in c22 is:

[0093] D1. Establish an XY coordinate axis based on the lower left corner of the image data as the starting point. Establish the X axis from the starting point toward the bottom edge of the image data. Establish the Y axis from the starting point toward the side edge of the image data. The X axis and Y axis intersect at right angles at the starting point.

[0094] D2. Based on the coordinate axis, the coordinates of reference points P1 and P2 in the image data are (x1, y1) and (x2, y2), respectively, while the coordinates of reference points Q1 and Q2 are (x3, y3) and (x4, y4), respectively.

[0095] D3. Calculate the deviation between reference point P1 and reference point P2 and reference point Q2. If the value is equal to 0, the product feature is normal, and the corresponding equipment and sensor are also normal. Otherwise, if the value is not equal to 0, the product feature position is abnormal, and the corresponding equipment is abnormal.

[0096] In the embodiment of the present invention, the formula for calculating the deviation value in D3 is:

[0097] F=∑[|x1-x3|+|y1-y3|+|x2-x4|+|y2-y4|];

[0098] F represents the calculated deviation value, |x1-x3| is the absolute value of the deviation between reference point P1 and reference point Q1 on the X-axis, |y1-y3| is the absolute value of the deviation between reference point P1 and reference point Q1 on the Y-axis, |x2-x4| is the absolute value of the deviation between reference point P2 and reference point Q2 on the X-axis, and |y2-y4| is the absolute value of the deviation between reference point P2 and reference point Q2 on the Y-axis;

[0099] If F=0, the corresponding device and sensor are normal; if F>0, the corresponding device is abnormal.

[0100] In the embodiment of the present invention, the abnormality diagnosis module performs the following steps to trace and locate a specific problem:

[0101] E1. After discovering an abnormality, it is diagnosed that the fault is caused by production line equipment or sensor;

[0102] E2. Based on the sorting and positioning of the image data, trace the location of the corresponding device or sensor where the abnormal image data appears, and perform maintenance and repair operations by personnel;

[0103] The early warning transmission module will use the corresponding device or sensor location found as the center point and send out a signal. The signal will be sent first to the person holding the device terminal closest to the center point, informing the person to provide timely processing instructions.

[0104] Among them, the segmentation operation of product features and background features is realized by grayscale processing of the selected image data and the result image data, and reference points are set for the product features of the selected image data and the result image data. According to the direction of the background features, the corner points of the two image data are aligned to realize the overlay operation of the image data, and the positions of the corresponding reference points on the two image data are compared. In this way, the equipment abnormality is diagnosed based on the abnormality of the product features, and the source can be traced to inform the nearest person holding the equipment terminal, so as to complete the fast and effective processing operation.

[0105] Example 2: The difference from Example 1 is that the present invention also designs a comparative experiment, in which the existing industrial production line management and the industrial doctor enterprise comprehensive service platform of the present invention are applied to multiple production lines of a certain industry, and the number of problems generated by industrial equipment during the cycle time monitoring and management process, as well as the time for problem handling and production line downtime are recorded. The specific results are shown in Table 1:

[0106] Table 1 Result record table

[0107]

[0108] Comprehensive service platform

[0109] To sum up, when the Industrial Doctor Enterprise Comprehensive Service Platform of the present invention is applied, more problems can be discovered during the monitoring process within the cycle time, and the average time required to solve the problems is shorter, and the production line downtime required is shorter, so it can be better applied to actual operations.

[0110] Meanwhile, the contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0111] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0112] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. The Industrial Doctor Enterprise Comprehensive Service Platform is characterized by: include: The data acquisition module uses multiple sensors and combines IoT transmission to realize real-time collection of operating status data of industrial production line equipment; The data preprocessing and storage module receives and stores the collected data in the database, and simultaneously implements the classification and screening operations of the data to form the required initial image data set, which is marked as J; The abnormality diagnosis module establishes an abnormality diagnosis model and introduces data analysis to determine whether there are abnormalities in the data of each sensor or production line equipment on the production line. It then traces and locates the specific problem based on the abnormality and generates diagnostic results for data transmission. The early warning transmission module sends abnormal instructions to the nearest device terminal held by personnel and marks the abnormal location with a flashing red light; User visualization and touch management interface provide users with intuitive data display, transmit commands according to on-site conditions, and perform integrated management and control operations; The establishment operation of the abnormal diagnosis model is: B1. Based on actual equipment operation data, a simulation model similar to the actual industrial production line equipment is introduced and constructed; B2. Extract historical data of industrial production line equipment from the database based on product image data at the end of each process, and aggregate them into a result image dataset labeled K; B3. Introducing the result image dataset K into the simulation model for data integration to establish the required abnormality diagnosis model; The steps of the abnormality diagnosis module to analyze whether there are abnormalities in the data of each sensor or production line equipment on the production line are as follows: C1. Introduce the initial image dataset J into the abnormality diagnosis model and divide the image data corresponding to the sensor in the initial image dataset J into image datasets corresponding to each different process in the industrial production line, labeled as j. m (j1, j2, ..., j m ), first realize the image data group j m The traversal operation; C2, through the image data set j m (j1, j2, ..., j m ) Perform feature comparison operation with the result image dataset K in the abnormal diagnosis model; C3. If the feature data in the corresponding image data are different, the currently detected image data is abnormal and a mark extraction is performed. Conversely, if the feature data in the corresponding image data are consistent, the currently detected image data is not abnormal; The specific steps of the feature comparison operation in C2 are: c21. Select image data group j m (j1, j2, ..., j m ) in which the last image data in a single process is used as the image to be compared; c22, then performing pixel conversion on the selected image data to form the same size as the image data corresponding to the result image dataset K; Grayscale the selected image data and the result image data to achieve segmentation of product features and background features. Reference points P1 and P2 are set for the product features of the selected image data, and reference points Q1 and Q2 are set at the same positions corresponding to the product features of the result image data. According to the direction of the background features, the corner points of the two image data are aligned to achieve the overlay operation of the image data, and the positions of the corresponding reference points on the two image data are compared; c23, and extract and mark the parts where there are differences in the feature data during comparison; The operation of comparing the positions of the corresponding reference points on the two image data in c22 is: D1. Establish an XY coordinate axis based on the lower left corner of the image data as the starting point. Establish the X axis from the starting point toward the bottom edge of the image data. Establish the Y axis from the starting point toward the side edge of the image data. The X axis and Y axis intersect at right angles at the starting point. D2. Based on the coordinate axis, the coordinates of the reference points P1 and P2 in the image data are (x1, y1) and (x2, y2), respectively, while the coordinates of the reference points Q1 and Q2 are (x3, y3) and (x4, y4), respectively. D3. Calculate the deviation between reference point P1 and reference point P2 and reference point Q2. If the value is equal to 0, the product feature is normal, and the corresponding equipment and sensor are also normal. Otherwise, if the value is not equal to 0, the product feature position is abnormal, and the corresponding equipment is abnormal.

2. The industrial doctor enterprise comprehensive service platform according to claim 1 is characterized by: The operations for implementing data classification and screening in the data preprocessing and storage module are: A1. Frame the collected video data to form the required image data; A2. Set the required data template, extract the corresponding image category name from the data template, and classify the image data name according to the corresponding image category name; A3. Classify the image data corresponding to the sensors in the industrial production line process sequence to form an initial image dataset J.

3. The industrial doctor enterprise comprehensive service platform according to claim 2 is characterized by: The framing operation for the video data collected in A1 is as follows: a11. Take a segment of video data and set the processing frame period according to the production line's operating rate. Then, segment the video data from the beginning based on the frame period. Images separated by corresponding frame periods form sequential image data of the video data. a12. Compare the sequential image data. If the adjacent image data in the sequential image data are the same, remove the preceding image data and retain the subsequent image data. If the adjacent image data in the sequential image data are different, retain both the preceding and subsequent image data.

4. The industrial doctor enterprise comprehensive service platform according to claim 1 is characterized by: The image data set j in C1 m The traversal operation is: c11, realize the image data group j m The number of inspection operations is performed, and the image data group j is divided according to the actual number of sensors. m The number of sensors is compared with the number of actual sensors. If the two numbers are consistent, the image data set j m The number of is normal. If the two numbers are inconsistent, some image data are missing in the image data group jm, and there are problems with the recognition of some sensor devices; c12. Realize image data set j m The product features are checked in the image data, and the image data is identified in turn. If there are product features that do not exist in the image data, the device that generated the image data is directly traced back to determine that there is a problem with the production line equipment here.

5. The industrial doctor enterprise comprehensive service platform according to claim 1 is characterized by: The formula for calculating the deviation value in D3 is: ； F represents the calculated deviation value, |x1-x3| is the absolute value of the deviation between reference point P1 and reference point Q1 on the X-axis, |y1-y3| is the absolute value of the deviation between reference point P1 and reference point Q1 on the Y-axis, |x2-x4| is the absolute value of the deviation between reference point P2 and reference point Q2 on the X-axis, and |y2-y4| is the absolute value of the deviation between reference point P2 and reference point Q2 on the Y-axis; If F=0, the corresponding device and sensor are normal; if F>0, the corresponding device is abnormal.

6. The industrial doctor enterprise comprehensive service platform according to claim 1 is characterized by: The abnormality diagnosis module performs the following steps to trace and locate specific problems: E1. After discovering an abnormality, it is diagnosed that the fault is caused by production line equipment or sensor; E2. Based on the sorting and positioning of the image data, trace the location of the corresponding device or sensor where the abnormal image data appears, and perform maintenance and repair operations by personnel; The early warning transmission module will use the corresponding device or sensor location found as the center point and send out a signal. The signal will be sent first to the person holding the device terminal closest to the center point, informing the person to provide timely processing instructions.

Images