Machine learning based industrial production area digitization data transmission system and method

Abstract

The application discloses a machine learning-based industrial production area digital data transmission system and method, relates to the technical field of data transmission, and comprises an industrial data extraction module, a target data screening module, a priority analysis module, a data transmission feature link construction module and a data pushing module; the industrial data extraction module is used for extracting industrial data recorded by an industrial production area for different demand areas; the target data screening module is used for screening target data associated with the demand of the industrial production area; the priority analysis module analyzes the target data priority of each type of industrial production area corresponding management personnel based on operation data; the data transmission feature link construction module is used for extracting all operation data recording core data, and constructing a data transmission feature link; the data pushing module judges the management authority of real-time personnel when the system responds to real-time personnel accessing industrial data, and pushes all industrial data related to the target data based on the management authority.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data transmission, in particular to an industrial production area digitization data transmission system and method based on machine learning. BACKGROUND

[0002] In the field of industrial production area digitization, digital twin technology can establish a virtual model for production equipment, production lines, or even entire factories. By collecting data in real-time during the actual production process, the virtual model is driven to run synchronously with the actual production system, enabling real-time monitoring, prediction, and optimization of the production process. Engineers can conduct various tests and optimizations on the virtual model, identifying and resolving production issues in advance, thereby reducing production costs and risks. Different industrial production area managers have different management permissions, and the key core data required in the vast amount of industrial data also has deviations. Therefore, it is worth analyzing the data based on the virtual model combined with data events, and when there are temporary permission access personnel, how to quickly and effectively transmit the required data of temporary permission personnel based on the existing access data transmission footprint for searching. SUMMARY

[0003] The present application aims to provide an industrial production area digitization data transmission system and method based on machine learning to solve the problems in the prior art.

[0004] To achieve the above-mentioned purpose, the present application provides the following technical solution: an industrial production area digitization data transmission method based on machine learning, the method comprising the following specific steps:

[0005] Step S100: Extract industrial data recorded by the industrial production area for different demand areas, where the demand area refers to the area corresponding to the production demand performed in the industrial production process; the industrial data record corresponds to the management personnel given data access permission in the demand area and the operation data performed by the management personnel based on the system; based on the operation data, filter the target data associated with the industrial production area;

[0006] Step S200: Analyze the target data priority of each type of industrial production area corresponding management personnel based on operation data;

[0007] Step S300: Take the first target data in the target data priority as the core data of the corresponding industrial production area, extract all operation data recording the core data, and construct a data transmission feature link;

[0008] Step S400: When the system responds to real-time personnel accessing industrial data, determine the management permission of the real-time personnel, and based on the management permission, transmit and push all industrial data related to the target data.

[0009] Further, the screening of the target data associated with the demand of the industrial production area includes the following specific steps:

[0010] Step S110: The operation data is stored according to the operation link in the industrial production process, and is divided into input data and output data; the input data is the data required by the current operation link, and the output data is the data output by the current operation link;

[0011] Step S120: The input data and the output data of each operation link form a data pair, the input data and the output data can include multiple types of data, and each type of data in the input data and each type of data in the output data generate a one-to-one corresponding association structure;

[0012] Step S130: The corresponding output data response abnormality association structure when the input data is abnormal in the historical record is the key association structure, the input data is abnormal refers to the difference between the input data before and after modification is greater than the difference threshold, and the output data response abnormality refers to the record data returns to the operation to reacquire the output process;

[0013] The analysis of the key association structure is to judge whether the input data in the corresponding operation link plays a role in influencing the link data;

[0014] Step S140: Traverse the association structure of each operation link, capture the same type of data recorded in different association structures, and the operation data belonging to the output data and the input data in the corresponding association structure is the first target data; compare the input data in the key association structure with the first target data in all association structures, match the intersection association structure as a group of input and output data, and form the target data associated with the demand of the industrial production area.

[0015] Further, the analysis of the target data priority of each type of industrial production area based on the operation data of the corresponding management personnel includes the following:

[0016] Step S210: Obtain the adjustment event recorded in the target data, the adjustment event refers to the event of returning to the determined output data due to the abnormal input data; extract the deviation value N of the input data adjustment and the deviation value M of the adjacent output data after adjustment, and the number of adjustments K; use the formula: Q=[(N / M)max-(N / M)min] / K, to calculate the adjustment intensity index Q of the target data, wherein (N / M)max represents the maximum value of N / M in the same adjustment event, and (N / M)min represents the minimum value of N / M in the same adjustment event;

[0017] Step S220: judging whether the output data in the target data belongs to the first target data, when belonging to the first target data, obtaining the response number U of other abnormal operation links in the adjustment event record of the output data in the target data, the abnormal operation link refers to an operation link that cannot respond to operation by using the output data as input data; calculating the fluctuation index Z of the output data in the target data, Z=U / V, wherein V represents the number of all operation links recorded in the same industrial production area; when not belonging to the first target data, outputting Z=0;

[0018] Step S230: calculating the preferred index F of each target data, F=a1*Q+a2*Z; wherein a1 and a2 represent corresponding reference coefficients; sorting each target data in order from small to large according to the numerical value of the preferred index F, to generate the priority of the target data.

[0019] The greater the preferred index is, the greater the data influence of the corresponding target data in the corresponding industrial production area is, and the greater the demand of user access is.

[0020] Further, step S300 includes the following specific steps:

[0021] The all operation data of the core data refers to the data recorded by the operation link containing the core data; taking the operation link corresponding to the core data as a root node, dividing a forward child node and a backward child node, the forward child node refers to the operation link experienced by the generation and transmission of the core data, the backward child node refers to the operation link experienced by the application and transmission of the core data, when the adjacent two nodes are the same, marking the data type to distinguish the different nodes;

[0022] Based on the root node, the forward child node and the backward child node, the nodes are linked by using the data transmission direction; generating the first data link of the industrial production area where the core data is located;

[0023] Obtaining the first data link in all industrial production areas, finding other industrial production areas having a data transmission relationship with the current analysis first data link as target production areas, and adding target nodes to the first data link in the data transmission direction, the target nodes being other target production areas having a data transmission relationship;

[0024] Recording the updated first data link as the data transmission characteristic link of the core data.

[0025] Further, step S400 includes the following:

[0026] When the real-time personnel has the management authority corresponding to the industrial production area, the core data of the industrial production area record where the management authority is located is acquired and pushed; the core data is pushed because the core data is the key attention data of the current industrial production area, and it is the most necessary for the retrieval requirement, and the abnormality of the core data has the greatest influence on other data of the whole production area, and the improvement of the attention degree can effectively monitor the data abnormality, so as to facilitate the management personnel to quickly find the data;

[0027] When the real-time personnel is a temporary permission access user, the data transmission characteristics of each industrial production area corresponding to the real-time access data type are extracted as a control group, the real-time access data type is taken as real-time feature data, the first data link in the control group is compared with the real-time feature data, the first similarity is calculated, and the first similarity refers to the proportion that the real-time feature data and the first data link record the same feature data type;

[0028] The type of the industrial production area where the real-time access feature data is located is the real-time production area type, the real-time production area type is compared with all the industrial production area types recorded in the control group, the second similarity is calculated, and the second similarity refers to the proportion that the real-time production area type is the same as all the industrial production area types in the control group;

[0029] The first similarity and the second similarity are summed to obtain the control similarity, and the core data recorded by the data transmission characteristics corresponding to the maximum control similarity is extracted and pushed.

[0030] In the present application, all the data that can be transmitted and pushed are the data allowed under the system security protocol, and the data pushing is to improve the transmission efficiency and the efficient practicability of the user.

[0031] The industrial production area digital data transmission system based on machine learning, the system comprises an industrial data extraction module, a target data screening module, a priority analysis module, a data transmission feature link construction module and a data pushing module;

[0032] The industrial data extraction module is used for extracting industrial data of different demand areas recorded by the industrial production area;

[0033] The target data screening module is used for screening target data associated with the industrial production area;

[0034] The priority analysis module is used for analyzing the target data priority of each type of industrial production area corresponding to the management personnel based on the operation data;

[0035] The data transmission feature link construction module is used for extracting all operation data recording the core data, and constructing the data transmission feature link;

[0036] The data pushing module is configured to determine the management authority of the real-time personnel when the system responds to real-time personnel accessing industrial data, and transmit and push all industrial data related to the target data based on the management authority.

[0037] Further, the target data screening module includes a data division unit, an associated structure generation unit, and a target data output unit.

[0038] The data division unit is configured to store the operation data according to the operation links in the industrial production process, and divide the operation data into input data and output data.

[0039] The associated structure generation unit is configured to form a data pair of the input data and the output data for each operation link, each type of data in the input data and each type of data in the output data can form a one-to-one associated structure.

[0040] The target data output unit is configured to traverse the associated structure of each operation link, capture the same type of data recorded in different associated structures, and the operation data belonging to the output data and the input data in the corresponding associated structure as the first target data; compare the input data in the key associated structure with the first target data with all associated structures, match the intersection associated structure as a group of input and output data, and form the target data associated with the demand of the industrial production area.

[0041] Further, the priority analysis module includes an adjustment intensity index calculation unit, a fluctuation index calculation unit, and a preferred index calculation unit.

[0042] The adjustment intensity index calculation unit is configured to calculate the adjustment intensity index of the target data.

[0043] The fluctuation index calculation unit is configured to calculate the fluctuation index of the output data in the target data.

[0044] The preferred index calculation unit is configured to calculate the preferred index of each target data, and sort each target data in order from small to large according to the numerical value of the preferred index F, and generate the target data priority.

[0045] Further, the data transmission feature link construction module includes a first data link generation unit and a link update unit.

[0046] The first data link generation unit is configured to link each node by using the data transmission direction based on the root node, the forward child node, and the backward child node; and generate the first data link of the industrial production area where the core data is located.

[0047] The link update unit is used to find other industrial production areas that have data transmission relationships with the currently analyzed first data link as target production areas, and add target nodes to the first data link in the direction of data transmission, and record the updated first data link as the data transmission feature link of the core data.

[0048] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention rationally plans data categories based on industrial production areas within a complex industrial production data system. Furthermore, after granting access permissions to a specific type of manager, the system can quickly push the highest-priority industrial data to them, preventing managers from being unable to promptly and effectively select the target data from a large amount of industrial data. Secondly, this application also implements data transmission path prediction analysis for temporary access users, determining data transmission characteristics and enabling personalized feedback transmission when these users operate on large amounts of industrial data. This improves the system's effective feedback and intelligence when temporary access users request data. It avoids data security issues caused by temporary access users' unfamiliarity with the system's industrial data, as well as problems such as untimely data acquisition, complex transmission paths, and channel occupancy. Attached Figure Description

[0049] Figure 1 This is a schematic diagram of the structure of the machine learning-based digital data transmission method for industrial production areas according to the present invention. Detailed Implementation

[0050] Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0051] Example: Figure 1 As shown, this invention provides a technical solution for a machine learning-based digital data transmission system and method for industrial production areas. The machine learning-based digital data transmission method for industrial production areas includes the following specific steps:

[0052] Step S100: Extract industrial data from different demand areas of the industrial production area record. The demand area refers to the area corresponding to the production demand executed during the industrial production process. The industrial data record records the management personnel who have been granted system data access permissions in the corresponding demand area, as well as the operation data executed by the management personnel based on the system. Based on the operation data, filter the target data that is related to the demand of the industrial production area.

[0053] Step S200: Analyze the target data priorities of the managers corresponding to each type of industrial production area based on operational data;

[0054] Step S300: The first target data in the target data priority is the core data corresponding to the industrial production area, all operation data recording the core data is extracted, and a data transmission feature link is constructed;

[0055] Step S400: When the system responds to real-time personnel accessing industrial data, the management authority of the real-time personnel is judged, and all industrial data related to the target data is transmitted and pushed based on the management authority.

[0056] The target data associated with the demand of the industrial production area includes the following specific steps:

[0057] Step S110: The operation data is stored according to the operation link in the industrial production process, and is divided into input data and output data; the input data is the data required by the current operation link, and the output data is the data output by the current operation link;

[0058] Step S120: The input data and output data of each operation link form a data pair, the input data and the output data can contain multiple types of data, and each type of data in the input data and each type of data in the output data generate a one-to-one corresponding association structure;

[0059] Step S130: The association structure in which the corresponding output data responds to the abnormality when the input data is abnormal is the key association structure, the input data is abnormal refers to the difference between the input data before and after modification is greater than the difference threshold, and the output data responds to the abnormality refers to the process of returning the operation to reacquire the output; when there is a single corresponding relationship between multiple input and output, the influence is not great;

[0060] The analysis of the key association structure is to judge whether the input data in the corresponding operation link plays a role in the influence link data;

[0061] Step S140: Traverse the association structure of each operation link, capture the same type of data recorded in different association structures, and the operation data belonging to the output data and the input data in the corresponding association structure is the first target data; compare the input data in the key association structure with the first target data in all association structures, match the intersection association structure as a group of input and output data, and form the target data associated with the demand of the industrial production area.

[0062] As shown in the embodiment: the quality supervision area, recording operation link 1 and operation link 2;

[0063] Operation link 1 acquires input data a and output data b, and generates output data c;

[0064] Operation link 2 acquires input data c to generate output data d;

[0065] The association structure that can be constituted in the operation link 1 is ac and bc.

[0066] The association structure that can be constituted in the operation link 2 is cd.

[0067] From the above, it can be seen that in the quality supervision area, c is the first target data, and the association structure ac is the key association structure. Therefore, the input data in the key association structure is compared with the first target data ac, and all the association structures {ac, bc, cd} are matched to obtain the intersection association structure as a set of input and output data, that is, ac∩{ac, bc, cd}=ac. Therefore, the output ac is the target data that exists in the industrial production area and needs to be associated.

[0068] The target data priority of each type of industrial production area based on the operation data of the management personnel includes the following:

[0069] Step S210: Obtain the adjustment event corresponding to the recorded target data. The adjustment event refers to an event in which the input data is abnormally returned to the determined output data. Extract the deviation value N of the input data adjustment and the deviation value M of the adjacent output data after adjustment, as well as the number of adjustments K in the adjustment event. Calculate the adjustment intensity index Q of the target data using the formula: Q=[(N / M)max-(N / M)min] / K, wherein (N / M)max represents the maximum value of N / M in all adjustments recorded in the same adjustment event, and (N / M)min represents the minimum value of N / M in all adjustments recorded in the same adjustment event.

[0070] Step S220: Determine whether the output data in the target data belongs to the first target data. When it belongs to the first target data, obtain the response number U of other abnormal operation links under the adjustment event record of the output data in the target data. The abnormal operation link refers to an operation link that cannot respond to the operation when the output data is used as the input data. Calculate the fluctuation index Z of the output data in the target data, Z=U / V, wherein V represents the number of all operation links recorded in the same industrial production area. When it does not belong to the first target data, output Z=0.

[0071] Step S230: Calculate the preferred index F of each target data, F=a1*Q+a2*Z; wherein a1 and a2 represent the corresponding reference coefficients. Sort each target data in order from small to large according to the numerical value of the preferred index F to generate the target data priority.

[0072] The larger the preferred index is, the greater the data influence of the corresponding target data in the corresponding industrial production area, and the greater the demand of the user access.

[0073] Step S300 includes the following specific steps:

[0074] All operation data recording core data refers to data recording operation links containing core data; taking the operation link corresponding to the core data as the root node, dividing the forward child node and the backward child node, the forward child node refers to the operation link experienced by the generation and transmission of the core data, the backward child node refers to the operation link experienced by the application and transmission of the core data, when the adjacent two nodes are the same, mark the data type to distinguish the different nodes;

[0075] Based on the root node, the forward child node and the backward child node, link each node by using the data transmission direction; generate the first data link of the industrial production area where the core data is located;

[0076] Obtain the first data link in all industrial production areas, find other industrial production areas having data transmission relationship with the current analysis first data link as target production area, and add target nodes to the first data link in data transmission direction, the target node is other target production area having data transmission relationship;

[0077] Record the updated first data link as the data transmission characteristic link of the core data.

[0078] As shown in the embodiment: when analyzing the industrial production area as the quality monitoring area, the core data is recorded as the yield rate, which is taken as the core data;

[0079] For the quality monitoring area, the forward child node includes the quantity extraction link corresponding to the quantity of test products, and the quality detection link corresponding to the quality detection result of test products; the backward child node includes the screening link and the verification link; then the above links constitute the first data link of the core data;

[0080] At the same time, there is also an analysis of the yield rate in other industrial production areas such as packaging area, so the first data link needs to add the packaging area after updating.

[0081] Step S400 includes the following:

[0082] When real-time personnel have management authority corresponding to the industrial production area, obtain the core data recorded in the industrial production area where the management authority is located and push it; the core data is pushed because it is the key attention data of the current industrial production area, and it is the most necessary for retrieval demand, and the abnormality of the core data has the greatest influence on other data of the whole production area, so the improvement of its attention degree can effectively monitor the data anomaly, which is convenient for the management personnel to quickly find the data;

[0083] When the real-time personnel is a temporary access user, the data transmission characteristics corresponding to each industrial production area of the real-time access data type are extracted as a control group, the real-time access data type is taken as real-time feature data, the real-time feature data is compared with the first data link in the control group, a first similarity is calculated, the first similarity refers to the proportion that the real-time feature data and the first data link record the same feature data type;

[0084] The type of the industrial production area where the real-time access feature data is located is taken as a real-time production area type, the real-time production area type is compared with all the industrial production area types recorded in the control group, a second similarity is calculated, the second similarity refers to the proportion that the real-time production area type is the same as all the industrial production area types in the control group;

[0085] The first similarity and the second similarity are summed to obtain a control similarity, and the core data recorded by the data transmission characteristics corresponding to the maximum control similarity is pushed.

[0086] In the present application, all the data that can be transmitted and pushed are data allowed under the system security protocol, and the pushing of the data is to improve the transmission efficiency and the efficient practicability of the user.

[0087] The industrial production area digital data transmission system based on machine learning includes an industrial data extraction module, a target data screening module, a priority analysis module, a data transmission feature link construction module and a data pushing module;

[0088] The industrial data extraction module is used to extract industrial data of different demand areas recorded by the industrial production area;

[0089] The target data screening module is used to screen target data associated with the industrial production area;

[0090] The priority analysis module is used to analyze the target data priority of each type of industrial production area corresponding to the management personnel based on operation data;

[0091] The data transmission feature link construction module is used to extract all operation data recording core data and construct a data transmission feature link;

[0092] The data pushing module is used to judge the management authority of the real-time personnel when the system responds to the real-time personnel accessing industrial data, and to transmit and push all industrial data related to the target data based on the management authority.

[0093] The target data screening module includes a data division unit, an association structure generation unit and a target data output unit;

[0094] The data division unit is configured to store operation data according to operation links in an industrial production process, and divide the operation data into input data and output data.

[0095] The association structure generation unit is configured to form a data pair of input data and output data for each operation link, each type of data in the input data and each type of data in the output data can form a one-to-one association structure, and each type of data in the input data and each type of data in the output data form a one-to-one association structure.

[0096] The target data output unit is configured to traverse the association structure of each operation link, capture operation data of the same type recorded in different association structures and belonging to output data and input data in the corresponding association structure as first target data, and compare the input data in the key association structure with the first target data with all association structures, match the intersection association structure as a group of input and output data, and form target data associated with the demand of the industrial production area.

[0097] The priority analysis module includes an adjustment intensity index calculation unit, a fluctuation index calculation unit, and a preferred index calculation unit.

[0098] The adjustment intensity index calculation unit is configured to calculate the adjustment intensity index of the target data.

[0099] The fluctuation index calculation unit is configured to calculate the fluctuation index of the output data in the target data.

[0100] The preferred index calculation unit is configured to calculate the preferred index of each target data, and sort each target data in order from small to large according to the numerical value of the preferred index F, and generate a target data priority.

[0101] The data transmission feature link construction module includes a first data link generation unit and a link update unit.

[0102] The first data link generation unit is configured to link each node based on the root node, the forward child node and the backward child node by using the data transmission direction; and generate a first data link of the industrial production area where the core data is located.

[0103] The link update unit is configured to find other industrial production areas having a data transmission relationship with the first data link as target production areas, and add target nodes to the first data link in the data transmission direction, and record the updated first data link as the data transmission feature link of the core data.

[0104] Finally, it should be noted that the above only describes the preferred embodiments of the present application and is not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art will appreciate that the technical solutions described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalent ones. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for machine learning based digital data transmission in industrial production areas, characterized in that: The method comprises the following specific steps: Step S100: Extract industrial data of different demand areas of the industrial production area, wherein the demand area refers to an area corresponding to a production demand performed in the industrial production process; and the industrial data records a management personnel who is given a system data access permission in the corresponding demand area and operation data performed by the management personnel based on the system; Based on the operation data, target data associated with the demand of the industrial production area is screened; Step S200: Analyze the priority of the target data based on the operation data of each type of management personnel corresponding to the industrial production area; The analysis of the priority of the target data based on the operation data of each type of management personnel corresponding to the industrial production area comprises the following: Step S210: Obtain an adjustment event recorded in the target data, wherein the adjustment event refers to an event of returning to a determined output data due to an abnormal input data; extract a deviation value N of the input data adjustment and a deviation value M of the adjacent output data after the adjustment, and the number K of adjustments; calculate the adjustment intensity index Q of the target data by using the formula: Q=[(N / M)max-(N / M)min] / K, wherein (N / M)max represents the maximum value of N / M in all adjustments recorded in the same adjustment event, and (N / M)min represents the minimum value of N / M in all adjustments recorded in the same adjustment event; Step S220: Determine whether the output data in the target data belongs to first target data, when it belongs to the first target data, obtain the response number U of other abnormal operation links recorded in the adjustment event of the output data in the target data, wherein the abnormal operation link refers to an operation link that cannot respond to the operation by using the output data as the input data; calculate the fluctuation index Z of the output data in the target data, Z=U / V, wherein V represents the number of all operation links recorded in the same industrial production area; when it does not belong to the first target data, output Z=0; Step S230: Calculate the preferred index F of each target data, F=a1*Q+a2*Z; wherein a1 and a2 represent corresponding reference coefficients; sort each target data in the order from small to large according to the numerical value of the preferred index F to generate a target data priority; Step S300: Take the first target data in the target data priority as the core data of the corresponding industrial production area, extract all operation data recorded in the core data, and construct a data transmission feature link; Step S400: When the system responds to real-time personnel accessing industrial data, determine the management permission of the real-time personnel, and transmit and push all industrial data related to the target data based on the management permission.

2. The machine learning based industrial production area digitization data transmission method according to claim 1, characterized in that: The screening of the target data associated with the demand of the industrial production area comprises the following specific steps: Step S110: Store the operation data according to the operation links in the industrial production process, and divide the operation data into input data and output data; the input data is the data required by the current operation link, and the output data is the data output by the current operation link; Step S120: forming a data pair for the input data and the output data of each operation link, the input data and the output data can include multiple types of data, and each type of data in the input data and each type of data in the output data generate a one-to-one corresponding association structure; Step S130: extracting the association structure corresponding to the abnormal response of the output data when the input data is abnormal as the key association structure, the input data being abnormal refers to the difference between the input data before and after modification being greater than a difference threshold, and the output data responding abnormally refers to the process of returning the operation data to reacquire the output; Step S140: traversing the association structure of each operation link, capturing the operation data of the same type of data recorded in different association structures and belonging to the output data and the input data in the corresponding association structure as the first target data; Comparing the input data in the key association structure with the first target data in all association structures, matching the intersection association structure as a group of input and output data, and forming the target data associated with the demand of the industrial production area.

3. The machine learning based industrial production area digitization data transmission method according to claim 1, characterized in that: The step S300 includes the following specific steps: All operation data of the record core data refers to the data recorded by the operation link containing the core data; taking the operation link corresponding to the core data as the root node, dividing the forward child node and the backward child node, the forward child node refers to the operation link experienced by the generation and transmission of the core data, the backward child node refers to the operation link experienced by the application and transmission of the core data, and when the adjacent two nodes are the same, the data type is marked to distinguish the different nodes; Based on the root node, the forward child node and the backward child node, the nodes are linked by using the data transmission direction; Generating the first data link of the industrial production area where the core data is located; Obtaining the first data link in all industrial production areas, finding other industrial production areas having a data transmission relationship with the current analysis first data link as the target production area, and adding target nodes to the first data link in the data transmission direction, the target node is the other target production area having a data transmission relationship; Recording the updated first data link as the data transmission feature link of the core data.

4. The machine learning based industrial production area digitization data transmission method according to claim 3, characterized in that: The step S400 includes the following: When the real-time personnel has the management authority of the corresponding industrial production area, obtaining the core data recorded by the industrial production area where the management authority is located to push; When the real-time personnel is a temporary permission access user, extracting the data transmission features of each industrial production area corresponding to the real-time access data type as a control group, taking the real-time access data type as the real-time feature data, comparing the real-time feature data with the first data link in the control group, calculating the first similarity, the first similarity refers to the proportion of the same feature data type between the real-time feature data and the first data link; Obtaining the industrial production area type where the real-time access feature data is located as the real-time production area type, comparing the real-time production area type with all the industrial production area types recorded in the control group, calculating the second similarity, the second similarity refers to the proportion of the same type between the real-time production area type and all the industrial production area types in the control group; Sum the first similarity and the second similarity to obtain a control similarity, and push the core data corresponding to the data transmission feature recorded when the control similarity is maximum.

5. A machine learning based industrial production area digitized data transmission system using the machine learning based industrial production area digitized data transmission method of any one of claims 1-4, characterized by: The system comprises an industrial data extraction module, a target data screening module, a priority analysis module, a data transmission feature link construction module, and a data pushing module. The industrial data extraction module is configured to extract industrial data recorded by different demand areas in an industrial production area. The target data screening module is configured to screen target data associated with the industrial production area in terms of demand. The priority analysis module is configured to analyze the priority of target data based on operation data of a management personnel corresponding to each type of industrial production area. The data transmission feature link construction module is configured to extract all operation data recording core data and construct a data transmission feature link. The data pushing module is configured to determine the management authority of a real-time personnel when the system responds to real-time personnel accessing industrial data, and push all industrial data related to target data based on the management authority.

6. The machine learning based industrial production area digitization data transmission system according to claim 5, characterized in that: The target data screening module comprises a data division unit, an association structure generation unit, and a target data output unit. The data division unit is configured to store operation data corresponding to operation links in an industrial production process and divide the operation data into input data and output data. The association structure generation unit is configured to form data pairs of input data and output data for each operation link, wherein each type of data in the input data and each type of data in the output data generate a one-to-one association structure. The target data output unit is configured to traverse the association structure of each operation link, capture operation data of the same type recorded in different association structures and belonging to output data and input data in the corresponding association structure as first target data. The input data in the focus association structure and the first target data are compared with all association structures, and the intersection association structure is matched as a set of input and output data to form target data associated with the industrial production area in terms of demand.

7. The machine learning based industrial production area digitization data transmission system according to claim 6, characterized in that: The priority analysis module comprises an adjustment intensity index calculation unit, a fluctuation index calculation unit, and a preferred index calculation unit. The adjustment intensity index calculation unit is configured to calculate the adjustment intensity index of the target data. The fluctuation index calculation unit is configured to calculate the fluctuation index of the output data in the target data. The preferred index calculation unit is configured to calculate the preferred index of each target data and sort each target data in order from small to large according to the numerical value of the preferred index F to generate a target data priority.

8. The machine learning based industrial production area digitization data transmission system of claim 6, wherein: The data transmission feature link construction module comprises a first data link generation unit and a link update unit. The first data link generation unit is configured to link nodes based on root nodes, forward child nodes, and backward child nodes using data transmission directions. The first data link generation unit is configured to link nodes based on root nodes, forward child nodes, and backward child nodes using data transmission directions. The first data link generation unit is configured to link nodes based on root nodes, forward child nodes, and backward child nodes using data transmission directions. The link updating unit is configured to find other industrial production areas having a data transmission relationship with the first data link as target production areas, and add target nodes to the first data link in a data transmission direction, and record the updated first data link as a data transmission characteristic link of the core data.

Images