Industrial internet security monitoring system and method

Abstract

The invention discloses an industrial internet security monitoring system and method, and relates to the field of industrial monitoring management, and the system comprises a management module which is used for connecting a production intranet, carrying out the identity verification, and carrying out the safety monitoring according to the current production period and the operation state of specific production equipment, acquiring the control authority of each image acquisition component and the operation acquisition component, and editing and issuing a control instruction for each functional module; the acquisition unit is used for dividing acquisition targets of operation areas of production equipment according to production targets, setting a moving path scheme of an image acquisition component and an acquisition frequency of the operation acquisition component, and extracting images and operation data acquired in each area according to a preset time period; through accurate data analysis, the accuracy of anomaly detection is greatly improved, more complex state data can be processed, anomaly can be identified, dynamic analysis can be carried out according to historical data, and the monitoring frequency of a target area and the moving path of image acquisition equipment can be dynamically adjusted.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial monitoring and management, and specifically to an industrial Internet security monitoring system and method. Background Art

[0002] Industrial Internet security monitoring aims to ensure the security of equipment, systems and data in the industrial Internet environment. With the widespread application of technologies such as the Internet of Things, big data, and artificial intelligence, the industrial Internet is developing rapidly. This development has promoted the intelligence, connectivity, and automation of equipment, improved production efficiency and flexibility, and further identified, prevented, and responded to security threats, ensuring the safe and stable operation of industrial production.

[0003] Many existing methods lack efficiency in data collection and processing and are unable to respond to abnormal situations in a timely manner. Traditional abnormal monitoring methods often rely on static rules, are weak in adaptability and flexibility, and are difficult to quickly adjust to new production conditions or equipment status changes. When an abnormality occurs, it is difficult for existing methods to adjust the monitoring resource configuration in a timely manner, resulting in insufficient monitoring of potential problems. Previous methods lack the ability to deeply mine and dynamically associate historical data when planning image acquisition paths and monitoring equipment operation data, resulting in unsafe movement of image acquisition equipment under abnormal conditions and low accuracy in the collection of equipment operation data. Summary of the invention

[0004] In view of the above-mentioned shortcomings of the prior art, the present invention provides an industrial Internet security monitoring system and method, which can effectively solve the problems of the prior art.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions.

[0006] The present invention discloses an industrial Internet security monitoring system, comprising: The management module is used to connect to the production intranet. After identity authentication, it obtains the control authority of each image acquisition component and operation acquisition component according to the current production cycle and the operating status of specific production equipment, and edits and issues control instructions to each functional module; The acquisition unit is used to divide the acquisition targets of the production equipment operation area according to the production target, set the moving path plan of the image acquisition component and the acquisition frequency of the operation acquisition component, and extract the images and operation data collected in each area according to the preset time period; A model building module for training a prediction model based on historical image acquisition component movement path data and operation acquisition component frequency data using a convolutional neural network deep learning algorithm; The state analysis unit is used to input the currently acquired image and operation data state into the trained prediction model, obtain the output result of the prediction model for judgment, and capture abnormal data; The path adjustment module is used to trigger when abnormal data appears. The prediction model analyzes the path data of the current image acquisition component, identifies the event history similar to the current abnormal data, combines the associated historical image data with the current image acquisition component state, and generates a moving path plan for the image monitoring component; The frequency adjustment module is used to trigger when abnormal data occurs. The prediction model analyzes the severity and type of abnormal data and automatically generates a monitoring frequency plan based on the equipment operation history and the frequency of abnormal events. The comprehensive analysis module is used to evaluate the availability of the solutions submitted by the path adjustment module and the frequency adjustment module. When the preset threshold is reached, it is submitted to the acquisition unit and the feedback data of the adjusted status analysis unit is obtained in the next cycle. If there is no abnormal feedback from the status analysis unit, it means that the adjustment and availability evaluation are successful. The trend prediction module is used to integrate all the collection and analysis results, and output the equipment operation trends and potential abnormal events within the future specified period through the prediction model.

[0007] Furthermore, the acquisition unit is deployed with submodules at the lower level, including: a region division module, a path planning module, a collection module, a preprocessing module and a setting module, and the collection module is interactively connected with the region division module, the path planning module, the preprocessing module and the setting module through a wireless network, wherein: The area division module is used to define and divide the working areas of several production equipment according to the production process and production equipment layout of the current production cycle, form various acquisition target areas, and store area information; The path planning module is used to formulate the moving path of the image acquisition component through obstacle monitoring and the shortest path algorithm based on the area division data and the location data of the target production equipment; The acquisition module is used to collect production images and equipment operation data based on the target area and movement path submitted by the area division module, and synchronize the data to the cloud database; The pre-processing module is used to remove noise, standardize and format the collected production images and equipment operation data; The setting module is used to customize the scheduled data collection tasks of each area, send scheduled task instructions to the image acquisition component and the operation acquisition component, and execute the regular collection of image and sensor data.

[0008] Furthermore, the definition and division basis of the area division module includes: production equipment type, image acquisition component setting parameters, operation acquisition component setting parameters, production process flow and risk assessment.

[0009] Furthermore, the path planning module is connected to the path adjustment module through an electrical medium. The path adjustment module receives the moving path plan of the image monitoring component evaluated by the comprehensive analysis module, and after reading the plan, adjusts the operating posture, moving process and moving speed of several image acquisition components.

[0010] Furthermore, the setting module is connected to the frequency adjustment module through an electrical medium. The frequency adjustment module receives the frequency adjustment plan evaluated by the comprehensive analysis module and submits it to each production equipment to adjust the monitoring time, interval and part target of the collection components running in the equipment.

[0011] Furthermore, the state analysis unit is deployed with submodules at the lower level, including: a judgment module, a capture module and an alarm module, the judgment module and the capture module are interactively connected via a wireless network, and the capture module and the alarm module are interactively connected via a wireless network, wherein: The judgment module is used to input the collected real-time data into the prediction model for analysis to determine whether it meets the normal operation standards; The capture module is used to compare the image data with the standard atlas data, and the equipment operation settings with the standard equipment data in real time, and to capture abnormal data; The alarm module will execute the preset alarm mechanism and feedback to the management module if any suspected abnormality is found.

[0012] Furthermore, the management module is interactively connected to the collection unit through a wireless network, the collection unit is interactively connected to the model building module through a wireless network, the state analysis unit is interactively connected to the path adjustment module and the frequency adjustment module through a wireless network, and the comprehensive analysis module is interactively connected to the path adjustment module, the frequency adjustment module and the trend prediction module through a wireless network.

[0013] An industrial Internet security monitoring method comprises the following steps: Step 1: During the current production cycle, obtain access rights to the equipment sensors, divide the equipment operation areas according to production goals, and set the collection targets in each area; Step 2: Based on the equipment's operating efficiency, acquisition frequency, and safety, a moving path plan for the image acquisition equipment is developed through a path planning algorithm and an obstacle avoidance algorithm; Step 3: Within a preset period, regularly extract the data collected by sensors from each area, clean the extracted data and convert it into a format suitable for the prediction model; Step 4: Extract features from the collected data, select the convolutional neural network algorithm to build a prediction model, input path and operation data samples, and train the prediction model; Step 5: Compare the image data collected in the current cycle with the standard atlas data, and compare the equipment operation setting data with the standard equipment data to identify data anomalies. Based on the comparison results, determine whether the data status is normal, and use anomaly detection algorithms to capture and mark abnormal data; Step 6: If there is suspected abnormal data, output the monitoring frequency adjustment plan for the control target area, increase the monitoring of the abnormal area, analyze the moving path of the image monitoring device based on the correlation of historical abnormal data, obtain the tracking path, and adjust the moving path of the image acquisition device, and output the monitoring frequency adjustment plan and path adjustment plan; Step 7: Evaluate and analyze the implementation of the monitoring frequency adjustment plan and the path adjustment plan, debug the equipment to adapt, obtain image data and operation data again after implementation, and evaluate the data accuracy of the adjustment plan; Step 8: Based on the output of the prediction model, using the existing image data and operation data in step 7, output the potential abnormal data of the equipment operation trend in the future cycle, and summarize the abnormal events and submit them to the management end.

[0014] Furthermore, in the process of adjusting the moving path of the image acquisition device in step 6, the adjustment calculation formula of the moving path is set as: ; In the formula, represents the adjusted path point, Represents the current path point position, represents the weight factor for position adjustment, Represents the distance difference between the current abnormal data and the historical abnormal data. Represents the impact weight of historical abnormal data, Represents the correlation index from the current path point to the related historical anomaly point, represents the environmental adaptability weight, represents the environmental factor adjustment parameter, represents the safety correction factor, Represents external environmental factors.

[0015] Furthermore, the distance difference It is obtained by calculating the Euclidean distance between the current path point and the historical outlier point. The calculation formula is: ; In the formula, represents the horizontal coordinate of the target position, Represents the horizontal coordinate of the current path point. The vertical coordinate representing the target location is usually the reference coordinate corresponding to the historical anomaly data. Represents the ordinate of the current path point.

[0016] Compared with the known prior art, the technical solution provided by the present invention has the following beneficial effects: 1. The prediction model built by the deep learning algorithm analyzes the collected data, compares the image data with the standard map data, and compares the equipment setting data with the standard equipment data. Through precise data analysis, the accuracy of anomaly detection is greatly improved. Compared with the traditional rule engine, the intelligent algorithm can handle more complex status data and reduce false positives and negatives. The prediction model can not only identify anomalies, but also perform dynamic analysis and pattern recognition based on historical data, realizing the intelligence and automation of the monitoring mechanism, reducing the need for manual intervention, and enabling the system to make adjustments autonomously.

[0017] 2. By obtaining the real-time data acquisition permissions of the image acquisition component and the operation acquisition component, dynamic data acquisition is performed according to the preset time period and targeted acquisition plan adjustments are made based on abnormal factors. According to the analysis results of real-time data and historical data, the monitoring frequency of the target area and the moving path of the image acquisition device are dynamically adjusted to ensure that the equipment can avoid potential dangerous areas during monitoring and operation, monitor the equipment status in real time, and then quickly respond to potential anomalies, obtain more additional abnormal data, and then improve the timeliness of fault detection, thereby reducing equipment downtime and losses, allowing the monitoring system to adapt to different production environments and states, improving the targetedness and efficiency of monitoring, and helping to optimize resource allocation.

[0018] Establish a full-process monitoring system from data collection to exception handling, conduct a comprehensive assessment of the equipment operating status, and conduct comprehensive analysis based on historical data and real-time data, which improves the effectiveness of risk management. Combined with the future equipment operation trends output by the model, potential abnormal events can be identified in advance. Through effective prediction of future trends, preventive measures can be taken in advance to reduce the impact of potential failures and enhance the safety and stability of overall production. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work: Figure 1 It is a schematic diagram of the framework of an industrial Internet security monitoring system in the present invention; Figure 2 It is a flow chart of an industrial Internet security monitoring method in the present invention.

[0020] The numbers in the figure represent respectively, 1. Management module; 2. Acquisition unit; 21. Area division module; 22. Path planning module; 23. Acquisition module; 24. Preprocessing module; 25. Setting module; 3. Model building module; 4. State analysis unit; 41. Judgment module; 42. Capture module; 43. Alarm module; 5. Path adjustment module; 6. Frequency adjustment module; 7. Comprehensive analysis module; 8. Trend prediction module. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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 work are within the scope of protection of the present invention.

[0022] The present invention will be further described below in conjunction with the embodiments.

[0023] Example 1.

[0024] An industrial Internet security monitoring system of this embodiment, such as Figure 1 As shown, including: Management module 1 is used to connect to the production intranet. After identity authentication, it obtains the control authority of each image acquisition component and operation acquisition component according to the current production cycle and the operating status of specific production equipment, and edits and issues control instructions to each functional module. It ensures that only authorized personnel can access system functions through identity authentication, and intelligently allocates control authority of image acquisition components and operation acquisition components according to the production cycle and equipment status, thereby improving the safety and flexibility of production operations; The acquisition unit 2 is used to divide the acquisition targets of the production equipment operation area according to the production target, set the moving path plan of the image acquisition component and the acquisition frequency of the operation acquisition component, extract the images and operation data collected in each area according to the preset time period, ensure the efficient extraction of the required images and data within the preset time period, and improve the monitorability of the production process and the reliability of the data; The collection unit 2 is deployed with submodules at the lower level, including: an area division module 21, a path planning module 22, a collection module 23, a preprocessing module 24 and a setting module 25. The collection module 23 is interactively connected with the area division module 21, the path planning module 22, the preprocessing module 24 and the setting module 25 through a wireless network, wherein: The area division module 21 is used to define and divide the working areas of several production equipment according to the production process and production equipment layout of the current production cycle, form various acquisition target areas, and store area information; the definition and division basis includes: production equipment type, image acquisition component setting parameters, operation acquisition component setting parameters, production process flow and risk assessment, to ensure the comprehensiveness and accuracy of data collection; The path planning module 22 is used to formulate the moving path of the image acquisition component through obstacle monitoring and the shortest path algorithm based on the area division data and the target production equipment location data; the path planning module 22 is connected to the path adjustment module 5 through an electrical medium, and the path adjustment module 5 receives the moving path plan of the image monitoring component evaluated by the comprehensive analysis module 7, and after reading the plan, adjusts the running posture, moving process and moving speed of several image acquisition components, thereby improving the intelligence and adaptability of the path; The acquisition module 23 is used to acquire production images and equipment operation data based on the target area and movement path submitted by the area division module 21, and synchronously store the data in the cloud database; The pre-processing module 24 is used to perform denoising, standardization and formatting on the collected production images and equipment operation data, thereby improving the quality and usability of the data; The setting module 25 is used to customize the timed data collection tasks of each area, send the timed task instructions to the image collection component and the operation collection component, and perform the regular collection of image and sensor data; the setting module 25 is connected to the frequency adjustment module 6 through an electrical medium, and the frequency adjustment module 6 receives the frequency adjustment plan evaluated by the comprehensive analysis module 7 and submits it to each production equipment to adjust the monitoring time, interval and part target of the operation collection component in the equipment. The issuance of the timed task instruction ensures the regularity of data collection and improves the automation level of the system; Model building module 3 is used to use a convolutional neural network deep learning algorithm to train a prediction model based on historical image acquisition component movement path data and operation acquisition component frequency data, so that the system can infer possible future events based on historical patterns, thereby improving the scientific nature of decision-making; The state analysis unit 4 is used to input the currently collected image and operation data state into the trained prediction model, obtain the output result of the prediction model for judgment, and capture abnormal data; The state analysis unit 4 has submodules deployed at the lower level, including: a judgment module 41, a capture module 42 and an alarm module 43. The judgment module 41 and the capture module 42 are interactively connected via a wireless network, and the capture module 42 and the alarm module 43 are interactively connected via a wireless network, wherein: The judgment module 41 is used to input the collected real-time data into the prediction model for analysis to determine whether it meets the normal operation standards and ensure the stability of the production process; The capture module 42 is used to compare the image data with the standard atlas data, and the equipment operation settings with the standard equipment data in real time, capture abnormal data, and promptly discover abnormal conditions in the equipment operation; The alarm module 43, if a suspected abnormality is found, will execute the preset alarm mechanism and feedback to the management module 1; The path adjustment module 5 is used to be triggered when abnormal data appears. The prediction model analyzes the path data of the current image acquisition component, identifies the event history similar to the current abnormal data, combines the associated historical image data with the current image acquisition component state, generates a moving path plan for the image monitoring component, and generates a new path plan by analyzing the historical data, so that the image acquisition component can respond to abnormal situations more effectively; Frequency adjustment module 6, used to trigger when abnormal data occurs, predict the model to analyze the severity and type of abnormal data, automatically generate monitoring frequency plan based on the equipment operation history and the frequency of abnormal events, optimize the monitoring strategy of the equipment based on the characteristics of abnormal data, and improve the intelligence of monitoring work; The comprehensive analysis module 7 is used to evaluate the availability of the schemes submitted by the path adjustment module 5 and the frequency adjustment module 6. When the preset threshold is reached, it is submitted to the acquisition unit 2, and the feedback data of the adjusted state analysis unit 4 is obtained in the next cycle. If there is no abnormal feedback from the state analysis unit 4, it means that the adjustment and availability evaluation are successful. The availability evaluation is performed on the generated monitoring and frequency schemes to ensure that the system can still operate stably after the adjustment, thereby enhancing the reliability and responsiveness of the system. The trend prediction module 8 is used to integrate all the collected and analyzed results, and output the equipment operation trend and potential abnormal events in the future specified period through the prediction model, which helps to identify problems in advance and ensure smooth production.

[0025] As an implementation method in this embodiment, Figure 1 As shown, the management module 1 is interactively connected with the collection unit 2 through a wireless network, the collection unit 2 is interactively connected with the model building module 3 through a wireless network, the state analysis unit 4 is interactively connected with the path adjustment module 5 and the frequency adjustment module 6 through a wireless network, and the comprehensive analysis module 7 is interactively connected with the path adjustment module 5, the frequency adjustment module 6 and the trend prediction module 8 through a wireless network.

[0026] In the specific implementation of this example, based on the current production cycle, the management module 1 obtains the collection authority of the image and operation data, divides the collection target of the equipment operation area according to the production target through the area division module 21, sets the moving path plan of the image acquisition component through the path planning module 22, presets the time period through the setting module 25, and the collection module 23 extracts the collection data of each area according to the preset time period, and the pre-processing module 24 filters and transforms the collected data; The model building module 3 builds a prediction model through a deep learning algorithm, and inputs the obtained data into the prediction model. The judgment module 41 compares the image data with the standard atlas data, and the equipment operation setting data with the standard equipment data, and determines whether the current data status is normal. The capture module 42 captures abnormal data. When abnormal data appears, an alarm is sent to the management personnel through the alarm module 43; If there is suspected abnormal data in the current data, the frequency adjustment module 6 outputs the monitoring frequency adjustment plan for the control target area, and analyzes the moving path plan of the image detection device based on the associated historical history of similar abnormalities through the path adjustment module 5, obtains the precise tracking path of the image detection device and executes it, outputs the moving path adjustment plan of the image acquisition device and the frequency adjustment plan of the operation acquisition component, and the plan is applied after evaluation and analysis by the comprehensive analysis module 7, and then the image data and operation data are re-acquired, the feasibility of the plan adjustment is analyzed, and the potential abnormal events of the equipment operation trend in the future period are output through the trend prediction module 8 and submitted to the management module 1.

[0027] Example 2.

[0028] At other levels, this embodiment also provides an industrial Internet security monitoring method, such as Figure 2 As shown, the following steps are included: Step 1: During the current production cycle, obtain access rights to the equipment sensors, divide the equipment operation areas according to production goals, and set the collection targets in each area; Step 2: Based on the equipment's operating efficiency, acquisition frequency, and safety, a moving path plan for the image acquisition equipment is developed through a path planning algorithm and an obstacle avoidance algorithm; Step 3: Within a preset period, regularly extract the data collected by sensors from each area, clean the extracted data and convert it into a format suitable for the prediction model; Step 4: Extract features from the collected data, select the convolutional neural network algorithm to build a prediction model, input path and operation data samples, and train the prediction model; Step 5: Compare the image data collected in the current cycle with the standard atlas data, and compare the equipment operation setting data with the standard equipment data to identify data anomalies. Based on the comparison results, determine whether the data status is normal, and use anomaly detection algorithms to capture and mark abnormal data; Step 6: If there is suspected abnormal data, output the monitoring frequency adjustment plan for the control target area, increase the monitoring of the abnormal area, analyze the moving path of the image monitoring device based on the correlation of historical abnormal data, obtain the tracking path, and adjust the moving path of the image acquisition device, and output the monitoring frequency adjustment plan and path adjustment plan; Step 7: Evaluate and analyze the implementation of the monitoring frequency adjustment plan and the path adjustment plan, debug the equipment to adapt, obtain image data and operation data again after implementation, and evaluate the data accuracy of the adjustment plan; Step 8: Based on the output of the prediction model, using the existing image data and operation data in step 7, output the potential abnormal data of the equipment operation trend in the future cycle, and summarize the abnormal events and submit them to the management end.

[0029] Compared with the prior art, this embodiment can timely reflect the operation status of the equipment by acquiring the equipment sensor data in real time and performing regular extraction and analysis, ensuring dynamic response to potential abnormalities, and optimizing the moving path of the image acquisition device by using path planning and obstacle avoidance algorithms, making the acquisition process more efficient and flexible, and able to adapt to different monitoring needs in a changing production environment; By selecting the convolutional neural network algorithm, feature extraction and anomaly detection can be performed more accurately. Compared with traditional methods, it has stronger pattern recognition and prediction capabilities. Combined with the comparison of standard equipment data and real-time data, it can quickly identify abnormal conditions and output dynamic monitoring frequency and path adjustment plans, which improves the pertinence and effectiveness of monitoring. After implementing the monitoring frequency and path adjustment plan, the monitoring strategy can be continuously optimized through effective evaluation of the data obtained again, ensuring that the management end always obtains the latest and most accurate information; Based on the established predictive model, it is possible to identify potential anomalies in future cycles in advance, provide information for predictive maintenance, reduce downtime, and improve production efficiency. The summary of abnormal events and the submission of their analysis results can provide comprehensive decision-making support for management and help more efficient resource allocation and risk management.

[0030] Example 3.

[0031] In this embodiment, an adjustment calculation formula for setting the moving path is provided, specifically: ;pass Reflects the impact of historical abnormal data and its correlation, through appropriate weighting and To adjust the current position to get closer to the target position, Add the influence of environmental factors to ensure that the equipment can adapt to complex working environments. Safety correction, as a safety correction item, ensures that the equipment will not move to an unsafe position in the event of obstacles or other dangerous situations. The specific adjustment parameters of the image detection equipment in the moving path can be systematically calculated, which not only considers the impact of historical abnormal data, but also integrates environmental factors and safety requirements, thereby ensuring the efficiency and safety of production equipment operation; In the formula, represents the adjusted path point, Represents the current path point position, which is the coordinate of the current position of the device. Represents the weight factor of position adjustment, which is used to control the adjustment amplitude of the position. Represents the distance difference between the current abnormal data and the historical abnormal data, indicating the change in distance between the target location and the current path point. It is used to evaluate the distance of the current device location relative to the abnormal data so as to make appropriate decisions when adjusting the path. Represents the impact weight of historical abnormal data, which is used to adjust the impact of abnormal data on the moving path. The correlation index representing the current path point to the related historical anomaly point is based on the matching degree of image features. Represents the environmental adaptability weight, reflecting the ability of the device to adapt to environmental changes. Represents the environmental factor adjustment parameter, specifically the comprehensive coefficient of temperature, humidity and equipment load. This parameter can affect the flexibility and safety of equipment operation. Represents the safety correction factor, which is used to prevent the equipment from deviating from the optimal path due to obstacles and human interference. Represents the external environmental factor, which is the distance from the obstacle to the current path point. The range is obtained through the position sensor data; Distance Difference It is obtained by calculating the Euclidean distance between the current path point and the historical outlier point. The calculation formula is: ; By calculating the Euclidean distance between the current position of the image acquisition component and the target position, it provides an intuitive way to quantify the spatial relationship between the device and the historical anomaly data. In this way, it is possible to evaluate how much adjustment is needed to get closer to the target position; In the formula, The horizontal coordinate representing the target position is usually a reference coordinate corresponding to the historical abnormal data, which identifies the X coordinate of the target position that the device needs to reach during the path adjustment. The horizontal coordinate of the current path point represents the X coordinate of the current position of the image detection device, which serves as the basis for calculating the distance difference and helps determine the difference between the device and the target position. The ordinate representing the target position is usually a reference coordinate corresponding to historical abnormal data, and identifies the Y coordinate of the target position that the device needs to reach during path adjustment. Represents the ordinate of the current path point, indicating the Y coordinate of the current position of the image detection device, which serves as the basis for calculating the distance difference and helps determine the difference between the device and the target position.

[0032] In summary, the present invention ensures that sensor data can be converted into usable information in a timely manner through dynamic data collection and real-time data cleaning, improves the speed and efficiency of data processing, enables real-time monitoring, introduces deep learning algorithms such as convolutional neural networks for feature extraction and model training, forms an intelligent anomaly detection mechanism, enables the system to autonomously identify abnormal patterns, reduces the risk of false positives and false negatives, introduces the influence of multiple environmental factors in path adjustment, ensures that the movement and operation of the equipment can adapt to complex working environments, and thus improves the flexibility and safety of the equipment; Combining historical abnormal data and current equipment location, the equipment's movement path is optimized through a weight adjustment algorithm to ensure that the equipment can avoid potential dangerous areas during monitoring and operation. After identifying abnormal data, the monitoring frequency and resource allocation are automatically adjusted to conduct more frequent monitoring of abnormal areas to ensure that problems are caught in a timely manner. A full-process monitoring system from data collection to abnormality processing is established, and a comprehensive assessment of the equipment's operating status is conducted. A comprehensive analysis is conducted in combination with historical data and real-time data, which improves the effectiveness of risk management. Based on the predictive model, potential abnormal data for future production cycles is output to help managers identify and respond to possible problems in advance.

[0033] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An industrial Internet security monitoring system, characterized in that: include: The management module (1) is used to connect to the production intranet, and after identity authentication, obtain the control authority of each image acquisition component and operation acquisition component according to the current production cycle and the operating status of specific production equipment, and edit and issue control instructions to each functional module; The acquisition unit (2) is used to divide the acquisition targets of the production equipment operation area according to the production target, set the moving path plan of the image acquisition component and the acquisition frequency of the operation acquisition component, and extract the images and operation data collected in each area according to the preset time period; A model building module (3) is used to train a prediction model based on historical image acquisition component movement path data and operation acquisition component frequency data using a convolutional neural network deep learning algorithm; A state analysis unit (4) is used to input the currently collected image and operation data state into the trained prediction model, obtain the output result of the prediction model for judgment, and capture abnormal data; A path adjustment module (5) is used to be triggered when abnormal data appears, and the prediction model analyzes the path data of the current image acquisition component, identifies the event history similar to the current abnormal data, combines the associated historical image data with the current image acquisition component state, and generates a moving path plan for the image monitoring component; A frequency adjustment module (6) is used to be triggered when abnormal data occurs, and the prediction model analyzes the severity and type of the abnormal data, and automatically generates a monitoring frequency plan based on the equipment operation history and the frequency of abnormal events; The comprehensive analysis module (7) is used to evaluate the availability of the solutions submitted by the path adjustment module (5) and the frequency adjustment module (6). When a preset threshold is reached, the solution is submitted to the acquisition unit (2) and the feedback data of the adjusted state analysis unit (4) is obtained in the next cycle. If there is no abnormal feedback from the state analysis unit (4), it indicates that the adjustment and availability evaluation are successful. The trend prediction module (8) is used to integrate all the collected and analyzed results and output the equipment operation trend and potential abnormal events within a future specified period through a prediction model.

2. An industrial Internet security monitoring system according to claim 1, characterized in that: The acquisition unit (2) is provided with submodules at a lower level, including: an area division module (21), a path planning module (22), an acquisition module (23), a preprocessing module (24) and a setting module (25); the acquisition module (23) is interactively connected with the area division module (21), the path planning module (22), the preprocessing module (24) and the setting module (25) via a wireless network, wherein: A region division module (21) is used to define and divide the working areas of a number of production equipment according to the production process and production equipment layout of the current production cycle, form various acquisition target areas, and store area information; A path planning module (22) is used to formulate a moving path of the image acquisition component through obstacle monitoring and a shortest path algorithm based on the area division data and the target production equipment location data; A collection module (23) is used to collect production images and equipment operation data based on the target area and movement path submitted by the area division module (21), and to synchronously store the data in a cloud database; A preprocessing module (24) is used to perform noise removal, standardization and formatting on the collected production images and equipment operation data; The setting module (25) is used to customize the timed data collection tasks of each area, send timed task instructions to the image collection component and the operation collection component, and execute the regular collection of image and sensor data.

3. An industrial Internet security monitoring system according to claim 2, characterized in that: The definition and division basis of the area division module (21) includes: production equipment type, image acquisition component setting parameters, operation acquisition component setting parameters, production process flow and risk assessment.

4. An industrial Internet security monitoring system according to claim 2, characterized in that: The path planning module (22) is connected to the path adjustment module (5) via an electrical medium. The path adjustment module (5) receives the image monitoring component movement path plan evaluated by the comprehensive analysis module (7) and, after reading the plan, adjusts the operating posture, movement process and movement speed of a number of image acquisition components.

5. An industrial Internet security monitoring system according to claim 2, characterized in that: The setting module (25) is connected to the frequency adjustment module (6) via an electrical medium. The frequency adjustment module (6) receives the frequency adjustment scheme evaluated by the comprehensive analysis module (7) and submits it to each production equipment to adjust the monitoring time, interval and part target of the collection components running in the equipment.

6. An industrial Internet security monitoring system according to claim 1, characterized in that: The state analysis unit (4) is provided with submodules at a lower level, including: a judgment module (41), a capture module (42) and an alarm module (43); the judgment module (41) and the capture module (42) are interactively connected via a wireless network; the capture module (42) and the alarm module (43) are interactively connected via a wireless network, wherein: A judgment module (41) is used to input the collected real-time data into the prediction model for analysis to determine whether it meets the normal operation standard; A capture module (42) is used to compare the image data with the standard atlas data, and the equipment operation settings with the standard equipment data in real time, and to capture abnormal data; The alarm module (43) will execute a preset alarm mechanism if a suspected abnormality is found, and provide feedback to the management module (1).

7. An industrial Internet security monitoring system according to claim 1, characterized in that: The management module (1) is interactively connected to the acquisition unit (2) via a wireless network, the acquisition unit (2) is interactively connected to the model building module (3) via a wireless network, the state analysis unit (4) is interactively connected to the path adjustment module (5) and the frequency adjustment module (6) via a wireless network, and the comprehensive analysis module (7) is interactively connected to the path adjustment module (5), the frequency adjustment module (6) and the trend prediction module (8) via a wireless network.

8. An industrial Internet security monitoring method, the method is an implementation method of an industrial Internet security monitoring system based on any one of claims 1-5, characterized in that: The following steps are involved: Step 1: During the current production cycle, obtain access rights to the equipment sensors, divide the equipment operation areas according to production goals, and set the collection goals in each area; Step 2: Based on the equipment's operating efficiency, acquisition frequency, and safety, a moving path plan for the image acquisition equipment is developed through a path planning algorithm and an obstacle avoidance algorithm; Step 3: Within a preset period, regularly extract the data collected by sensors from each area, clean the extracted data and convert it into a format suitable for the prediction model; Step 4: Extract features from the collected data, select the convolutional neural network algorithm to build a prediction model, input path and operation data samples, and train the prediction model; Step 5: Compare the image data collected in the current cycle with the standard atlas data, and compare the equipment operation setting data with the standard equipment data to identify data anomalies. Based on the comparison results, determine whether the data status is normal, and use anomaly detection algorithms to capture and mark abnormal data; Step 6: If there is suspected abnormal data, output the monitoring frequency adjustment plan for the control target area, increase the monitoring of the abnormal area, analyze the moving path of the image monitoring device based on the correlation of historical abnormal data, obtain the tracking path, and adjust the moving path of the image acquisition device, and output the monitoring frequency adjustment plan and path adjustment plan; Step 7: Evaluate and analyze the implementation of the monitoring frequency adjustment plan and the path adjustment plan, debug the equipment to adapt, obtain image data and operation data again after implementation, and evaluate the data accuracy of the adjustment plan; Step 8: Based on the output of the prediction model, using the existing image data and operation data in step 7, output the potential abnormal data of the equipment operation trend in the future cycle, and summarize the abnormal events and submit them to the management end.

9. An industrial Internet security monitoring method according to claim 8, characterized in that: In the process of adjusting the moving path of the image acquisition device in step 6, the adjustment calculation formula of the moving path is set as: ； In the formula, represents the adjusted path point, Represents the current path point position, represents the weight factor for position adjustment, Represents the distance difference between the current abnormal data and the historical abnormal data. Represents the impact weight of historical abnormal data, Represents the correlation index from the current path point to the related historical anomaly point, represents the environmental adaptability weight, represents the environmental factor adjustment parameter, represents the safety correction factor, Represents external environmental factors.

10. An industrial Internet security monitoring method according to claim 9, characterized in that: The distance difference It is obtained by calculating the Euclidean distance between the current path point and the historical outlier point. The calculation formula is: ； In the formula, represents the horizontal coordinate of the target position, Represents the horizontal coordinate of the current path point, The vertical coordinate representing the target location is usually the reference coordinate corresponding to the historical anomaly data. Represents the ordinate of the current path point.

Images