Automatic emergency management method for intelligent industrial production

Abstract

The invention discloses an automatic emergency management method for intelligent industrial production, and relates to the field of industrial management, and the method comprises the steps: calculating the production capacity range of a production line; constructing a real-time command platform; the production capacity of the production line is monitored in real time, the real-time production capacity is obtained, whether the real-time production capacity belongs to the production capacity range or not is judged, and operation environment state data of the abnormal production line are obtained; determining production capacity abnormity reasons of the abnormal production line; determining the emergency level of the abnormal production line; outputting an alarm signal of a corresponding degree; and smooth communication among the command center, field personnel and related departments is judged. The method has the advantages that the abnormal reason of the abnormal production line is rapidly determined through the digital twinborn model, an emergency measure execution scheme for historically processing the reason is given, closed-loop feedback is completed, meanwhile, the real-time command platform detects signals in the emergency operation execution process, smooth communication of all departments is guaranteed, and the emergency operation execution efficiency is improved. And the collaboration efficiency of all departments is improved.

Description

Technical Field

[0001] The present invention relates to the field of industrial management, and in particular to an automated emergency management method for intelligent industrial production. Background Art

[0002] The traditional manual emergency response method mainly relies on operators to discover abnormal situations by observing the operating status of the production line, equipment indicator lights, alarm signals, etc., and report the discovered abnormal situations to superior managers or maintenance personnel. When the maintenance personnel arrive at the scene, they diagnose the cause of the fault by checking the equipment, checking the logs, and asking the operators. The maintenance personnel or managers formulate emergency response plans based on the diagnosis results, which results in a long time from the discovery of the fault to the arrival of the maintenance personnel at the scene for diagnosis and treatment, resulting in prolonged production interruption time and increased economic losses. In addition, fault diagnosis relies on the experience and skills of the maintenance personnel and is easily affected by subjective factors, which may lead to misjudgment and missed judgment, affecting the effectiveness of emergency response. At the same time, it is impossible to monitor the operating status and environmental data of the production line in real time, unable to timely discover potential abnormal situations, and unable to achieve early warning and prevention of abnormal situations. Summary of the invention

[0003] In order to solve the above technical problems, an automated emergency management method for intelligent industrial production is provided. This technical solution solves the problem in the above background technology that it is impossible to monitor the operating status and environmental data of the production line in real time, and to detect potential abnormal situations in time. It takes a long time from the discovery of the fault to the arrival of maintenance personnel at the scene for diagnosis and processing. At the same time, fault diagnosis depends on the experience and skills of the maintenance personnel, and is easily affected by subjective factors, which may lead to misjudgment and missed judgment, affecting the effectiveness of emergency treatment.

[0004] In order to achieve the above purpose, the technical solution adopted by the present invention is: An automated emergency management method for intelligent industrial production, obtaining at least one production line in intelligent industrial production, and calculating the production volume range of the production line based on order requirements and production resources and in combination with a scheduling algorithm; Build a visual and intelligent emergency command system through environmental sensors, build a real-time command platform, and monitor the production line; Based on the real-time command platform, the production volume of the production line is monitored in real time to obtain the real-time production volume, and to determine whether the real-time production volume is within the production volume range. If so, it is determined that the production line meets the production demand. If not, it is determined that the production line cannot meet the production demand, and the operating environment status data of the abnormal production line is obtained; Obtain the standard operating parameters of the production line, and combine them with the operating environment status data of the abnormal production line to determine the cause of the abnormal production volume of the abnormal production line; Based on several historical maintenance parameters of abnormal production volume, conduct a comprehensive analysis of the maintenance status of the abnormal production line and determine the emergency level of the abnormal production line; Based on the emergency level of the production line, output alarm signals of corresponding severity; Track the implementation of emergency measures in real time to determine whether the communication between the command center and on-site personnel and relevant departments is smooth. If so, no output is made; if not, an operation communication alarm signal is output.

[0005] Preferably, the calculation of the production volume range of the production line based on order demand and production resources in combination with the scheduling algorithm specifically includes: Based on the standard production data and historical production volume data of the production line, a production database is established to calculate the upper and lower control limits of production; A production volume range is formed, and the two endpoints of the production volume range are the upper control limit and the lower control limit.

[0006] Based on the production database and combined with the algorithm of dynamically adjusting control limits, a dynamic adjustment platform is constructed. The real-time command platform can input the remaining production of the abnormal production line into the production database, and the dynamic adjustment model automatically updates the upper and lower control limits of production according to the real-time production database.

[0007] Preferably, the establishing of a production database based on the standard production data and historical production volume data of the production line, and the calculation of the production upper control limit and the production lower control limit specifically include: Obtain all data in the production database and divide all data in the production database into several subgroups; Calculate the subgroup mean based on the mean function in the production database; Calculate the subgroup standard deviation based on the standard deviation function in the production database; Based on the subgroup mean and subgroup standard deviation, the production upper control limit and production lower control limit are calculated through linear equations.

[0008] Preferably, the linear equation is: , Where: i is the subscript, UCL i is the upper control limit of the ith subgroup, LCL i is the lower control limit of the ith subgroup, US is the subgroup mean, σ is the subgroup standard deviation, N is the total number of subgroups, and X is the control limit standard multiple; Calculate the production upper control limit based on the first mean value formula, denoted as UCL; Calculate the production lower control limit based on the second mean value formula, denoted as LCL; The first average formula is: , The second average formula is: , Preferably, the construction of a visualized and intelligent emergency command system through environmental sensors and the construction of a real-time command platform specifically include: Obtain edge computing devices, synchronize data and communications between edge computing devices and the real-time command platform through the database system of the real-time command platform, and be controlled by the real-time command platform; Synchronize the model deployment of the real-time command platform to the edge computing device; The real-time command platform uses monitoring slices to quickly slice the information collected by sensors. Based on the communication needs of each slice, the real-time communication needs and network conditions are used to adjust the resources of each slice for allocation. Obtain the standard signal strength and historical signal strength based on the communication equipment, and set the minimum value of communication signal strength allowed in the real-time command platform; The real-time signal on the production line is transmitted to the real-time command platform and edge computing equipment through the signal detection equipment. The real-time command platform and edge computing equipment monitor the production line signal in real time and issue an alarm when the production line signal is lower than the minimum communication signal strength value; Based on the real-time command platform database, the weights of information collected by different sensors are divided, and a dynamic allocation mechanism is established. The real-time command platform dynamically allocates edge computing and its own signal processing tasks.

[0009] Preferably, the step of obtaining the standard operating parameters of the abnormal production line and determining the cause of the abnormal production volume of the abnormal production line in combination with the abnormal production volume feedback report and the operating environment status data specifically includes: Obtain the operating parameters of the production line at different time points to form a data set containing spatiotemporal information, extract useful time features, analyze the spatial relationship between the equipment on the production line, and extract spatial features; The extracted temporal and spatial features are integrated to form comprehensive spatiotemporal features, and feature selection and dimensionality reduction are performed; The spatiotemporal characteristics are divided into multiple quasi-measurement groups, and the historical fault causes are divided into multiple influencing factor groups; By combining several data of the criterion group and the influencing factor group in pairs, a data sample is formed; The average data in the screening data samples is combined with the production line standard data to form the calibration data, and the remaining data samples constitute the test data; Based on the test data and combined with the prediction model, a digital twin model is constructed and trained; Based on the correction data, run the digital twin model, compare the predicted value of the digital twin model with the standard value in the correction data, and make regular corrections to the digital twin model; Input the historical fault causes, fault data and emergency measures execution plan into the digital twin model, run the digital twin model, and quickly simulate the historical abnormal production line. Correct the consistency between the predicted fault data in the digital twin model and the actual historical fault data. If they are consistent, they are recorded in the digital twin model database. If they are inconsistent, correct the digital twin model. When a production line fails, the fault data is input into the digital twin model, the digital twin model is run, the abnormal production line is simulated, and the historical fault data in the digital twin model database is combined. The operation data before and after the abnormal moment simulated by the digital twin model is compared with the benchmark data under normal conditions. Combined with the causes of historical faults in the digital twin model database, the differences in key parameters are analyzed to determine the cause of the abnormality. The causes and possible factors of the abnormality are automatically arranged according to the possibility based on the algorithm of the digital twin model, and synchronized to the real-time command platform to determine the emergency level of the abnormal production line and output an alarm signal of the corresponding degree; The real-time command platform notifies maintenance personnel to check the causes of the abnormality and possible factors one by one, determine the final cause of the failure, and feed it back to the digital twin model, realizing closed-loop feedback of the spatiotemporal characteristics in the digital twin model database.

[0010] Preferably, the digital twin model is constructed based on the test data and combined with the prediction model, and the training of the digital twin model specifically includes: The prediction model is: , Where: Y is the simulation value of the digital twin model, ω k is the influence of the kth data, ψ is the total influence coefficient, X k is the kth test data, and ε is the error.

[0011] Preferably, the real-time tracking of the implementation of emergency measures specifically includes: The emergency measures implementation plan is divided into multiple key nodes. The real-time command platform captures the monitoring information of key nodes and feeds the monitoring information back to the digital twin model. The prediction model in the digital twin model is used to predict the real-time status of key nodes and the execution of emergency measures, and the prediction value is fed back to the real-time command platform; The real-time command platform will compare and analyze the real-time status prediction value of key nodes with the signal fluctuation time in the communication status data. If the trigger condition is met, it will immediately issue an early warning to the communication technicians and the emergency command center and adjust the signal transmission part of the emergency measures execution plan. If the trigger condition is not met, no output response will be made. The real-time command platform adjusts the production line part of the emergency measures implementation plan according to the predicted value of the emergency measures implementation plan, and updates the production database at the same time to adjust the upper control limit and the lower control limit of the normal production line.

[0012] Preferably, the triggering condition includes: The emergency measures execution progress condition, communication status condition and data transmission packet loss rate, set the first threshold of the emergency measures execution progress condition, set the second threshold of the communication status condition and set the third threshold of the data transmission packet loss rate, and the real-time command platform's emergency measures execution progress condition, communication status condition and data transmission packet loss rate real-time data, any one of which is lower than the corresponding threshold, is triggered; Based on the real-time command platform database storing emergency measures execution data and communication status data, a synchronous task scheduling mechanism is built to achieve unified scheduling of emergency measures execution tasks and communication status monitoring tasks; Based on the task scheduling mechanism, set the priorities of emergency measures execution tasks and communication status monitoring tasks.

[0013] Preferably, setting the first threshold of the emergency measure execution progress condition specifically includes: Based on the emergency measures execution plan, the digital twin model is used to quantitatively analyze the emergency measures execution plan, obtain the total emergency measures execution amount, and use the first emergency formula to calculate the first threshold; The first emergency formula is as follows: , Among them: Q1 is the set threshold of the emergency measures execution progress condition, TEO is the total emergency measures execution amount, TPC is the conversion rate between engineering progress, and EC is the emergency level; Setting the second threshold of the communication state condition and setting the third threshold of the data transmission packet loss rate specifically include: Obtain the standard signal strength of the communication equipment, obtain the allowable loss rate of signal transmission based on the real-time command platform database and the digital twin model database, and use the second emergency formula to calculate the first threshold; The second emergency formula is as follows: , Among them: Q2 is the communication status condition threshold, Q3 is the threshold of data transmission packet loss rate, RS is the standard signal strength, SL is the allowable loss rate, and EC is the emergency level.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The present invention proposes an automated emergency management method for intelligent industrial production. When a production line fails, the fault data is input into a digital twin model, and the digital twin model is used to simulate the abnormal production line. The operating data before and after the abnormal moment of the simulation data is compared with the benchmark data under normal conditions. Combined with the causes of historical faults in the digital twin model database, the difference in key parameters is analyzed, and the cause of the abnormality is determined quickly and objectively, shortening the overall response, analysis and judgment steps. The real-time command platform then notifies maintenance personnel to check one by one according to the causes of the abnormality and possible factors, determine the final cause of the fault, and feed it back to the digital twin model, so as to realize the closed-loop feedback of the spatiotemporal characteristics in the digital twin model database, and achieve the effect of continuous optimization and improvement. The present invention proposes an automated emergency management method for intelligent industrial production. When an abnormal production line occurs, the real-time command platform uses monitoring slices to quickly slice the information collected by the sensor, and quickly distributes the detection signals of the normal production line to the edge computing device. The edge computing device detects the normal production line, and reserves a confidence channel for the real-time command platform to detect the abnormal production line, ensuring that the real-time command platform can monitor the signals and on-site images of the abnormal production line in real time, improving the signal transmission stability between the command personnel in the real-time command platform and the maintenance personnel of the abnormal production line, and enhancing the coordination ability of various parts during emergency operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram of the process of the automated emergency management method for intelligent industrial production of the present invention; Figure 2 It is a flow chart of calculating the production range of a production line based on order requirements and production resources in combination with a scheduling algorithm according to the present invention; Figure 3 A schematic diagram of a process for establishing a production database and calculating an upper control limit and a lower control limit of production based on standard production data and historical production volume data of a production line according to the present invention; Figure 4 A schematic diagram of the process of constructing a visual and intelligent emergency command system and a real-time command platform through environmental sensors of the present invention; Figure 5 A flow chart of obtaining standard operating parameters of an abnormal production line according to the present invention, and determining the cause of abnormal production volume of the abnormal production line in combination with abnormal production volume feedback report and operating environment status data; Figure 6 A schematic diagram of a flow chart for real-time tracking of the implementation of emergency measures of the present invention; Figure 7 It is a flow chart of the triggering conditions of the present invention. DETAILED DESCRIPTION

[0016] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0017] Reference Figure 1 As shown, an automated emergency management method for intelligent industrial production, An automated emergency management method for intelligent industrial production, comprising: Obtain at least one production line in intelligent industrial production, and calculate the production volume range of the production line based on order requirements and production resources and in combination with a scheduling algorithm; Build a visual and intelligent emergency command system through environmental sensors, build a real-time command platform, and monitor the production line; Based on the real-time command platform, the production volume of the production line is monitored in real time to obtain the real-time production volume, and to determine whether the real-time production volume is within the production volume range. If so, it is determined that the production line meets the production demand. If not, it is determined that the production line cannot meet the production demand, and the operating environment status data of the abnormal production line is obtained; Obtain the standard operating parameters of the production line, and combine them with the operating environment status data of the abnormal production line to determine the cause of the abnormal production volume of the abnormal production line; Based on several historical maintenance parameters of abnormal production volume, conduct a comprehensive analysis of the maintenance status of the abnormal production line and determine the emergency level of the abnormal production line; Based on the emergency level of the production line, output alarm signals of corresponding severity; Track the implementation of emergency measures in real time to determine whether the communication between the command center and on-site personnel and relevant departments is smooth. If so, no output is made; if not, an operation communication alarm signal is output.

[0018] When this solution is used, it can directly combine the order demand and production resources of industrial production with the scheduling algorithm to calculate the production range of the production line, and monitor the production line through various sensors on the production line, monitor the actual daily production of the production line, and ensure that the production line can reach the preset production range. When an abnormal production line occurs, the sensors on the production line can transmit the data of the abnormal production line to the real-time command platform, and the real-time command platform detects the signals of the production line and the real-time monitoring screen through the sensors, so as to determine the cause of the abnormal production volume, and formulate corresponding emergency measures implementation plans. According to the emergency measures implementation plan, the production tasks delayed by the abnormal production line are fed back to the production database, and the production range of the normal production line is recalculated to ensure the effective completion of the production tasks.

[0019] During the automated emergency response period of this solution, the real-time command platform can effectively and continuously monitor the images and signals of the production line, thereby ensuring the stability of the signal between the production line and the command personnel, increasing the coordination effect between departments, and shortening the overall emergency operation process.

[0020] Reference Figure 2 As shown in the figure, based on order demand and production resources, the production range of the production line is calculated in combination with the scheduling algorithm, including: Based on the standard production data and historical production volume data of the production line, a production database is established to calculate the upper and lower control limits of production; A production volume range is formed, and the two endpoints of the production volume range are the upper control limit and the lower control limit.

[0021] Based on the production database and combined with the algorithm of dynamically adjusting control limits, a dynamic adjustment platform is constructed. The real-time command platform can input the remaining production of the abnormal production line into the production database, and the dynamic adjustment model automatically updates the upper and lower control limits of production according to the real-time production database.

[0022] In this solution, the standard production data of the production line, such as the maximum production speed, minimum production speed, average production speed, equipment utilization rate, and the actual production volume data of the historical production line, can be integrated into a database to form a comprehensive production data set. Based on the production data set, the real-time command platform can receive real-time data from the production line, process and analyze the real-time production data, extract key information, and automatically update the upper production control limit and the lower production control limit based on the algorithm for dynamically adjusting the control limit. When an abnormal situation occurs in the production line, the real-time command platform can input the remaining production volume of the abnormal production line into the production database. These data can be used to dynamically adjust the upper production control limit and the lower production control limit to adapt to new production needs and resource conditions, so as to ensure the normal completion of production tasks.

[0023] Reference Figure 3 As shown in the figure, based on the standard production data and historical production volume data of the production line, a production database is established, and the calculation of the upper control limit and the lower control limit of production specifically includes: Obtain all data in the production database and divide all data in the production database into several subgroups; Calculate the subgroup mean based on the mean function in the production database; Calculate the subgroup standard deviation based on the standard deviation function in the production database; Based on the subgroup mean and subgroup standard deviation, the production upper control limit and production lower control limit are calculated through linear equations.

[0024] In this solution, all data in the production database are grouped and calculated, which can effectively remove outliers, missing values ​​and duplicate values, ensure the accuracy and completeness of the data, and realize the control and optimization of the production process. At the same time, the setting of upper control limits and lower control limits can timely discover and deal with abnormal situations in the production process, reduce equipment downtime, and improve equipment utilization. Through the setting and adjustment of control limits, the quality variation in the production process can be effectively controlled to improve product quality.

[0025] The linear equation is: , Where: i is the subscript, UCL i is the upper control limit of the ith subgroup, LCL i is the lower control limit of the ith subgroup, US is the subgroup mean, σ is the subgroup standard deviation, N is the total number of subgroups, and X is the control limit standard multiple; Calculate the production upper control limit based on the first mean value formula, denoted as UCL; Calculate the production lower control limit based on the second mean value formula, denoted as LCL; The first average formula is: , The second average formula is: , The set control limits can not only effectively identify abnormal variations in the production process, but also control the false alarm rate according to the X value to ensure the sensitivity of the control limits. At the same time, the control limits are calculated using historical data to ensure the scientificity and rationality of the control limits. The linear equation can also clearly define the control limits to help quality control personnel make targeted quality improvements and process optimizations. In addition, the control limits can be updated according to real-time data in combination with the dynamic adjustment mechanism to adapt to changes in the production process.

[0026] Reference Figure 4 As shown in the figure, a visual and intelligent emergency command system is built through environmental sensors, and the construction of a real-time command platform specifically includes: Obtain edge computing devices, synchronize data and communications between edge computing devices and the real-time command platform through the database system of the real-time command platform, and be controlled by the real-time command platform; Synchronize the model deployment of the real-time command platform to the edge computing device; The real-time command platform uses monitoring slices to quickly slice the information collected by sensors. Based on the communication needs of each slice, the real-time communication needs and network conditions are used to adjust the resources of each slice for allocation. Obtain the standard signal strength and historical signal strength based on the communication equipment, and set the minimum value of communication signal strength allowed in the real-time command platform; The real-time signal on the production line is transmitted to the real-time command platform and edge computing equipment through the signal detection equipment. The real-time command platform and edge computing equipment monitor the production line signal in real time and issue an alarm when the production line signal is lower than the minimum communication signal strength value; Based on the real-time command platform database, the weights of information collected by different sensors are divided, and a dynamic allocation mechanism is established. The real-time command platform dynamically allocates edge computing and its own signal processing tasks.

[0027] In this solution, by interacting with the communication equipment, the real-time command platform obtains the standard signal strength and historical signal strength data based on the communication equipment, and sets a minimum value of the communication signal strength allowed in the real-time command platform based on the analysis of the production environment and communication requirements. The setting of this minimum value usually takes into account the communication quality required for the normal operation of the production equipment, the signal interference situation, and the signal fluctuation range in the historical data, so as to set a clear threshold for the communication signal strength of the production line, which is used as the basis for judging whether the communication of the production line is normal, and provides a quantitative standard for subsequent real-time monitoring and alarm, ensuring that the communication equipment on the production line operates within the range where the signal strength can meet the production requirements, avoiding data transmission errors or interruptions due to weak signals, and affecting the normal production. At the same time, through the database system, the data consistency between the edge computing device and the real-time command platform is ensured, so that the platform can obtain field data in real time. At the same time, the edge computing device processes the data at a location close to the data source, reducing the delay and bandwidth occupation of data transmission. When monitoring the production line, the edge computing device performs preliminary processing on the data, reducing the amount of data transmission, thereby reducing the communication delay. At the same time, the edge computing device can quickly process the field data and improve the data processing efficiency.

[0028] In this solution, the real-time command platform first divides the massive information collected by the sensor into multiple slices with different communication requirements through monitoring slice technology. Then it continuously monitors the network status, including parameters such as bandwidth, delay, and packet loss rate. According to the communication requirements of each slice and the current network status, the real-time command platform dynamically adjusts the network resources available to each slice to ensure that the data of each slice can be transmitted and processed in a timely and accurate manner. It divides the weights of different types of information and continuously monitors the load of itself and edge computing devices, including CPU usage, memory usage, network bandwidth usage, etc. According to the information weight and the current load, the real-time command platform dynamically determines which signal processing tasks are assigned to the edge computing device and which are processed by itself to achieve the optimal allocation of computing resources. During its use, the signal processing tasks can be reasonably allocated according to the importance of the information and the load of the equipment to ensure that important information can be processed in a timely and accurate manner, while making full use of the computing resources of the edge computing device and the real-time command platform to avoid excessive load on one side.

[0029] Reference Figure 5 As shown, the standard operating parameters of the abnormal production line are obtained, and combined with the abnormal production feedback report and the operating environment status data, the abnormal production reasons of the abnormal production line are determined, including: Obtain the operating parameters of the production line at different time points to form a data set containing spatiotemporal information, extract useful time features, analyze the spatial relationship between the equipment on the production line, and extract spatial features; The extracted temporal and spatial features are integrated to form comprehensive spatiotemporal features, and feature selection and dimensionality reduction are performed; The spatiotemporal characteristics are divided into multiple quasi-measurement groups, and the historical fault causes are divided into multiple influencing factor groups; By combining several data of the criterion group and the influencing factor group in pairs, a data sample is formed; The average data in the screening data samples is combined with the production line standard data to form the calibration data, and the remaining data samples constitute the test data; Based on the test data and combined with the prediction model, a digital twin model is constructed and trained; Based on the correction data, run the digital twin model, compare the predicted value of the digital twin model with the standard value in the correction data, and make regular corrections to the digital twin model; Input the historical fault causes, fault data and emergency measures execution plan into the digital twin model, run the digital twin model, and quickly simulate the historical abnormal production line. Correct the consistency between the predicted fault data in the digital twin model and the actual historical fault data. If they are consistent, they are recorded in the digital twin model database. If they are inconsistent, correct the digital twin model. When a production line fails, the fault data is input into the digital twin model, the digital twin model is run, the abnormal production line is simulated, and the historical fault data in the digital twin model database is combined. The operation data before and after the abnormal moment simulated by the digital twin model is compared with the benchmark data under normal conditions. Combined with the causes of historical faults in the digital twin model database, the differences in key parameters are analyzed to determine the cause of the abnormality. The causes and possible factors of the abnormality are automatically arranged according to the possibility based on the algorithm of the digital twin model, and synchronized to the real-time command platform to determine the emergency level of the abnormal production line and output an alarm signal of the corresponding degree; The real-time command platform notifies maintenance personnel to check the causes of the abnormality and possible factors one by one, determine the final cause of the failure, and feed it back to the digital twin model, realizing closed-loop feedback of the spatiotemporal characteristics in the digital twin model database.

[0030] In this solution, for time features, signal processing and statistical analysis methods are used, such as calculating the rate of change and periodicity of parameters. For spatial features, the positional relationship, connection relationship, and material flow relationship of each device are analyzed to determine the spatial topological structure between devices. Time features can reflect the dynamic changes of the production process, and spatial features can show the relationship between devices, providing a rich data basis for subsequent analysis. Accurately extracted spatiotemporal features can also help to quickly locate equipment and links that may have problems when a fault occurs, thereby improving the accuracy of fault diagnosis. The comprehensive spatiotemporal features formed, feature selection and dimensionality reduction reduce the amount of data, improve the training speed and accuracy of subsequent models, make the data more stable, and the model is more sensitive to data changes, which can better capture abnormal situations in the production process. The trained digital twin model can accurately simulate the operating status of the production line by cooperating with the test data, providing strong support for subsequent fault analysis and prediction. By running the digital twin model with correction data, the predicted value of the model is compared with the standard value in the correction data, and the model parameters are adjusted according to the difference to achieve timed correction. By continuously comparing and correcting with the actual data, the digital twin model can more accurately reflect the actual operation of the production line, improve the accuracy of fault prediction and simulation, and the processing of historical data and model correction process can help the model accumulate fault knowledge and better deal with various complex fault situations.

[0031] In this solution, when a production line fails, the fault data is input into the model to simulate the abnormal production line. The operating data before and after the abnormal moment simulated by the model is compared with the normal baseline data. The key parameter differences are analyzed in combination with the historical fault causes to determine the cause of the abnormality. According to the analysis results, the abnormal causes and possibility factors are arranged according to the possibility according to the algorithm of the digital twin model and transmitted to the real-time command platform. The platform determines the emergency level according to the preset rules and issues an alarm. At the same time, the real-time command platform notifies the maintenance personnel to investigate and determine the final cause of the fault and feedback it to the digital twin model, and updates the spatiotemporal characteristics in the model database, so as to achieve the effect of quickly simulating abnormal situations, accurately analyzing the cause of the fault, providing maintenance personnel with a clear investigation direction, shortening the fault handling time, and reducing production losses. At the same time, automatically judging the emergency level and issuing an alarm will help enterprises to reasonably allocate resources, take appropriate emergency measures, and improve the efficiency and scientificity of emergency management. Finally, through closed-loop feedback, the digital twin model can continuously learn and optimize, improve the prediction and analysis capabilities of production line failures, and achieve continuous improvement of emergency management.

[0032] Based on the test data and combined with the prediction model, a digital twin model is constructed and trained, including: The prediction model is: , Where: Y is the simulation value of the digital twin model, ω k is the influence of the kth data, ψ is the total influence coefficient, X k is the kth test data, and ε is the error term.

[0033] Reference Figure 6 As shown in the figure, real-time tracking of the implementation of emergency measures specifically includes: The emergency measures implementation plan is divided into multiple key nodes. The real-time command platform captures the monitoring information of key nodes and feeds the monitoring information back to the digital twin model. The prediction model in the digital twin model is used to predict the real-time status of key nodes and the execution of emergency measures, and the prediction value is fed back to the real-time command platform; The real-time command platform will compare and analyze the real-time status prediction value of key nodes with the signal fluctuation time in the communication status data. If the trigger condition is met, it will immediately issue an early warning to the communication technicians and the emergency command center and adjust the signal transmission part of the emergency measures execution plan. If the trigger condition is not met, no output response will be made. The real-time command platform adjusts the production line part of the emergency measures implementation plan according to the predicted value of the emergency measures implementation plan, and updates the production database at the same time to adjust the upper control limit and the lower control limit of the normal production line.

[0034] In this solution, by monitoring key nodes, various status data during the execution of emergency measures can be obtained in time, providing a basis for subsequent analysis and decision-making. Based on these feedback information, the digital twin model can simulate and predict the execution of emergency measures more accurately. At the same time, the prediction model uses data analysis and machine learning technologies based on historical data and current monitoring information to estimate the future status and execution of key nodes, helping to understand the direction of emergency measures in advance. By comparison, the correlation between the status of key nodes and the fluctuation of communication signals can be found. If an abnormal situation reaches the preset trigger condition, it means that there may be potential problems that affect the execution of emergency measures. Timely warning and adjustment of the signal transmission part are required to ensure smooth communication and the smooth execution of emergency measures. Adjustment of the production line part according to the prediction results can make the production line better adapt to emergency situations and ensure the stability and continuity of production. Updating the production database and adjusting the control limit can help to achieve dynamic management and optimization of the production process. In this process, through real-time monitoring and status prediction of key nodes, problems in the execution of emergency measures can be quickly discovered, and early warnings and adjustments can be made in time to avoid the expansion of problems, improve the efficiency and effectiveness of emergency response, and adjust the emergency measures execution plan according to the predicted value, so that it can be continuously optimized according to the actual situation, improve the rationality and feasibility of the plan, and better respond to various complex emergency situations. The key node status and communication signal fluctuations are compared and analyzed to ensure the coordination of communication and emergency measures execution, and avoid the impact of communication problems on the development of emergency work. At the same time, the adjustment of the production line and the update of the production database ensure the stable operation of production in an emergency state, reduce production losses, and form a closed-loop management system from monitoring, prediction, analysis to adjustment in the overall process, which helps to continuously accumulate experience, improve the overall emergency management level of intelligent industrial production, and enhance the ability of enterprises to respond to emergencies.

[0035] Reference Figure 7 As shown, the trigger conditions include: The emergency measures execution progress condition, communication status condition and data transmission packet loss rate, set the first threshold of the emergency measures execution progress condition, set the second threshold of the communication status condition and set the third threshold of the data transmission packet loss rate, and the real-time command platform's emergency measures execution progress condition, communication status condition and data transmission packet loss rate real-time data, any one of which is lower than the corresponding threshold, is triggered; Based on the real-time command platform database storing emergency measures execution data and communication status data, a synchronous task scheduling mechanism is built to achieve unified scheduling of emergency measures execution tasks and communication status monitoring tasks; Based on the task scheduling mechanism, set the priorities of emergency measures execution tasks and communication status monitoring tasks.

[0036] In this scheme, the threshold is set to determine a standard for judging whether the system has abnormal conditions. By comparing the real-time monitoring data with the threshold, it is possible to timely discover possible problems in the emergency management process, and promptly start the corresponding processing process to avoid further deterioration of the problem and reduce possible losses. At the same time, through the construction of the task scheduling mechanism, the data in the database can be used as a basis to fully understand the actual situation of emergency measures execution and communication status. Based on these data, the task scheduling mechanism reasonably arranges the execution order and time of the emergency measures execution task and the communication status monitoring task to ensure that the two tasks can be coordinated and avoid task conflicts or unreasonable resource allocation. At the same time, different priorities are assigned to different tasks according to their importance and urgency. When resources are limited or multiple tasks need to be executed at the same time, tasks with high priority can obtain resources and execution opportunities first, ensuring that key tasks can be processed in a timely manner, which helps managers make more scientific and reasonable decisions and improve the overall level of emergency management.

[0037] The first threshold for setting the progress condition of the emergency measures execution specifically includes: Based on the emergency measures execution plan, the digital twin model is used to quantitatively analyze the emergency measures execution plan, obtain the total emergency measures execution amount, and use the first emergency formula to calculate the first threshold; The first emergency formula is as follows: , Among them: Q1 is the set threshold of the emergency measures execution progress condition, TEO is the total emergency measures execution amount, TPC is the conversion rate between engineering progress, and EC is the emergency level; Setting the second threshold of the communication state condition and setting the third threshold of the data transmission packet loss rate specifically include: Obtain the standard signal strength of the communication equipment, obtain the allowable loss rate of signal transmission based on the real-time command platform database and the digital twin model database, and use the second emergency formula to calculate the first threshold; The second emergency formula is as follows: , Among them: Q2 is the communication status condition threshold, Q3 is the threshold of data transmission packet loss rate, RS is the standard signal strength, SL is the allowable loss rate, and EC is the emergency level.

[0038] In this solution, by calculating the threshold value through the formula, key indicators such as the progress of emergency measures execution, communication status and data transmission packet loss rate can be accurately quantified, so that emergency management can be transformed from vague qualitative judgment to precise quantitative management, improving the accuracy and scientificity of management. At the same time, relying on the digital twin model and the data in the database for threshold calculation ensures the reliability and comprehensiveness of the decision-making basis. Managers can make decisions that are more in line with the actual situation based on these accurate thresholds, avoiding errors caused by subjective assumptions and empiricism, and improving the quality of emergency management decisions. Reasonable threshold settings provide clear operating boundaries and judgment criteria for the emergency management system. When various indicators approach or exceed the threshold, early warnings can be issued in time to remind relevant personnel to take measures, thereby ensuring the stable operation of the emergency management system and the entire intelligent industrial production system, and reducing the risk of system failures or accidents due to abnormal indicators.

[0039] In summary, the advantages of the present invention are: through the digital twin model, the abnormal cause of the abnormal production line can be quickly determined, the emergency measures execution plan for historically handling the cause is given, and closed-loop feedback is completed. At the same time, the real-time command platform detects the signal of the emergency operation execution process to ensure smooth communication between departments and improve the efficiency of coordination among departments.

[0040] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. An automated emergency management method for intelligent industrial production, characterized in that: include: Obtain at least one production line in intelligent industrial production, and calculate the production volume range of the production line based on order requirements and production resources and in combination with a scheduling algorithm; Build a visual and intelligent emergency command system through environmental sensors, build a real-time command platform, and monitor the production line; Based on the real-time command platform, the production volume of the production line is monitored in real time to obtain the real-time production volume, and to determine whether the real-time production volume is within the production volume range. If so, it is determined that the production line meets the production demand. If not, it is determined that the production line cannot meet the production demand, and the operating environment status data of the abnormal production line is obtained; Obtain the standard operating parameters of the production line, and combine them with the operating environment status data of the abnormal production line to determine the cause of the abnormal production volume of the abnormal production line; Based on several historical maintenance parameters of abnormal production volume, conduct a comprehensive analysis of the maintenance status of the abnormal production line and determine the emergency level of the abnormal production line; Based on the emergency level of the production line, output alarm signals of corresponding severity; Track the implementation of emergency measures in real time to determine whether the communication between the command center and on-site personnel and relevant departments is smooth. If so, no output is made; if not, an operation communication alarm signal is output.

2. An automated emergency management method for intelligent industrial production according to claim 1, characterized in that: The production range of the production line calculated based on order demand and production resources in combination with the scheduling algorithm specifically includes: Based on the standard production data and historical production volume data of the production line, a production database is established to calculate the upper and lower control limits of production; The production volume range is formed, and the two end points of the production volume range are the production upper control limit and the production lower control limit; Based on the production database and combined with the algorithm of dynamically adjusting control limits, a dynamic adjustment platform is constructed. The real-time command platform can input the remaining production of the abnormal production line into the production database, and the dynamic adjustment model automatically updates the upper and lower control limits of production according to the real-time production database.

3. An automated emergency management method for intelligent industrial production according to claim 2, characterized in that: The establishment of a production database based on the standard production data and historical production volume data of the production line and the calculation of the production upper control limit and the production lower control limit specifically include: Obtain all data in the production database and divide all data in the production database into several subgroups; Calculate the subgroup mean based on the mean function in the production database; Calculate the subgroup standard deviation based on the standard deviation function in the production database; Based on the subgroup mean and subgroup standard deviation, the production upper control limit and production lower control limit are calculated through linear equations.

4. An automated emergency management method for intelligent industrial production according to claim 3, characterized in that: The linear equation is: , Where: i is the subscript, UCL i is the upper control limit of the ith subgroup, LCL i is the lower control limit of the ith subgroup, US is the subgroup mean, σ is the subgroup standard deviation, N is the total number of subgroups, and X is the control limit standard multiple; Calculate the production upper control limit based on the first mean value formula, denoted as UCL; Calculate the production lower control limit based on the second mean value formula, denoted as LCL; The first average formula is: , The second average formula is: .

5. An automated emergency management method for intelligent industrial production according to claim 4, characterized in that: The construction of a visualized and intelligent emergency command system through environmental sensors and a real-time command platform specifically includes: Obtain edge computing devices, synchronize data and communications between edge computing devices and the real-time command platform through the database system of the real-time command platform, and be controlled by the real-time command platform; Synchronize the model deployment of the real-time command platform to the edge computing device; The real-time command platform uses monitoring slices to quickly slice the information collected by sensors. Based on the communication needs of each slice, the real-time communication needs and network conditions are used to adjust the resources of each slice for allocation. Obtain the standard signal strength and historical signal strength based on the communication equipment, and set the minimum value of communication signal strength allowed in the real-time command platform; The real-time signal on the production line is transmitted to the real-time command platform and edge computing equipment through the signal detection equipment. The real-time command platform and edge computing equipment monitor the production line signal in real time and issue an alarm when the production line signal is lower than the minimum communication signal strength value; Based on the real-time command platform database, the weights of information collected by different sensors are divided, and a dynamic allocation mechanism is established. The real-time command platform dynamically allocates edge computing and its own signal processing tasks.

6. An automated emergency management method for intelligent industrial production according to claim 5, characterized in that: The method of obtaining the standard operating parameters of the abnormal production line and determining the cause of the abnormal production volume of the abnormal production line in combination with the abnormal production volume feedback report and the operating environment status data specifically includes: Obtain the operating parameters of the production line at different time points to form a data set containing spatiotemporal information, extract useful time features, analyze the spatial relationship between the equipment on the production line, and extract spatial features; The extracted temporal and spatial features are integrated to form comprehensive spatiotemporal features, and feature selection and dimensionality reduction are performed; The spatiotemporal characteristics are divided into multiple quasi-measurement groups, and the historical fault causes are divided into multiple influencing factor groups; By combining several data of the criterion group and the influencing factor group in pairs, a data sample is formed; The average data in the screening data samples is combined with the production line standard data to form the calibration data, and the remaining data samples constitute the test data; Based on the test data and combined with the prediction model, a digital twin model is constructed and trained; Based on the correction data, run the digital twin model, compare the predicted value of the digital twin model with the standard value in the correction data, and make regular corrections to the digital twin model; Input the historical fault causes, fault data and emergency measures execution plan into the digital twin model, run the digital twin model, and quickly simulate the historical abnormal production line. Correct the consistency between the predicted fault data in the digital twin model and the actual historical fault data. If they are consistent, they are recorded in the digital twin model database. If they are inconsistent, correct the digital twin model. When a production line fails, the fault data is input into the digital twin model, the digital twin model is run, the abnormal production line is simulated, and the historical fault data in the digital twin model database is combined. The operation data before and after the abnormal moment simulated by the digital twin model is compared with the benchmark data under normal conditions. Combined with the causes of historical faults in the digital twin model database, the differences in key parameters are analyzed to determine the cause of the abnormality. The causes and possible factors of the abnormality are automatically arranged according to the possibility based on the algorithm of the digital twin model, and synchronized to the real-time command platform to determine the emergency level of the abnormal production line and output an alarm signal of the corresponding degree; The real-time command platform notifies maintenance personnel to check the causes of the abnormality and possible factors one by one, determine the final cause of the failure, and feed it back to the digital twin model, realizing closed-loop feedback of the spatiotemporal characteristics in the digital twin model database.

7. An automated emergency management method for intelligent industrial production according to claim 6, characterized in that: The digital twin model is constructed based on the test data and combined with the prediction model, and the training of the digital twin model specifically includes: The prediction model is: , Where: Y is the simulation value of the digital twin model, ω k is the influence of the kth data, ψ is the total influence coefficient, X k is the kth test data, and ε is the error term.

8. An automated emergency management method for intelligent industrial production according to claim 7, characterized in that: The real-time tracking of the implementation of emergency measures specifically includes: The emergency measures implementation plan is divided into multiple key nodes. The real-time command platform captures the monitoring information of key nodes and feeds the monitoring information back to the digital twin model. The prediction model in the digital twin model is used to predict the real-time status of key nodes and the execution of emergency measures, and the prediction value is fed back to the real-time command platform; The real-time command platform will compare and analyze the real-time status prediction value of key nodes with the signal fluctuation time in the communication status data. If the trigger condition is met, it will immediately issue an early warning to the communication technicians and the emergency command center and adjust the signal transmission part of the emergency measures execution plan. If the trigger condition is not met, no output response will be made. The real-time command platform adjusts the production line part of the emergency measures implementation plan according to the predicted value of the emergency measures implementation plan, and updates the production database at the same time to adjust the upper control limit and the lower control limit of the normal production line.

9. An automated emergency management method for intelligent industrial production according to claim 8, characterized in that: The trigger conditions include: The emergency measures execution progress condition, communication status condition and data transmission packet loss rate, set the first threshold of the emergency measures execution progress condition, set the second threshold of the communication status condition and set the third threshold of the data transmission packet loss rate, and the real-time command platform's emergency measures execution progress condition, communication status condition and data transmission packet loss rate real-time data, any one of which is lower than the corresponding threshold, is triggered; Based on the real-time command platform database storing emergency measures execution data and communication status data, a synchronous task scheduling mechanism is built to achieve unified scheduling of emergency measures execution tasks and communication status monitoring tasks; Based on the task scheduling mechanism, set the priorities of emergency measures execution tasks and communication status monitoring tasks.

10. An automated emergency management method for intelligent industrial production according to claim 9, characterized in that: The first threshold value for setting the progress condition of the emergency measures execution specifically includes: Based on the emergency measures execution plan, the digital twin model is used to quantitatively analyze the emergency measures execution plan, obtain the total emergency measures execution amount, and use the first emergency formula to calculate the first threshold; The first emergency formula is as follows: , Among them: Q1 is the set threshold of the emergency measures execution progress condition, TEO is the total emergency measures execution amount, TPC is the conversion rate between engineering progress, and EC is the emergency level; Setting the second threshold of the communication state condition and setting the third threshold of the data transmission packet loss rate specifically include: Obtain the standard signal strength of the communication equipment, obtain the allowable loss rate of signal transmission based on the real-time command platform database and the digital twin model database, and use the second emergency formula to calculate the first threshold; The second emergency formula is as follows: , Among them: Q2 is the communication status condition threshold, Q3 is the threshold of data transmission packet loss rate, RS is the standard signal strength, SL is the allowable loss rate, and EC is the emergency level.

Images