A method and system for collaborative management of industrial production lines and data center facilities based on digital twins

Abstract

This invention relates to the field of intelligent production control technology, specifically including a method and system for collaborative management of industrial production lines and data center facilities based on digital twins. The method includes: configuring a hybrid network architecture with control, non-control, and management information partitions according to the industrial production line communication environment; constructing a multi-partition equipment digital twin mapping model to form a cross-partition database; generating a collaborative strategy set; and transmitting and distributing the set after simulation and iterative optimization to achieve dynamic collaborative management. This invention solves the technical problems of complex environments in which industrial production lines and data center facilities are located, constantly changing states of these facilities, and the inability to capture real-time response needs, leading to resource waste or insufficient supply. By using a semantic association rule base to achieve intelligent parsing of equipment data, it provides accurate state input to the digital twin model and dynamically adjusts collaborative strategies based on real-time data, achieving efficient collaborative management of production lines and data center facilities.

Description

Technical Field

[0001] This invention relates to the field of intelligent production control technology, specifically to a method and system for collaborative management of industrial production lines and computer room facilities based on digital twins. Background Technology

[0002] Driven by the wave of industrial digital transformation, the level of intelligence in industrial production lines is constantly improving. As a key infrastructure for ensuring power supply, data storage and environmental control, data room facilities are crucial to the coordinated operation with industrial production lines. However, under the traditional industrial management model, production lines and data room facilities operate independently, lacking efficient data interaction and coordination mechanisms.

[0003] Currently, there are many problems in the collaborative management of industrial production lines and data center facilities. There is a lack of systematic collaborative strategies. Traditional operation and maintenance rely heavily on experience and rules, and cannot dynamically adjust the energy supply and environmental protection strategies of data center facilities according to the operating status of the production line. This can easily lead to resource waste or insufficient supply. For example, the overlap of peak electricity consumption of the production line and cooling demand of the data center can cause the power load to exceed the limit.

[0004] In summary, existing technologies suffer from the following problems: the industrial production lines and computer room facilities are located in complex environments, and their status is constantly changing, making it impossible to capture the real-time response of the computer room facilities to demand, resulting in resource waste or insufficient supply. Summary of the Invention

[0005] This application provides a collaborative management method and system for industrial production lines and data center facilities based on digital twins. It aims to solve the technical problems in the existing technology where industrial production lines and data center facilities are located in complex environments, their status is constantly changing, and they cannot capture the response needs of data center facilities in real time, resulting in resource waste or insufficient supply.

[0006] In view of the above problems, the technical solution to achieve the present application is as follows:

[0007] This application provides a method for collaborative management of industrial production lines and data center facilities based on digital twins. The method includes: configuring an adaptive hybrid network architecture based on the communication environment information of the target industrial production line, wherein the adaptive hybrid network architecture includes a control partition, a non-control partition, and a management information partition; constructing a digital twin mapping model of the equipment status in multiple partitions, synchronously mapping the operating data of key industrial production line equipment in the control partition, the data center facility status data in the non-control partition, and the operation and maintenance instructions in the management information partition to the digital twin space, forming a cross-partition status association database; generating a collaborative strategy set containing security constraints and efficiency optimization based on the cross-partition status association database, wherein the collaborative strategy set predefines cross-partition interaction rules, data interaction protocols, and security verification mechanisms; iteratively adjusting the collaborative strategy set according to the simulation execution effect of the digital twin model, and transmitting the optimized applicable collaborative strategies back to the digital twin model and distributing them to the control terminal for dynamic collaborative management of the target industrial production line and data center facilities.

[0008] Preferably, the control partition and the non-control partition are each equipped with a vertical encryption unit; a firewall is provided between the control partition and the non-control partition; the control partition and the non-control partition are connected to the monitoring host; the management information partition is equipped with a publishing server; and the monitoring host and the publishing server are isolated in a forward / reverse manner.

[0009] Preferably, the monitoring host collects device status data from the control partition and non-control partition in real time, filters it through the firewall, and synchronizes it to the digital twin model to trigger a matching of the collaborative policy library. If a suitable collaborative policy is matched, the execution command corresponding to the suitable collaborative policy is issued to the management information partition through the publishing server, and the execution log is updated synchronously.

[0010] Preferably, the system collects equipment operation data from the target industrial production line and real-time monitoring data from the computer room facilities; based on the fusion of BIM model and IoT data, it constructs a three-dimensional geometric model corresponding to the target industrial production line containing equipment spatial coordinates and connection relationships, and a physical attribute model corresponding to the computer room facilities containing equipment power and heat dissipation efficiency; and sets a semantic association rule base according to the equipment operation data and three-dimensional geometric model, the real-time monitoring data and physical attribute model.

[0011] Preferably, based on the semantic association rule base, the collected device operation data and real-time monitoring data are semantically parsed and labeled, and the parsed and labeled data are associated with the corresponding elements in the digital twin model according to the preset mapping rules.

[0012] Preferably, the annotated data is associated with the corresponding elements in the digital twin model according to the preset mapping rules. At the same time, a causal relationship model is trained based on the semantic association rule library to identify the lagging impact of production line load fluctuations on the power demand of the computer room. The association mapping table is dynamically updated to support the automatic access and relationship matching of parameters of newly added equipment.

[0013] Preferably, reinforcement learning is used to iteratively adjust the strategy parameters and combine them with an expert experience rule base for correction; the strategy execution unit of the digital twin model is updated, and the optimized applicable collaborative strategy is sent to the PLC controller and BMS management system of the control terminal.

[0014] Preferably, the analysis focuses on the collaborative needs of the target industrial production line and the data center facilities: identifying resource conflict scenarios; locating efficiency bottlenecks; and determining collaborative optimization points. The resource conflict scenarios characterize the total load exceeding limits when peak electricity demand overlaps with data center cooling power consumption. The efficiency bottlenecks characterize production stagnation when the production line waits for the data center air conditioning to cool to a set temperature before restarting. The collaborative optimization points are the time difference between adjusting the production line start-up / shutdown time and the data center equipment operation mode.

[0015] Preferably, based on the resource conflict scenario, a power load joint scheduling simulation unit is constructed in the digital twin model. The dynamic change curves of power consumption in the production line and power consumption for cooling in the computer room are simulated using Monte Carlo simulation to generate multiple load allocation schemes. Based on the multiple load allocation schemes, with the objective function of minimizing total energy consumption and load fluctuation, and combined with the peak-valley electricity pricing mechanism of the power grid, the power resource optimization allocation strategy under the resource conflict scenario is determined and converted into control commands that are synchronously sent to the production line distribution cabinet and the computer room UPS system for dynamic peak-shaving allocation of power resources.

[0016] In another aspect, this application provides a collaborative management system for industrial production lines and data center facilities based on digital twins. The system includes: a network architecture configuration module for configuring an adaptive hybrid network architecture based on the communication environment information of the target industrial production line. This adaptive hybrid network architecture includes a control partition, a non-control partition, and a management information partition. A synchronization mapping module for constructing a digital twin mapping model of the multi-partition equipment status, synchronously mapping the operating data of key industrial production line equipment in the control partition, the data center facility status data in the non-control partition, and the operation and maintenance instructions in the management information partition to the digital twin space, forming a cross-partition status association database. A strategy set generation module for generating a collaborative strategy set containing security constraints and efficiency optimizations based on the cross-partition status association database. This collaborative strategy set predefines cross-partition interaction rules, data interaction protocols, and security verification mechanisms. A collaborative management module for iteratively adjusting the collaborative strategy set based on the simulation execution effect of the digital twin model, transmitting the optimized applicable collaborative strategies back to the digital twin model, and distributing them to the control terminal for dynamic collaborative management of the target industrial production line and data center facilities.

[0017] In summary, one or more technical solutions provided in this application achieve the technical effect of secure isolation and controllable interaction of cross-partition data through a hybrid network architecture and vertical encryption units, intelligent parsing of equipment data through a semantic association rule base, providing accurate status input for the digital twin model, and dynamically adjusting collaborative strategies based on real-time data, thereby achieving efficient collaborative management of production lines and data center facilities. Attached Figure Description

[0018] Figure 1 This application provides a flowchart illustrating a collaborative management method for industrial production lines and data center facilities based on digital twins.

[0019] Figure 2 This application provides a schematic diagram of the structure of a collaborative management system for industrial production lines and data center facilities based on digital twins.

[0020] Explanation of reference numerals in the attached diagram: Network architecture configuration module M100, synchronization mapping module M200, policy set generation module M300, and collaborative management module M400. Detailed Implementation

[0021] Example 1

[0022] The present application will now be described in detail with reference to the accompanying drawings, such as... Figure 1 As shown, this application provides a method for collaborative management of industrial production lines and data center facilities based on digital twins, wherein the method includes:

[0023] S1: Based on the communication environment information of the target industrial production line, configure an adaptive hybrid network architecture, which includes a control partition, a non-control partition, and a management information partition; S2: Construct a digital twin mapping model of the equipment status of multiple partitions, and synchronously map the operating data of key industrial production line equipment in the control partition, the data center facility status data of the non-control partition, and the operation and maintenance instructions of the management information partition to the digital twin space to form a cross-partition status association database.

[0024] Specifically, a hybrid network architecture refers to dividing an industrial network into control zones, non-control zones, and management information zones. Control zones are used to directly control the operation of industrial equipment, non-control zones are used for monitoring and auxiliary control, and management information zones are used to process management-related data, such as operation and maintenance instructions. A digital twin mapping model is a technology that reflects the status of physical equipment in real time through a digital model. By synchronously mapping data from different zones to a digital twin space, a cross-zone status association database is formed, which enables centralized management and correlation analysis of the operation data of key equipment in industrial production lines, data center facilities, and operation and maintenance instructions.

[0025] Execution Steps: Based on the communication environment information of the target industrial production line, configure an adapted hybrid network architecture, including control zones, non-control zones, and management information zones. This process allows for flexible configuration of the network structure according to the actual situation of the industrial production line, ensuring the security and compatibility of data interaction. For example, in an industrial production line with multiple devices, configuring a hybrid network architecture reduces data transmission latency between different zones, thereby improving the efficiency of collaborative management. Construct a digital twin mapping model of the multi-zone device status, synchronously mapping the operating data of key industrial production line equipment in the control zone, the status data of data center facilities in the non-control zone, and the operation and maintenance instructions in the management information zone to the digital twin space, forming a cross-zone status association database. This mapping process enables deep integration of physical entities and virtual models, providing accurate data support for subsequent collaborative strategy formulation.

[0026] S3: Based on the cross-partition state association database, generate a collaborative strategy set that includes security constraints and efficiency optimization. The collaborative strategy set predefines cross-partition interaction rules, data interaction protocols, and security verification mechanisms. S4: Based on the simulation execution effect of the digital twin model, iteratively adjust the collaborative strategy set, and send the optimized applicable collaborative strategies back to the digital twin model and distribute them to the control terminal to perform dynamic collaborative management of the target industrial production line and computer room facilities.

[0027] Specifically, the collaborative strategy set refers to a set of strategies generated based on data in a cross-partition state association database, which includes security constraints and efficiency optimizations. Security constraints refer to the security rules and restrictions that must be followed during data interaction and collaborative management, such as data encryption and access control. Efficiency optimization refers to improving the efficiency and response speed of collaborative management by optimizing algorithms and strategies. Cross-partition interaction rules refer to the rules for data interaction between different partitions, such as data transmission priority and format requirements. Data interaction protocols refer to the protocols that need to be followed during data transmission, such as TCP / IP and MQTT protocols. Security verification mechanisms refer to methods for security verification of interactive data and collaborative operations, such as digital signatures and key verification. The simulation execution effect of the digital twin model refers to using the digital twin model to simulate and test the collaborative strategies and observe their execution effects, such as system response time and resource utilization. Iterative adjustment refers to optimizing and adjusting the collaborative strategy set based on the simulation execution effect to improve the performance of collaborative management. Dynamic collaborative management refers to dynamically adjusting the collaborative management strategy based on real-time data and simulation results to achieve efficient collaboration between industrial production lines and data center facilities.

[0028] Execution Steps: Based on a cross-partition state association database, a set of collaborative strategies, including security constraints and efficiency optimizations, is generated. This process analyzes data in the cross-partition state association database, predefines cross-partition interaction rules, defines data interaction protocols, and establishes a security verification mechanism to generate the collaborative strategy set. The generation of the collaborative strategy set improves the reliability of cross-partition data interaction and performs security verification on collaborative operations, ensuring the security and standardization of collaborative management. Based on the simulation execution effect of the digital twin model, the collaborative strategy set is iteratively adjusted. The collaborative strategies are simulated and tested using the digital twin model, and optimized based on the simulation results. The preferred strategy, found through simulation execution to have a long response time when handling data center facility failures, is shortened through iterative adjustments. The optimized applicable collaborative strategies are then fed back to the digital twin model and distributed to the control terminal for dynamic collaborative management of the target industrial production line and data center facilities. This dynamic collaborative management enables real-time monitoring and dynamic adjustment of industrial production lines and data center facilities, improving the overall system's collaborative efficiency.

[0029] Furthermore, the method of this application includes:

[0030] The control partition and the non-control partition are each equipped with a vertical encryption unit; a firewall is provided between the control partition and the non-control partition; the control partition and the non-control partition are connected to the monitoring host; the management information partition is equipped with a publishing server; and the monitoring host and the publishing server are isolated in a forward / reverse manner.

[0031] Specifically, a vertical encryption unit is a device used to encrypt data, typically deployed in both control and non-control partitions to ensure the confidentiality and integrity of data during transmission. A firewall is a network security system used to monitor and control data flows into and out of the network, preventing unauthorized access and malicious attacks. Here, the firewall is positioned between the control and non-control partitions to isolate network traffic between them, preventing potential security threats from spreading between partitions. A monitoring host is a system used to monitor and manage the status of industrial production line equipment in real time. It connects to both control and non-control partitions and can collect and analyze equipment operating data. A publishing server is used to send data and instructions to the management information partition. It is located in the management information partition and connected to the monitoring host via forward / reverse isolation to ensure the security and unidirectionality of data transmission. Forward / reverse isolation is a network security technology used to restrict the direction of data transmission and prevent the leakage of sensitive information.

[0032] Execution Steps: The control and non-control partitions are each equipped with vertical encryption units. This effectively protects data security within each partition, preventing data theft or tampering during transmission. For example, in an industrial scenario with multiple production lines and data center facilities, vertical encryption units can encrypt the operational data of critical equipment, ensuring that only authorized monitoring hosts can decrypt and access this data, thus reducing the risk of data leakage. A firewall is installed between the control and non-control partitions to prevent unauthorized access and malicious attacks from spreading between partitions, improving the overall network security. For example, in an industrial network, a firewall can prevent potential malicious attacks from the non-control partition from entering the control partition, protecting the normal operation of critical equipment. The control and non-control partitions are connected to the monitoring host. The management information partition is equipped with a publishing server, and the monitoring host is forward / reverse isolated from the publishing server. This architecture ensures that the monitoring host can collect and analyze equipment status data from each partition in real time and securely transmit relevant data and instructions to the management information partition through the publishing server, while preventing the leakage of sensitive information and improving the security and reliability of collaborative management of industrial production lines and data center facilities.

[0033] Furthermore, the method of this application includes:

[0034] The monitoring host collects device status data in real time from the control partition and non-control partition, filters it through the firewall, and synchronizes it to the digital twin model to trigger a matching of the collaborative policy library. If a suitable collaborative policy is matched, the execution command corresponding to the suitable collaborative policy is issued to the management information partition through the publishing server, and the execution log is updated synchronously.

[0035] Specifically, a monitoring host refers to a computer system installed in the control center, used to monitor and manage the operational status of industrial production lines and computer room facilities in real time. It connects to equipment in various zones via a network and can collect equipment status data such as temperature, humidity, and power consumption. Firewall filtering refers to the firewall filtering data packets according to preset security rules, blocking packets that do not conform to the rules, thereby protecting the internal network from external threats. A digital twin model is a virtual model used to reflect the operational status and behavior of physical equipment in real time. A collaborative policy library is a database storing various collaborative management policies, which define how to coordinate the operation of industrial production lines and computer room facilities under different circumstances. A publishing server is a server used to send data and instructions to the management information zone, connected to the monitoring host, and ensuring the security of data transmission through forward / reverse isolation. An execution log is a detailed log recording the system's operations, used to track and audit the system's operation.

[0036] Execution Steps: The monitoring host collects real-time device status data from both the control and non-control zones. After filtering by the firewall, this data is synchronized to the digital twin model, triggering a matching of the collaborative policy library. This process ensures that only security-filtered data enters the digital twin model, thereby improving data security and reliability. For example, in an industrial production line, the monitoring host can collect multiple device status data points per second. After firewall filtering, potentially malicious data is effectively prevented from entering the digital twin model, ensuring the model's accuracy and security. If an applicable collaborative policy is matched, the publishing server issues the corresponding execution command to the management information zone and simultaneously updates the execution log. This ensures the timely execution and recording of the collaborative policy, facilitating subsequent auditing and optimization. Furthermore, in data center facilities, when the power load exceeds a threshold, a corresponding energy-saving policy is automatically matched. The publishing server issues an execution command to the management information zone to adjust the equipment's operating mode, and the operation is recorded in the execution log, improving power resource utilization and ensuring traceability. This, in turn, enhances the efficiency and security of collaborative management between industrial production lines and data center facilities.

[0037] Furthermore, after being filtered by a firewall and synchronized to the digital twin model, the method of this application also includes:

[0038] Collect equipment operation data and real-time monitoring data of computer room facilities from the target industrial production line; based on the fusion of BIM model and IoT data, construct a three-dimensional geometric model of the target industrial production line containing equipment spatial coordinates and connection relationships, and a physical attribute model of the computer room facilities containing equipment power and heat dissipation efficiency; set a semantic association rule base according to the equipment operation data and three-dimensional geometric model, the real-time monitoring data and physical attribute model.

[0039] Specifically, firewall filtering refers to a firewall screening data packets according to preset security rules, blocking packets that do not conform to the rules from passing through, thereby protecting the internal network from external threats. A digital twin model is a virtual model used to reflect the real-time operating status and behavior of physical equipment. Equipment operation data refers to the operating status information of equipment in an industrial production line, such as temperature, pressure, and speed; real-time monitoring data of computer room facilities refers to the operating status information of equipment in the computer room, such as power consumption, temperature, and humidity; a BIM model is a building information model used to represent the three-dimensional structure and related information of a building or industrial facility; IoT data fusion refers to integrating IoT data from different sensors and devices to obtain more comprehensive equipment status information. A three-dimensional geometric model is a three-dimensional model representing the spatial location and connectivity of equipment; a physical attribute model is a model representing the physical characteristics of equipment, such as power and heat dissipation efficiency. A semantic association rule base is a database storing semantic association rules used to associate equipment operation data and monitoring data with elements in the digital twin model.

[0040] Execution steps: After being filtered by a firewall, the data is synchronized to the digital twin model. The method also includes collecting equipment operation data from the target industrial production line and real-time monitoring data from the data center facilities. This process ensures that only data that has passed security filtering can enter the digital twin model, thereby improving data security and reliability. Based on the fusion of BIM model and IoT data, a 3D geometric model containing equipment spatial coordinates and connection relationships is constructed for the target industrial production line, and a physical attribute model containing equipment power and heat dissipation efficiency is constructed for the data center facilities. The construction of the physical attribute model can provide a more intuitive display of equipment status and a more accurate analysis of physical characteristics. Specifically, in a data center facility, by fusing BIM model and IoT data, a 3D geometric model of the data center equipment can be constructed, intuitively displaying the spatial location and connection relationships of the equipment. At the same time, combined with the physical attribute model, the power and heat dissipation efficiency of the equipment can be accurately analyzed, thereby improving equipment management efficiency.

[0041] Based on the equipment operation data and 3D geometric model, the real-time monitoring data and physical attribute model, a semantic association rule base is set up. Through the semantic association rule base, the collected data is associated with the elements in the digital twin model to realize the semantic and intelligent processing of the data. In the above steps, through the semantic association rule base, the temperature data of the equipment can be associated with the location of the equipment in the 3D geometric model, thereby realizing real-time monitoring and early warning of the equipment temperature and improving the operational safety and management efficiency of the equipment.

[0042] Furthermore, the method of this application includes:

[0043] Based on the semantic association rule base, the collected device operation data and real-time monitoring data are semantically parsed and labeled, and the parsed and labeled data are associated with the corresponding elements in the digital twin model according to the preset mapping rules.

[0044] Specifically, semantic parsing and annotation refers to performing semantic analysis and annotation on collected equipment operation data and real-time monitoring data, assigning specific meanings and labels to the data to make it more readable and understandable. For example, temperature data can be labeled as a high-temperature warning or within the normal range. Pre-defined mapping rules refer to a set of pre-set rules used to associate the parsed and annotated data with corresponding elements in the digital twin model. Corresponding elements in the digital twin model are virtual objects that represent actual equipment or facilities, possessing attributes and behaviors corresponding to the actual equipment.

[0045] Execution Steps: Based on the semantic association rule base, the collected equipment operation data and real-time monitoring data are semantically parsed and labeled. This semantic annotation transforms the raw data into information with clear semantics, thereby improving data usability and understandability. Specifically, through semantic annotation, equipment temperature data can be labeled as normal, warning, or dangerous, allowing managers to intuitively understand the equipment's operating status. The parsed and labeled data is then associated with corresponding elements in the digital twin model according to preset mapping rules. This association mechanism ensures that the digital twin model reflects the actual equipment status in real time, providing accurate data support for collaborative management. Specifically, in a data center facility, preset mapping rules can associate the power consumption data of the data center equipment with equipment elements in the digital twin model, enabling the model to display the equipment's power consumption in real time. This improves the efficiency and accuracy of collaborative management, making the collaboration between industrial production lines and data center facilities more intelligent and efficient.

[0046] Furthermore, the method of this application includes:

[0047] The annotated data is associated with the corresponding elements in the digital twin model according to the preset mapping rules. At the same time, a causal relationship model is trained based on the semantic association rule library to identify the lagging impact of production line load fluctuations on the power demand of the computer room. The association mapping table is dynamically updated to support the automatic access and relationship matching of parameters of newly added equipment.

[0048] Specifically, preset mapping rules refer to pre-defined rules used to associate semantically parsed and labeled data with corresponding elements in the digital twin model; corresponding elements in the digital twin model refer to virtual objects in the model that represent actual equipment or facilities, and these objects have attributes and behaviors corresponding to the actual equipment; the semantic association rule base is a database that stores semantic association rules to guide the semantic parsing and labeling process of data. A causal relationship model is a model used to identify and analyze causal relationships between different data, such as identifying the impact of production line load fluctuations on the power demand of the computer room; dynamically updating the association mapping table refers to a mechanism that automatically updates the association relationship between data and the digital twin model based on real-time data and system changes; automatic parameter access and relationship matching for newly added equipment means that when new equipment is added to the system, the equipment parameters can be automatically identified and matched and associated with corresponding elements in the digital twin model.

[0049] Execution Steps: The annotated data is associated with corresponding elements in the digital twin model according to preset mapping rules. Simultaneously, a causal relationship model is trained based on a semantic association rule base to identify the lagging impact of production line load fluctuations on the power demand of the data center. Preset mapping rules ensure accurate association between the data and the digital twin model. The causal relationship model is also used to analyze the relationship between production line load fluctuations and data center power demand. For example, the causal relationship model identifies a lagging increase in data center power demand after a dynamic time window when the production line speed increases, allowing for proactive adjustments to the data center power supply to avoid power shortages. The association mapping table is dynamically updated, supporting automatic parameter access and relationship matching for newly added equipment. This dynamic update mechanism ensures that when new equipment is added, the system automatically identifies its parameters and matches and associates them with corresponding elements in the digital twin model, improving the system's scalability and adaptability. Furthermore, the system automatically identifies its parameters and creates corresponding virtual elements in the digital twin model, while simultaneously updating the association mapping table, enabling new equipment to immediately participate in collaborative management and improving the overall system's collaborative efficiency.

[0050] Furthermore, based on the simulation execution effect of the digital twin model, the collaborative strategy set is iteratively adjusted, and the optimized applicable collaborative strategy is sent back to the digital twin model and distributed to the control terminal. This application also includes:

[0051] The strategy parameters are iteratively adjusted using reinforcement learning and corrected by combining them with an expert experience rule base; the strategy execution unit of the digital twin model is updated, and the optimized applicable collaborative strategy is sent to the PLC controller and BMS management system of the control terminal.

[0052] Specifically, reinforcement learning is a machine learning method that learns the optimal policy through the interaction between an agent and its environment, iteratively adjusting policy parameters to maximize cumulative rewards; an expert experience rule base refers to a set of validated rules and experiences summarized by domain experts to guide system decision-making and policy optimization; the policy execution unit of a digital twin model refers to the part of the digital twin model responsible for executing collaborative policies, controlling and adjusting the physical system according to received policy instructions; a PLC controller is a programmable logic controller, widely used in industrial automation to control the operation of mechanical equipment; a BMS management system refers to a building management system or battery management system used to monitor and manage the electromechanical equipment or battery status of a building.

[0053] Execution Steps: The strategy parameters are iteratively adjusted using reinforcement learning and refined using an expert rule base. The reinforcement learning algorithm continuously optimizes the parameters of the collaborative strategy, improving its adaptability and effectiveness. Simultaneously, the strategy is refined using the expert rule base to ensure it meets the needs of the actual industrial scenario. Preferably, the collaborative strategy is fine-tuned based on rules from the expert rule base, resulting in more precise temperature control in the computer room. The strategy execution unit of the digital twin model is updated, and the optimized applicable collaborative strategy is sent to the PLC controller and BMS management system at the control terminal. Updating the strategy execution unit of the digital twin model ensures that the model can accurately execute the optimized collaborative strategy. Furthermore, the updated strategy execution unit can precisely control the start-up and stop times of the production line according to the optimized collaborative strategy, avoiding excessive operation of computer room equipment and extending equipment lifespan. The optimized collaborative strategy is sent to the PLC controller and BMS management system to achieve precise control of the industrial production line and computer room facilities. Specifically, through the BMS management system, the operating mode of the computer room air conditioning can be adjusted according to the collaborative strategy, ensuring that the computer room temperature is always within the optimal range while reducing energy consumption. The above steps make the collaborative management of industrial production lines and computer room facilities more intelligent and efficient.

[0054] Furthermore, the method of this application includes dynamic collaborative management of target industrial production lines and computer room facilities:

[0055] Analyze the collaborative needs of the target industrial production line and the data center facilities: identify resource conflict scenarios; locate efficiency bottlenecks; determine collaborative optimization points; wherein, the resource conflict scenarios are used to characterize the total load exceeding the limit when peak electricity demand overlaps with data center cooling electricity demand; the efficiency bottlenecks are used to characterize the production line stagnation when waiting for the data center air conditioning to cool down to the set temperature before starting; the collaborative optimization points are the time difference between adjusting the start-up and shutdown time of the production line and the operating mode of the data center equipment.

[0056] Specifically, in the field of industrial collaborative management, resource conflict scenarios refer to resource competition caused by unreasonable resource allocation or overlapping peak demand during the operation of industrial production lines and data center facilities, such as exceeding power load limits; efficiency bottlenecks refer to situations where the overall collaborative efficiency is affected by the inefficiency of a certain link in the collaborative management process, such as a production line stopping while waiting for the data center air conditioning to cool down; collaborative optimization points refer to key points or links in collaborative management that can be optimized, such as adjusting the time difference between the start-up and shutdown time of the production line and the operating mode of the data center equipment; by identifying these scenarios and links, a basis can be provided for the formulation of subsequent collaborative strategies.

[0057] Execution Steps: Analyze the collaborative needs of the target industrial production line and data center facilities, identify resource conflict scenarios, pinpoint efficiency bottlenecks, and determine collaborative optimization points. Through system analysis, identify situations where peak electricity consumption overlaps with data center cooling power consumption, resulting in total load exceeding limits. Specifically, analysis reveals that during high-temperature periods in summer, peak electricity consumption of the industrial production line overlaps with peak electricity consumption of the data center cooling system, causing the total power load to exceed the power supply capacity and posing a risk of insufficient power supply. Simultaneously, identifying efficiency bottlenecks reveals that the production line waits for the data center air conditioning to cool to the set temperature before starting, leading to an average production downtime. Therefore, the collaborative optimization point is determined to be adjusting the time difference between the production line start-up / shutdown time and the data center equipment operating mode. Specifically, through optimization, the data center air conditioning system is started earlier for pre-cooling, enabling the production line to start up in a timely manner.

[0058] Furthermore, the method of this application also includes:

[0059] Based on the resource conflict scenario, a power load joint scheduling simulation unit is constructed in the digital twin model. The dynamic change curves of power consumption in the production line and power consumption for cooling in the computer room are simulated using Monte Carlo simulation to generate multiple load allocation schemes. Based on the multiple load allocation schemes, with the objective function of minimizing total energy consumption and load fluctuation, and combined with the peak-valley electricity pricing mechanism of the power grid, the power resource optimization allocation strategy under the resource conflict scenario is determined and converted into control commands that are synchronously sent to the production line distribution cabinet and the computer room UPS system for dynamic peak-shifting of power resources.

[0060] Specifically, the power load joint dispatch simulation unit refers to a module built in the digital twin model specifically for simulating and optimizing power load allocation. By simulating different power load allocation schemes, it helps identify the optimal power resource dispatch strategy. Monte Carlo simulation is a numerical simulation method based on probability statistics. It uses random sampling to simulate the dynamic change curves of power consumption in production lines and power consumption for cooling in computer rooms, thereby generating multiple load allocation schemes. The objective function refers to the goal to be achieved during the optimization process, such as minimizing total energy consumption and load fluctuation. The power grid peak-valley pricing mechanism refers to rationally arranging power consumption time according to the peak and off-peak periods of the power grid load to reduce the pressure on the power grid load. The UPS system is an uninterruptible power supply system used to provide temporary power support when the power supply is interrupted, ensuring the normal operation of critical equipment.

[0061] Execution steps: The resource conflict scenario is used to characterize the total load exceeding the limit when peak electricity demand overlaps with computer room cooling electricity demand. Further, based on the resource conflict scenario, a power load joint scheduling simulation unit is constructed in the digital twin model. The dynamic change curves of production line electricity demand and computer room cooling electricity demand are simulated through Monte Carlo simulation to generate multiple load allocation schemes. Different power load allocation schemes are simulated through the simulation unit to provide data support for subsequent optimization strategies. For example, through Monte Carlo simulation, multiple different load allocation schemes are generated, covering the changes in production line electricity demand and computer room cooling electricity demand in different time periods, thus providing a rich data foundation for optimizing power resource allocation.

[0062] Based on the aforementioned multiple load allocation schemes, with the objective functions of minimizing total energy consumption and load fluctuation, and in conjunction with the power grid peak-valley pricing mechanism, an optimized power resource allocation strategy for resource conflict scenarios is determined. This strategy is then converted into control commands and synchronously sent to the production line distribution cabinets and the computer room UPS system to achieve dynamic peak-shaving allocation of power resources, thereby improving the utilization efficiency of power resources. After being converted into control commands, the power resources are sent to the production line distribution cabinets and the computer room UPS system, realizing dynamic peak-shaving allocation of power resources, ensuring the stability and reliability of power supply, improving the utilization efficiency of power resources, and enhancing the overall performance of the system.

[0063] The efficiency bottleneck is used to characterize the production line stagnation when it waits for the computer room air conditioning to cool down to the set temperature before restarting. Furthermore, for the identified efficiency bottleneck, a dynamic correlation model between the production line start-up temperature threshold and the computer room air conditioning cooling efficiency is established based on real-time data feedback from the digital twin model. A grey prediction algorithm is used to predict the computer room cooling time based on historical data, and combined with the production line production plan, the air conditioning operating parameters are adjusted in advance to pre-lower the computer room temperature to the critical value. Simultaneously, the production line start-up time and energy consumption changes under different temperature thresholds are simulated in the digital twin model, and the optimal temperature control range is determined through sensitivity analysis, achieving seamless integration between the production line and computer room equipment.

[0064] The collaborative optimization point is the time difference between adjusting the production line start-up and shutdown time and the machine room equipment operation mode. Further, based on the determined collaborative optimization point, a cross-system linkage control mechanism is constructed. Collaborative control triggers are set in the digital twin model; when a change in the production line production plan is detected, the machine room equipment operation mode adjustment process is automatically triggered. Utilizing the simulation and deduction capabilities of the digital twin model, the collaborative effects of production line start-up and shutdown and machine room equipment operation mode switching under different time differences are simulated, generating a collaborative control scheme that includes equipment start-up and shutdown sequence and parameter adjustment range. Control commands are sent to the production line PLC controller and the machine room BMS management system via the OPC UA protocol to achieve precise collaboration between the production line and machine room facilities.

[0065] Subsequently, a collaborative management effectiveness evaluation system was established. A key performance indicator (KPI) monitoring panel was set up in the digital twin model to collect data in real time, including production line efficiency, data room energy consumption, and equipment failure rate. The entropy weight method was used to determine the weights of each KPI, and a comprehensive evaluation model was constructed to quantitatively score the implementation effect of the collaborative management strategy. When the score falls below a preset threshold, a strategy optimization process is automatically triggered. Combining the simulation results of the digital twin model with actual operating data, a genetic algorithm is used to iteratively optimize the collaborative strategy, continuously improving the collaborative management level of industrial production lines and data center facilities.

[0066] In summary, the beneficial effects of the embodiments of this application are:

[0067] By employing a hybrid network architecture adapted to the communication environment information of the target industrial production line, including control zones, non-control zones, and management information zones, a digital twin mapping model of multi-zone equipment status is constructed. This model synchronously maps the operating data of key industrial production line equipment in the control zone, the status data of data center facilities in the non-control zone, and the operation and maintenance instructions in the management information zone to the digital twin space, forming a cross-zone status association database. Based on this database, a collaborative strategy set containing security constraints and efficiency optimizations is generated. This strategy set predefines cross-zone interaction rules, data interaction protocols, and security verification mechanisms. The collaborative strategy set is iteratively adjusted based on the simulation execution effect of the digital twin model. The optimized and applicable collaborative strategies are then fed back to the digital twin model and distributed to the control terminal for dynamic collaborative management of the target industrial production line and data center facilities. This application provides a method and system for collaborative management of industrial production lines and data center facilities based on digital twins. Through a hybrid network architecture and vertical encryption units, it achieves secure isolation and controllable interaction of cross-zone data, intelligent parsing of equipment data through a semantic association rule base, providing accurate status input for the digital twin model, and dynamically adjusting collaborative strategies based on real-time data, thereby achieving the technical effect of efficient collaborative management of production lines and data center facilities.

[0068] Example 2

[0069] Based on the same inventive concept as the digital twin-based collaborative management method for industrial production lines and data center facilities in the foregoing embodiments, such as Figure 2 As shown in the figure, this application provides a collaborative management system for industrial production lines and data center facilities based on digital twins, wherein the system includes:

[0070] The network architecture configuration module M100 is used to configure an adaptive hybrid network architecture based on the communication environment information of the target industrial production line. The adaptive hybrid network architecture includes a control partition, a non-control partition, and a management information partition.

[0071] The synchronization mapping module M200 is used to construct a digital twin mapping model of the status of equipment in multiple zones. It synchronously maps the operating data of key equipment in the industrial production line of the control zone, the status data of the computer room facilities in the non-control zone, and the operation and maintenance instructions of the management information zone to the digital twin space, forming a cross-zone status association database.

[0072] The strategy set generation module M300 is used to generate a collaborative strategy set containing security constraints and efficiency optimizations based on the cross-partition state association database. The collaborative strategy set predefines cross-partition interaction rules, data interaction protocols, and security verification mechanisms.

[0073] The collaborative management module M400 is used to iteratively adjust the collaborative strategy set based on the simulation execution effect of the digital twin model, and send the optimized applicable collaborative strategies back to the digital twin model and distribute them to the control terminal to perform dynamic collaborative management of the target industrial production line and computer room facilities.

[0074] Furthermore, the digital twin-based industrial production line and data center facility collaborative management system is also used to perform the following methods:

[0075] The control partition and the non-control partition are each equipped with a vertical encryption unit; a firewall is provided between the control partition and the non-control partition; the control partition and the non-control partition are connected to the monitoring host; the management information partition is equipped with a publishing server; and the monitoring host and the publishing server are isolated in a forward / reverse manner.

[0076] Furthermore, the digital twin-based industrial production line and data center facility collaborative management system is also used to perform the following methods:

[0077] The monitoring host collects device status data in real time from the control partition and non-control partition, filters it through the firewall, and synchronizes it to the digital twin model to trigger a matching of the collaborative policy library. If a suitable collaborative policy is matched, the execution command corresponding to the suitable collaborative policy is issued to the management information partition through the publishing server, and the execution log is updated synchronously.

[0078] Furthermore, the digital twin-based industrial production line and data center facility collaborative management system is also used to perform the following methods:

[0079] Collect equipment operation data and real-time monitoring data of computer room facilities from the target industrial production line; based on the fusion of BIM model and IoT data, construct a three-dimensional geometric model of the target industrial production line containing equipment spatial coordinates and connection relationships, and a physical attribute model of the computer room facilities containing equipment power and heat dissipation efficiency; set a semantic association rule base according to the equipment operation data and three-dimensional geometric model, the real-time monitoring data and physical attribute model.

[0080] Furthermore, the digital twin-based industrial production line and data center facility collaborative management system is also used to perform the following methods:

[0081] Based on the semantic association rule base, the collected device operation data and real-time monitoring data are semantically parsed and labeled, and the parsed and labeled data are associated with the corresponding elements in the digital twin model according to the preset mapping rules.

[0082] Furthermore, the digital twin-based industrial production line and data center facility collaborative management system is also used to perform the following methods:

[0083] The annotated data is associated with the corresponding elements in the digital twin model according to the preset mapping rules. At the same time, a causal relationship model is trained based on the semantic association rule library to identify the lagging impact of production line load fluctuations on the power demand of the computer room. The association mapping table is dynamically updated to support the automatic access and relationship matching of parameters of newly added equipment.

[0084] Furthermore, the collaborative management module M400 is also used to perform the following methods:

[0085] The strategy parameters are iteratively adjusted using reinforcement learning and corrected by combining them with an expert experience rule base; the strategy execution unit of the digital twin model is updated, and the optimized applicable collaborative strategy is sent to the PLC controller and BMS management system of the control terminal.

[0086] Furthermore, the collaborative management module M400 is also used to perform the following methods:

[0087] Analyze the collaborative needs of the target industrial production line and the data center facilities: identify resource conflict scenarios; locate efficiency bottlenecks; determine collaborative optimization points; wherein, the resource conflict scenarios are used to characterize the total load exceeding the limit when peak electricity demand overlaps with data center cooling electricity demand; the efficiency bottlenecks are used to characterize the production line stagnation when waiting for the data center air conditioning to cool down to the set temperature before starting; the collaborative optimization points are the time difference between adjusting the start-up and shutdown time of the production line and the operating mode of the data center equipment.

[0088] Furthermore, the collaborative management module M400 is also used to perform the following methods:

[0089] Based on the resource conflict scenario, a power load joint scheduling simulation unit is constructed in the digital twin model. The dynamic change curves of power consumption in the production line and power consumption for cooling in the computer room are simulated using Monte Carlo simulation to generate multiple load allocation schemes. Based on the multiple load allocation schemes, with the objective function of minimizing total energy consumption and load fluctuation, and combined with the peak-valley electricity pricing mechanism of the power grid, the power resource optimization allocation strategy under the resource conflict scenario is determined and converted into control commands that are synchronously sent to the production line distribution cabinet and the computer room UPS system for dynamic peak-shifting of power resources.

[0090] In summary, any step can be stored as a computer instruction or program in an unrestricted computer memory and can be called and recognized by an unrestricted computer processor; no further restrictions are imposed here.

[0091] Furthermore, the above technical solutions only embody the preferred technical solutions of the embodiments of this application. Any changes that those skilled in the art may make to certain parts of these solutions embody the novel principles of the embodiments of this application. Obviously, those skilled in the art can make various modifications and variations to this application without departing from the scope of this application.

Claims

1. A collaborative management method for industrial production lines and data center facilities based on digital twins, characterized in that, The method includes: Based on the communication environment information of the target industrial production line, an adaptive hybrid network architecture is configured. The adaptive hybrid network architecture includes a control partition, a non-control partition, and a management information partition. The control partition is used to control the operation of industrial equipment, and the non-control partition is used for monitoring and auxiliary control. A digital twin mapping model for the status of equipment in multiple zones is constructed, and the operation data of key industrial production line equipment in the control zone, the status data of computer room facilities in the non-control zone, and the operation and maintenance instructions of the management information zone are synchronously mapped to the digital twin space to form a cross-zone status association database. Based on the cross-partition state association database, a collaborative strategy set containing security constraints and efficiency optimization is generated. The collaborative strategy set predefines cross-partition interaction rules, data interaction protocols, and security verification mechanisms. Cross-partition interaction rules refer to the rules for data interaction between different partitions. Based on the simulation results of the digital twin model, the set of collaborative strategies is iteratively adjusted, and the optimized applicable collaborative strategies are sent back to the digital twin model and distributed to the control terminal for dynamic collaborative management of the target industrial production line and the computer room facilities. The method for dynamic collaborative management of target industrial production lines and computer room facilities includes: Analyze the collaborative needs of the target industrial production line and data center facilities: identify resource conflict scenarios; pinpoint efficiency bottlenecks; determine collaborative optimization points; The resource conflict scenario is used to characterize the total load exceeding the limit when the peak electricity demand of the target industrial production line overlaps with the power consumption of the computer room cooling system; the efficiency bottleneck is used to characterize the production line stagnation when it waits for the computer room air conditioning to cool down to the set temperature before starting up; and the collaborative optimization point is to adjust the time difference between the production line start-up and shutdown time and the computer room equipment operation mode.

2. The collaborative management method for industrial production lines and data center facilities based on digital twins as described in claim 1, characterized in that, The control partition and the non-control partition are each equipped with a vertical encryption unit; A firewall is provided between the control partition and the non-control partition; the control partition and the non-control partition are connected to the monitoring host; the management information partition is equipped with a publishing server; and the monitoring host and the publishing server are isolated in a forward / reverse manner.

3. The collaborative management method for industrial production lines and data center facilities based on digital twins as described in claim 2, characterized in that, The monitoring host collects device status data in the control and non-control zones in real time, filters it through the firewall, and synchronizes it to the digital twin model to trigger matching of the collaborative policy library. If a suitable collaboration strategy is matched, the execution command corresponding to the applicable collaboration strategy is issued to the management information partition through the publishing server, and the execution log is updated synchronously.

4. The collaborative management method for industrial production lines and data center facilities based on digital twins as described in claim 3, characterized in that, After being filtered by a firewall, the data is synchronized to the digital twin model. The method further includes: Collect equipment operation data and real-time monitoring data of computer room facilities from the target industrial production line; Based on the fusion of BIM model and IoT data, a three-dimensional geometric model with equipment spatial coordinates and connection relationships is constructed for the target industrial production line, and a physical attribute model with equipment power and heat dissipation efficiency is constructed for the computer room facilities. A semantic association rule base is set up based on the equipment operation data and three-dimensional geometric model, the real-time monitoring data and physical attribute model.

5. The collaborative management method for industrial production lines and data center facilities based on digital twins as described in claim 4, characterized in that, Based on the semantic association rule base, the collected device operation data and real-time monitoring data are semantically parsed and labeled, and the parsed and labeled data are associated with the corresponding elements in the digital twin model according to the preset mapping rules.

6. The collaborative management method for industrial production lines and data center facilities based on digital twins as described in claim 5, characterized in that, The annotated data is associated with the corresponding elements in the digital twin model according to the preset mapping rules. At the same time, a causal relationship model is trained based on the semantic association rule library to identify the lagged impact of production line load fluctuations on the power demand of the computer room. The association mapping table is dynamically updated, supporting automatic parameter access and relationship matching for newly added devices.

7. The collaborative management method for industrial production lines and data center facilities based on digital twins as described in claim 3, characterized in that, Based on the simulation results of the digital twin model, the collaborative strategy set is iteratively adjusted, and the optimized applicable collaborative strategies are sent back to the digital twin model and distributed to the control terminal. The method further includes: The strategy parameters are iteratively adjusted using reinforcement learning and further refined using an expert rule base. The strategy execution unit of the digital twin model is updated, and the optimized applicable collaborative strategy is sent to the PLC controller and BMS management system of the control terminal.

8. The method for collaborative management of industrial production lines and data center facilities based on digital twins as described in claim 1, characterized in that, The method further includes: Based on the resource conflict scenario, a power load joint scheduling simulation unit is constructed in the digital twin model. The dynamic change curves of production line power consumption and computer room cooling power consumption are simulated by Monte Carlo simulation to generate multiple load allocation schemes. Based on the aforementioned multiple load allocation schemes, with the objective functions of minimizing total energy consumption and load fluctuation, and in conjunction with the power grid peak-valley pricing mechanism, an optimized power resource allocation strategy for resource conflict scenarios is determined and converted into control commands that are synchronously sent to the production line distribution cabinets and the computer room UPS system for dynamic peak-shifting of power resources.

9. A collaborative management system for industrial production lines and data center facilities based on digital twins, characterized in that: The system is used to implement the collaborative management method for industrial production lines and data center facilities based on digital twins as described in any one of claims 1-8, the system comprising: The network architecture configuration module is used to configure an adaptive hybrid network architecture based on the communication environment information of the target industrial production line. The adaptive hybrid network architecture includes a control partition, a non-control partition, and a management information partition. The control partition is used to control the operation of industrial equipment, and the non-control partition is used for monitoring and auxiliary control. The synchronization mapping module is used to construct a digital twin mapping model of the status of equipment in multiple zones. It synchronously maps the operation data of key equipment in the industrial production line of the control zone, the status data of the computer room facilities in the non-control zone, and the operation and maintenance instructions of the management information zone to the digital twin space, forming a cross-zone status association database. The strategy set generation module is used to generate a collaborative strategy set containing security constraints and efficiency optimizations based on the cross-partition state association database. The collaborative strategy set predefines cross-partition interaction rules, data interaction protocols and security verification mechanisms. Cross-partition interaction rules refer to the rules for data interaction between different partitions. The collaborative management module is used to iteratively adjust the collaborative strategy set based on the simulation execution effect of the digital twin model, and send the optimized applicable collaborative strategies back to the digital twin model and distribute them to the control terminal to perform dynamic collaborative management of the target industrial production line and computer room facilities.

Images