Intelligent factory safety management method and system based on digital twinning

Abstract

This application relates to a smart factory safety management method and system based on digital twins, involving the fields of intelligent manufacturing and industrial safety, and information technology. It includes: acquiring factory equipment operating status data; using a digital twin model for mapping and analysis to generate an equipment operating status diagram containing equipment nodes and their operating status annotations; then identifying abnormal areas based on the changing trends of key parameters of equipment nodes in the diagram; next, dividing the abnormal areas into risk assessment zones according to the attribute information of equipment nodes; calculating a risk index considering the correlation between equipment operating parameters and environmental parameters in the risk assessment zone; and finally generating a risk control scheme based on the risk index ranking and combined with a preset strategy library. This scheme can achieve comprehensive monitoring and dynamic analysis of equipment operating status based on digital twin technology, thereby improving the accuracy and efficiency of smart factory safety management, minimizing safety risks, and ensuring stable factory operation.

Description

Technical Field

[0001] This application relates to the fields of intelligent manufacturing and industrial safety, and information technology, and in particular to a method and system for intelligent factory safety management based on digital twins. Background Technology

[0002] With the deepening of industrial intelligence and the transformation and upgrading of the manufacturing industry, the demand for monitoring and managing the operating status of factory equipment is increasing. Accurate equipment operation data can not only provide a scientific basis for production scheduling and maintenance, but also play an important role in risk warning and safety assurance. Therefore, improving the efficiency and safety of equipment operation management is of great significance.

[0003] Currently, the monitoring and management of factory equipment operation status typically relies on a combination of traditional sensor data acquisition and manual inspection. For example, sensors are installed to monitor key equipment parameters, and regular manual inspections are used to identify potential problems. However, this approach fails to fully consider the real-time correlation analysis of multi-dimensional data during equipment operation, which may lead to delays or omissions in the identification of potential risks.

[0004] Therefore, how to achieve comprehensive monitoring and dynamic analysis of equipment operating status based on digital twin technology, so as to effectively identify potential risks and formulate targeted countermeasures, has become an urgent problem to be solved. Summary of the Invention

[0005] The main purpose of this application is to provide a smart factory safety management method and system based on digital twins. This system can achieve comprehensive monitoring and dynamic analysis of equipment operating status based on digital twin technology, dynamically identify potential risks and formulate targeted countermeasures, thereby improving the safety and efficiency of factory equipment operation.

[0006] To achieve the above objectives, embodiments of the present invention provide a smart factory safety management method based on digital twins, the method comprising: Obtain equipment operating status data from factory equipment; Based on the digital twin model, the equipment operation status data is mapped and analyzed to generate an equipment operation status diagram. The equipment operation status diagram includes at least the equipment nodes corresponding to the factory equipment and the labeling information of the equipment nodes. The labeling information indicates the operation status of the factory equipment. Based on the changing trends of key parameters of equipment nodes in the equipment operation status diagram, identify areas of abnormal equipment operation; Based on the attribute information of the device nodes in the abnormal equipment operation area, the abnormal equipment operation area is divided into multiple risk assessment areas by level. Based on the correlation between equipment operating parameters and environmental parameters in the risk assessment area, calculate the risk index for each risk assessment area; Risk assessment areas are prioritized based on risk indices, and risk control plans are generated for the ranked risk assessment areas based on a pre-set risk response strategy library.

[0007] Accordingly, embodiments of this application also provide a smart factory safety management system based on digital twins, including: The acquisition module is used to acquire equipment operating status data of factory equipment; The digital twin module is used to perform mapping analysis on equipment operation status data based on the digital twin model to generate an equipment operation status diagram. The equipment operation status diagram includes at least the equipment nodes corresponding to the factory equipment and the labeling information of the equipment nodes. The labeling information indicates the operation status of the factory equipment. The identification module is used to identify abnormal areas of equipment operation based on the changing trends of key parameters of equipment nodes in the equipment operation status diagram. The region division module is used to divide the abnormal equipment operation region into multiple risk assessment regions by level according to the attribute information of the equipment nodes in the abnormal equipment operation region. The risk assessment module is used to calculate the risk index of each risk assessment zone based on the correlation between equipment operating parameters and environmental parameters in the risk assessment zone; The control scheme module is used to prioritize risk assessment areas based on risk indices and generate risk control schemes corresponding to the ranked risk assessment areas based on a preset risk response strategy library.

[0008] In summary, the technical solution of this application, by acquiring factory equipment operating status data and using digital twin model mapping analysis to generate an equipment operating status diagram containing equipment nodes and their operating status annotations, helps to intuitively grasp the overall operating status of the equipment. Then, based on the changing trends of key parameters of equipment nodes in the diagram, abnormal areas can be identified, enabling precise location of potentially problematic areas. Next, risk assessment zones are obtained by hierarchically dividing the equipment node attribute information in the abnormal areas, achieving detailed division of risk areas. Furthermore, considering the correlation between equipment operating parameters and environmental parameters in the risk assessment zones, a risk index is calculated, making the risk assessment more comprehensive and scientific. Finally, risk control schemes are generated based on risk index ranking and combined with a preset strategy library, enabling effective management and control of different risk areas according to priority. This solution can achieve comprehensive monitoring and dynamic analysis of equipment operating status based on digital twin technology, thereby improving the accuracy and efficiency of smart factory safety management, minimizing safety risks, and ensuring stable factory operation. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a schematic diagram of a scenario for a smart factory safety management method based on digital twins, as described in an embodiment of this application. Figure 2 A flowchart is provided for an embodiment of this application to illustrate a smart factory safety management method based on digital twins; Figure 3 This is a schematic flowchart illustrating the generation of the device operation status diagram provided in the embodiments of this application. Figure 4 A schematic diagram illustrating the process of abnormal region identification provided in an embodiment of this application; Figure 5 A schematic diagram illustrating the process of risk assessment area division provided in this application embodiment; Figure 6 A schematic diagram illustrating the process of calculating the risk index provided in this application embodiment; Figure 7 A flowchart illustrating the risk control scheme generated for the embodiments of this application; Figure 8 A schematic diagram illustrating the template update process for embodiments of this application. Figure 9 A schematic diagram of the structure of a digital twin-based smart factory safety management system provided in an embodiment of this application; Figure 10 Another structural schematic diagram of the intelligent factory safety management system based on digital twin provided in this application embodiment. Figure 11 A schematic diagram of the structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0011] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0012] This application provides a smart factory safety management method and system based on digital twins, which will be described in detail below.

[0013] Digital twins are a technology that maps physical entities and their behaviors to a digital space. In smart factories, digital twin models accurately replicate various aspects of the factory's physical layout, equipment structure, and technological processes. It is not merely a static digital model, but rather a system capable of reflecting the state, behavior, and interactions of physical entities in real time. For example, every piece of equipment in the factory, from large production machines to small sensors, has a corresponding virtual entity in the digital twin model, and the parameters of these virtual entities (such as operating parameters like temperature, speed, and pressure) are updated in real time according to the actual operating status of the physical equipment.

[0014] Smart factory safety management encompasses multiple aspects, including personnel safety, equipment safety, production process safety, and environmental safety. This includes preventing personnel from being injured during production, ensuring equipment operates normally to avoid accidents caused by malfunctions, guaranteeing the smooth operation of production processes to prevent production interruptions or product quality issues, and maintaining the factory environment in accordance with safety standards (such as preventing leaks of harmful gases and fire hazards).

[0015] In this embodiment, the smart factory safety management method based on digital twins is an innovative management approach that integrates digital twin technology with the safety requirements of smart factories. Digital twins construct a virtual model that precisely corresponds to the physical smart factory, encompassing information such as factory layout, equipment structure, and operational logic. For safety management, a sensor network is first used to collect data on equipment operating status (e.g., temperature, pressure, vibration), environmental data (e.g., humidity, gas concentration), and personnel-related data (e.g., location, operation), which are then mapped into the digital twin model. Next, this data is analyzed in real time based on the model. By monitoring the changing trends of equipment operating parameters, the impact of environmental factors, and the degree to which personnel behavior conforms to safety regulations, potential safety risks are accurately identified, such as equipment failure risks, hazardous environmental conditions, and risks associated with personnel violations. Finally, the identified risks are quantitatively assessed according to preset risk assessment standards to determine the risk level. Based on the risk level and specific circumstances, corresponding safety management plans are generated from a pre-defined response strategy library, such as equipment maintenance plans, environmental control measures, or personnel training and supervision strategies. At the same time, emergency drills can be simulated using digital twin models to optimize emergency plans, ultimately achieving comprehensive, efficient, and dynamic safety management in smart factories and improving the overall safety and reliability of the factory.

[0016] As shown in Figure 1, a scenario for intelligent factory safety management based on digital twins is provided. The scenario mainly includes factory equipment, data processing equipment, risk assessment equipment, and solution generation equipment. The equipment interacts with each other through a network.

[0017] Data processing and acquisition equipment first acquires key operational parameters of the equipment through sensors deployed on the factory equipment. These parameters include temperature, vibration frequency, current intensity, and pressure. Sensors are installed in critical parts of the equipment, such as motor housings, bearing housings, and pipe joints, and are connected to the central processing unit via signal lines. The data collected by the sensors is transmitted to the data acquisition equipment via a communication interface. The data acquisition equipment then inputs this data into a pre-built digital twin model. Based on the equipment's physical characteristics and operational logic rules, the digital twin model performs mapping analysis on the input data to generate an equipment operation status diagram. This diagram not only marks the operational status of each equipment node but also clarifies the relationships between these nodes and upstream and downstream equipment. This relationship is represented by a topological structure, where each node represents a device or subsystem, and the connections between nodes represent the physical connections and logical dependencies between devices.

[0018] The risk assessment equipment is used to identify abnormal equipment operation areas based on the changing trends of key parameters of equipment nodes in the equipment operation status diagram; to divide the abnormal equipment operation areas into multiple risk assessment areas by level according to the attribute information of equipment nodes in the abnormal equipment operation areas; and to calculate the risk index of each risk assessment area based on the correlation between equipment operation parameters and environmental parameters in the risk assessment areas.

[0019] The solution generation equipment prioritizes risk assessment areas based on a risk index and generates risk control solutions for each area based on a pre-set risk response strategy library. Strategy templates include specific content such as equipment maintenance plans, environmental control recommendations, and emergency shutdown plans. The equipment maintenance plan includes specific maintenance steps, required tools, and personnel arrangements; environmental control recommendations cover measures such as temperature and humidity regulation, ventilation improvement, and dust removal; and the emergency shutdown plan specifies shutdown conditions, shutdown sequence, and the inspection process for resuming operation. The risk assessment equipment can sequentially match the corresponding response strategy templates according to the risk priority list and adjust the strategy details based on the specific parameters of the risk assessment area to generate the final risk control solution. For example, for a high-risk assessment area, if its main risk stems from equipment overheating, the solution generation equipment will add steps for cooling system inspection and lubricant replacement to the equipment maintenance plan and propose specific measures to reduce the ambient temperature in the environmental control recommendations.

[0020] refer to Figure 2 , Figure 2This is a flowchart illustrating a smart factory security management method based on digital twins provided in this application embodiment. The execution subject of this method can be a computer device, which can be a single computer device or a cluster of multiple computer devices. The computer device can be a terminal device or a server, etc. The smart factory security management method based on digital twins provided in this application embodiment specifically includes: Step S10: Obtain equipment operating status data of the factory equipment.

[0021] Factory equipment refers to mechanical and electrical equipment used in a smart factory environment for various functions such as production, processing, transportation, and testing. For example, in an automobile manufacturing plant, factory equipment includes stamping presses, welding robots, conveyor belts, and testing equipment. These devices each have different functions and working principles, and they cooperate or are interconnected throughout the entire production process.

[0022] Equipment operating status data refers to a collection of information reflecting the operation of factory equipment. This information covers various aspects of equipment operation, such as mechanical condition, electrical condition, and operating efficiency. Taking motor equipment as an example, equipment operating status data includes, but is not limited to, motor speed, current intensity, temperature, and vibration frequency. Speed ​​reflects the motor's operating speed, current intensity reflects the motor's power consumption, temperature can indicate whether the motor is overheating, and vibration frequency can indicate the stability of the motor's internal mechanical structure.

[0023] In this application, acquiring equipment operating status data is fundamental to the entire smart factory safety management methodology. A comprehensive understanding of equipment operation requires the collection of various data types. Equipment operating status data serves as the basis for subsequent operations such as digital twin model analysis, anomaly area identification, and risk assessment. Without accurate and comprehensive equipment operating status data, subsequent analysis and management cannot be effectively carried out. Only by acquiring sufficient data can a complete understanding of the equipment's operating status be achieved, thereby enabling the accurate identification and management of potential risks.

[0024] In one embodiment, during the design phase of factory equipment, key operating parameters to be monitored are determined based on factors such as the equipment's function, structure, and operating principles. For each parameter, a matching sensor is selected. For example, to monitor equipment temperature, thermocouple sensors or thermistor sensors can be used, installed on heat-generating parts of the equipment, such as near motor windings or the surface of heat exchangers. For vibration monitoring, accelerometers are a suitable choice, installed on the equipment's bearing housings or support structures. To ensure the stability and timeliness of data transmission, sensors can transmit the collected equipment operating status data to a data acquisition center in real time via wired (e.g., industrial Ethernet) or wireless (e.g., ZigBee protocol) communication methods. This approach, through careful selection of sensors and appropriate communication methods, enables comprehensive and accurate acquisition of equipment operating status data, laying a solid foundation for subsequent analysis and management, and effectively avoiding risk analysis errors caused by missing or inaccurate data.

[0025] Step S20: Based on the digital twin model, perform mapping analysis on the equipment operation status data to generate an equipment operation status diagram. The equipment operation status diagram includes at least the equipment nodes corresponding to the factory equipment and the labeling information of the equipment nodes. The labeling information indicates the operation status of the factory equipment.

[0026] A digital twin is a digital model built upon physical equipment. It accurately represents the various characteristics of the physical equipment in digital form, including its structure, function, and operating logic. For example, for a complex automated production line, a digital twin model can simulate in detail the connection relationships between the various devices in the production line, the working principle of each device, and the material transfer logic between devices.

[0027] In a digital twin model, a device node is a digital abstraction of factory equipment, with each device node corresponding to an actual existing piece of factory equipment. For example, in a smart factory containing multiple machine tools, conveyor belts, control cabinets, and other equipment, each machine tool, conveyor belt, control cabinet, etc., will have a corresponding device node in the digital twin model.

[0028] The labeling information is a detailed description of the operating status of the equipment represented by the device node. It may include whether the equipment is operating normally, whether there are potential risks, the operating efficiency of the equipment, and the status of the key parameters of the equipment. For example, for a motor equipment corresponding to a device node, the labeling information may be "operating normally, temperature within the normal range, vibration frequency stable, and operating efficiency high" or "abnormal operation, temperature too high, with potential failure risk".

[0029] In this embodiment, inputting the acquired equipment operating status data into the digital twin model for mapping analysis is a crucial step. The digital twin model associates the equipment operating status data with equipment nodes within the model based on pre-defined rules and algorithms. Each equipment node has its corresponding operating status data; through this mapping relationship, the actual operating status of each device can be clearly identified. The generated equipment operating status diagram can intuitively display the overall operating status of all equipment in the factory, where equipment nodes represent individual devices, and the annotation information clearly indicates the operating status of each device. This provides an intuitive and detailed basis for subsequently identifying abnormal equipment operating areas.

[0030] In one embodiment, a digital twin model is first constructed based on the factory equipment's design drawings, physical characteristics, operating principles, and other information. Sub-models are then built for each component of the equipment, and these sub-models are combined into a complete digital twin model according to the equipment's actual structure and connection relationships. Within the digital twin model, a device node is defined for each device or component, and a mapping relationship is established between the device node and actual operating status data. Upon receiving equipment operating status data, the data is allocated to the corresponding device node according to the mapping relationship. Next, based on the equipment's normal operating parameter range and fault diagnosis algorithms, annotation information is generated for each device node. For example, if the temperature of a device node is within the normal range, the annotation information is "Normal Operation - Normal Temperature"; if the temperature exceeds the normal range, the annotation information might be "Abnormal - Excessively High Temperature, Potential Risk." The equipment operating status diagram constructed in this way accurately reflects the equipment's operating status, helps to promptly identify potential risks, and improves the efficiency and accuracy of equipment management.

[0031] In one embodiment, to improve the accuracy of the device operating state diagram, such as Figure 3 As shown, the device operating status diagram can be generated in the following manner: Step S201: Input the device operation status data into the pre-constructed digital twin model, and generate a device operation status diagram based on the preset device operation logic rules in the digital twin model.

[0032] The pre-defined equipment operation logic rules in a digital twin model refer to a series of rules established during the construction of the digital twin model based on the physical principles, working mechanisms, and logical relationships of the equipment during normal operation. These rules guide the digital twin model in generating accurate equipment operation status diagrams based on input equipment operation status data. For example, for a production equipment containing multiple processing steps, its operation logic rules might include the sequence of each step, the range of influence of each step on equipment parameters, and the mutual constraints between different parameters. If the equipment is a CNC machine tool, its operation logic rules will involve the relationship between the tool's motion trajectory and cutting parameters, such as the correlation between cutting speed and tool wear, as well as the motion sequence and speed limits of each axis in different processing steps.

[0033] In this embodiment, inputting equipment operating status data into a pre-built digital twin model is the initial operation for generating an equipment operating status diagram. The digital twin model processes this data according to internally preset equipment operating logic rules. These rules are a digital abstraction of the actual operating logic of the equipment, ensuring that the model accurately simulates the operating state of the equipment. Only based on correct logic rules can the model reasonably analyze and process the input data, thereby generating a status diagram that matches the actual equipment operating state. If there are deviations in the logic rules, the generated equipment operating status diagram will not accurately reflect the true situation of the equipment, thus affecting subsequent evaluation and management of the equipment operating status. For example, on an automated production line, if the equipment operating logic rules do not correctly set the material transfer time and speed relationships between the various devices, the equipment operating status diagram may incorrectly show material accumulation or insufficient supply, leading to a misjudgment of the equipment operating status.

[0034] In one embodiment, rule-based expert system technology is employed when constructing the digital twin model. First, domain experts conduct a detailed analysis and review of the equipment's operational logic, transforming the equipment's physical principles, workflow, and parameter relationships into explicit rules. For example, for a complex chemical production equipment, experts determine rules based on the principles of chemical reactions, the flow sequence of materials, and the effects of temperature and pressure on the reaction. Then, these rules are encoded into the digital twin model in the form of a rule-based language (such as CLIPS). When equipment operational status data is input into the model, the model's inference engine reasones and processes the data according to these rules, thereby generating an equipment operational status diagram. The advantages of this approach are that the rules are clear, easy to understand and maintain, and can accurately generate status diagrams based on preset rules.

[0035] In one embodiment, rule-learning algorithms from machine learning can also be utilized. By learning from a large amount of data on normal and fault states of equipment, the algorithm automatically discovers potential rules between equipment operating status data and equipment operating logic. For example, for a power transmission device, data such as voltage, current, and power factor under different loads, as well as the device's normal operating status labels, can be collected. Rule-learning algorithms (such as the RIPPER algorithm) are used to learn patterns in this data, and the learned rules are applied to a digital twin model. When new equipment operating status data is input, the model generates an equipment operating status diagram based on the learned rules. The advantage of this approach is that it can automatically discover rules from actual data, reducing the workload of manually formulating rules, and it can adapt to the operating logic of equipment under different operating conditions.

[0036] Step S202: Mark the operating status of each device node and its relationship with upstream and downstream devices in the device operation status diagram.

[0037] The operating status of a device node refers to the working condition of the device at a specific moment, including different states such as normal operation, fault state, and performance degradation. For example, for a motor device node, the normal operating state may be characterized by stable speed, current within the normal range, and normal temperature; the fault state may be characterized by the motor suddenly stopping, excessive current, or excessive temperature.

[0038] The upstream and downstream equipment relationship refers to the relationship between equipment and its directly connected upstream and downstream counterparts in the entire production process. This relationship includes material transfer, energy transfer, and signal interaction. For example, in an electronics assembly line, a pick-and-place machine is an upstream device of a soldering machine. The pick-and-place machine mounts electronic components onto circuit boards, which are then transferred to the soldering machine for soldering. This is the upstream and downstream equipment relationship between the pick-and-place machine and the soldering machine, including the material (circuit board) transfer relationship. In an automated control system, a sensor, as an upstream device of the controller, transmits the collected signals to the controller, which is a signal interaction relationship.

[0039] In this embodiment, it is crucial to label the operating status of each device node and its relationships with upstream and downstream devices in the device operation status diagram. Labeling the operating status of device nodes provides a clear understanding of the working condition of each device, which is essential for quickly locating potentially problematic equipment. Furthermore, clarifying the relationships between devices and their upstream and downstream counterparts helps analyze the scope of the impact of equipment failure on the entire production process. For example, when a critical device fails, its relationships with upstream and downstream devices allow identification of which devices will be directly affected, enabling appropriate countermeasures such as adjusting the operating parameters of upstream and downstream devices or suspending the operation of some devices to prevent further spread of the fault. Without accurately labeling this information, it is difficult to comprehensively assess the impact on the entire production system when equipment malfunctions, potentially leading to untimely fault handling or inappropriate measures taken.

[0040] In one embodiment, this can be achieved by establishing a device connection matrix within the digital twin model. When constructing the model, the upstream and downstream devices of each device node are determined based on the actual physical connections and process flow, and these relationships are stored in the model in matrix form. For example, for a production system with n device nodes, an n×n matrix is ​​established, where the elements represent the connection relationships between devices. When generating the device operating state diagram, based on the operating state data of the device nodes and this connection matrix, the operating state of each device node and its association with upstream and downstream devices are labeled in the state diagram. The advantage of this approach is that the relationships are clearly represented, easy to query and maintain, and can accurately label relevant information in the state diagram.

[0041] In one embodiment, the devices and their relationships can be constructed into a graph structure, where device nodes are the nodes in the graph, and the relationships between devices are the edges. The operating status information of each device node is stored in a graph database. When it is necessary to annotate the operating status of a device node and its relationships with upstream and downstream devices in the device operating status graph, the relevant information is retrieved from the graph database and annotated. For example, in a smart factory's device management system, the Neo4j graph database is used to construct a graph structure for each device. When the operating status of a device changes, the corresponding node information in the graph database is updated. Then, when generating the device operating status graph, the annotation information is directly retrieved from the graph database for annotation. This method can efficiently handle complex device relationships and facilitates dynamic updates of device relationships and status information.

[0042] Step S30: Identify abnormal areas of equipment operation based on the changing trends of key parameters of equipment nodes in the equipment operation status diagram.

[0043] Key parameters refer to those that significantly impact the operating status of equipment. Different types of equipment have different key parameters. For example, for engines, key parameters might include fuel pressure, oil temperature, and engine speed; for CNC machine tools, key parameters might include tool wear, cutting force, and the displacement accuracy of the coordinate axes. Changes in these key parameters directly reflect whether the equipment is operating normally or is about to malfunction.

[0044] The trend of key parameter changes refers to the pattern of change of key parameters over time. It can be monotonically increasing, monotonically decreasing, periodic, or irregular. For example, if the temperature of equipment continues to rise over a period of time, this is a monotonically increasing trend; if the vibration amplitude of equipment shows periodic fluctuations, this is a periodic trend.

[0045] For example, key parameters of equipment nodes, such as current intensity, vibration frequency, temperature, and pressure, exhibit patterns of change over time that constitute key parameter trends. For instance, the temperature of a normally operating motor may remain relatively stable over a period of time; this is a stable trend in temperature parameter variation. However, if the motor malfunctions, the temperature may rise continuously or fluctuate more significantly; this is an abnormal trend in temperature parameter variation.

[0046] In this embodiment, the changing trends of key parameters of equipment nodes are an important basis for determining whether equipment is malfunctioning. In the complex production environment of a factory, equipment is interconnected and influences each other; an anomaly in one device may affect surrounding equipment. By analyzing the changing trends of key parameters of equipment nodes in the equipment operation status diagram, it is possible to identify equipment nodes whose parameter changes do not follow normal patterns. These nodes may indicate operational anomalies in the equipment or its surrounding area. Identifying areas of equipment malfunction requires comprehensive consideration of the changing trends of key parameters of multiple equipment nodes, because anomalies in localized equipment may spread and affect surrounding equipment, thus forming a relatively large anomaly area.

[0047] In one embodiment, key parameters of each equipment node in the equipment operation status diagram can be monitored and analyzed in real time. Short-term and long-term change thresholds are set for each key parameter. Taking temperature as an example, the short-term change threshold is set to an increase of 5°C per hour, and the long-term change threshold is set to an increase of 10°C per day. When the change of a key parameter of an equipment node exceeds the short-term threshold, the equipment node is marked as a preliminary abnormal node. Subsequently, correlation analysis is performed on the equipment nodes surrounding the preliminary abnormal node, considering the physical connections and process correlations between the equipment. If the key parameters of surrounding equipment nodes also show similar abnormal change trends, or if the change of a single parameter does not exceed the threshold, but multiple parameters collectively show an abnormal correlation, then the area where these equipment nodes are located is determined as an abnormal equipment operation area. This technical implementation can identify abnormal equipment operation areas in a timely and accurate manner, avoiding the oversight of the overall operational risk due to misjudgment of a single equipment node, and improving the accuracy and comprehensiveness of abnormal area identification.

[0048] Step S40: Based on the attribute information of the device nodes in the abnormal device operation area, the abnormal device operation area is divided into multiple risk assessment areas.

[0049] The attribute information of equipment nodes can include the number of devices, their type (such as motors, sensors, conveyors, etc.), functions (such as power supply, data acquisition, material handling, etc.), importance level (classified according to the criticality of the equipment in the production process, such as core equipment, key auxiliary equipment, general auxiliary equipment, etc.), and their position in the production process. For example, in an automobile manufacturing plant, welding robots are core equipment with a high importance level, their function is to weld car bodies, and they are located in the body assembly stage of the production process; while lighting equipment is general auxiliary equipment, its function is to provide lighting, and its position in the production process is relatively independent.

[0050] In this embodiment, the equipment nodes in the abnormal equipment operation area have different attributes, resulting in varying degrees of risk impact on the entire area. Dividing the abnormal equipment operation area hierarchically based on the attribute information of the equipment nodes allows for a more detailed and accurate assessment of the risk status of different areas. First, an initial division is made according to equipment type, grouping similar types of equipment nodes together, as the same type of equipment may have similar risk characteristics. Then, further division is made based on the importance level of the equipment; areas containing equipment nodes with higher importance levels tend to have higher risks. Finally, the initial division is adjusted and refined by comprehensively considering factors such as the location of the equipment in the production process, resulting in multiple risk assessment zones. This division helps to formulate more precise risk control measures for different risk situations.

[0051] In one embodiment, a device node attribute database can be created to store the attribute information of each device node within an area of ​​abnormal device operation. During the partitioning process, the attribute information of the device nodes is read from the database. First, the device nodes are grouped according to their device type; for example, all motor device nodes are grouped into one group, and all sensor device nodes into another. Next, each device node is assigned a weight value based on its importance level, with higher-importance devices receiving larger weight values. For example, the weight value is set to 0.8 for core processing equipment on the production line, and 0.2 for auxiliary lighting equipment. Based on the group to which the device node belongs and its weight value, the comprehensive risk value for each group is calculated. Groups with similar comprehensive risk values ​​and closely related positions of the device nodes in the production process are merged into a single risk assessment area. In this way, multiple risk assessment areas partitioned based on device node attribute information can be obtained. This partitioning method can more accurately reflect the risk status of different areas, providing a more accurate basis for subsequent risk index calculations.

[0052] Step S50: Calculate the risk index of each risk assessment zone based on the correlation between equipment operating parameters and environmental parameters in the risk assessment zone.

[0053] Environmental parameters refer to parameters related to the external environment in which equipment operates, and may include humidity, dust concentration, noise level, etc. For example, in an electronic equipment manufacturing workshop, excessive humidity may affect the performance of electronic components, and excessive dust concentration may cause short circuits or accelerated wear of equipment.

[0054] A correlation is a relationship between equipment operating parameters and environmental parameters that influence each other. For example, a high humidity environment may reduce the insulation performance of electrical equipment, thereby affecting the equipment's electrical parameters and increasing the risk of equipment operation; excessive dust concentration may affect the heat dissipation efficiency of equipment, leading to an increase in equipment temperature. This is an example of a correlation between equipment operating parameters and environmental parameters.

[0055] In this embodiment, the correlation between equipment operating parameters and environmental parameters is considered when calculating the risk index for each risk assessment zone. This is because in a real factory environment, equipment operation is not isolated; the external environment affects equipment operation, and changes in equipment operation may also have a reciprocal effect on the environment. By quantifying equipment operating parameters and environmental parameters, analyzing their correlation, and determining their influence weights, the risk level of each risk assessment zone can be more accurately assessed, thus providing a basis for developing reasonable risk control plans.

[0056] In one embodiment, a risk index calculation model can be constructed, comprising an equipment operating parameter analysis module and an environmental parameter analysis module. The equipment operating parameter analysis module is responsible for quantifying and analyzing the equipment operating parameters within the risk assessment area, such as calculating statistical indicators like the mean, standard deviation, and rate of change of the equipment operating parameters. The environmental parameter analysis module is responsible for performing the same quantification on environmental parameters, such as calculating their average and extreme values. Then, a correlation matrix is ​​established to analyze the correlation between equipment operating parameters and environmental parameters. Based on historical data and expert experience, different weight values ​​are assigned to different correlations between equipment operating parameters and environmental parameters. For example, the weight of humidity's influence on equipment temperature is determined to be 0.3, and the weight of dust concentration's influence on equipment vibration frequency is determined to be 0.2. Finally, the quantified indicators of equipment operating parameters, the quantified indicators of environmental parameters, and their weight values ​​are substituted into the risk index calculation formula to calculate the risk index for the risk assessment area. This method can scientifically and accurately calculate the risk index for the risk assessment area, fully considering the correlation between equipment operating parameters and environmental parameters, making the risk index more reflective of the actual risk situation, and providing a reliable basis for the subsequent generation of risk control plans.

[0057] Step S60: Prioritize the risk assessment areas based on the risk index, and generate risk control plans corresponding to the ranked risk assessment areas based on the preset risk response strategy library.

[0058] The risk index is a comprehensive quantitative indicator used to measure the level of risk in a risk assessment area. The higher the risk index, the greater the risk in the area, and the more urgent it needs to be addressed.

[0059] The pre-built risk response strategy library is a knowledge base containing various risk response strategies, which are tailored to different risk situations. For example, for the risk of excessive equipment temperature, strategies might include adding heat dissipation equipment or adjusting the equipment workload; for the risk of excessive dust concentration, strategies might include improving ventilation and dust removal or installing dust filtration devices.

[0060] In this embodiment, prioritizing risk assessment areas based on risk indices is to determine which areas require priority handling. Areas with higher risk indices have greater potential risks and may have a more severe impact on production, therefore they should be prioritized. Then, based on the priority order of the risk assessment areas, appropriate strategies are selected from a pre-set risk response strategy library and adjusted according to the specific circumstances of each risk assessment area to generate targeted risk control solutions. This effectively reduces equipment operation risks and ensures the safety and normal production of the smart factory.

[0061] In one embodiment, a risk assessment zone priority ranking mechanism is established, arranging risk assessment zones in descending order according to their risk index. Simultaneously, a risk control scheme generation module is established, connected to a risk response strategy library. When generating a risk control scheme, the module first reads the priority ranking results of the risk assessment zones. For the highest-priority risk assessment zone, a matching response strategy template is searched in the risk response strategy library based on the zone's equipment operating parameters, environmental parameters, and risk type. For example, if the main risks of the risk assessment zone are excessively high equipment temperature and high dust concentration, response strategy templates targeting these two issues are found in the risk response strategy library. Then, based on the specific parameters of the risk assessment zone, such as the specific equipment model and location, the searched response strategy templates are adjusted and optimized to generate the final risk control scheme. This process is repeated for subsequent risk assessment zones until corresponding risk control schemes are generated for all risk assessment zones. This technical implementation can reasonably determine the priority order based on the actual risk status of the risk assessment zones and generate targeted risk control schemes, effectively reducing equipment operation risks and improving the safety and production efficiency of the smart factory.

[0062] In one embodiment, such as Figure 4 As shown, step S30 may include the following steps: Step S301: Extract the key parameter change curves of each device node in the device operation status diagram.

[0063] A critical parameter variation curve refers to the curve showing how key parameters of a device node (such as current intensity, vibration frequency, temperature, and pressure) change over time. This curve visually illustrates the fluctuation of the critical parameter over a period of time. For example, for the critical parameter of temperature of a device, the variation curve may show that the temperature gradually rises during the initial startup phase, then fluctuates around a relatively constant value after reaching a stable operating state. If a malfunction occurs in the device, the temperature may suddenly rise or continue to rise. These changes can all be reflected in the temperature critical parameter variation curve.

[0064] In this embodiment, extracting the key parameter change curves of each device node in the device operation status diagram is a crucial foundational step for identifying abnormal operating areas. These change curves are a visual representation of the device's operating status, containing key parameter information at different points in time. By analyzing these curves, a deeper understanding of the device's operational dynamics can be gained. Without accurately extracting these change curves, it is impossible to further analyze the changing trends of key parameters at device nodes, thus affecting the accurate identification of abnormal operating areas. For example, in a large industrial refrigeration system, the key parameter change curves of the compressor are crucial for judging the overall operating status of the refrigeration system. If the pressure and temperature key parameter change curves of the compressor cannot be extracted, it is difficult to determine whether the compressor is in normal working condition, and consequently, it is impossible to determine whether there are abnormal operating areas in the entire refrigeration system.

[0065] In one embodiment, a Supervisory Control and Data Acquisition (SCADA) system can be used. Specifically, during equipment operation, the SCADA system continuously collects key parameter data from equipment nodes and stores it in chronological order. Then, data processing software converts this stored data into visualized key parameter change curves. For example, for a transformer in a power substation, the SCADA system collects key parameters such as transformer oil temperature and winding temperature at regular intervals (e.g., every minute). After a period of time (e.g., a day), this data is imported into professional data analysis software (e.g., Matlab). The software's plotting function is used to plot the changes in key parameters such as oil temperature and winding temperature over time, thereby extracting the key parameter change curves.

[0066] Step S302: Calculate the rate of change of each device node from the normal range based on the change curve of the key parameters.

[0067] The rate of change deviating from the normal range refers to the quantitative indicator of how quickly a key parameter of a device changes compared to its normal operating range. The normal operating range is determined based on the device's design specifications, historical operating data, and industry standards. For example, for a normally operating motor with a vibration frequency between 10 and 50 Hz, if the vibration frequency curve shows a rapid increase from 30 Hz to 60 Hz within a certain period, then this rate of change from the normal range to exceeding the normal range is the rate of change deviating from the normal range.

[0068] In this embodiment, calculating the rate of change of each device node from the normal range is a crucial step in identifying abnormal nodes. This rate of change accurately reflects how quickly the key parameters of a device node deviate from their normal state, providing a better indication of the device's operational trend than simply observing whether key parameters exceed the normal range. If a device node's key parameters are temporarily within the normal range, but the rate of change from the normal range is high, this may indicate that the device is about to malfunction or is already in the early stages of abnormal operation. For example, in a high-precision optical processing device, the levelness of its processing platform is a key parameter, with a normal range within ±0.01 mm. If the levelness change curve shows a rapid change from 0.005 mm to exceeding the normal range within a short period (e.g., 1 hour), even if it is not yet outside the normal range, this high rate of change indicates a potential problem with the device, requiring further inspection.

[0069] In one embodiment, the upper and lower limits of the normal range of key parameters of the equipment node can be determined first. Then, for adjacent data points on the change curve of the key parameter, the rate of change is calculated according to the numerical difference formula. For example, for the change curve of the equipment temperature parameter, assuming the normal temperature range is [20℃, 30℃], there are two adjacent data points (t1, T1) and (t2, T2) on the change curve, where t1 and t2 are time points and T1 and T2 are the corresponding temperature values. Then the rate of change can be calculated by the formula: (T2 - T1) / (t2 - t1). The calculated rate of change between each data point is compared with the rate of change threshold set according to the normal range to determine the rate of change of the equipment node from the normal range.

[0070] Step S303: Designate the device nodes whose rate of change exceeds the preset threshold as abnormal nodes, and select the upstream and downstream device nodes directly related to the abnormal nodes to form an abnormal area.

[0071] An abnormal node is a device node in the equipment operation status diagram whose key parameters deviate from the normal range at a rate exceeding a preset threshold. These nodes indicate that the corresponding equipment may be experiencing operational abnormalities. For example, in a complex automated production line, if the rate of change of a key parameter (such as the processing accuracy of the equipment) of a certain equipment node exceeds a set threshold, then this equipment node is identified as an abnormal node, meaning that the processing accuracy of the equipment may be rapidly declining, affecting product quality.

[0072] Directly related upstream and downstream equipment nodes refer to the equipment nodes in the production process that are directly connected to the equipment corresponding to the abnormal node, specifically the upstream and downstream equipment. In a production system, there are material, energy, or information transfer relationships between equipment, which determine the upstream and downstream relationships. For example, in an electronic product assembly line, a circuit board printer is an upstream device of a circuit board insertion machine because the circuit boards are first printed on the printer before being sent to the insertion machine for electronic component insertion; conversely, the insertion machine is an upstream device of the soldering machine, and the soldering machine is a downstream device of the insertion machine.

[0073] In this embodiment, identifying equipment nodes with a rate of change exceeding a preset threshold as abnormal nodes, and selecting directly related upstream and downstream equipment nodes centered on these nodes to form an abnormal region, is based on the interconnectivity of equipment within the production system. In actual factory production environments, equipment is interconnected and mutually influential; an abnormality in one piece of equipment can significantly impact the operational status of its directly connected upstream and downstream devices. By constructing such an abnormal region, the potential impact of equipment abnormalities can be comprehensively considered, rather than being limited to a single abnormal equipment node. For example, in a chemical production process, if the temperature control device of a reactor is identified as an abnormal node (because its temperature change rate exceeds a threshold), then the equipment nodes corresponding to the feed pump (upstream equipment) and discharge pipeline valve (downstream equipment) directly connected to this reactor should also be included in the abnormal region consideration. This is because an abnormal reactor temperature may affect the feed flow rate and discharge status, thereby affecting the stability of the entire chemical production process.

[0074] In one embodiment, when constructing the equipment model of the production system, the upstream and downstream equipment relationship information of each equipment can be stored in a database. When a certain equipment node is determined to be an abnormal node, the database can be queried to quickly obtain its directly associated upstream and downstream equipment node information, thereby forming an abnormal region. For example, in an automobile manufacturing plant, for each piece of equipment (such as a stamping press, welding robot, painting equipment, etc.), its upstream and downstream equipment relationships are recorded in detail in the database. When the rate of change of a certain key parameter of the stamping press exceeds a threshold and is determined to be an abnormal node, the database is queried to obtain the equipment nodes corresponding to the previous steel plate conveying equipment (upstream equipment) and the next forming mold (downstream equipment) directly associated with the stamping press, and these three equipment nodes constitute an abnormal region.

[0075] In one embodiment, the equipment nodes in the entire production system can be viewed as nodes in a graph, and the connections between equipment as edges, thus constructing an equipment relationship graph. When an abnormal node is identified, a graph traversal algorithm (such as breadth-first search) is used to search for directly connected upstream and downstream equipment nodes, thereby determining the abnormal region. For example, in a semiconductor manufacturing plant, there are numerous complex pieces of equipment (such as lithography machines, etching machines, diffusion furnaces, etc.). By constructing an equipment relationship graph from these devices, when the rate of change of a key parameter of the lithography machine exceeds a threshold, becoming an abnormal node, a breadth-first search algorithm is used to search for the directly connected silicon wafer transport device (upstream equipment) and photoresist coating device (downstream equipment) from the lithography machine node, forming the abnormal region together with the lithography machine node.

[0076] In one embodiment, reference Figure 5 The attribute information of the device nodes may include: the number of device nodes and their distribution density; at this time, step S40 may specifically include: Step S401: Set the size of the grid according to the number and distribution density of device nodes in the abnormal device operation area.

[0077] The number of equipment nodes refers to the number of equipment nodes contained in an area where equipment operation is abnormal. For example, in a factory workshop containing various types of equipment, if there are equipment nodes corresponding to 5 machine tools, 3 conveyor belt devices, and 2 control cabinets in an abnormal area, then the number of equipment nodes in this abnormal area is 10.

[0078] Distribution density refers to the degree of dispersion of equipment nodes within an area of ​​abnormal equipment operation. It can be calculated as the ratio of the number of equipment nodes to the area of ​​the abnormal region. For example, within a rectangular abnormal region, if equipment nodes are concentrated in one corner, the distribution density is relatively low; if the equipment nodes are evenly distributed throughout the region, the distribution density is relatively high.

[0079] In this embodiment, setting the grid size based on the number and distribution density of device nodes in the abnormal equipment operation area is a crucial starting point for hierarchical region subdivision. The number of device nodes directly affects the fineness of the subdivision; a larger number may require more detailed subdivision, while a smaller number may not necessitate such fine subdivision. Distribution density is equally important. High distribution density indicates closer interrelationships between device nodes, potentially requiring a smaller grid size to distinguish different sub-regions; conversely, low distribution density allows for a larger grid size. Arbitrarily setting the grid size without considering the number and distribution density of device nodes may result in unreasonable sub-regions—either too fragmented, leading to excessive complexity in subsequent analysis, or too coarse, failing to accurately reflect the risk differences within the region.

[0080] In one embodiment, the number of equipment nodes in the abnormal equipment operation area is first counted, and the area of ​​the abnormal area is measured. Then, a basic density coefficient is set based on the equipment type and historical experience. For example, for a precision manufacturing workshop, the density coefficient is set to 3 equipment nodes per square meter. The theoretical grid area is calculated by dividing the number of equipment nodes by (the area of ​​the abnormal area multiplied by the density coefficient). Finally, the calculated grid area is adjusted appropriately according to the actual situation (such as the importance distribution of equipment, process flow, etc.) to determine the size of the grid.

[0081] Step S402: Divide the abnormal equipment operation area into multiple sub-regions according to the dimensions.

[0082] A sub-region is a smaller unit obtained by dividing the area of ​​abnormal equipment operation according to a set grid size. Each sub-region contains several device nodes, which are spatially concentrated within that sub-region. For example, a large area of ​​abnormal equipment operation can be divided into multiple square sub-regions, each of which acts as an independent small unit, and the device nodes within that unit are interconnected.

[0083] In this embodiment, dividing the abnormal equipment operation area into multiple sub-regions according to a set size is a necessary step for further detailed analysis. By dividing into sub-regions, a large and complex abnormal area can be decomposed into relatively simple and homogeneous smaller areas. This helps to more specifically analyze the characteristics and risk status of equipment nodes within each smaller area in subsequent steps. If such a division is not performed and the entire abnormal area is analyzed directly, it will be difficult to accurately grasp the risk characteristics of each local area due to the large size of the area and the large number of equipment nodes, and some local risk information may be missed.

[0084] In one embodiment, a spatial partitioning algorithm from Geographic Information System (GIS) technology can be utilized. The area experiencing equipment malfunction is treated as a geographic spatial region. Using the spatial partitioning function of GIS software, this region is divided into multiple sub-regions of regular shapes (such as rectangles or squares) according to a set grid size. Each sub-region has clear boundaries and limits, and device nodes within each sub-region can be easily marked and managed.

[0085] Step S403: Perform cluster analysis on the device nodes in each sub-region, and merge adjacent sub-regions with similar changing trends into a risk assessment area based on the clustering results.

[0086] Cluster analysis is a statistical analysis method that groups data objects (in this case, device nodes) based on the similarity of their characteristics (such as the changing trends of key parameters). For example, for the key parameter of temperature of device nodes, if the temperature change trend of some device nodes is gradually increasing, while the temperature change trend of other device nodes is gradually decreasing, then cluster analysis can group device nodes with the same temperature change trend into different groups.

[0087] Similar trends refer to key parameters of equipment nodes exhibiting similar patterns of change over a period of time, such as simultaneous increases, simultaneous decreases, or similar fluctuation amplitudes and frequencies. For example, if the vibration frequencies of several equipment nodes all exhibit periodic small fluctuations over a period of time, this indicates that they have similar trends.

[0088] In this embodiment, cluster analysis is performed on the device nodes within each sub-region, and adjacent sub-regions with similar changing trends are merged into a single risk assessment area to more accurately assess risks. Cluster analysis can uncover deeper relationships between device nodes within a sub-region, grouping devices with similar operational characteristics into one category. Merging adjacent sub-regions with similar changing trends into a single risk assessment area is because, in actual production environments, adjacent sub-regions with similar changing trends in their device nodes often exhibit similarities in risk propagation and impact, and should be considered as a whole. This reduces the complexity of risk assessment while improving its accuracy.

[0089] In one embodiment, the K-means algorithm can be used to group device nodes within a sub-region, and the clustering results are determined based on the similarity of the changing trends of node parameters; adjacent sub-regions with similar changing trends are merged into a risk assessment area.

[0090] The K-means algorithm is a clustering algorithm that aims to divide a dataset into K distinct clusters such that the sum of the distances from each data point to the center of its cluster is minimized. In this scenario, the dataset represents the device nodes within a sub-region, and the distance is measured based on the similarity of the changing trends of node parameters.

[0091] The similarity of node parameter change trends refers to the degree of similarity in the patterns of change of key parameters (such as temperature, vibration frequency, etc.) of equipment nodes over time. For example, if the temperature parameters of two equipment nodes both show a trend of first rising and then falling over a period of time, and the magnitude and rate of rise and fall are similar, then these two equipment nodes have a high degree of similarity in node parameter change trends.

[0092] In this embodiment, using the K-means algorithm to group equipment nodes within a sub-region is an effective data analysis method. The K-means algorithm iteratively divides equipment nodes into different groups. In this process, determining the clustering results based on the similarity of the changing trends of node parameters is crucial. This is because the changing trends of the operating status of equipment nodes directly reflect the potential risk characteristics of the equipment. If clustering is not based on this similarity, equipment nodes with significantly different operating statuses may be grouped together, making it impossible to accurately identify equipment groups with different risk characteristics. For example, in a sub-region of a machining workshop, different trends in the vibration frequency of equipment nodes may indicate different equipment health conditions. Without clustering based on similarity, it is difficult to manage equipment risks in a targeted manner. For example, for a group of machine tool equipment nodes, vibration frequency and machining accuracy are selected as key parameters. Then, the changing trend data of the key parameters of each equipment node are standardized to eliminate the influence of dimensions. Next, the number of clusters K is set (which can be determined based on experience or preliminary data analysis). The standardized equipment node data is then used as input to run the K-means algorithm. During the algorithm's operation, the Euclidean distance of the node parameter change trend is used as the similarity metric. That is, the Euclidean distance between device nodes in terms of the key parameter change trend is calculated, and device nodes that are close to each other are grouped together.

[0093] Adjacent sub-regions refer to sub-regions that are spatially close to each other. In the previous process of dividing the equipment malfunction area into sub-regions, these sub-regions were geographically adjacent. For example, in an equipment malfunction area divided according to a rectangular grid, sub-regions that are adjacent in the horizontal or vertical direction are considered adjacent sub-regions.

[0094] In this embodiment, merging adjacent sub-regions with similar changing trends into a single risk assessment area is based on the interconnectedness of equipment and the risk propagation characteristics in the actual production environment. Equipment within adjacent sub-regions may influence each other due to factors such as physical connections, material transfer, or process linkages. When equipment nodes within these adjacent sub-regions exhibit similar changing trends, it indicates consistency in their risk characteristics, and they should be assessed as a whole. This simplifies the complexity of risk assessment while providing a more comprehensive and accurate understanding of the overall risk situation in the region. For example, in a localized area of ​​a chemical production process, if the temperature change trends of equipment in several adjacent sub-regions are similar, this may indicate an overall thermal management risk. Merging these sub-regions into a single risk assessment area helps in developing a unified and effective risk control strategy.

[0095] In one embodiment, reference Figure 6 Step S50 may specifically include the following steps: Step S501: Obtain time series data of equipment operating parameters in the risk assessment area, and calculate the fluctuation amplitude and frequency of equipment operating parameters.

[0096] Time series data of equipment operating parameters refers to a data sequence consisting of the values ​​of equipment operating parameters recorded in chronological order. For example, for a motor, its current intensity is recorded every certain time interval (e.g., every 5 minutes). These current intensity values ​​arranged in chronological order constitute the time series data of the equipment operating parameter, current intensity.

[0097] Fluctuation range refers to the range of change of equipment operating parameters over a period of time, that is, the difference between the maximum and minimum values. For example, if the highest temperature of a piece of equipment in a day is 50°C and the lowest temperature is 30°C, then its temperature fluctuation range is 20°C.

[0098] Fluctuation frequency refers to the number of times a device's operating parameters fluctuate within a unit of time. For example, if a device's vibration frequency fluctuates from 10Hz to 20Hz and back to 10Hz three times in one hour, then its fluctuation frequency in that hour is 3 times / hour.

[0099] In this embodiment, obtaining time-series data of equipment operating parameters in the risk assessment zone is the foundation for calculating the risk index. This data reflects the dynamic characteristics of equipment operation, including information on changes in the equipment's operating status over time. Calculating the fluctuation amplitude and frequency provides a quantitative description of the changing characteristics of the equipment operating parameters. Fluctuation amplitude reflects the stability of equipment operation; a large fluctuation amplitude indicates unstable equipment operation and potentially significant risks. For example, in a chemical production facility, large pressure fluctuations within the reactor could lead to uncontrolled chemical reactions or damage to equipment seals. Fluctuation frequency is equally important; a high fluctuation frequency may indicate periodic disturbances or internal structural instability. By calculating these two indicators, a preliminary assessment of the equipment's operational risk can be made.

[0100] In one embodiment, a database management system can be used to store and retrieve time-series data of equipment operating parameters. During equipment operation, various equipment operating parameters are recorded in the database at regular time intervals. For example, for a large smart factory, a relational database (such as MySQL) or a time-series database (such as InfluxDB) can be used to store the equipment operating parameter data. When calculating the fluctuation amplitude, the maximum and minimum values ​​of the equipment operating parameters within a specific time period are obtained by querying the database, and then subtracted to obtain the fluctuation amplitude. For calculating the fluctuation frequency, a difference algorithm can be used to calculate the difference between adjacent data points in the time-series data, count the number of times the difference exceeds a certain threshold (indicating fluctuation), and then divide by the time length to obtain the fluctuation frequency.

[0101] Step S502: Obtain environmental parameter data in the risk assessment area and calculate the correlation coefficient between the environmental parameter data and the equipment operating parameters.

[0102] Environmental parameter data refers to data reflecting various parameters of the environment in which equipment operates, such as humidity, dust concentration, and noise level. For example, in an electronic equipment manufacturing workshop, the humidity level affects the performance of electronic components; humidity value is a type of environmental parameter data.

[0103] The correlation coefficient is a statistical indicator used to measure the strength and direction of the linear relationship between two variables (in this case, environmental parameter data and equipment operating parameters). The correlation coefficient ranges from -1 to 1, where -1 indicates a perfect negative correlation, 1 indicates a perfect positive correlation, and 0 indicates no correlation. For example, if higher humidity results in lower insulation resistance in equipment, then there is a negative correlation between humidity and the equipment's insulation resistance, and its correlation coefficient is close to -1.

[0104] In this embodiment, acquiring environmental parameter data from the risk assessment area and calculating its correlation coefficient with equipment operating parameters is a crucial step in considering the impact of the environment on equipment operation. Environmental parameters have a significant impact on equipment operation, and different environmental parameters may have varying degrees of influence. For example, in a machining workshop, excessively high dust concentrations may lead to accelerated wear and tear on equipment, affecting its precision and lifespan. By calculating the correlation coefficient, the degree of association between environmental parameters and equipment operating parameters can be clearly defined. If the absolute value of the correlation coefficient is large, it indicates that the environmental parameters have a strong impact on the equipment operating parameters and should be given priority consideration when assessing risks.

[0105] In one embodiment, environmental parameter data for the risk assessment area can be obtained from environmental monitoring equipment, while corresponding equipment operation parameter data can be obtained from the equipment operation data acquisition system. These two sets of data are then organized into a format suitable for analysis (such as a data matrix). The correlation coefficient between the environmental parameter data and the equipment operation parameters is calculated using the `pearsonr` function (used to calculate the Pearson correlation coefficient) or the `spearmanr` function (used to calculate the Spearman correlation coefficient, suitable for nonlinear relationships) from the SciPy library.

[0106] Step S503: Calculate the risk index based on the fluctuation amplitude, fluctuation frequency and correlation coefficient to obtain the risk index of the risk assessment area.

[0107] In this embodiment, calculating the risk index based on fluctuation amplitude, fluctuation frequency, and correlation coefficient is a comprehensive quantification of the risk level of the risk assessment area. Fluctuation amplitude and fluctuation frequency reflect the instability of the equipment's operation, while the correlation coefficient reflects the degree of environmental influence on equipment operation. Considering these factors comprehensively allows for a more complete and accurate risk assessment. For example, a risk assessment area with large fluctuation amplitude and frequency of equipment operating parameters, and where environmental parameters are strongly correlated with equipment operating parameters, will have a high risk index, indicating that the area has significant potential risks and requires priority attention.

[0108] In one embodiment, different weights can be assigned to the fluctuation amplitude, fluctuation frequency, and correlation coefficient according to their numerical values; the fluctuation amplitude, fluctuation frequency, and correlation coefficient are weighted according to the weights to obtain the risk index of the risk assessment area.

[0109] Weight is a numerical value that represents the relative importance of each factor (volatility amplitude, volatility frequency, correlation coefficient) in calculating a risk index. Different weights reflect the different degrees of contribution of each factor to risk assessment. For example, if volatility amplitude is considered to have the most significant impact on equipment risk, then it will be given a larger weight.

[0110] In one embodiment, assuming the risk index is RI, the fluctuation range is A, the fluctuation frequency is F, and the correlation coefficient is C, the formula can be set as: RI = a × A + b × F + c × |C|, where a, b, and c are weighting coefficients determined based on experience or expert evaluation. For example, based on previous studies on the relationship between equipment failure and risk, a = 0.4, b = 0.3, and c = 0.3 are determined. Substituting the calculated fluctuation range, fluctuation frequency, and correlation coefficient into the formula, the risk index of the risk assessment area can be calculated.

[0111] In one embodiment, reference Figure 7 Step S60 may include the following steps: Step S601: Sort all risk assessment areas from high to low according to the risk index to generate a risk priority list.

[0112] A risk priority list is a list sorted by the risk index of each risk assessment area, clearly indicating the order of risk level. Areas with higher risk indices appear higher on the list, indicating that they pose a greater risk and require priority attention.

[0113] In this embodiment, sorting all risk assessment areas according to their risk index from high to low to generate a risk priority list is a crucial preliminary step for risk control. The risk index reflects the overall risk level of the risk assessment areas, and sorting them allows for a clear identification of which areas pose the most pressing risks. This enables priority action to be taken on high-risk areas within limited resources and time, preventing further escalation of risks. For example, in a large smart factory with multiple risk assessment areas, randomly processing risk areas without prioritization could lead to uncontrolled risks in high-risk areas, thereby impacting the overall production safety of the factory.

[0114] For example, a bubble sort algorithm can be used. First, treat all risk assessment zones as array elements, each containing relevant information such as a risk index. Then, compare the risk indices of adjacent risk assessment zones; if the risk index of a preceding zone is lower than that of a following zone, swap their positions. After multiple rounds of comparisons and swaps, the zones are finally sorted from highest to lowest risk index, thus generating a risk priority list.

[0115] Step S602: Retrieve the response strategy template corresponding to the risk assessment area from the risk response strategy library. The response strategy template includes at least an equipment maintenance plan, environmental control recommendations, and an emergency shutdown plan.

[0116] The risk response strategy library is a pre-built knowledge base that stores various response strategy templates. These templates are developed based on different risk types and equipment and environmental conditions.

[0117] An equipment maintenance plan is a plan for maintenance work during equipment operation, including the maintenance cycle, the content of maintenance (such as inspection and replacement of equipment parts), and the arrangement of maintenance personnel.

[0118] Environmental control recommendations are suggestions for improving environmental conditions based on the impact of the equipment operating environment on equipment risks, such as adjusting humidity and reducing dust concentration.

[0119] An emergency shutdown plan is a shutdown procedure to be taken when equipment faces an emergency risk situation. It includes the shutdown triggering conditions, the shutdown sequence, and the safety assurance measures after shutdown.

[0120] In this embodiment, retrieving the response strategy template corresponding to the risk assessment area from the risk response strategy library is a crucial step in formulating a risk control plan. Since different risk assessment areas may face different types and degrees of risk, it is necessary to obtain matching response strategy templates from the strategy library. For example, if the main risk in a risk assessment area is equipment overheating, the corresponding response strategy template might include an equipment maintenance plan for inspecting and maintaining the cooling system, environmental control recommendations for lowering the ambient temperature, and an emergency shutdown plan for immediately stopping equipment operation and performing cooling treatment when the temperature exceeds a certain critical value.

[0121] In one embodiment, an indexing mechanism can be established to link risk assessment areas and response strategy templates. In the risk response strategy library, each response strategy template is assigned a relevant risk label, which corresponds to different risk types, equipment types, and other information. When a response strategy template needs to be retrieved, query conditions are generated based on relevant information from the risk assessment area (such as major risk factors, equipment types, etc.). By querying this indexing mechanism, the corresponding response strategy template can be quickly located and retrieved.

[0122] Step S603: Match the corresponding response strategy templates sequentially according to the risk priority list. Matching refers to the process of linking the risk assessment areas in the risk priority list with the response strategy templates retrieved from the risk response strategy library according to certain rules, so as to ensure that a suitable response strategy template can be found for each risk assessment area.

[0123] In this embodiment, matching the corresponding response strategy templates sequentially according to the risk priority list is to systematically assign appropriate response strategies to each risk assessment area. Matching according to risk priority ensures that high-risk areas receive appropriate response strategies first. This helps to allocate resources rationally and improve the efficiency of risk control. For example, if there are three risk assessment areas, A, B, and C, with corresponding response strategy templates 1, 2, and 3 respectively, then risk assessment area A is first matched with template 1, then B with template 2, and C with template 3, thus carrying out risk response in an orderly manner.

[0124] In one embodiment, a cyclic matching algorithm is employed. A loop is set up, starting with the first risk assessment area in the risk priority list, and sequentially matching each risk assessment area with a response strategy template retrieved from the risk response strategy library. During the matching process, the success of the match is determined by whether the key characteristics of the risk assessment area (such as the main risk type, equipment type, etc.) match the applicable scope of the response strategy template. If the match is successful, the matching proceeds to the next risk assessment area; if unsuccessful, a more suitable response strategy template is searched again from the risk response strategy library.

[0125] In one embodiment, an equipment maintenance plan template corresponding to the risk assessment area can be retrieved from the risk response strategy library based on the risk priority list; the steps and tools in the equipment maintenance plan template can be adjusted according to the specific parameters of the risk assessment area to generate the final response strategy template.

[0126] The equipment maintenance plan template is a standardized, template-based planning framework specifically designed for equipment maintenance work. This template includes the basic steps of equipment maintenance, such as daily equipment inspections, periodic component replacement, and equipment calibration; it also includes tools that may be used during maintenance, such as testing instruments and repair tools, and specifies general content such as the approximate cycle and sequence of maintenance work.

[0127] In this embodiment, the risk priority list clarifies the processing order of risk assessment areas. Prioritizing high-risk areas can minimize potential losses. The equipment maintenance plan templates in the risk response strategy library form the basic framework for addressing equipment risks. Since different risk assessment areas may involve different types of equipment or different risk conditions, it is necessary to accurately retrieve the matching template based on the characteristics of each risk assessment area. For example, in a complex production workshop, there are multiple different types of equipment risk assessment areas, such as machining equipment areas, electrical equipment areas, and hydraulic equipment areas. The equipment in each area has different working principles and risk characteristics. For machining equipment, the focus may be more on risks such as tool wear and the stability of mechanical structures, while for electrical equipment, the focus is more on risks such as electrical insulation and circuit overload. Therefore, accurately retrieving the appropriate equipment maintenance plan template for each risk assessment area according to the risk priority list is an important prerequisite for ensuring the effectiveness of equipment maintenance work. If the template cannot be retrieved accurately, the maintenance work may lack focus, failing to effectively address the equipment's risk conditions, thereby affecting the normal operation and service life of the equipment.

[0128] The specific parameters of a risk assessment area refer to various parameters that describe in detail the equipment condition and operating environment within the risk assessment area. These parameters include the specific model of the equipment, its operating time, its load condition, the temperature and humidity of the environment where the equipment is located, and the spatial layout around the equipment. For example, for a specific model of injection molding machine, its operating time reaches 5000 hours, the load is often above 80%, and the temperature and humidity of the environment where the equipment is located are high; these are all specific parameters of a risk assessment area.

[0129] The final response strategy template is a comprehensive, adjusted, and optimized template that includes not only a detailed plan for equipment maintenance but also recommendations for environmental control related to equipment operation, emergency shutdown plans, and other aspects. This template is customized to the specific circumstances of the risk assessment area, making it more targeted and practical, and designed to comprehensively and effectively address the risk situation within the risk assessment area.

[0130] In this embodiment, it is essential to adjust the steps and tools in the equipment maintenance plan template according to the specific parameters of the risk assessment zone to generate the final response strategy template. The specific parameters of the risk assessment zone are key factors reflecting the actual operating conditions and specific needs of the equipment. For example, different models of equipment may differ in structure, performance, and vulnerable components, requiring adjustments to the maintenance steps based on the specific model. The operating time of the equipment is also an important factor; equipment operating for extended periods may require more frequent inspections of critical components or earlier replacement of certain vulnerable parts. Workload conditions affect the wear rate and performance stability of the equipment; equipment operating under high loads may require additional performance testing steps and enhanced cooling and lubrication maintenance measures. Furthermore, environmental conditions, such as high temperature and high humidity environments, may necessitate adding moisture-proof and heat dissipation-related operations to the maintenance steps and may require the use of specialized maintenance tools. Only by fully considering these specific parameters and accurately adjusting the equipment maintenance plan template can a final response strategy template that meets the actual needs of the equipment be generated, thereby more effectively controlling risks and ensuring the normal operation of the equipment.

[0131] Step S604: Adjust the response strategy template based on the parameters of the risk assessment area to generate the final risk control plan.

[0132] The parameters of the risk assessment area refer to the various characteristic parameters of the risk assessment area, including equipment operating parameters (such as equipment power, speed, etc.), environmental parameters (such as humidity, temperature, etc.), and equipment layout and structural parameters.

[0133] In this embodiment, the final risk control plan is generated by adjusting the response strategy template based on the parameters of the risk assessment zone to make the response strategy more targeted. Although the templates retrieved from the risk response strategy library provide general guidance, each risk assessment zone has its unique parameter conditions, which need to be adjusted accordingly. For example, for a risk assessment zone with high equipment power, the equipment maintenance plan may need to add more detection points and more frequent detection cycles; if the ambient temperature is low, the environmental control recommendations may not require cooling measures, but rather focus on equipment insulation measures, etc.

[0134] In one embodiment, the equipment maintenance plan can be adjusted in terms of maintenance frequency and specific maintenance content based on parameters such as equipment power and speed; the environmental control recommendations can be adjusted in terms of target values ​​and control methods based on parameters such as ambient temperature and humidity; and the emergency shutdown plan can be adjusted in terms of shutdown sequence and safety measures based on equipment layout and structural parameters. Finally, the adjusted equipment maintenance plan, environmental control recommendations, and emergency shutdown plan are combined to generate the final risk control scheme.

[0135] In one embodiment, reference Figure 8 The method of this application may also include: Step S70: Obtain the equipment operating status data after the risk control plan is implemented, and compare and analyze it with the equipment operating status data before implementation.

[0136] Post-implementation equipment operation status data refers to the collection of various status information generated during equipment operation after the equipment has been managed and operated in accordance with the risk control plan. This data may be similar in type to the data before implementation, such as including parameters like equipment temperature, vibration frequency, and current intensity, but the values ​​may change due to the implementation of risk control measures.

[0137] Comparative analysis is a method for comparing and studying two sets of data (equipment operating status data before and after execution). It aims to identify differences, trends, and correlations between the two sets of data, thereby assessing the impact of risk control schemes on equipment operating status.

[0138] In this embodiment, acquiring equipment operating status data after the implementation of the risk control plan and comparing it with data before implementation is a crucial step in evaluating the effectiveness of the risk control plan. By comparing these two sets of data, the changes in equipment operating status before and after the implementation of the risk control plan can be clearly seen. For example, in a scenario with the risk of equipment overheating, the risk control plan may include measures such as adding heat dissipation equipment and optimizing equipment workload. Before implementation, the equipment temperature data often approaches or exceeds the safety threshold, while after implementation, if the temperature data drops significantly and remains within the safe range, this indicates that the risk control plan may be effective in temperature control. This comparative analysis helps us to comprehensively understand the impact of the risk control plan on all aspects of equipment operation, rather than being limited to a specific parameter. Without such comparative analysis, it is difficult to determine whether the risk control plan has truly improved equipment operating status and reduced risk.

[0139] Step S80: Calculate the variation difference of key parameters in the equipment operating status data, and evaluate the implementation effect of the risk control plan based on the variation difference.

[0140] The difference in key parameter variation refers to the difference in the values ​​of key parameters (such as temperature, vibration frequency, and current intensity mentioned earlier) in the equipment operating status data before and after the implementation of the risk control plan. This difference can quantitatively reflect the degree of impact of the risk control plan on the key operating parameters of the equipment.

[0141] Evaluating the effectiveness of risk control measures involves comprehensively assessing whether the measures have achieved their intended goals of reducing risk and improving equipment operating conditions based on the differences in changes in key parameters. For example, if the changes in key parameters move in a direction conducive to stable equipment operation (such as lower temperature or reduced vibration), the measures are considered effective; otherwise, further adjustments to the plan may be necessary.

[0142] In this embodiment, calculating the difference in changes of key parameters in the equipment operating status data and evaluating the effectiveness of the risk control scheme based on this difference is a method for quantitatively analyzing the effectiveness of risk control. The difference in changes of key parameters provides specific numerical basis for the evaluation. For example, for a motor, if its vibration frequency is 15Hz before the risk control scheme is implemented and becomes 10Hz after implementation, then the difference in vibration frequency is -5Hz. This difference indicates that the risk control scheme has an improving effect on the vibration of the motor. However, simply calculating the difference in changes of a single parameter is insufficient; it is also necessary to comprehensively consider the changes of multiple key parameters to fully evaluate the implementation effect. Because the operation of equipment is a complex system, the improvement of one parameter may be accompanied by changes in other parameters, requiring a balance of overall benefits and drawbacks. For example, although the vibration frequency decreases, if the current intensity increases significantly, this may mean that the risk control scheme has introduced other potential risks while solving one problem, requiring further analysis.

[0143] For example, for each key parameter, the values ​​before and after the implementation of the risk control plan are read from the database. Then, the difference in value is obtained by subtracting the value before implementation from the value after implementation. To evaluate the effectiveness of the implementation, some evaluation rules can be set. For example, if the difference in value of the key parameter is within a pre-defined effective range (e.g., a temperature difference between -5℃ and 5℃ is considered normal), the risk control effect of that key parameter is considered good; if it exceeds this range, further analysis is needed to determine whether it exceeds positively (indicating over-control or the emergence of new problems) or exceeds negatively (indicating insufficient risk control). Based on these evaluation results, a preliminary judgment can be made on the effectiveness of the risk control plan.

[0144] Step S90: Update and optimize the strategy templates in the risk response strategy library based on the execution results.

[0145] The implementation effect is a comprehensive evaluation of whether the risk control plan has achieved its expected goals in actual application, including considerations such as the degree of improvement in equipment operation status and the degree of risk reduction.

[0146] In this embodiment, updating and optimizing the strategy templates in the risk response strategy library based on execution results is a process of continuous improvement of risk response strategies. Execution results reflect the adaptability and effectiveness of the current strategy templates in practical applications. If the execution results show deficiencies in certain aspects, such as the risk control plan failing to effectively reduce a certain risk of the equipment, then the relevant content in the strategy template needs to be adjusted. For example, if it is found that existing heat dissipation measures are ineffective in actual implementation for the overheating risk of a certain type of equipment, then the heat dissipation-related steps in the equipment maintenance plan of the strategy template need to be optimized, which might involve replacing the heat dissipation equipment with more efficient equipment or adjusting the operating parameters of the heat dissipation equipment. By continuously updating the strategy templates based on execution results, the accuracy and practicality of the risk response strategy library can be improved, enabling it to better address different equipment risk situations.

[0147] In one embodiment, a feedback mechanism and update process can be established. After evaluating the effectiveness of the risk control plan, the evaluation results are fed back to the risk response strategy library management system in a structured form (such as a report containing performance indicators of key parameters and information on the overall risk reduction). The management system includes a dedicated analysis module that identifies areas for improvement in the strategy template based on the feedback results. For example, if the environmental control recommendations in a risk control plan for a certain device are found to be ineffective, the analysis module will locate the relevant environmental control content in the strategy template. Then, experts or data analysis will determine specific optimization measures, such as adjusting the control range of ambient temperature or increasing the frequency of humidity regulation, and these optimized measures will be updated in the strategy template.

[0148] In one embodiment, the method of this application embodiment may further include: Obtain historical operating data and a prediction model for the equipment in the risk assessment area, and train the prediction model based on the historical data. Extract key parameter features of equipment operation from the historical operating data; The extracted key parameter features are input into the prediction model, and the time series prediction value of the equipment operating status is generated by the time series analysis algorithm in the prediction model. A device operation prediction map is plotted based on time series forecast values, and the predicted state change range of each device node is marked on the device operation prediction map. Based on the predicted state change areas, identify the equipment nodes that may experience abnormalities in the equipment operation prediction map, and formulate preventive maintenance plans in advance.

[0149] Historical operational data of equipment within the risk assessment zone refers to various status data recorded during the operation of equipment within the risk assessment zone over a past period, including equipment operating parameters (such as temperature, pressure, flow rate, etc.), operating time, fault records, and other information. This data reflects the past operational behavior and status change trends of the equipment.

[0150] A predictive model is a model built using mathematical algorithms and machine learning techniques. It learns from and analyzes historical operating data to predict the future operating status of equipment. For example, it can be a time series analysis-based model, such as the ARIMA model, or a neural network-based model, such as the LSTM (Long Short-Term Memory) model.

[0151] In this embodiment, acquiring historical operating data and a predictive model for the equipment in the risk assessment area is fundamental for predicting equipment operating status. Historical operating data serves as the input data source for the predictive model, containing various patterns and rules governing equipment operation. The predictive model is a predictive tool constructed using information from this data. For example, in the risk assessment area of ​​a chemical production facility, historical operating data includes changes in parameters such as temperature and pressure at different production stages. By training the predictive model using this data, the model can learn the patterns of temperature, pressure, and other parameters changing over time and through the production process, thereby predicting future equipment operating status. Without accurate historical operating data or a suitable predictive model, it is impossible to effectively predict equipment operating status, thus affecting subsequent preventative maintenance planning and other related work.

[0152] Key parameters of equipment operation are crucial information extracted from historical operating data that reflects the equipment's operating status and behavior patterns. These characteristics can be numerical, such as the average, standard deviation, maximum, and minimum values ​​of equipment operating parameters; or they can be time-series based, such as periodicity and trends. For example, for a motor, the average and standard deviation of its current intensity can reflect the motor's power consumption level and stability during normal operation, while periodic changes in current intensity may be related to periodic changes in the motor's load.

[0153] In this embodiment, extracting key parameter features from historical operating data aims to transform the large amount of raw historical operating data into more valuable and representative information for better use in predictive models. Raw historical operating data often contains a wealth of detailed information; directly using it for predictive models may lead to overly complex models, low training efficiency, or poor prediction results. By extracting key parameter features, the essential characteristics of equipment operation can be captured, reducing data dimensionality and noise. For example, analyzing the historical operating data of a complex automated production line may involve various parameter data collected by multiple sensors, but not all data points are equally important for predicting the equipment's operating status. Extracting key parameter features such as the average temperature of key equipment components and the standard deviation of vibration frequencies of key equipment allows for a more concise description of the equipment's operating status, improving the accuracy and efficiency of the predictive model.

[0154] Time series analysis algorithms are used in predictive models to process time series data, predicting future values ​​based on historical patterns in the data. Common time series analysis algorithms include moving average algorithms, exponential smoothing algorithms, and algorithms based on the Autoregressive Integrated Moving Average (ARIMA) model. These algorithms generate predicted values ​​by analyzing the trend, seasonality, and periodicity characteristics of time series data.

[0155] Time-series forecasts of equipment operating status are numerical predictions of the future operating status of equipment, calculated by a prediction model based on key input parameters and using time-series analysis algorithms. These forecasts are arranged in chronological order, such as predicting the temperature value of the equipment every hour for the next day or the daily vibration frequency of the equipment for the next week.

[0156] In this embodiment, the core step in predicting equipment operating status is to input the extracted key parameter features into the prediction model and generate time-series predicted values ​​of equipment operating status using a time-series analysis algorithm. The prediction model uses the input key parameter features as a basis to predict the future operating status of the equipment through its internal time-series analysis algorithm. Since the operating status of equipment typically changes over time, the time-series analysis algorithm can capture this temporal correlation. For example, in predicting the operation of a power equipment, key parameter features extracted from historical operating data (such as the average voltage, standard deviation, and periodic variation characteristics) are input into the prediction model. The time-series analysis algorithm in the model (such as the algorithm in the ARIMA model) predicts the voltage changes of the power equipment over a future period based on these features and the time-series patterns in the historical data. These predicted values ​​can provide important basis for preventative maintenance and risk management of the equipment.

[0157] An equipment operation prediction chart is a visual representation used to show the predicted operating status of equipment over a future period. It uses time as the horizontal axis and a certain parameter of equipment operation (such as temperature, vibration frequency, etc.) as the vertical axis, and intuitively displays the changing trend of equipment operating status over time by plotting curves.

[0158] The predicted state change range refers to the range of possible operating states predicted for each device node in the device operation prediction map. This range reflects the uncertainty of the prediction; because predictions have a certain degree of error, a range is used to represent the future operating state range of the device node, rather than a precise value.

[0159] In this embodiment, the predicted state change range provides the expected range of equipment operating status. If the predicted value of a certain equipment node exceeds the normal change range, then this node may have an abnormal risk. For example, on an automated production line, the equipment operation prediction diagram shows that the predicted state change range of a certain robotic arm's movement speed is 10-15 operations per minute, but in a certain period of time, the predicted value reaches 20 operations. This indicates that the robotic arm may be abnormal, possibly due to wear of mechanical parts, program errors, or other factors. Once the equipment node that may be abnormal is identified, a maintenance plan can be formulated based on the characteristics of the equipment, past maintenance experience, and current production plans. For robotic arms that may be abnormal, a comprehensive inspection can be planned during production breaks, including checks on the wear of mechanical parts, the flexibility of joints, and the accuracy of the control program. Preventive maintenance plans can not only avoid production interruptions and losses caused by sudden equipment failures, but also extend the service life of the equipment and improve its overall reliability.

[0160] Accordingly, to better implement the above methods, this application also provides a smart factory safety management system based on digital twins. For example... Figure 9 As shown, the intelligent factory safety management system 80 based on digital twins includes an acquisition module 801, a digital twin module 802, an identification module 803, a region division module 804, a risk assessment module 805, and a control scheme module 806, as detailed below: The acquisition module 801 is used to acquire equipment operating status data of factory equipment; The digital twin module 802 is used to perform mapping analysis on equipment operation status data based on the digital twin model to generate an equipment operation status diagram. The equipment operation status diagram includes at least the equipment nodes corresponding to the factory equipment and the labeling information of the equipment nodes. The labeling information indicates the operation status of the factory equipment. The identification module 803 is used to identify abnormal areas of equipment operation based on the changing trends of key parameters of equipment nodes in the equipment operation status diagram. The region division module 804 is used to divide the abnormal equipment operation region step by step according to the attribute information of the equipment nodes in the abnormal equipment operation region to obtain multiple risk assessment regions; The risk assessment module 805 is used to calculate the risk index of each risk assessment zone based on the correlation between equipment operating parameters and environmental parameters in the risk assessment zone; The control scheme module 806 is used to prioritize risk assessment areas based on risk indices and generate risk control schemes corresponding to the ranked risk assessment areas based on a preset risk response strategy library.

[0161] In one embodiment, the digital twin module 802 is used for: The device operation status data is input into a pre-constructed digital twin model, and a device operation status diagram is generated based on the preset device operation logic rules in the digital twin model. The operating status of each device node and its relationship with upstream and downstream devices are marked in the device operation status diagram.

[0162] In one embodiment, the identification module 803 is specifically used for: Extract the key parameter change curves for each device node in the device operation status diagram; Calculate the rate of change of each device node from the normal range based on the change curve of the key parameters; Device nodes whose rate of change exceeds a preset threshold are designated as abnormal nodes, and the upstream and downstream device nodes directly associated with the abnormal node are selected to form an abnormal region centered on the abnormal node.

[0163] In one embodiment, the region division module 804 is used for: The size of the grid is set according to the number and distribution density of device nodes in the area of ​​abnormal device operation; The abnormal equipment operation area is divided into multiple sub-regions according to the dimensions described above; Cluster analysis is performed on the device nodes in each sub-region, and adjacent sub-regions with similar changing trends are merged into a risk assessment area based on the clustering results.

[0164] In one embodiment, the risk assessment module 805 is used for: Obtain time-series data of equipment operating parameters in the risk assessment area, and calculate the fluctuation amplitude and frequency of the equipment operating parameters; Obtain environmental parameter data in the risk assessment area, and calculate the correlation coefficient between the environmental parameter data and the equipment operating parameters; The risk index of the risk assessment area is obtained by calculating the risk index based on the fluctuation amplitude, fluctuation frequency and correlation coefficient.

[0165] In one embodiment, the control scheme module 806 is used for: Sort all risk assessment areas from high to low according to their risk index to generate a risk priority list; Retrieve response strategy templates corresponding to the risk assessment area from the risk response strategy library. The response strategy templates include at least equipment maintenance plans, environmental control recommendations, and emergency shutdown plans. Match the corresponding response strategy templates sequentially based on the risk priority list; The response strategy template is adjusted based on the parameters of the risk assessment area to generate the final risk control plan.

[0166] In one embodiment, the control scheme module 806 is used for: Retrieve the equipment maintenance plan template corresponding to the risk assessment area from the risk response strategy library based on the risk priority list; Adjust the steps and tools in the equipment maintenance plan template according to the specific parameters of the risk assessment area to generate the final response strategy template.

[0167] In one embodiment, reference Figure 10 The intelligent factory safety management system may also include: a template update module 807, specifically: Obtain equipment operating status data after the risk control plan is implemented, and compare and analyze it with the equipment operating status data before implementation; Calculate the variation difference of key parameters in the equipment operation status data, and evaluate the implementation effect of the risk control plan based on the variation difference; Update and optimize the strategy templates in the risk response strategy library based on the implementation results.

[0168] In one embodiment, the control scheme module 806 can also be used for: Obtain historical operating data and a prediction model for the equipment in the risk assessment area, and train the prediction model based on the historical data. Extract key parameter features of equipment operation from the historical operating data; The extracted key parameter features are input into the prediction model, and the time series prediction value of the equipment operating status is generated by the time series analysis algorithm in the prediction model. Based on the time series predicted values, a device operation prediction map is drawn, and the predicted state change range of each device node is marked in the device operation prediction map; Based on the predicted state change areas, identify the equipment nodes that may experience abnormalities in the equipment operation prediction map, and formulate preventive maintenance plans in advance.

[0169] The implementation details of each module are provided in the preceding method embodiments and will not be repeated here. The technical effects achieved by each module and device are described in the foregoing method embodiments.

[0170] It should be noted that, in practical implementation, the above modules can be arbitrarily combined and integrated into one or more modules, or implemented as independent entities. Furthermore, the above modules can be implemented in hardware or as software functional modules. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. The aforementioned storage medium can be a read-only memory, a hard disk, or an optical disk, etc.

[0171] like Figure 11 As shown, this application embodiment also provides a computer device 90, characterized in that it includes a processor 901 and a memory 902, wherein the memory 902 stores a computer program, and when the computer program is executed by the processor 901, the processor 901 performs the steps of any of the methods described above.

[0172] In the several embodiments provided in this application, it should be understood that the disclosed systems and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0173] The above-disclosed embodiments are merely preferred embodiments of this application and should not be construed as limiting the scope of this application. Therefore, any equivalent variations made in accordance with the claims of this application shall still fall within the scope of this application.

Claims

1. A smart factory safety management method based on digital twins, characterized in that, include: Obtain equipment operating status data from factory equipment; Based on the digital twin model, the equipment operation status data is mapped and analyzed to generate an equipment operation status diagram. The equipment operation status diagram includes at least the equipment nodes corresponding to the factory equipment and the labeling information of the equipment nodes. The labeling information indicates the operation status of the factory equipment. Based on the changing trends of key parameters of equipment nodes in the equipment operation status diagram, identify areas of abnormal equipment operation; Based on the attribute information of the device nodes in the abnormal equipment operation area, the abnormal equipment operation area is divided into multiple risk assessment areas by level. Based on the correlation between equipment operating parameters and environmental parameters in the risk assessment area, calculate the risk index for each risk assessment area; Risk assessment areas are prioritized based on risk indices, and risk control plans are generated for the ranked risk assessment areas based on a pre-set risk response strategy library.

2. The method according to claim 1, characterized in that, Based on the digital twin model, the device operating status data is mapped and analyzed to generate a device operating status diagram, including: The device operation status data is input into a pre-constructed digital twin model, and a device operation status diagram is generated based on the preset device operation logic rules in the digital twin model. The operating status of each device node and its relationship with upstream and downstream devices are marked in the device operation status diagram.

3. The method according to claim 2, characterized in that, Based on the changing trends of key parameters of equipment nodes in the equipment operation status diagram, abnormal equipment operation areas are identified, including: Extract the key parameter change curves for each device node in the device operation status diagram; Calculate the rate of change of each device node from the normal range based on the change curve of the key parameters; Device nodes whose rate of change exceeds a preset threshold are designated as abnormal nodes, and the upstream and downstream device nodes directly associated with the abnormal node are selected to form an abnormal region centered on the abnormal node.

4. The method according to claim 3, characterized in that, Based on the attribute information of the device nodes in the abnormal equipment operation area, the abnormal equipment operation area is divided into multiple risk assessment areas, including: The size of the grid is set according to the number and distribution density of device nodes in the area of ​​abnormal device operation; The abnormal equipment operation area is divided into multiple sub-regions according to the dimensions described above; Cluster analysis is performed on the device nodes in each sub-region, and adjacent sub-regions with similar changing trends are merged into a risk assessment area based on the clustering results.

5. The method according to claim 3, characterized in that, Based on the correlation between equipment operating parameters and environmental parameters in the risk assessment area, the risk index for each risk assessment area is calculated, including: Obtain time-series data of equipment operating parameters in the risk assessment area, and calculate the fluctuation amplitude and frequency of the equipment operating parameters; Obtain environmental parameter data in the risk assessment area, and calculate the correlation coefficient between the environmental parameter data and the equipment operating parameters; The risk index of the risk assessment area is obtained by calculating the risk index based on the fluctuation amplitude, fluctuation frequency and correlation coefficient.

6. The method according to claim 5, characterized in that, The process of prioritizing risk assessment zones based on risk indices and generating risk control schemes corresponding to the prioritized risk assessment zones based on a pre-set risk response strategy library includes: Sort all risk assessment areas from high to low according to their risk index to generate a risk priority list; Retrieve response strategy templates corresponding to the risk assessment area from the risk response strategy library. The response strategy templates include at least equipment maintenance plans, environmental control recommendations, and emergency shutdown plans. Match the corresponding response strategy templates sequentially based on the risk priority list; The response strategy template is adjusted based on the parameters of the risk assessment area to generate the final risk control plan.

7. The method according to claim 6, characterized in that, Match the corresponding response strategy templates sequentially based on the risk priority list, including: Retrieve the equipment maintenance plan template corresponding to the risk assessment area from the risk response strategy library based on the risk priority list; Adjust the steps and tools in the equipment maintenance plan template according to the specific parameters of the risk assessment area to generate the final response strategy template.

8. The method according to any one of claims 1-7, characterized in that, Also includes: Obtain equipment operating status data after the risk control plan is implemented, and compare and analyze it with the equipment operating status data before implementation; Calculate the variation difference of key parameters in the equipment operation status data, and evaluate the implementation effect of the risk control plan based on the variation difference; Update and optimize the strategy templates in the risk response strategy library based on the implementation results.

9. The method according to any one of claims 1-7, characterized in that, Also includes: Obtain historical operating data and a prediction model for the equipment in the risk assessment area, and train the prediction model based on the historical data. Extract key parameter features of equipment operation from the historical operating data; The extracted key parameter features are input into the prediction model, and the time series prediction value of the equipment operating status is generated by the time series analysis algorithm in the prediction model. A device operation prediction map is plotted based on time series forecast values, and the predicted state change range of each device node is marked on the device operation prediction map. Based on the predicted state change areas, identify the equipment nodes that may experience abnormalities in the equipment operation prediction map, and formulate preventive maintenance plans in advance.

10. A smart factory safety management system based on digital twins, characterized in that, include: The acquisition module is used to acquire equipment operating status data of factory equipment; The digital twin module is used to perform mapping analysis on equipment operation status data based on the digital twin model to generate an equipment operation status diagram. The equipment operation status diagram includes at least the equipment nodes corresponding to the factory equipment and the labeling information of the equipment nodes. The labeling information indicates the operation status of the factory equipment. The identification module is used to identify abnormal areas of equipment operation based on the changing trends of key parameters of equipment nodes in the equipment operation status diagram. The region division module is used to divide the abnormal equipment operation region into multiple risk assessment regions by level according to the attribute information of the equipment nodes in the abnormal equipment operation region. The risk assessment module is used to calculate the risk index of each risk assessment zone based on the correlation between equipment operating parameters and environmental parameters in the risk assessment zone; The control scheme module is used to prioritize risk assessment areas based on risk indices and generate risk control schemes corresponding to the ranked risk assessment areas based on a preset risk response strategy library.

Images