Digital operation management system and method

Abstract

The invention discloses a digital operation management system and method, and relates to the technical field of digital operation management, and the system comprises an operation monitoring module which is used for carrying out the communication with a distributed industrial Internet of Things node through a network interface, collecting production environment parameters and process data, and forming equipment baseline parameters; the knowledge graph engine module is used for constructing an equipment relation network and process flow mapping based on the equipment baseline parameters; the predictive maintenance module is used for carrying out equipment life evaluation and fault early warning according to the relational data of the knowledge graph engine module; the decision support module is used for generating a resource scheduling scheme based on equipment life evaluation and fault early warning in combination with a historical decision effect, and transmitting the resource scheduling scheme to the digital twinning display module; and the digital twinning display module is used for displaying the real-time operation state of the factory by utilizing a three-dimensional visualization technology. The data processing efficiency, the resource utilization rate and the flexibility in the production process are improved, and global monitoring and intelligent optimization of the production process are achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of digital operation management, and in particular to a digital operation management system and method. Background Art

[0002] In the traditional industrial production environment, equipment management mostly adopts a passive maintenance strategy, that is, a failure-repair mode, lacking systematic management of the entire life cycle of equipment. In recent years, the integrated application of industrial Internet of Things, big data analysis and artificial intelligence technologies has brought revolutionary changes to the manufacturing industry, making data-driven intelligent decision-making possible. However, existing industrial management systems mostly exhibit fragmented characteristics, and data circulation between subsystems is blocked, resulting in the widespread existence of information islands. At the same time, although traditional monitoring systems can achieve basic parameter collection and display, they lack in-depth understanding of complex process flows and system-level correlation analysis capabilities, and are unable to build a dynamic knowledge network reflecting the interdependent relationship between equipment, which severely restricts the realization of predictive maintenance and intelligent decision support functions. In addition, existing systems mostly adopt a centralized architecture in data processing, resulting in insufficient real-time performance, making it difficult to cope with emergencies in the production environment, and lacking edge computing capabilities, unable to achieve local processing and caching of key data.

[0003] In the prior art, although some advanced systems have begun to attempt to introduce digital twin technology for visual display, its application depth is limited, mostly staying at the level of static 3D models, lacking dynamic mapping and interaction functions with actual process flows. At the same time, in terms of decision support, existing systems mostly rely on preset rules or simple threshold judgments, and are unable to perform global optimization based on the equipment relationship network, resulting in low resource scheduling efficiency. Especially in complex production environments, due to the lack of accurate modeling of the causal relationship between equipment, existing systems are difficult to accurately identify the root cause of failures and predict the risk of cascading failures, resulting in unreasonable allocation of maintenance resources and a decline in production efficiency. On the other hand, existing systems generally lack a unified knowledge graph engine as the core architecture, and are unable to achieve deep integration and intelligent analysis of equipment data, process flows and maintenance decisions, which makes it difficult for the system to adapt to the dynamic changes in the production environment and the continuous optimization of process requirements.

[0004] In summary, existing industrial operation management technologies have problems such as information islands, lack of system-level correlation analysis, insufficient predictive maintenance capabilities, simplistic decision support, and non-optimal resource scheduling. The present invention proposes a digital operation management system and method, which realizes the accurate mapping of the relationship network between equipment and process flows, solves the problems of real-time data processing and abnormal judgment, and achieves intuitive visualization of the production environment through digital twin technology. Summary of the Invention

[0005] In view of the problems existing in the existing digital operation management system, such as inaccurate equipment health prediction, insufficient analysis of the dependency relationship between devices, and lack of flexibility in resource scheduling optimization, the present invention is proposed.

[0006] Therefore, the problem to be solved by the present invention is how to improve the accuracy of equipment health prediction, optimize the analysis of the dependency relationship between devices, and achieve more flexible and refined management in resource scheduling by introducing a knowledge graph engine, a predictive maintenance module, and digital twin display technology, so as to improve production efficiency, reduce the failure rate, and optimize resource allocation.

[0007] To solve the above technical problems, the present invention provides the following technical solutions:

[0008] In a first aspect, an embodiment of the present invention provides a digital operation management system, which includes an operation monitoring module for communicating with distributed industrial Internet of Things nodes through a network interface, collecting production environment parameters and process data, and forming device baseline parameters; a knowledge graph engine module for constructing a device relationship network and a process flow mapping based on the device baseline parameters; a predictive maintenance module for evaluating the device life and warning of faults according to the relationship data of the knowledge graph engine module; a decision support module for generating a resource scheduling plan based on the device life evaluation and fault warning, in combination with historical decision-making effects, and transmitting it to the digital twin display module; and a digital twin display module for presenting the real-time operation status of the factory using three-dimensional visualization technology and supporting virtual roaming and multi-angle viewing of the production line status.

[0009] As a preferred solution of the digital operation management system of the present invention, wherein: the knowledge graph engine module is respectively connected to the operation monitoring module, the predictive maintenance module, and the digital twin display module; the predictive maintenance module is connected to the decision support module; the decision support module is connected to the digital twin display module; an edge computing unit and a local cache module are provided on the distributed industrial Internet of Things node; the edge computing unit is used to perform data screening and anomaly judgment; the local cache module is used to store key data when the network is interrupted.

[0010] As a preferred solution of the digital operation management system of the present invention, wherein: the operation monitoring module includes a multi-strategy scheduler; the multi-strategy scheduler is used to dynamically allocate computing resources according to the production task priority and resource utilization rate, and trigger a resource expansion mechanism under high load conditions.

[0011] In a second aspect, an embodiment of the present invention provides a digital operation management method, which includes: establishing a digital device file through the operation monitoring module, collecting device operation data collected by distributed industrial Internet of Things nodes, and forming device baseline parameters; based on the device baseline parameters, constructing a process flow knowledge graph, and mapping the dependency relationship between devices and the material flow path through the knowledge graph engine module; performing real-time status monitoring through the predictive maintenance module, comparing the current operation parameters with the device health index of the device baseline parameters, and generating a predictive maintenance plan; based on the predictive maintenance plan, generating a resource scheduling plan through the decision support module, and using a multi-strategy scheduler to balance production efficiency and device life, and optimizing the resource scheduling plan.

[0012] As a preferred solution of the digital operation management method of the present invention, wherein: the method for constructing the process flow knowledge graph is as follows: based on the dynamic operation parameters in the device baseline parameters, extracting the feature vectors of the device state, wherein the feature vectors are processed by dimensionality reduction through principal component analysis; based on the feature vectors, using the Granger causality test method to perform time series correlation analysis on the device operation historical data, and identifying the causal chain between devices; according to the device interaction data in the device baseline parameters and the causal chain, establishing a device relationship matrix, wherein the relationship matrix represents the transfer direction and intensity of the material flow, energy flow and information flow between devices; using the relationship matrix to construct a set of directed edges, assigning edge weight values, and forming an initial graph structure, wherein the edge weight values are comprehensively calculated according to the importance of the material flow, energy flow and information flow; using a community detection algorithm to perform module division on the initial graph structure, identifying the subsystem groups of devices, and combining with the process flow specification document, adding semantic labels and process stage attributes to the subsystem groups; integrating the subsystem groups, semantic labels and process stage attributes to form a hierarchical process flow knowledge graph.

[0013] As a preferred solution of the digital operation management method of the present invention, wherein: when the construction of the process flow knowledge graph is completed, then combining the functional characteristics of the subsystem group and the edge weight values of the relationship matrix, calculating the device health index S(t), and the specific formula is as follows:

[0014]

[0015] wherein, S(t) is the device health index at time t, α is the overall system calibration coefficient, γ i is the importance coefficient calculated based on the knowledge graph relationship matrix of the i-th device baseline parameter, λ i (t) is the time decay function of the i-th device baseline parameter, V i (t) is the normalized value of the i-th device baseline parameter at time t, μ iis the historical mean of the baseline parameters of the i-th device, β is the system degradation rate coefficient, n is the total number of device baseline parameters, and t is the time variable.

[0016] As a preferred solution of the digital operation management method described in the present invention, wherein: the method for generating the predictive maintenance plan is that when the warning condition of the device health index S(t) is triggered, the maintenance resource scheduling plan is evaluated through the resource coordination coefficient R, where the resource coordination coefficient R is calculated through the device importance and the failure probability; if the device health index S(t) satisfies the first condition and the resource coordination coefficient R ≤ 0.3, the online maintenance method is adopted without interrupting production, and the device operation data is collected through the predictive maintenance module and displayed in green in the digital twin display module; if the device health index S(t) satisfies the second condition and 0.3 < resource coordination coefficient R ≤ 0.7, it enters the warning state and is flashed and prompted in yellow in the digital twin display module, and the planned shutdown maintenance is arranged, and at the same time, the edge computing unit is started for real-time parameter analysis; if the device health index S(t) satisfies the third condition or the resource coordination coefficient R > 0.7, the emergency shutdown procedure is executed and the standby device is started, and it is flashed in red in the digital twin display module; if the maintenance count of a certain device is greater than or equal to the preset maintenance device count and the device health index S(t) continuously triggers a warning after maintenance, the device update evaluation process is started.

[0017] As a preferred solution of the digital operation management method described in the present invention, wherein: the optimization method of the resource scheduling plan is to generate the maintenance resource demand prediction and the maintenance strategy type through the multi-strategy scheduler based on the process stage attributes of the subsystem group; according to the maintenance resource demand prediction and the maintenance strategy type, the maintenance tasks are classified and the corresponding resource allocation plan is established; when resource competition occurs, the resource scheduling plan is dynamically optimized based on the hierarchical structure of the process flow knowledge graph and combined with the production task priority.

[0018] In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, where the memory stores a computer program, and wherein: when the computer program instructions are executed by the processor, the steps of the digital operation management system described in the first aspect of the present invention are implemented.

[0019] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and wherein: when the computer program instructions are executed by the processor, the steps of the digital operation management system described in the first aspect of the present invention are implemented.

[0020] The beneficial effects of the present invention are as follows: By integrating an operation monitoring module, a knowledge graph engine module, a predictive maintenance module, a decision support module, and a digital twin display module, real-time monitoring of equipment status, fault warning, and intelligent resource scheduling are achieved; by collecting production data and constructing baseline parameters of equipment, the system can accurately evaluate the health status of equipment, predict faults, and optimize maintenance plans; at the same time, combined with the process knowledge graph, the system dynamically adjusts resource allocation, improves production efficiency, and reduces downtime; the digital twin display provides intuitive real-time status feedback, helps managers make decisions quickly, and enhances the operational reliability and flexibility of the production line. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts. Among them:

[0022] Figure 1 It is a system framework diagram of the digital operation management system in Embodiment 1.

[0023] Figure 2 It is an architecture diagram of the operation monitoring module of the digital operation management system in Embodiment 1.

[0024] Figure 3 It is a flowchart of the digital operation management method in Embodiment 2. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0025] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following will describe the specific embodiments of the present invention in detail with reference to the drawings of the specification.

[0026] Many specific details are set forth in the following description in order to provide a thorough understanding of the present invention. However, the present invention may be implemented in other ways different from those described herein. Those skilled in the art can make similar promotions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0027] Secondly, the so-called "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation manner of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment that excludes other embodiments.

[0028] Embodiment 1

[0029] Refer to Figures 1 to 2, which is the first embodiment of the present invention. This embodiment provides a digital operation management system, including an operation monitoring module, a knowledge graph engine module, a predictive maintenance module, a decision support module, and a digital twin display module.

[0030] Specifically, the operation monitoring module is used to communicate with distributed industrial Internet of Things nodes through a network interface, collect production environment parameters and process data, and form device baseline parameters; the knowledge graph engine module constructs a device relationship network and a process flow mapping based on the device baseline parameters; the predictive maintenance module is used to evaluate the device life and give early warnings of faults according to the relationship data of the knowledge graph engine module; the decision support module generates a resource scheduling plan based on the device life evaluation and fault early warning, combined with historical decision-making effects, and transmits it to the digital twin display module; the digital twin display module is used to present the real-time operation status of the factory using three-dimensional visualization technology and support virtual roaming and multi-angle viewing of the production line status.

[0031] Furthermore, the knowledge graph engine module is respectively connected to the operation monitoring module, the predictive maintenance module, and the digital twin display module; the predictive maintenance module is connected to the decision support module; the decision support module is connected to the digital twin display module.

[0032] Even further, an edge computing unit and a local cache module are provided on the distributed industrial Internet of Things node; the edge computing unit is used to perform data screening and anomaly judgment; the local cache module is used to store key data when the network is interrupted.

[0033] Specifically, the operation monitoring module includes a multi-strategy scheduler; the multi-strategy scheduler is used to dynamically allocate computing resources according to the production task priority and resource utilization rate, and trigger a resource expansion mechanism under high load conditions.

[0034] Embodiment 2

[0035] Refer to Figure 3 , which is the second embodiment of the present invention. This embodiment also provides a digital operation management method, including:

[0036] S1: Establish a device digital file through the operation monitoring module, collect the device operation data collected by the distributed industrial Internet of Things nodes, and form device baseline parameters.

[0037] Specifically, the process of establishing the device digital file is to collect the device basic information and use a data template to structurally store the device basic information according to the ISO 14224 standard to ensure the standardization and traceability of the data; the device basic information includes the model specifications of the device, manufacturer information, installation date, and maintenance records.

[0038] Furthermore, deploy IoT nodes according to the importance of the equipment and monitoring requirements, adopting a hierarchical deployment strategy; for critical equipment, configure high-performance edge computing nodes with a sampling frequency that can reach the millisecond level; for auxiliary equipment, use standard nodes with a sampling frequency set at the second level; each node is equipped with a data preprocessing module to perform signal filtering, anomaly detection, and data compression in real time, effectively reducing the data transmission load.

[0039] Even further, collect equipment operation data through an industrial-grade sensor network and achieve unified access to the data through a protocol conversion gateway; based on the equipment operation data, use a dynamic adaptive algorithm to form baseline parameters of the equipment; at the same time, implement a hierarchical caching strategy during the data collection process, set up a local cache pool at the edge node, and ensure the continuity and integrity of the data when the network fluctuates.

[0040] S2: Based on the baseline parameters of the equipment, construct a process flow knowledge graph, and map the dependency relationships and material flow paths between equipment through the knowledge graph engine module.

[0041] Specifically, the method for constructing the process flow knowledge graph is to extract the feature vectors of the equipment status based on the dynamic operation parameters in the baseline parameters of the equipment, where the feature vectors are processed by dimensionality reduction through principal component analysis.

[0042] Furthermore, based on the feature vectors, use the Granger causality test method to perform time series correlation analysis on the historical operation data of the equipment to identify the causal chains between equipment; according to the interaction data between equipment and the causal chains in the baseline parameters of the equipment, establish a relationship matrix between equipment, where the relationship matrix represents the transfer directions and intensities of material flow, energy flow, and information flow between equipment.

[0043] It should be noted that each directed edge represents a specific type of relationship from the upstream equipment to the downstream equipment; add material transfer specifications, energy consumption, and quality impact coefficients to each relationship edge in the set of directed edges; the material transfer specifications record the material type, flow range, and ideal flow value; the energy consumption represents the energy conversion efficiency during the material transfer process; the quality impact coefficient quantifies the degree of influence of the change in the upstream equipment status on the quality of the downstream product.

[0044] Even further, use the relationship matrix to construct a set of directed edges, assign edge weight values, and form an initial graph structure, where the edge weight values are comprehensively calculated according to the importance of material flow, energy flow, and information flow.

[0045] Specifically, use the community detection algorithm to perform module partitioning on the initial graph structure, identify the subsystem groups of the equipment, and combine with the process flow specification document to add semantic labels and process stage attributes to the subsystem groups.

[0046] Further, integrate subsystem groups, semantic tags, and process stage attributes to form a hierarchical process flow knowledge graph.

[0047] Furthermore, when the process flow knowledge graph is constructed, combine the functional characteristics of the subsystem group and the edge weight values of the relationship matrix to calculate the equipment health index S(t). The specific formula is as follows:

[0048]

[0049] Where S(t) is the equipment health index at time t, α is the overall system calibration coefficient, γ i is the importance coefficient of the baseline parameter of the i-th equipment calculated based on the knowledge graph relationship matrix, λ i (t) is the time decay function of the baseline parameter of the i-th equipment, V i (t) is the normalized value of the baseline parameter of the i-th equipment at time t, μ i is the historical mean of the baseline parameter of the i-th equipment, β is the system degradation rate coefficient, n is the total number of equipment baseline parameters, and t is the time variable.

[0050] S3: Execute real-time status monitoring through the predictive maintenance module, compare the equipment health index of the current operating parameters and the equipment baseline parameters, and generate a predictive maintenance plan.

[0051] Specifically, the method for generating the predictive maintenance plan is that when the warning condition of the equipment health index S(t) is triggered, evaluate the maintenance resource scheduling plan through the resource coordination coefficient R, where the resource coordination coefficient R is calculated through the equipment importance and the failure probability.

[0052] Further, if the equipment health index S(t) meets the first condition and the resource coordination coefficient R ≤ 0.3, then adopt the online maintenance method without interrupting production, collect the equipment operation data through the predictive maintenance module, and display it in green in the digital twin display module; if the equipment health index S(t) meets the second condition and 0.3 < resource coordination coefficient R ≤ 0.7, then enter the warning state, and flash and prompt in yellow in the digital twin display module, arrange planned shutdown maintenance, and at the same time start the edge computing unit for real-time parameter analysis; if the equipment health index S(t) meets the third condition or the resource coordination coefficient R > 0.7, then execute the emergency shutdown procedure and start the standby equipment, and flash in red in the digital twin display module.

[0053] It should be noted that the first condition is that the equipment health index S(t) is in a slightly deviated state, that is, 0.85 ≤ equipment health index S(t) < 0.95; the specific manifestation is that some monitoring parameters of the equipment begin to deviate from the baseline value, but the amplitude of the equipment health index is small and there is no rapid deterioration trend; the second condition is that the equipment health index S(t) is in a medium warning state, that is, 0.70 ≤ equipment health index S(t) < 0.85; the specific manifestation is that multiple key monitoring parameters are abnormal at the same time, or a single important parameter continuously deviates from the baseline value and shows a gradually deteriorating trend; the third condition is that the equipment health index S(t) is in a severe warning state, that is, the equipment health index S(t) < 0.70; the specific manifestation is that the core parameter seriously deviates from the baseline value, or there is a collaborative deterioration phenomenon of multiple parameters.

[0054] Furthermore, if the maintenance count of a certain equipment is greater than or equal to the preset maintenance equipment count and the equipment health index S(t) continues to trigger a warning after maintenance, the equipment update evaluation process is started.

[0055] S4: Based on the predictive maintenance plan, generate a resource scheduling plan through the decision support module, and use the multi-strategy scheduler to balance production efficiency and equipment life to optimize the resource scheduling plan.

[0056] Specifically, the optimization method of the resource scheduling plan is to generate a prediction of maintenance resource requirements and maintenance strategy types through the multi-strategy scheduler based on the process stage attributes of the subsystem group; according to the prediction of maintenance resource requirements and maintenance strategy types, classify the maintenance tasks and establish a corresponding resource allocation plan.

[0057] Further, when resource competition occurs, the resource scheduling plan is dynamically optimized based on the hierarchical structure of the process flow knowledge graph and combined with the production task priority.

[0058] This embodiment also provides a computer device applicable to the digital operation management system, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the digital operation management system proposed in the above embodiment.

[0059] The computer device may be a terminal, which includes a processor, a memory, a communication interface, a display screen, and an input device connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with external terminals in a wired or wireless manner, and the wireless manner can be achieved through WIFI, operator networks, NFC (Near Field Communication), or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball, or touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0060] This embodiment also provides a storage medium, on which a computer program is stored. When the program is executed by a processor, the following steps are implemented: an operation monitoring module, which is used to communicate with distributed industrial Internet of Things nodes through a network interface, collect production environment parameters and process data, and form device baseline parameters; a knowledge graph engine module, which constructs a device relationship network and a process flow mapping based on the device baseline parameters; a predictive maintenance module, which is used to evaluate the device life and give a fault warning according to the relationship data of the knowledge graph engine module; a decision support module, which generates a resource scheduling plan based on the device life evaluation and fault warning, combined with historical decision-making effects, and transmits it to the digital twin display module; the digital twin display module, which is used to present the real-time operation state of the factory using three-dimensional visualization technology, and support virtual roaming and multi-angle viewing of the production line state.

[0061] In summary, through the integration of the operation monitoring module, the knowledge graph engine module, the predictive maintenance module, the decision support module, and the digital twin display module, the present invention realizes the real-time monitoring of device status, fault warning, and intelligent resource scheduling; by collecting production data and constructing device baseline parameters, the system can accurately evaluate the device health status, predict faults, and optimize the maintenance plan; at the same time, combined with the process flow knowledge graph, the system dynamically adjusts resource allocation, improves production efficiency, and reduces downtime; the digital twin display provides intuitive real-time status feedback, helps managers make decisions quickly, and improves the operation reliability and flexibility of the production line.

[0062] Embodiment 3

[0063] Referring to Table 1, this is the third embodiment of the present invention. This embodiment provides a digital operation management system. In order to verify the beneficial effects of the present invention, scientific demonstrations are carried out through economic benefit calculations and simulation experiments.

[0064] Specifically, industrial IoT nodes with edge computing capabilities are deployed on each key device. Each node is configured with an Intel Core i5 processor and 32GB of local storage space for performing real-time data processing and temporary storage. The operation monitoring module establishes a secure channel with these nodes through the industrial Ethernet to collect 28 key process parameters including equipment vibration, temperature, pressure, and flow rate, etc., with a sampling frequency of 100ms. The system adopts a hierarchical architecture, and the edge nodes perform preliminary data filtering and anomaly detection, and transmit the processed data to the central server.

[0065] Furthermore, the knowledge graph engine module constructs a device relationship network using the Neo4j graph database, and reduces the original 28-dimensional parameters to 8 main feature vectors through the principal component analysis method. On this basis, an improved Granger causality test algorithm (confidence threshold set to 0.95) is used to analyze the correlation between devices, and finally a device relationship graph containing 267 nodes and 1,384 edges is constructed; the system uses the Louvain community detection algorithm to divide the devices into 12 functional subsystems.

[0066] Even further, the predictive maintenance module constructs a fault prediction model based on the long short-term memory network. This model is trained on 3 years of historical data, and the cross-validation method is used to evaluate the model performance. The decision support module integrates a resource scheduling algorithm based on reinforcement learning, which can dynamically generate a scheduling plan according to the device health index, production task priority, and maintenance resource availability. In addition, the system also deploys a digital twin visualization platform developed based on the Unity3D engine, which supports millisecond-level status updates and multi-dimensional data display.

[0067] Specifically, during the system deployment process, 6 typical working conditions in the production line are selected for comparative testing, namely: normal production working condition, equipment performance slightly degraded working condition, equipment fluctuating working condition, single equipment failure working condition, multi-equipment collaborative failure working condition, and extreme working condition (such as sudden power outage). 72 hours of continuous operation data is collected under each working condition, and the warning accuracy, decision response time, and resource scheduling efficiency of the system are analyzed through comparison.

[0068] Furthermore, as shown in Table 1, for the early warning accuracy analysis: The system demonstrates an extremely high early warning accuracy rate (99.8%) under normal production conditions, and can still maintain an accuracy rate of 91.5% even under extreme conditions. This is due to the innovative integration of the equipment relationship network and the process flow mapping in the knowledge graph engine module, which can accurately capture the complex dependencies between equipment. For the system response performance analysis: The average response time increases with the increase in the complexity of the working conditions, from 127 ms under normal conditions to 245 ms under extreme conditions, but it is still far lower than the industry's general requirement of 500 ms response time. This is mainly attributed to the local processing ability of the edge computing unit and the dynamic resource allocation mechanism of the multi-strategy scheduler.

[0069] Table 1 Test Data Sheet

[0070]

[0071] Furthermore, the maintenance cost reduction rate shows a significant upward trend, reaching a maximum of 32.7% under extreme conditions. This indicates that the predictive maintenance module of the present invention can accurately predict equipment failures and significantly reduce unnecessary maintenance expenditures through an optimized resource scheduling plan. For the production efficiency improvement: The equipment utilization rate and production efficiency have increased by 8.7% - 13.2% and 6.9% - 10.7% respectively. This improvement is due to the innovative mechanism of the decision support module to dynamically optimize the resource scheduling plan based on historical decision-making effects.

[0072] Specifically, for the energy efficiency optimization: By using digital twin technology to monitor and optimize the production process in real time, the system has achieved a 7.2% - 14.5% reduction in energy consumption. This effect is most significant under extreme conditions, proving that the present invention can maintain efficient energy management under complex working conditions. For the downtime management: The system has significantly reduced the unplanned downtime, with a reduction rate of 12.5% - 28.9%. This improvement is mainly due to the accurate assessment of the equipment life and the fault early warning mechanism of the predictive maintenance module, combined with the optimized scheduling plan generated by the decision support module, to achieve precise arrangement of maintenance activities.

[0073] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not restrictive. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A digital operation management system, characterized in that: include, Operation monitoring module, which is used to communicate with distributed industrial IoT nodes through network interfaces, collect production environment parameters and process data, and form equipment baseline parameters; The knowledge graph engine module builds equipment relationship networks and process flow mapping based on equipment baseline parameters; A predictive maintenance module, used to perform equipment life assessment and fault warning based on the relational data of the knowledge graph engine module; The decision support module generates resource scheduling plans based on equipment life assessment and fault warning, combined with historical decision-making results, and transmits them to the digital twin display module; The digital twin display module is used to present the real-time operation status of the factory using 3D visualization technology, and supports virtual roaming and multi-angle viewing of the production line status.

2. The digital operation management system according to claim 1, characterized in that: The knowledge graph engine module is respectively connected to the operation monitoring module, the predictive maintenance module and the digital twin display module; the predictive maintenance module is connected to the decision support module; the decision support module is connected to the digital twin display module; the distributed industrial Internet of Things node is provided with an edge computing unit and a local cache module; the edge computing unit is used to perform data screening and abnormality judgment; the local cache module is used to store key data when the network is interrupted.

3. The digital operation management system according to claim 2, characterized in that: The operation monitoring module includes a multi-strategy scheduler; the multi-strategy scheduler is used to dynamically allocate computing resources according to production task priorities and resource utilization, and trigger a resource expansion mechanism under high load conditions.

4. A digital operation management method, based on the digital operation management system according to any one of claims 1 to 3, characterized in that: include, Establishing a digital file of equipment through the operation monitoring module, collecting equipment operation data collected by distributed industrial Internet of Things nodes, and forming equipment baseline parameters; Based on the equipment baseline parameters, a process knowledge graph is constructed, and the dependencies between equipment and the material flow path are mapped through the knowledge graph engine module; Performing real-time condition monitoring through a predictive maintenance module, comparing the equipment health index of current operating parameters and baseline parameters of the equipment, and generating a predictive maintenance plan; Based on the predictive maintenance plan, a resource scheduling plan is generated through the decision support module, and a multi-strategy scheduler is used to balance production efficiency and equipment life to optimize the resource scheduling plan.

5. The digital operation management method according to claim 4, characterized in that: The method for constructing the process knowledge graph is as follows: Extracting a characteristic vector of the device state based on the dynamic operating parameters in the device baseline parameters, wherein the characteristic vector is subjected to dimensionality reduction processing by principal component analysis; Based on the characteristic vector, the Granger causality test method is used to perform time series correlation analysis on the historical data of equipment operation to identify the causal chain between equipment; Establishing an inter-device relationship matrix based on the inter-device interaction data in the device baseline parameters and the causal chain, wherein the relationship matrix describes the transmission direction and intensity of material flow, energy flow and information flow between devices; Using the relationship matrix to construct a directed edge set, assigning edge weight values, and forming an initial graph structure, wherein the edge weight values ​​are comprehensively calculated based on the importance of material flow, energy flow, and information flow; Using a community detection algorithm to divide the initial graph structure into modules, identify subsystem groups of equipment, and add semantic labels and process stage attributes to the subsystem groups in combination with process flow specification documents; The subsystem groups, semantic labels and process stage attributes are integrated to form a hierarchical process flow knowledge graph.

6. The digital operation management method according to claim 5, characterized in that: When the process knowledge graph is constructed, the equipment health index S(t) is calculated by combining the functional characteristics of the subsystem group and the edge weight value of the relationship matrix. The specific formula is as follows: Among them, S(t) is the equipment health index at time t, α is the overall system calibration coefficient, and γ i is the importance coefficient of the baseline parameter of the i-th device calculated based on the knowledge graph relationship matrix, λ i (t) is the time decay function of the baseline parameter of the i-th device, V i (t) is the standardized value of the baseline parameter of the ith device at time t, μ i is the historical mean of the baseline parameter of the ith device, β is the system degradation rate coefficient, n is the total number of device baseline parameters, and t is the time variable.

7. The digital operation management method according to claim 5 or 6, characterized in that: The method for generating the predictive maintenance plan is: When the warning condition of the equipment health index S(t) is triggered, the maintenance resource scheduling plan is evaluated by the resource coordination coefficient R, wherein the resource coordination coefficient R is calculated by the equipment importance and the failure probability; If the equipment health index S(t) meets the first condition and the resource coordination coefficient R≤0.3, online maintenance is adopted without interrupting production, and the equipment operation data is collected through the predictive maintenance module and displayed in green in the digital twin display module; If the equipment health index S(t) meets the second condition and 0.3<resource coordination coefficient R≤0.7, it will enter the warning state, and a yellow flashing reminder will be displayed in the digital twin display module. Planned downtime maintenance will be arranged, and the edge computing unit will be started for real-time parameter analysis. If the equipment health index S(t) meets the third condition or the resource coordination coefficient R>0.7, the emergency shutdown procedure is executed and the backup equipment is started, and the red mark flashes in the digital twin display module; If the maintenance count of a device is greater than or equal to the preset maintenance device count and the device health index after maintenance If S(t) continues to trigger warnings, the equipment update evaluation process will be initiated.

8. The digital operation management method according to claim 7, characterized in that: The optimization method of the resource scheduling scheme is: Based on the process stage attributes of the subsystem group, the maintenance resource demand forecast and maintenance strategy type are generated through the multi-strategy scheduler; According to the maintenance resource demand forecast and the maintenance strategy type, the maintenance tasks are classified and corresponding resource allocation plans are established; When resource competition occurs, the resource scheduling plan is dynamically optimized based on the hierarchical structure of the process knowledge graph and combined with the production task priority.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the digital operation management system according to any one of claims 1 to 3 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the digital operation management system according to any one of claims 1 to 3 are implemented.

Images