An AI intelligent operation and maintenance system and method based on digital twins

Abstract

The present invention discloses an AI intelligent operation and maintenance system and method based on digital twins, which belongs to the field of intelligent operation and maintenance technology. The present invention constructs a three-dimensional map, collects energy data and equipment data; constructs an energy digital twin model and an equipment digital twin model, and integrates them into the three-dimensional map; determines whether the current energy data and current equipment data are abnormal; marks the location of abnormal equipment or abnormal energy nodes on the three-dimensional map, performs correlation analysis on the digital twin model, and obtains the abnormal energy node source; when the energy output of the abnormal energy node source reaches a bottleneck, constructs an energy and equipment relationship diagram; analyzes the correlation path and closeness between the abnormal energy node source and the abnormal equipment through the shortest path algorithm to obtain a graph structure analysis result; determines the priority of the abnormal equipment, analyzes its adjustable parameters for abnormal equipment of different priorities, and preferentially adjusts the parameters of abnormal equipment with low equipment priority.

Description

Technical Field

[0001] The present invention relates to the field of intelligent operation and maintenance technology, and specifically to an AI intelligent operation and maintenance system and method based on digital twins. Background Art

[0002] As the core of modern transportation, high-speed rail stations host massive passenger flows and complex functions within their concourses. Within these concourses, power supply, heating, ventilation, and other energy systems are intertwined, and numerous devices, including lighting, air conditioning units, and escalators, operate continuously. The stable and efficient operation of these energy systems and equipment is crucial to the smooth operation of high-speed rail stations, creating a comfortable waiting environment and ensuring safe and convenient travel for passengers.

[0003] Under the existing technology system, energy data and equipment data are independently collected and managed by different specialized systems. This results in a lack of real-time interaction and deep integration between the two types of data, making it impossible to gain macro-level insight into the inherent connections between energy and equipment. Traditional methods often use general, simple mathematical models to construct energy and equipment models, failing to fully consider the complex actual operating conditions and equipment characteristics of high-speed rail station concourses. When dealing with abnormal equipment, existing technologies lack a comprehensive consideration of the equipment's importance and the impact of failures on the production process. This can lead to excessive maintenance of low-priority equipment and waste of resources. Summary of the Invention

[0004] The purpose of the present invention is to provide an AI intelligent operation and maintenance system and method based on digital twins to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] In a first aspect, the present invention provides an AI intelligent operation and maintenance method based on digital twins, comprising the following steps:

[0007] Determine the 3D map scope of the high-speed railway station square, construct the 3D map, and collect energy and equipment data. Based on the collected data, use the multi-layer perceptron algorithm to construct energy digital twin models and equipment digital twin models and integrate them into the 3D map.

[0008] Based on the digital twin model in the 3D map, determine whether there are anomalies in the current energy data and current equipment data. If anomalies exist, mark the location of the abnormal equipment or abnormal energy node on the 3D map, and perform correlation analysis on the digital twin model to obtain the source of the abnormal energy node.

[0009] When it is confirmed that the energy output of an abnormal energy node source has reached a bottleneck, a graph database is used to construct an energy and equipment relationship graph based on the location of the abnormal energy node source in the three-dimensional map. The shortest path algorithm is used to analyze the association path and closeness between the abnormal energy node source and the abnormal equipment to obtain the graph structure analysis results.

[0010] Based on the results of graph structure analysis, combined with the importance of the equipment and the impact of equipment failure on the production process, the priority of abnormal equipment is determined; for abnormal equipment of different priorities, their adjustable parameters are analyzed separately, and an abnormal equipment parameter adjustment table is generated, with parameter adjustment of abnormal equipment with low equipment priority being given priority.

[0011] In conjunction with the first aspect, in a first implementation of the first aspect of the present application, determining the three-dimensional map range of the high-speed railway station square, constructing the three-dimensional map, and collecting energy data and equipment data include:

[0012] Determine the 3D map scope of the high-speed railway station concourse, including the main area and surrounding areas of the concourse; the main area includes the passenger waiting area, ticket hall, entrance and exit passages, underground parking lot, bus transfer area, and taxi transfer area; the surrounding areas specifically include areas related to energy supply and equipment; mark the outlines and relative positions of each area;

[0013] Using a laser scanner, scanning sites are planned according to the layout of the high-speed railway station square. Three-dimensional point cloud data is acquired at each scanning site. Point cloud processing software is used to stitch and fuse the point cloud data from multiple sites to generate a three-dimensional point cloud model. Using a drone, a flight route is planned and camera parameters are set based on the range of the three-dimensional map. During the drone flight, the flight status and image acquisition are monitored in real time. The acquired image data is imported into three-dimensional reconstruction software, and a three-dimensional map is generated by combining the image stitching and three-dimensional reconstruction algorithms with the three-dimensional point cloud model. SketchUp software is used to model the equipment within the high-speed railway station square, including power supply equipment and power-consuming equipment. After modeling, the equipment is placed on the three-dimensional map based on its actual position and orientation in the square.

[0014] The energy data includes energy nodes and energy consumption, and the equipment data includes equipment operation data, equipment control instructions and equipment basic information.

[0015] In combination with the first aspect, in a second implementation of the first aspect of this application, the energy digital twin model and the equipment digital twin model are constructed based on the collected data using a multi-layer perceptron algorithm and integrated into the three-dimensional map, including:

[0016] When building energy digital twin models and equipment digital twin models, the number of neurons in their respective input layers is determined based on the number of features of the energy data and equipment data after feature engineering. Grid search and cross-validation are used to debug different combinations of the number of neurons.

[0017] The weights of the digital twin model are initialized using the Xavier initialization method. The training set data is input in batches, and the batch size is determined through experiments. For each batch of data, the predicted output is calculated through forward propagation, and the loss value is calculated compared with the true label. The gradient of the loss value with respect to the digital twin model parameters is calculated using the backpropagation algorithm. The parameters are updated through the optimizer, and the performance of the digital twin model is evaluated using the validation set.

[0018] By creating a custom data source, the digital twin model is encapsulated as an entity and added to the data source, and the data source is integrated into the three-dimensional map.

[0019] In combination with the first aspect, in a third implementation of the first aspect of the present application, determining whether there are abnormalities in current energy data and current device data based on the digital twin model in the three-dimensional map includes:

[0020] For energy data, historical data is collected and combined with energy system design standards to calculate statistics, including mean and standard deviation. Based on these statistics, normal ranges for various energy parameters in the energy data are set. For equipment data, the allowable deviation ranges for equipment parameters are determined for different types of equipment according to manufacturer settings.

[0021] Input the current energy data and current equipment data into the digital twin model in the three-dimensional map, and output the current energy parameters and current equipment parameters; compare the current energy parameters with the normal range of each energy parameter, and compare the current equipment parameters with the allowable deviation range of the equipment parameters to determine whether there are any abnormalities in the current energy data and current equipment data.

[0022] In combination with the first aspect, in a fourth implementation of the first aspect of the present application, when an anomaly occurs, the location of the abnormal device or abnormal energy node is marked on the three-dimensional map, and the digital twin model is subjected to correlation analysis to obtain the abnormal energy node source, including:

[0023] When it is determined that energy data or equipment data is abnormal, the coordinate information of the equipment or energy node involved in the three-dimensional map is extracted from the abnormal data record and marked in the three-dimensional map;

[0024] Extract all energy data and equipment data within the time period related to the current anomaly from the digital twin model; construct data association relationships based on the logical relationship descriptions of the existing energy nodes and equipment in the digital twin model; starting from the abnormal device or energy node, use a combination of forward reasoning and reverse tracing to analyze the possible propagation path of the anomaly based on the established data association relationship; for each energy node that may be related to the anomaly, calculate the probability of the anomaly source based on the closeness of its association with the abnormal device and the severity of the data anomaly; verify the possible abnormal energy node sources, and when a power supply equipment failure is found in an energy node during an on-site inspection, and the failure matches the abnormality analyzed in the digital twin model, determine that the energy node is an abnormal energy node source; mark the abnormal energy node source on the three-dimensional map.

[0025] In conjunction with the first aspect, in a fifth implementation of the first aspect of the present application, when it is confirmed that the energy output of the abnormal energy node source has reached a bottleneck, based on the location of the abnormal energy node source in the three-dimensional map, a graph database is used to construct an energy and equipment relationship graph, including:

[0026] When it is confirmed that the energy output of an abnormal energy node source has reached a bottleneck, the energy data of the abnormal energy node source is collected with the abnormal energy node source location marked in the three-dimensional map as the center, and the energy data and equipment data directly or indirectly related to the abnormal energy node source are collected;

[0027] The graph database uses Neo4j, and defines abnormal energy node sources and other related energy nodes and devices as graph database nodes; each energy node and device is given a unique identifier to determine the attributes of the graph database node;

[0028] Define energy transmission edges to describe the energy transmission relationship between different energy nodes; define device connection edges to describe the physical connection relationship between devices; define control relationship edges, and define them as edges when there is a control relationship between devices; call the ForceAtlas2 algorithm to calculate the layout of the graph database nodes, store the calculated graph database node position coordinates in the node attributes, and construct an energy and device relationship graph.

[0029] In combination with the first aspect, in a sixth implementation of the first aspect of the present application, the analysis of the association path and closeness between the abnormal energy node source and the abnormal device using the shortest path algorithm to obtain a graph structure analysis result includes:

[0030] Extract a subgraph related to abnormal energy node sources and abnormal devices from the energy and device relationship graph; use query statements to find the nodes corresponding to the abnormal energy node sources and abnormal devices, and obtain all nodes and edges with connected relationships in the corresponding nodes; set weights for the edges in the subgraph to reflect the closeness between the abnormal energy node sources and abnormal devices;

[0031] The shortest path algorithm adopts the Dijkstra algorithm to calculate the shortest path from the abnormal energy node source to the abnormal device on the constructed subgraph; based on the calculated shortest path, the association path between the abnormal energy node source and the abnormal device is analyzed, and through the nodes and edges in the association path, it is determined how the energy is transmitted from the abnormal energy node source to the abnormal device, as well as the energy nodes and devices connected in the middle; the closeness between the abnormal energy node source and the abnormal device is evaluated based on the length of the shortest path; and the graph structure analysis results are obtained, including the node sequence, path length and closeness of the shortest path.

[0032] In conjunction with the first aspect, in a seventh implementation of the first aspect of the present application, determining the priority of abnormal equipment based on the graph structure analysis results, combined with the importance of the equipment and the impact of the equipment failure on the production process, includes:

[0033] Set evaluation indicators, including the functional criticality, frequency of use, and difficulty of maintenance of the equipment, to determine the importance of the equipment; distinguish between core equipment, important equipment, and auxiliary equipment in terms of functional criticality; classify frequency of use and difficulty of maintenance into high, medium, and low levels; set a quantitative score range for each evaluation indicator; determine a weight for each evaluation indicator, and perform a weighted calculation on the quantitative scores of the evaluation indicators to determine the equipment importance score;

[0034] Define the specific role and location of each device in the operational process of the high-speed railway station square, and assess the scope and duration of the fault impact. Set weights for the impact scope and impact duration, and perform a weighted calculation of the impact scope and impact duration to obtain a score for the impact degree of the equipment fault.

[0035] The equipment importance score, fault impact score, and graph structure analysis results are comprehensively considered to perform weighted calculation and determine the priority of abnormal equipment.

[0036] In combination with the first aspect, in an eighth implementation of the first aspect of the present application, for abnormal devices of different priorities, respectively analyzing their adjustable parameters, generating an abnormal device parameter adjustment table, and preferentially adjusting parameters for abnormal devices with low device priorities, including:

[0037] According to the determined abnormal device priority scores and sorting results, priority grouping rules are set; all abnormal devices are traversed, and each device is classified into a corresponding priority group according to the priority grouping rules to obtain a priority device list;

[0038] Analyze the adjustable parameters of each priority abnormal device, design the abnormal device parameter adjustment table framework, extract device data from the priority device list in sequence, fill the device's adjustable parameters and device data into the abnormal device parameter adjustment table; give priority to adjusting parameters of abnormal devices with low device priority.

[0039] In a second aspect, the present invention provides an AI intelligent operation and maintenance system based on digital twins, comprising:

[0040] 3D map generation module: This module includes a 3D map construction unit and a digital twin model construction unit. The 3D map construction unit determines the 3D map range of the high-speed railway station square, constructs the 3D map, and collects energy data and equipment data. The digital twin model construction unit uses a multi-layer perceptron algorithm based on the collected data to construct energy digital twin models and equipment digital twin models, integrating them into the 3D map.

[0041] Abnormal energy node source location module: includes: a device data anomaly judgment unit and an abnormal energy node source location unit; wherein, the device data anomaly judgment unit determines whether there are anomalies in the current energy data and current device data based on the digital twin model in the three-dimensional map; when an anomaly exists, the abnormal energy node source location unit marks the location of the abnormal device or abnormal energy node on the three-dimensional map, and performs correlation analysis on the digital twin model to obtain the abnormal energy node source;

[0042] Graph structure analysis module: includes: a relationship graph construction unit and a graph structure analysis unit; wherein, when the relationship graph construction unit confirms that the energy output of an abnormal energy node source has reached a bottleneck, it uses the graph database to construct an energy and equipment relationship graph based on the location of the abnormal energy node source in the three-dimensional map; the graph structure analysis unit uses the shortest path algorithm to analyze the association path and closeness between the abnormal energy node source and the abnormal equipment to obtain the graph structure analysis results;

[0043] Abnormal equipment parameter adjustment module: includes: abnormal equipment priority determination unit, parameter adjustment table generation unit and abnormal equipment parameter adjustment unit; among them, the abnormal equipment priority determination unit determines the priority of the abnormal equipment based on the graph structure analysis results, combined with the importance of the equipment and the impact of equipment failure on the production process; the parameter adjustment table generation unit analyzes the adjustable parameters of abnormal equipment with different priorities, generates an abnormal equipment parameter adjustment table, and the abnormal equipment parameter adjustment unit gives priority to adjusting the parameters of abnormal equipment with low equipment priority.

[0044] Compared with the prior art, the present invention has the following beneficial effects:

[0045] 1. This invention builds a three-dimensional map to accurately associate energy data and equipment data with the map. Through data fusion technology, it breaks down data silos and organically integrates energy and equipment data, providing a rich and accurate data foundation for subsequent analysis and modeling.

[0046] 2. Based on the digital twin model, the present invention can accurately determine in real time whether there are anomalies in current energy data and equipment data. Through in-depth analysis of large amounts of historical data and real-time data, it can make anomaly judgments, discover various anomalies, and issue early warnings for potential faults.

[0047] 3. The present invention comprehensively considers the importance of the equipment, the impact of the fault on the production process, and the results of the graph structure analysis to determine the priority of the abnormal equipment and formulate personalized processing strategies for each. It gives priority to adjusting the parameters of low-priority equipment to optimize equipment performance and energy utilization efficiency without affecting the production process. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 This is a schematic diagram of the steps of an AI intelligent operation and maintenance method based on digital twins of the present invention;

[0049] Figure 2 This is a system structure diagram of an AI intelligent operation and maintenance system based on digital twins in the present invention. DETAILED DESCRIPTION

[0050] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0051] Example: Figure 1-Figure 2 As shown, the present invention provides a technical solution.

[0052] like Figure 1 As shown in the schematic diagram of the steps of an AI intelligent operation and maintenance method based on digital twins, the present invention provides an AI intelligent operation and maintenance method based on digital twins, comprising the following steps:

[0053] Step S100: Determine the 3D map range of the high-speed railway station square, construct the 3D map, and collect energy data and equipment data. Based on the collected data, use the multi-layer perceptron algorithm to construct energy digital twin models and equipment digital twin models, and integrate them into the 3D map.

[0054] Specifically, determine the three-dimensional map scope of the high-speed railway station square, including the main area and surrounding areas of the square; the main area includes the passenger waiting area, ticket hall, entrance and exit passages, underground parking lot, bus transfer area, and taxi transfer area; the surrounding areas are specifically areas related to energy supply and equipment; mark the outline and relative position relationship of each area;

[0055] Using a laser scanner, scanning sites are planned according to the layout of the high-speed railway station square. Three-dimensional point cloud data is acquired at each scanning site. Point cloud processing software is used to stitch and fuse the point cloud data from multiple sites to generate a three-dimensional point cloud model. Using a drone, a flight route is planned and camera parameters are set based on the range of the three-dimensional map. During the drone flight, the flight status and image acquisition are monitored in real time. The acquired image data is imported into three-dimensional reconstruction software, and a three-dimensional map is generated by combining the image stitching and three-dimensional reconstruction algorithms with the three-dimensional point cloud model. SketchUp software is used to model the equipment within the high-speed railway station square, including power supply equipment and power-consuming equipment. After modeling, the equipment is placed on the three-dimensional map based on its actual position and orientation in the square.

[0056] The energy data includes energy nodes and energy consumption, and the equipment data includes equipment operation data, equipment control instructions and equipment basic information.

[0057] Furthermore, when constructing energy digital twin models and equipment digital twin models, the number of neurons in their respective input layers is determined based on the number of features of energy data and equipment data after feature engineering. Through grid search and cross-validation, debugging is performed under different combinations of the number of neurons.

[0058] The weights of the digital twin model are initialized using the Xavier initialization method. The training set data is input in batches, and the batch size is determined through experiments. For each batch of data, the predicted output is calculated through forward propagation, and the loss value is calculated compared with the true label. The gradient of the loss value with respect to the digital twin model parameters is calculated using the backpropagation algorithm. The parameters are updated through the optimizer, and the performance of the digital twin model is evaluated using the validation set.

[0059] By creating a custom data source, the digital twin model is encapsulated as an entity and added to the data source, and the data source is integrated into the three-dimensional map.

[0060] In one specific example, a high-speed rail station concourse was used as the target. The main area of ​​the concourse, spanning 150,000 square meters, includes a typical passenger waiting area, ticket hall, entrance and exit passages, underground parking, bus and taxi transfer areas. Surrounding areas related to energy supply and equipment include three substations, 10 distribution boxes, and various energy transmission lines.

[0061] Given the complex layout of the square, a total of 60 scanning stations were planned to ensure comprehensive coverage of all areas. Twenty stations were set up in the passenger waiting area, 12 in the ticket hall, 10 in the entrance and exit passages, 15 in the underground parking lot, five each in the bus and taxi transfer areas, and eight in the surrounding energy areas. The laser scanner used was a RIEGL VZ-400i, with a scanning accuracy of ±0.03 meters. 3D point cloud data was collected at each station, averaging approximately 600,000 points. PolyWorks point cloud processing software was used for stitching and fusion, with a stitching error within ±0.08 meters. The resulting 3D point cloud model had a resolution of 0.08 meters by 0.08 meters by 0.08 meters.

[0062] Using a DJI Phantom 4 RTK drone, 12 flight routes were planned based on the 3D map to ensure full coverage of the square. Camera parameters were set to a resolution of 5472×3648 pixels, a focal length of 12 mm, and a shooting interval of 4 seconds. During the flight, 1,000 images were captured. These images were imported into Agisoft Metashape 3D reconstruction software. By combining image stitching and 3D reconstruction algorithms with a 3D point cloud model, the resulting 3D map achieved over 92% accuracy in reproducing building surface textures.

[0063] SketchUp was used to model the square's power supply equipment (such as three substations and 10 distribution boxes) and power-consuming equipment (2,000 lighting fixtures, 50 air conditioning units, and 30 escalators). Once modeled, on-site measurements were conducted to accurately position the equipment on the 3D map based on its actual location and orientation within the square, with a positioning error of less than 0.15 meters.

[0064] In the surrounding energy region, three major energy nodes were monitored, including two substations and one large distribution box. Energy consumption data was collected using smart meters and other equipment. Over a seven-day monitoring period, total daily energy consumption fluctuated between 12,000 and 14,000 kWh. Peak hours (9:00-11:00 a.m. and 6:00-8:00 p.m.) accounted for 42% of total energy consumption.

[0065] Operational data was collected for 120 typical devices (such as lighting fixtures, air conditioning units, and escalators). The average daily operating time for lighting fixtures was 13 hours, the average operating power of air conditioning units was between 6 and 9 kilowatts, and the escalators operated approximately 550 times per day. 1,200 pieces of device control command data were collected, covering commands such as start / stop and mode switching. Basic device information included model, manufacturer, and rated power, covering 25 different device models.

[0066] After feature engineering of the energy data, 18 key features were extracted, and the number of input layer neurons was determined to be 18. Through grid search and cross-validation, we tested the hidden layer neuron number combinations [30, 60, 90] and ultimately settled on a hidden layer structure of [60, 90]. After feature engineering of the device data, 20 key features were obtained, and the number of input layer neurons was set to 20. Similarly, through testing, the hidden layer structure was determined to be [70, 100].

[0067] The weights of the digital twin model were initialized using the Xavier initialization method. The training data was input in batches, with a batch size of 64 determined experimentally. During training, the Adam optimizer was used, with an initial learning rate of 0.001. For each batch of data, the predicted output was calculated using forward propagation, the loss was compared with the true labels, and the gradient was calculated using the backpropagation algorithm to update the parameters.

[0068] On the validation set, the root mean square error (RMSE) of the energy digital twin model stabilized at around 80 kWh after 60 rounds of training, and the accuracy of the equipment digital twin model in judging the equipment operating status (normal, warning, fault) reached 93%.

[0069] By creating a custom data source, encapsulate the digital twin model as an entity and add it to the data source.

[0070] Step S200: Based on the digital twin model in the three-dimensional map, determine whether there are any anomalies in the current energy data and current equipment data; if there are any anomalies, mark the location of the abnormal equipment or abnormal energy node on the three-dimensional map, and perform correlation analysis on the digital twin model to obtain the source of the abnormal energy node;

[0071] Specifically, for energy data, historical data is collected and combined with energy system design standards to calculate statistics, including mean and standard deviation. Based on these statistics, normal ranges for various energy parameters in the energy data are set. For equipment data, the allowable deviation ranges for equipment parameters are determined for different types of equipment according to manufacturer settings.

[0072] Input the current energy data and current equipment data into the digital twin model in the three-dimensional map, and output the current energy parameters and current equipment parameters; compare the current energy parameters with the normal range of each energy parameter, and compare the current equipment parameters with the allowable deviation range of the equipment parameters to determine whether there are any abnormalities in the current energy data and current equipment data.

[0073] Furthermore, when it is determined that the energy data or equipment data is abnormal, the coordinate information of the equipment or energy node involved in the three-dimensional map is extracted from the abnormal data record and marked in the three-dimensional map;

[0074] Extract all energy data and equipment data within the time period related to the current anomaly from the digital twin model; construct data association relationships based on the logical relationship descriptions of the existing energy nodes and equipment in the digital twin model; starting from the abnormal device or energy node, use a combination of forward reasoning and reverse tracing to analyze the possible propagation path of the anomaly based on the established data association relationship; for each energy node that may be related to the anomaly, calculate the probability of the anomaly source based on the closeness of its association with the abnormal device and the severity of the data anomaly; verify the possible abnormal energy node sources, and when a power supply equipment failure is found in an energy node during an on-site inspection, and the failure matches the abnormality analyzed in the digital twin model, determine that the energy node is an abnormal energy node source; mark the abnormal energy node source on the three-dimensional map.

[0075] In one specific embodiment, energy data from the high-speed railway station square over the past year was collected. For voltage data, the calculated mean was 382V, with a standard deviation of 5V, and the normal range was set to 372V-392V (mean ± 2 times the standard deviation). For current data, the statistical mean was 200A, with a standard deviation of 15A, and the normal range was set to 170A-230A. Regarding energy consumption, daily energy consumption data was calculated to have a mean of 12,500 kWh, a standard deviation of 1,000 kWh, and a normal range of 10,500-14,500 kWh. At a certain monitoring moment, the current voltage collected was 365V, below the lower limit of the normal range, and the voltage data was determined to be abnormal. The current was 240A, exceeding the upper limit of the normal range, indicating abnormal current data. The energy consumption at that moment was 15,000 kWh, exceeding the upper limit of the normal range, indicating abnormal energy consumption data.

[0076] Taking escalators as an example, the manufacturer sets their normal operating speed at 0.5m / s, with an allowable deviation range of ±0.05m / s. For air conditioning units, the manufacturer stipulates that the allowable deviation of the cooling temperature set point is ±2°C. The normal operating current of lighting fixtures is 0.5A, with an allowable deviation range of ±0.1A. The current operating speed of an escalator is monitored to be 0.6m / s, which exceeds the allowable deviation range, and the escalator speed parameter is determined to be abnormal. The current cooling temperature of an air conditioning unit is 20°C, while the set value should be 24°C, which exceeds the allowable deviation range. The temperature parameter of the air conditioning unit is abnormal. The current operating current of a lighting fixture is 0.7A, which exceeds the allowable deviation range. The current parameter of the lighting fixture is abnormal.

[0077] Once anomalies are identified in the energy and equipment data, coordinate information is extracted from the anomaly data record. For example, an energy node experiencing a voltage anomaly has coordinates (100, 200, 5) on the 3D map and is marked with a flashing red icon. An escalator experiencing an abnormal speed has coordinates (300, 400, 0) on the 3D map and is also marked with a flashing red icon.

[0078] All energy and equipment data from the two hours prior to the anomaly was extracted from the digital twin model. After establishing data associations, the anomaly propagation path was analyzed through a combination of forward reasoning and backward tracing. For example, analysis of the voltage anomaly revealed that a transformer failure at a substation caused the output voltage to drop. When calculating the anomaly source probability, a higher probability of anomaly source was assigned to energy nodes that were closely associated with the anomaly device and exhibited severe data anomalies. For example, the substation node mentioned above had an 80% probability of anomaly source due to its association with multiple anomaly devices and the severity of the voltage anomaly. During an on-site inspection, it was discovered that the transformer at the substation indeed had a winding short-circuit fault, which matched the anomaly detected by the digital twin model. Ultimately, the substation was identified as the source of the anomaly energy node and highlighted with a larger, flashing red icon on the 3D map.

[0079] Step S300: When it is confirmed that the energy output of the abnormal energy node source has reached a bottleneck, a graph database is used to construct an energy and equipment relationship graph based on the location of the abnormal energy node source in the three-dimensional map; the association path and closeness between the abnormal energy node source and the abnormal equipment are analyzed using the shortest path algorithm to obtain a graph structure analysis result;

[0080] Specifically, when it is confirmed that the energy output of the abnormal energy node source has reached a bottleneck, the energy data of the abnormal energy node source is collected with the abnormal energy node source position marked in the three-dimensional map as the center, and the energy data and equipment data directly or indirectly related to the abnormal energy node source are collected;

[0081] The graph database uses Neo4j, and defines abnormal energy node sources and other related energy nodes and devices as graph database nodes; each energy node and device is given a unique identifier to determine the attributes of the graph database node;

[0082] Define energy transmission edges to describe the energy transmission relationship between different energy nodes; define device connection edges to describe the physical connection relationship between devices; define control relationship edges, and define them as edges when there is a control relationship between devices; call the ForceAtlas2 algorithm to calculate the layout of the graph database nodes, store the calculated graph database node position coordinates in the node attributes, and construct an energy and device relationship graph.

[0083] Furthermore, a subgraph related to abnormal energy node sources and abnormal devices is extracted from the energy and device relationship graph. Through query statements, the nodes corresponding to the abnormal energy node sources and abnormal devices are found, and all nodes and edges with connected relationships in the corresponding nodes are obtained. Weights are set for the edges in the subgraph to reflect the closeness between the abnormal energy node sources and abnormal devices.

[0084] The shortest path algorithm adopts the Dijkstra algorithm to calculate the shortest path from the abnormal energy node source to the abnormal device on the constructed subgraph; based on the calculated shortest path, the association path between the abnormal energy node source and the abnormal device is analyzed, and through the nodes and edges in the association path, it is determined how the energy is transmitted from the abnormal energy node source to the abnormal device, as well as the energy nodes and devices connected in the middle; the closeness between the abnormal energy node source and the abnormal device is evaluated based on the length of the shortest path; and the graph structure analysis results are obtained, including the node sequence, path length and closeness of the shortest path.

[0085] In one specific embodiment, the abnormal energy source (substation) is defined as a node with a unique identifier of "substation_01." Attributes include type (substation), location coordinates (coordinates in a 3D map are [150, 250, 10]), rated output power (10,000 kW), and current output power (9,500 kW). Three transmission lines are defined as nodes, each with unique identifiers such as "transmission_line_01," "transmission_line_02," and "transmission_line_03." Attributes include transmission capacity, current transmission power, and transmission loss. For example, the transmission capacity of "transmission_line_01" is 3,000 kW, the current transmission power is 2,800 kW, and the transmission loss is 5%. Five distribution boxes are also defined as nodes, with unique identifiers such as "distribution_box_01" and attributes such as rated capacity and current load power. For example, the rated capacity of "distribution_box_01" is 500 kW, and the current load power is 450 kW. The 50 electrical devices are also defined as nodes, with unique identifiers such as "lighting_01" and "air_conditioner_01." Their attributes include device type, rated power, and current operating status. For example, the device type for "lighting_01" is a lighting fixture, with a rated power of 100W and a current operating status of On.

[0086] Attributes for defining an energy transmission edge, such as the edge from a substation to "transmission_line_01," include the transmitted energy type (electricity), transmission capacity (3000kW), current transmission power (2800kW), and transmission loss (5%). Attributes for a device connection edge, such as the edge from "distribution_box_01" to "lighting_01," include the connection line specifications (e.g., RVV2.5) and length (20 meters). If a control relationship exists, such as a control edge from a control system to "air_conditioner_01," attributes include the control signal type (on / off signal) and control priority (high).

[0087] The ForceAtlas2 algorithm in Neo4j is used to calculate the layout of the graph database nodes. After 1000 iterations, the coordinates of each node are obtained. For example, the coordinates of "substation_01" are updated to [155, 255, 10], and the coordinates of "distribution_box_01" are updated to [180, 270, 0]. These coordinates are stored in the node attributes, completing the construction of the energy and equipment relationship graph.

[0088] From the constructed energy and device relationship graph, extract the subgraph related to the abnormal energy node source (substation) and an abnormal escalator (unique identifier "escalator_01"). Use a query statement to find the corresponding node and the connected nodes and edges. Assign weights to the edges in the subgraph. For example, the weight of the energy transmission edge is calculated based on the transmission loss and the current transmission power ratio. The higher the transmission loss and the greater the current transmission power ratio to the transmission capacity, the higher the weight. For example, if the transmission loss of an energy transmission edge is 8% and the current transmission power accounts for 90% of the transmission capacity, the calculated weight is 0.8. The weight of the device connection edge is determined by the length of the connection line and the importance of the device. The longer the length and the less important the device, the higher the weight. For example, if a connection line is 30 meters long and the connected device is a standard lighting fixture, the weight is set to 0.6.

[0089] The Dijkstra algorithm was used to calculate the shortest path from the anomalous energy source (substation) to the abnormal escalator on the subgraph. The resulting shortest path node sequence was: "substation_01" -> "transmission_line_02" -> "distribution_box_03" -> "escalator_01." The path length, calculated as the sum of the edge weights, was 2.2. Analysis of this path reveals that energy is transmitted from the substation via "transmission_line_02" to "distribution_box_03," which then supplies power to "escalator_01." The shortest path length is used to assess closeness. A shorter path indicates a high degree of closeness between the anomalous energy source and the abnormal device, and a relatively direct energy transmission path with fewer intermediate links. The final graph structure analysis result is: the node sequence of the shortest path is ["substation_01", "transmission_line_02", "distribution_box_03", "escalator_01"], the path length is 2.2, and the degree of compactness is relatively high.

[0090] Step S400: Based on the results of the graph structure analysis, combined with the importance of the equipment and the impact of the equipment failure on the production process, the priority of the abnormal equipment is determined; for abnormal equipment of different priorities, their adjustable parameters are analyzed separately, and an abnormal equipment parameter adjustment table is generated, with the parameters of abnormal equipment with low equipment priority being adjusted first.

[0091] Specifically, evaluation indicators are set, including the functional criticality, frequency of use, and maintenance difficulty of the equipment, to obtain the importance of the equipment; in terms of functional criticality, core equipment, important equipment, and auxiliary equipment are distinguished; in terms of frequency of use and maintenance difficulty, they are divided into high, medium, and low levels; a quantitative score range is set for each evaluation indicator; a weight is determined for each evaluation indicator, and the quantitative scores of the evaluation indicators are weighted and calculated to obtain the equipment importance score;

[0092] Define the specific role and location of each device in the operational process of the high-speed railway station square, and assess the scope and duration of the fault impact. Set weights for the impact scope and impact duration, and perform a weighted calculation of the impact scope and impact duration to obtain a score for the impact degree of the equipment fault.

[0093] The equipment importance score, fault impact score, and graph structure analysis results are comprehensively considered to perform weighted calculation and determine the priority of abnormal equipment.

[0094] Furthermore, priority grouping rules are set based on the determined abnormal device priority scores and ranking results; all abnormal devices are traversed, and each device is classified into a corresponding priority group according to the priority grouping rules to obtain a priority device list;

[0095] Analyze the adjustable parameters of each priority abnormal device, design the abnormal device parameter adjustment table framework, extract device data from the priority device list in sequence, fill the device's adjustable parameters and device data into the abnormal device parameter adjustment table; give priority to adjusting parameters of abnormal devices with low device priority.

[0096] In one specific example, escalators are considered critical equipment and are assigned a quantitative score of 7 points; lighting fixtures are considered auxiliary equipment and are assigned a score of 4 points; and air conditioning units are considered core equipment and are assigned a score of 10 points. The escalator is frequently used and is assigned an 8-point score; the lighting fixture is moderately used and is assigned a 5-point score; and the air conditioning unit is frequently used in the summer and is assigned an 8-point score. The escalator is considered difficult to maintain and is assigned a 9-point score; the lighting fixture is considered easy to maintain and is assigned a 3-point score; and the air conditioning unit is considered moderately difficult to maintain and is assigned a 6-point score.

[0097] Assign weights of 0.5 for functional criticality, 0.3 for frequency of use, and 0.2 for maintenance difficulty, respectively. Escalator importance score = 7 × 0.5 + 8 × 0.3 + 9 × 0.2 = 7.7 points. Lighting importance score = 4 × 0.5 + 5 × 0.3 + 3 × 0.2 = 4.1 points. Air conditioning unit importance score = 10 × 0.5 + 8 × 0.3 + 6 × 0.2 = 8.6 points.

[0098] If the fault is regional (affecting passenger access in some areas of the waiting area), the quantitative score is 7 points; the estimated repair time is 4 hours, and the impact time score is 6 points. If the fault is local (affecting only lighting in a specific area), the score is 4 points; the estimated repair time is 1 hour, and the score is 3 points. If the fault is systemic (affecting the comfort of the entire waiting area), the score is 10 points; the estimated repair time is 8 hours, and the score is 9 points.

[0099] Assign weights of 0.6 for impact range and 0.4 for impact duration, respectively. The impact score for an escalator failure is 7 × 0.6 + 6 × 0.4 = 6.6 points. The impact score for a lighting fixture failure is 4 × 0.6 + 3 × 0.4 = 3.6 points. The impact score for an air conditioning unit failure is 10 × 0.6 + 9 × 0.4 = 9.6 points.

[0100] Combined with the results of the graph structure analysis (the converted score of the closeness between escalators and abnormal energy node sources is 0.8, the lighting fixtures is 0.4, and the air-conditioning units is 0.7), the weight of the equipment importance score is set to 0.4, the weight of the fault impact score is set to 0.4, and the weight of the graph structure analysis result score is set to 0.2.

[0101] The comprehensive priority score of escalator = 7.7×0.4+6.6×0.4+0.8×0.2=7.04 points.

[0102] The comprehensive priority score of lighting fixtures = 4.1×0.4+3.6×0.4+0.4×0.2=3.8 points.

[0103] The comprehensive priority score of the air-conditioning unit = 8.6×0.4+9.6×0.4+0.7×0.2=9.02 points.

[0104] Set the priority grouping rules: the top 30% of the scores are high priority, 30%-70% are medium priority, and the bottom 30% are low priority. After sorting, air conditioners are high priority, escalators are medium priority, and lighting fixtures are low priority. These devices are grouped into their corresponding priority groups to create a priority device list.

[0105] Design the abnormal device parameter adjustment table framework, including columns such as device name, device identifier, priority, adjustable parameter 1, parameter 1 current value, parameter 1 adjustment range, parameter 1 adjustment impact, adjustable parameter 2...

[0106] Extract lighting fixture data from the priority device list, such as "lighting_05." Adjustable parameters include brightness level, currently set to 8, with an adjustment range of 1-10. Increasing brightness increases energy consumption, while decreasing brightness saves energy but may affect lighting performance. Enter this data into the abnormal device parameter adjustment table.

[0107] Prioritize parameter adjustments for low-priority lighting fixtures. For example, adjust the brightness level of "lighting_05" from level 8 to level 6. After the adjustment, monitoring shows that the energy consumption in this area has decreased by 10%, and the lighting effect can still meet basic needs.

[0108] like Figure 2 As shown, the present invention provides an AI intelligent operation and maintenance system based on digital twins, including:

[0109] 3D map generation module: This module includes a 3D map construction unit and a digital twin model construction unit. The 3D map construction unit determines the 3D map range of the high-speed railway station square, constructs the 3D map, and collects energy data and equipment data. The digital twin model construction unit uses a multi-layer perceptron algorithm based on the collected data to construct energy digital twin models and equipment digital twin models, integrating them into the 3D map.

[0110] Abnormal energy node source location module: includes: a device data anomaly judgment unit and an abnormal energy node source location unit; wherein, the device data anomaly judgment unit determines whether there are anomalies in the current energy data and current device data based on the digital twin model in the three-dimensional map; when an anomaly exists, the abnormal energy node source location unit marks the location of the abnormal device or abnormal energy node on the three-dimensional map, and performs correlation analysis on the digital twin model to obtain the abnormal energy node source;

[0111] Graph structure analysis module: includes: a relationship graph construction unit and a graph structure analysis unit; wherein, when the relationship graph construction unit confirms that the energy output of an abnormal energy node source has reached a bottleneck, it uses the graph database to construct an energy and equipment relationship graph based on the location of the abnormal energy node source in the three-dimensional map; the graph structure analysis unit uses the shortest path algorithm to analyze the association path and closeness between the abnormal energy node source and the abnormal equipment to obtain the graph structure analysis results;

[0112] Abnormal equipment parameter adjustment module: includes: abnormal equipment priority determination unit, parameter adjustment table generation unit and abnormal equipment parameter adjustment unit; among them, the abnormal equipment priority determination unit determines the priority of the abnormal equipment based on the graph structure analysis results, combined with the importance of the equipment and the impact of equipment failure on the production process; the parameter adjustment table generation unit analyzes the adjustable parameters of abnormal equipment with different priorities, generates an abnormal equipment parameter adjustment table, and the abnormal equipment parameter adjustment unit gives priority to adjusting the parameters of abnormal equipment with low equipment priority.

[0113] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. An AI intelligent operation and maintenance method based on digital twins, characterized by: The following steps are involved: Determine the 3D map scope of the high-speed railway station square, construct the 3D map, and collect energy and equipment data. Based on the collected data, use the multi-layer perceptron algorithm to construct energy digital twin models and equipment digital twin models and integrate them into the 3D map. Based on the digital twin model in the 3D map, determine whether there are any anomalies in the current energy data and current equipment data; When an anomaly occurs, the location of the abnormal equipment or abnormal energy node is marked on the 3D map, and the digital twin model is correlated and analyzed to obtain the source of the abnormal energy node; Extract all energy data and equipment data within the time period related to the current anomaly from the digital twin model; construct data association relationships based on the logical relationship descriptions of energy nodes and equipment already in the digital twin model; starting from the abnormal device or energy node, analyze the possible propagation path of the anomaly using a combination of forward reasoning and reverse tracing based on the established data association relationships; calculate the probability of the anomaly source for each energy node that may be related to the anomaly based on the closeness of its association with the abnormal device and the severity of the data anomaly; verify the possible abnormal energy node sources; if a power supply equipment failure is found in an energy node during an on-site inspection, and the failure matches the abnormality analyzed in the digital twin model, the energy node is determined to be the abnormal energy node source; mark the abnormal energy node source on the three-dimensional map; When it is confirmed that the energy output of an abnormal energy node source has reached a bottleneck, a graph database is used to construct an energy and equipment relationship diagram based on the location of the abnormal energy node source in the three-dimensional map; The shortest path algorithm is used to analyze the association path and closeness between abnormal energy node sources and abnormal equipment, and the graph structure analysis results are obtained; Based on the results of the graph structure analysis, the priority of abnormal equipment is determined in combination with the importance of the equipment and the impact of the equipment failure on the production process. For abnormal equipment of different priorities, their adjustable parameters are analyzed separately, and a parameter adjustment table for abnormal equipment is generated. Parameter adjustment is performed first on abnormal equipment with low equipment priority. Set evaluation indicators, including the functional criticality, frequency of use, and difficulty of maintenance of the equipment, to determine the importance of the equipment; within the functional criticality, distinguish between core equipment, important equipment, and auxiliary equipment; In terms of frequency of use and difficulty of maintenance, they are divided into high, medium and low levels; Set a quantitative score range for each evaluation indicator; Determine the weight for each evaluation indicator, perform weighted calculation on the quantitative scores of the evaluation indicators, and obtain the equipment importance score; Define the specific role and location of each device in the operational process of the high-speed railway station square, and assess the scope and duration of the fault impact. Set weights for the impact scope and impact duration, and perform a weighted calculation of the impact scope and impact duration to obtain a score for the impact degree of the equipment fault. The device importance score, fault impact score, and graph structure analysis results are comprehensively considered to perform weighted calculations and determine the priority of abnormal devices. Set priority grouping rules based on the determined abnormal device priority scores and sorting results; Traverse all abnormal devices and classify each device into the corresponding priority group according to the priority grouping rules to obtain a priority device list; Analyze the adjustable parameters of each priority abnormal device, design the abnormal device parameter adjustment table framework, extract device data from the priority device list in sequence, fill the device's adjustable parameters and device data into the abnormal device parameter adjustment table; give priority to adjusting parameters of abnormal devices with low device priority.

2. The AI ​​intelligent operation and maintenance method based on digital twin according to claim 1 is characterized in that: Determining the three-dimensional map range of the high-speed railway station square, constructing the three-dimensional map, and collecting energy data and equipment data include: Determine the 3D map scope of the high-speed railway station concourse, including the main area and surrounding areas of the concourse; the main area includes the passenger waiting area, ticket hall, entrance and exit passages, underground parking lot, bus transfer area, and taxi transfer area; the surrounding areas specifically include areas related to energy supply and equipment; mark the outlines and relative positions of each area; Using a laser scanner, scanning sites are planned according to the layout of the high-speed railway station square. Three-dimensional point cloud data is acquired at each scanning site. Point cloud processing software is used to stitch and fuse the point cloud data from multiple sites to generate a three-dimensional point cloud model. Using a drone, a flight route is planned and camera parameters are set based on the range of the three-dimensional map. During the drone flight, the flight status and image acquisition are monitored in real time. The acquired image data is imported into three-dimensional reconstruction software, and a three-dimensional map is generated by combining the image stitching and three-dimensional reconstruction algorithms with the three-dimensional point cloud model. SketchUp software is used to model the equipment within the high-speed railway station square, including power supply equipment and power-consuming equipment. After modeling, the equipment is placed on the three-dimensional map based on its actual position and orientation in the square. The energy data includes energy nodes and energy consumption, and the equipment data includes equipment operation data, equipment control instructions and equipment basic information.

3. The AI ​​intelligent operation and maintenance method based on digital twin according to claim 1 is characterized in that: Based on the collected data, the energy digital twin model and the equipment digital twin model are constructed using a multi-layer perceptron algorithm and integrated into the three-dimensional map, including: When building energy digital twin models and equipment digital twin models, the number of neurons in their respective input layers is determined based on the number of features of the energy data and equipment data after feature engineering. Grid search and cross-validation are used to debug different combinations of the number of neurons. The weights of the digital twin model are initialized using the Xavier initialization method. The training set data is input in batches, and the batch size is determined through experiments. For each batch of data, the predicted output is calculated through forward propagation, and the loss value is calculated compared with the true label. The gradient of the loss value with respect to the digital twin model parameters is calculated using the backpropagation algorithm. The parameters are updated through the optimizer, and the performance of the digital twin model is evaluated using the validation set. By creating a custom data source, the digital twin model is encapsulated as an entity and added to the data source, and the data source is integrated into the three-dimensional map.

4. The AI ​​intelligent operation and maintenance method based on digital twin according to claim 1 is characterized in that: The digital twin model in the three-dimensional map is used to determine whether there are any anomalies in the current energy data and the current equipment data, including: For energy data, historical data is collected and combined with energy system design standards to calculate statistics, including mean and standard deviation. Based on these statistics, normal ranges for various energy parameters in the energy data are set. For equipment data, the allowable deviation ranges for equipment parameters are determined for different types of equipment according to manufacturer settings. Input the current energy data and current equipment data into the digital twin model in the three-dimensional map, and output the current energy parameters and current equipment parameters; compare the current energy parameters with the normal range of each energy parameter, and compare the current equipment parameters with the allowable deviation range of the equipment parameters to determine whether there are any abnormalities in the current energy data and current equipment data.

5. The AI ​​intelligent operation and maintenance method based on digital twin according to claim 1 is characterized in that: When an anomaly occurs, the location of the abnormal equipment or abnormal energy node is marked on the 3D map, and the digital twin model is analyzed for correlation to obtain the source of the abnormal energy node, including: When it is determined that energy data or equipment data is abnormal, the coordinate information of the equipment or energy node involved in the three-dimensional map is extracted from the abnormal data record and marked in the three-dimensional map.

6. The AI ​​intelligent operation and maintenance method based on digital twins according to claim 1 is characterized in that: When it is confirmed that the energy output of the abnormal energy node source has reached a bottleneck, based on the location of the abnormal energy node source in the three-dimensional map, a graph database is used to construct an energy and equipment relationship graph, including: When it is confirmed that the energy output of an abnormal energy node source has reached a bottleneck, the energy data of the abnormal energy node source is collected with the abnormal energy node source location marked in the three-dimensional map as the center, and the energy data and equipment data directly or indirectly related to the abnormal energy node source are collected; The graph database uses Neo4j, and defines abnormal energy node sources and other related energy nodes and devices as graph database nodes; each energy node and device is given a unique identifier to determine the attributes of the graph database node; Define energy transmission edges to describe the energy transmission relationship between different energy nodes; define device connection edges to describe the physical connection relationship between devices; define control relationship edges, and define them as edges when there is a control relationship between devices; call the ForceAtlas2 algorithm to calculate the layout of the graph database nodes, store the calculated graph database node position coordinates in the node attributes, and construct an energy and device relationship graph.

7. The AI ​​intelligent operation and maintenance method based on digital twin according to claim 1 is characterized in that: The shortest path algorithm is used to analyze the association paths and closeness between abnormal energy node sources and abnormal devices to obtain graph structure analysis results, including: Extract a subgraph related to abnormal energy node sources and abnormal devices from the energy and device relationship graph; use query statements to find the nodes corresponding to the abnormal energy node sources and abnormal devices, and obtain all nodes and edges with connected relationships in the corresponding nodes; set weights for the edges in the subgraph to reflect the closeness between the abnormal energy node sources and abnormal devices; The shortest path algorithm adopts the Dijkstra algorithm to calculate the shortest path from the abnormal energy node source to the abnormal device on the constructed subgraph; based on the calculated shortest path, the association path between the abnormal energy node source and the abnormal device is analyzed, and through the nodes and edges in the association path, it is determined how the energy is transmitted from the abnormal energy node source to the abnormal device, as well as the energy nodes and devices connected in the middle; the closeness between the abnormal energy node source and the abnormal device is evaluated based on the length of the shortest path; and the graph structure analysis results are obtained, including the node sequence, path length and closeness of the shortest path.

8. An AI intelligent operation and maintenance system based on digital twins, using an AI intelligent operation and maintenance method based on digital twins according to any one of claims 1 to 7, characterized in that: include: 3D map generation module: This module includes a 3D map construction unit and a digital twin model construction unit. The 3D map construction unit determines the 3D map range of the high-speed railway station square, constructs the 3D map, and collects energy data and equipment data. The digital twin model construction unit uses a multi-layer perceptron algorithm based on the collected data to construct energy digital twin models and equipment digital twin models, integrating them into the 3D map. Abnormal energy node source location module: includes: a device data anomaly judgment unit and an abnormal energy node source location unit; wherein, the device data anomaly judgment unit determines whether there are anomalies in the current energy data and current device data based on the digital twin model in the three-dimensional map; when an anomaly exists, the abnormal energy node source location unit marks the location of the abnormal device or abnormal energy node on the three-dimensional map, and performs correlation analysis on the digital twin model to obtain the abnormal energy node source; Graph structure analysis module: includes: a relationship graph construction unit and a graph structure analysis unit; wherein, when the relationship graph construction unit confirms that the energy output of an abnormal energy node source has reached a bottleneck, it uses the graph database to construct an energy and equipment relationship graph based on the location of the abnormal energy node source in the three-dimensional map; the graph structure analysis unit uses the shortest path algorithm to analyze the association path and closeness between the abnormal energy node source and the abnormal equipment to obtain the graph structure analysis results; Abnormal equipment parameter adjustment module: includes: abnormal equipment priority determination unit, parameter adjustment table generation unit and abnormal equipment parameter adjustment unit; among them, the abnormal equipment priority determination unit determines the priority of the abnormal equipment based on the graph structure analysis results, combined with the importance of the equipment and the impact of equipment failure on the production process; the parameter adjustment table generation unit analyzes the adjustable parameters of abnormal equipment with different priorities, generates an abnormal equipment parameter adjustment table, and the abnormal equipment parameter adjustment unit gives priority to adjusting the parameters of abnormal equipment with low equipment priority.

Images