Reservoir intelligent inspection system based on knowledge graph and digital twinning

Abstract

The invention discloses an intelligent reservoir inspection system based on a knowledge graph and digital twinning. The system comprises a sensing acquisition layer, a modeling simulation layer, a data processing and decision support layer, a communication layer and an interactive visualization layer, the sensing acquisition layer obtains image video data, water level data, rainfall data and temperature and humidity data of the reservoir and the surrounding environment through a camera and a sensor; the modeling simulation layer maps an actual reservoir and a surrounding environment into a virtual model, imports data acquired by the sensing acquisition layer into the virtual model, and realizes dynamic monitoring and analysis through a digital twinborn technology; the data processing and decision support layer is used for processing, analyzing and mining data and providing decision support; the communication layer is responsible for interconnection, remote control and automatic task execution of the system; and the interactive visualization layer provides an operation interface and data display.

Description

Technical Field

[0001] The present invention belongs to the technical field of reservoir inspection systems, and specifically relates to an intelligent reservoir inspection system based on knowledge graphs and digital twins. Background Art

[0002] Reservoirs are key facilities for water resource regulation, which can effectively store and distribute water to ensure the water demand of agriculture, industry and cities. They not only help prevent floods and reduce disasters, promote economic development, but also have far-reaching significance for ecological protection and social stability. In order to ensure the safe operation of reservoirs and maximize their benefits, regular reservoir inspections are essential.

[0003] Traditional reservoir inspections generally rely on manual inspections, which often require inspectors to visit each inspection point one by one. The workload is huge, time-consuming, and easily affected by external factors such as weather and terrain. The subjective factors of inspectors may also lead to omissions and low efficiency. In addition, inspectors often need to enter high-risk areas (such as high dams, landslide areas, deep water areas, etc.) for inspections, which poses a great safety hazard, especially in severe weather or sudden natural disasters, the life safety of personnel may be threatened. After each inspection, the data still needs to be manually recorded and sorted, and the information transmission is delayed, making it impossible to quickly grasp the status and potential risks of the reservoir area. The results of manual inspections are often limited to text records and photos, which make it difficult to intuitively present the severity or changing trends of the overall problems in the reservoir area. There is a lack of timely professional decision-making support, and it is impossible to make effective decisions in a timely manner.

[0004] At present, with the development of technology, some modern inspection technologies, such as drones, sensors, and remote monitoring systems, have been applied to reservoir inspections. These technologies can provide real-time data collection and monitoring, reduce the frequency of manual inspections, and effectively avoid manual entry into high-risk areas, thereby improving inspection safety.

[0005] However, existing technologies still have some shortcomings. Although large amounts of data can be obtained in real time, there is a lack of intelligent analysis and decision-making support for these data. Human intervention is still required to identify and judge problems. At the same time, the lack of intuitive visual display makes it difficult to quickly help managers identify potential risks or the severity of the overall problem, resulting in the overall efficiency and response speed of the system failing to reach the expected optimal level.

[0006] Therefore, a reservoir inspection system with a high degree of visualization and the ability to provide intelligent decision support is needed, and this application conducts further research in this direction. Summary of the invention

[0007] In response to the shortcomings in the prior art, the present application provides a reservoir intelligent inspection system based on knowledge graphs and digital twins. The system can reflect the dynamic changes of the reservoir area in real time through digital twin technology, intelligently judge potential hidden dangers through drone inspections, analyze problems based on knowledge graphs, query historical events and related data, and combine cross-domain (such as climate, environment, urban planning, etc.) knowledge graphs to comprehensively evaluate problems, help decision makers make more scientific decisions from a global perspective, and improve reservoir safety and management efficiency. That is, it can optimize the efficiency and accuracy of reservoir inspections and solve the problems of inefficient manual inspections and delayed decision-making in the prior art.

[0008] The technical solution of this application is as follows.

[0009] A reservoir intelligent inspection system based on knowledge graph and digital twin, the system includes a perception and acquisition layer, a modeling and simulation layer, a data processing and decision support layer, a communication layer, and an interactive visualization layer; the perception and acquisition layer obtains image and video data, water level data, rainfall data, temperature and humidity data of the reservoir and its surrounding environment through cameras and sensors; the modeling and simulation layer maps the actual reservoir and the surrounding environment into a virtual model, and imports the data obtained by the perception and acquisition layer into the virtual model, and realizes dynamic monitoring and analysis through digital twin technology; the data processing and decision support layer processes, analyzes, mines data, and provides decision support; the communication layer is responsible for the interconnection, remote control and automatic execution of tasks of the system; the interactive visualization layer provides an operation interface and data display.

[0010] Preferably, the data processing and decision support layer generates targeted and global operation and maintenance recommendations by integrating cross-domain knowledge graphs and combining historical data with real-time data to conduct in-depth analysis and intelligent decision-making.

[0011] Preferably, the cross-domain knowledge graph is integrated through the following method: S10: Construction of cross-domain knowledge graph: Based on the multiple fields to be integrated, a knowledge graph in the field of water conservancy and other related fields is constructed, and the knowledge graph includes at least one of entities, attributes, relationships and instances; through data mining, natural language processing NLP, and entity recognition NER technical means, the domain knowledge graph is automatically extracted and constructed from different data sources; S20: Standardized interface design: In order to ensure the interoperability of cross-domain data and knowledge, standardized data interfaces and exchange protocols are designed, and the interfaces support data transmission, access and update between knowledge graphs in different fields; S30: Construct a cross-domain graph neural network based on graph neural network and transfer learning, align and fuse multi-domain graph entities, and construct a unified cross-domain graph; S40: Introduction of blockchain technology: Use blockchain technology to ensure the data credibility, privacy protection and security of the cross-domain knowledge graph.

[0012] Preferably, the knowledge graph fields include: (1) Water conservancy engineering field: knowledge on the construction, design, operation and maintenance of reservoirs, including the design specifications, materials and construction methods of water conservancy projects; the knowledge graph in this field can provide support for the construction, maintenance and transformation of reservoirs, and help analyze the performance and durability of reservoir facilities; (2) Hydrometeorology field: precipitation, flow, water level, climate change factors involving weather, climate and rainfall related data; the knowledge graph in this field can help identify potential threats brought by severe weather, such as heavy rains and floods, during the scheduling and inspection of reservoirs; (3) Environmental protection field: including knowledge on water quality monitoring and ecological protection, involving water pollutants (such as heavy metals, organic matter, nutrients, etc.), ecological environment changes, species diversity and other contents; the knowledge graph in this field can provide a scientific basis for Ecological restoration or pollution prevention and control to ensure the water quality and ecological balance of the reservoir; (4) Urban planning field: including planning issues related to land use, infrastructure construction, urban development and reservoirs; the knowledge graph in this field can help identify the potential impact of urbanization on water quality and water resources, and evaluate the long-term impact of urban expansion on reservoirs; (5) Agricultural irrigation field: involving water resources management, agricultural water use, and irrigation technology; including agricultural irrigation demand forecasting and effective use of water resources; the knowledge graph in this field can help to reasonably dispatch water resources in reservoirs for crop growth; (6) Laws, regulations and policies field: relevant laws, regulations, policy requirements and compliance checks on reservoir management; the knowledge graph in this field provides real-time query of laws and regulations and policies related to reservoir management to ensure that inspection and operation processes meet government and industry standards.

[0013] Preferably, the knowledge graph is fused based on graph neural network and transfer learning, and the fusion method includes the following steps:

[0014] S31: For the graph neural network in each domain (assuming it is the graph of source domain A and target domain B), the node representation of each layer is represented by the update rule of the graph convolutional network GCN, as follows:

[0015]

[0016] in:

[0017] Indicates The node representation of the layer, That is the initial node feature;

[0018] is the normalized adjacency matrix, including self-connections;

[0019] It is The learned weight matrix of the layer;

[0020] It is an activation function, and the linear rectification function ReLU can be used;

[0021] In the task of cross-domain alignment, the graphs between the two domains can be convolved in this way to obtain node representations of the two graphs.

[0022] S32: The two domains are defined as source domain A and target domain B. After constructing the graph convolutional network, the source domain A is pre-trained: a GCN model is trained on the source domain A to learn the structure and entity feature representation of the source domain. At the same time, a regularization strategy is introduced. The GCN training loss of the source domain A is set to , the regularization term of the source domain A is ,as follows:

[0023]

[0024] in, is the pre-training loss function, and γ is a hyperparameter used to balance the regularization term Ra in the loss function and the loss La in the source domain.

[0025] S33: Migrate the pre-trained model to the target domain B, introduce an adversarial training mechanism to reduce the distribution difference between the source domain and the target domain, and perform fine-tuning and adversarial training on the target domain B at the same time: During fine-tuning, the GCN network structure of the source domain A can be fixed, and the node features and relationship mapping of the target domain can be adjusted; let the fine-tuning loss of the target domain B be , the adversarial training loss is , The regularization term of the target domain B, the total loss of the training process can be expressed as:

[0026]

[0027] in, , It is an important hyperparameter for adjusting transfer learning and is used to balance the loss of fine-tuning and adversarial training. is the adversarial training loss;

[0028] The goal of adversarial training is to achieve transfer learning by maximizing the similarity between the source domain and the target domain, which is achieved through an adversarial loss function as follows:

[0029]

[0030] in:

[0031] is the discriminator;

[0032] It is the node representation of the source domain and the target domain;

[0033] It’s about fighting against loss.

[0034] As a preferred method, the specific steps for introducing blockchain technology are as follows:

[0035] S41: Data on-chain: When new data (such as new nodes or relationships in the knowledge graph) is added to the knowledge graph, its data hash value is written to the blockchain; each data has a unique hash value, which represents the "fingerprint" of the data. Once the data is written to the blockchain, it cannot be modified or deleted, ensuring the data's immutability and historical traceability.

[0036] S42: Verification mechanism: All updates or modifications to cross-domain knowledge graphs need to be verified by consensus of multiple participants; for example, the update of certain data may need to be verified by multiple nodes (such as academic experts, data source providers, etc.). Only through the consensus mechanism can the data be confirmed and added to the knowledge graph. This ensures that the data is widely verified and avoids the addition of false or misleading data.

[0037] S43: Encryption technology: Blockchain uses encryption algorithms (such as public-private key encryption) to ensure that only authorized users can access and decrypt data. Even if the data is written to the blockchain, only authorized users can view the detailed content of the data instead of exposing it to everyone.

[0038] S44: Zero-knowledge proof: Zero-knowledge proof (ZKP) is a cryptographic technique that allows one party (the prover) to prove the authenticity of a piece of information to another party (the verifier) ​​without revealing the specific content of the information;

[0039] S45: Distributed storage: Data is not stored on a single server or central organization, but distributed across multiple nodes; even if one or more nodes fail or are attacked, copies on other nodes can still ensure the integrity and availability of the data.

[0040] S46: Smart Contract: A smart contract is a program that automatically executes an agreement when certain conditions are met. Through smart contracts, data access and operations in cross-domain knowledge graphs can be automatically managed and monitored through predefined rules, ensuring that only operations that meet specific conditions can be performed, avoiding unsafe operations.

[0041] S47: Permission management: Through the blockchain's permission management system, a strict access control mechanism can be set up to ensure that only authorized users can modify or access certain sensitive data.

[0042] In the prior art, the conventional reservoir inspection system mainly has the following problems:

[0043] (1) Limited data analysis capabilities: Existing sensors and monitoring equipment can collect large amounts of data in real time, but most inspection systems lack data processing and intelligent analysis capabilities, making it impossible to conduct in-depth analysis of data, automatically identify potential problems or predict risks, resulting in the failure to fully utilize the value of data.

[0044] (2) Low level of intelligence: The existing reservoir inspection system cannot automatically extract knowledge from multi-source data. Despite the large amount of real-time data, managers need to manually analyze and judge. The lack of intelligent decision-making support affects response speed and efficiency.

[0045] (3) Failure to provide a global perspective: Current reservoir inspections focus only on the water conservancy sector itself, lacking knowledge integration with the environment, climate change, urban planning and other fields. When dealing with problems, they may not be able to provide a global perspective, resulting in narrow-minded decision-making that lacks pertinence and long-term vision.

[0046] (4) Lack of dynamic monitoring capabilities: Current reservoir monitoring relies heavily on data from fixed sensors, which can only reflect the state of the reservoir at a certain moment and cannot provide comprehensive, real-time reservoir dynamic information. This makes it impossible for managers to dynamically observe the overall state of the reservoir, equipment failures, and potential risks and make real-time decisions.

[0047] In contrast, the present application provides an intelligent reservoir inspection system based on knowledge graphs and digital twins. The system can reflect the dynamic changes of the reservoir area in real time through digital twin technology, intelligently judge potential hidden dangers through drone inspections, analyze problems based on knowledge graphs, query historical events and related data, and combine cross-domain (such as climate, environment, urban planning, etc.) knowledge graphs to comprehensively evaluate problems, help decision makers make more scientific decisions from a global perspective, and improve reservoir safety and management efficiency. That is, it can optimize the efficiency and accuracy of reservoir inspections and solve the problems of inefficient manual inspections and delayed decision-making in existing technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 Overall flow chart of the reservoir intelligent inspection system based on knowledge graph and digital twin.

[0049] Figure 2 A cross-domain graph neural network graph fusion method is constructed based on graph neural network and transfer learning.

[0050] Figure 3 Flowchart of blockchain technology applied to the dynamic update scenario of knowledge graph. DETAILED DESCRIPTION

[0051] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0052] In the following embodiments, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0053] Referring to the accompanying drawings, the intelligent reservoir inspection system based on knowledge graph and digital twin involved in this application is mainly divided into perception and collection layer, modeling and simulation layer, data processing and decision support layer, communication layer, and interactive visualization layer.

[0054] Perception and collection layer: This layer is mainly responsible for real-time monitoring and data collection of the reservoir environment, facilities and equipment, using tools such as sensors, surveillance cameras and drones. Surveillance cameras are installed in the reservoir area, and the physical dam is equipped with a variety of sensors such as water level, rainfall, temperature and humidity, and liquid level to collect real-time data on the dam's operating status, structural health and surrounding environment. Drones are responsible for aerial patrols to obtain image and video information of the dam and surrounding areas, and all data is transmitted to the central system in real time.

[0055] Modeling and simulation layer: This layer maps the actual physical dam and surrounding environment into a virtual model, simulates the reservoir status based on the data obtained by the perception and collection layer, and realizes dynamic monitoring and analysis through digital twin technology.

[0056] Data processing and decision support layer: This layer mainly processes, analyzes, mines and provides decision support for data. It cleans, filters and verifies the raw data collected by sensors, and converts them into various data formats that meet business needs according to predetermined rules. By integrating cross-domain knowledge graphs, combining historical data and real-time data, it conducts in-depth analysis and intelligent decision-making to generate targeted and global operation and maintenance suggestions.

[0057] IoT communication layer: This layer is mainly responsible for the interconnection, remote control and automatic execution of tasks of the system. Wireless communication technology is used to ensure data transmission between sensors, control devices and monitoring centers. When an abnormality is found or the set threshold is reached, the system can automatically issue an alarm or perform preset automated operations (such as enabling backup equipment, adjusting water flow, etc.).

[0058] Interactive visualization layer: This layer is the inspection and dispatching center, which provides users with a friendly operation interface and data display to ensure that relevant personnel can intuitively and effectively understand the operation status of the reservoir. The real-time status, early warning information and inspection results of the reservoir are displayed through three-dimensional models and charts. It provides inspection task scheduling, equipment status monitoring, fault alarm and other functions.

[0059] The following is a specific embodiment of the present application.

[0060] An embodiment of the present invention provides a reservoir intelligent inspection system based on knowledge graph and digital twin. The various module layers and corresponding methods in the system are as follows.

[0061] Step S1: Install surveillance cameras in the reservoir area. The physical dam is equipped with various sensors such as water level, rainfall, temperature and humidity, and liquid level to collect real-time data on the dam's operating status, structural health, and surrounding environment, and transmit the data to the IoT center platform in real time. Collect the domain knowledge required to build the knowledge graph, including but not limited to water conservancy projects, hydrology and meteorology, environmental protection, urban planning, agricultural irrigation, laws, regulations, and policies, and focus on collecting information on inspection and operation and maintenance business in the field of water conservancy projects. Among them, the knowledge in each field and its role are as follows:

[0062] Water conservancy projects, knowledge on the construction, design, operation and maintenance of reservoirs, including the design specifications, materials and construction methods of water conservancy projects. The knowledge graph in this field can provide support for the construction, maintenance and renovation of reservoirs, and help analyze the performance and durability of reservoir facilities.

[0063] Hydrometeorology, data related to precipitation, flow, water level, climate change and other factors, involving weather, climate, rainfall, etc. The knowledge graph in this field can help identify potential threats brought by severe weather, such as heavy rain and floods, during the scheduling and inspection of reservoirs.

[0064] Environmental protection, including knowledge on water quality monitoring and ecological protection, involves water pollutants (such as heavy metals, organic matter, nutrients, etc.), ecological environment changes, species diversity, etc. The knowledge map in this field can provide a scientific basis for ecological restoration or pollution prevention and control to ensure the water quality and ecological balance of the reservoir.

[0065] Urban planning involves planning issues related to land use, infrastructure construction, urban development and reservoirs. The knowledge graph in this field can help identify the potential impact of urbanization on water quality and water resources, and assess the long-term impact of urban expansion on reservoirs.

[0066] Agricultural irrigation involves water resources management, agricultural water use, irrigation technology, etc. It includes the demand forecast of agricultural irrigation and the effective use of water resources. The knowledge graph in this field can help to reasonably dispatch water resources in reservoirs for crop growth.

[0067] Laws, regulations and policies: laws, regulations, policy requirements and compliance checks related to reservoir management. The knowledge graph in this field provides real-time query of laws and policies related to reservoir management to ensure that the inspection and operation process meets government and industry standards.

[0068] Step S2: Clean and verify the original data, and convert it into various data formats that meet business needs according to the rules. Build knowledge graphs in various fields based on knowledge graph construction technology, and integrate cross-domain knowledge graphs. Among them, a cross-domain graph neural network is constructed based on graph neural network and transfer learning for graph fusion method, please refer to Figure 2 :

[0069] Step S2.1.1, construct knowledge graphs in various fields based on knowledge graph construction technology, such as field A, field B, field C, etc.

[0070] Step S2.1.2, for the graph neural network in each domain (assuming it is the graph of domain A and domain B), the node representation of each layer (graph convolutional network GCN) can be represented by the following update rules:

[0071]

[0072] in:

[0073] Indicates The node representation of the layer, That is the initial node feature;

[0074] is the normalized adjacency matrix (including self-connections);

[0075] It is The learned weight matrix of the layer;

[0076] is the activation function, usually the rectified linear function (ReLU).

[0077] In the task of cross-domain alignment, the graphs between domains can be convolved in this way to obtain node representations of the two graphs.

[0078] Step S2.1.3, after building the graph convolutional network, pre-train the source domain A. Train a GCN model on the source domain A to learn the structure and entity feature representation of the source domain. At the same time, introduce a regularization strategy. Set the GCN training loss of the source domain A to , the regularization term of the source domain A is .

[0079]

[0080] in, is the pre-training loss function, and γ is a hyperparameter used to balance the regularization term Ra in the loss function and the loss La in the source domain.

[0081] Step S2.1.4, migrate the pre-trained model to the target domain B, introduce an adversarial training mechanism, reduce the distribution difference between the source domain and the target domain, and perform fine-tuning and adversarial training on domain B at the same time. During fine-tuning, the GCN network structure of the source domain can be fixed, but the node features and relationship mapping of the target domain can be adjusted. Suppose the fine-tuning loss of the target domain B is , the adversarial training loss is , is the regularization term of the target domain B, then the total loss of the training process can be expressed as:

[0082]

[0083] in, , It is an important hyperparameter for adjusting transfer learning and is used to balance the loss of fine-tuning and adversarial training. is the adversarial training loss. The goal of adversarial training is to achieve transfer learning by maximizing the similarity between the source domain and the target domain. This is achieved through the adversarial loss function as follows:

[0084]

[0085] in:

[0086] is the discriminator;

[0087] It is the node representation of the source domain and the target domain;

[0088] It’s about fighting against loss.

[0089] At the same time, blockchain technology is used to ensure the data credibility, privacy protection and security of cross-domain knowledge graphs. Figure 3 :

[0090] Step S2.2.1, add new knowledge graph data.

[0091] Step S2.2.2: Perform data preprocessing. Generate a unique hash value for the newly added data (such as entity relationships and attributes) using algorithms such as SHA-256 as the "fingerprint" of the data. Use asymmetric encryption (such as RSA) or homomorphic encryption for sensitive data to ensure that only authorized parties can decrypt it. Use zero-knowledge proof technology (such as zk-SNARKs) to verify the logical validity of the data without exposing the original data.

[0092] Step S2.2.3: Multi-party consensus verification. Nodes verify the authenticity of data through PBFT or PoA consensus mechanism.

[0093] Step S2.2.4, smart contract permission verification. The smart contract checks the requester's permissions and only allows operations that meet the conditions.

[0094] Step S2.2.5, distributed storage. The original data is stored in IPFS or a private database, and only the hash value and storage address are stored on the chain.

[0095] Step S2.2.6, blockchain. The signed transaction is broadcast to the node, and a new block is generated through consensus such as PoW / PoS to form a hash chain. The block hash chain structure ensures that historical data cannot be modified.

[0096] Step S2.2.7, trigger the smart contract rules. The on-chain event triggers the contract logic and automatically updates the knowledge graph state tree (such as adding new entity link relationships). If the execution fails (such as insufficient permissions), the contract rolls back to the previous state, but the transaction record is still retained.

[0097] Step S2.2.8, update the knowledge graph state tree. Write the new entity relationship into the Merkle tree to support fast retrieval and historical tracing.

[0098] Step S3: Use 3D GIS and UAV oblique photography technology to construct a 3D digital scene model of the reservoir area.

[0099] Step S3.1, collect basic geographic data of the reservoir, plan the UAV flight area, set the flight altitude, heading overlap rate and lateral overlap rate, and ensure the calibration of the equipment (multi-lens tilt camera, RTK positioning).

[0100] Step S3.2, obtain high-resolution images (including five perspectives: orthophoto, front, back, left, and right) through multi-angle oblique photography of the drone, synchronously record POS data (position and attitude), and fly in blocks over complex terrain.

[0101] Step S3.3, use software such as Pix4D and ContextCapture to perform aerial triangulation and generate a dense point cloud; construct a TIN triangulation network through the point cloud, generate a white model and automatically map the texture, and output a three-dimensional model in OSGB or 3D Tiles format.

[0102] Step S3.4, repair defects such as water surface holes and vegetation noise points, and simplify redundant triangles; import the model into a three-dimensional GIS platform (such as ArcGIS, SuperMap), overlay vector data such as terrain, hydrology, and engineering facilities, and construct a time-space integrated scene.

[0103] Step S4, overlay the constructed reservoir model data onto the GIS map for display, simulate the reservoir status according to the collected data, and display the environmental attribute information; realize the functions of maintenance plan recommendation and intelligent question and answer during the inspection process, and build a reservoir intelligent inspection system based on knowledge graph and digital twin.

[0104] The constructed system includes a cockpit, a three-dimensional visual monitoring subsystem, a safety inspection subsystem, an intelligent question-answering and knowledge retrieval subsystem, and a comprehensive early warning and plan deduction subsystem.

[0105] (1) The cockpit uses visual charts to display comprehensive data of multiple reservoirs, including macro indicators such as safety scores, flood control capabilities, and economic benefit analysis. It can integrate business data and generate reports with one click.

[0106] (2) The 3D visualization monitoring subsystem uses digital twin technology to build a holographic reservoir scene, dynamically displaying data such as topography, equipment distribution, and real-time water level / seepage. The 3D model is updated through sensor and drone inspection data, and key parameters such as water level, rainfall, and reservoir capacity are displayed simultaneously. Click on the monitoring point to view real-time monitoring.

[0107] (3) The safety inspection subsystem realizes inspection task management, hidden danger photo uploading and trajectory tracking, supports the real-time fusion of drone video and twin scenes, and superimposes the drone footage onto the three-dimensional scene in real time to assist in the precise positioning of leakage points or illegal intrusion areas.

[0108] (4) Intelligent question-answering and knowledge retrieval subsystem, which supports users to quickly obtain relevant knowledge, emergency plans and real-time data reports through text or voice commands. Based on multi-domain knowledge graphs, dynamic reasoning is performed in combination with the user's question context to generate cross-domain collaborative decision-making recommendations and risk warnings.

[0109] (5) The comprehensive early warning and emergency plan simulation subsystem, based on the "four predictions" (forecast, warning, rehearsal, and emergency plan) capabilities, dynamically marks risk areas through red / orange / yellow three-level warning signs.

[0110] Step S5: The user initiates a drone inspection task at the dispatch center and checks the equipment status monitoring and fault alarm information.

[0111] Step S6: For abnormal inspection results and fault alarm information, intelligent question and answer are used to provide suggestions; based on the suggestions, business personnel rectify the relevant problems.

[0112] Step S7: export business reports regularly based on historical data.

[0113] From the above description, it can be seen that the present application provides a reservoir intelligent inspection system based on knowledge graph and digital twin. The system can reflect the dynamic changes of the reservoir area in real time through digital twin technology, intelligently judge potential hidden dangers through drone inspections, analyze problems based on knowledge graphs, query historical events and related data, and combine cross-domain (such as climate, environment, urban planning, etc.) knowledge graphs to comprehensively evaluate problems, help decision makers make more scientific decisions from a global perspective, and improve reservoir safety and management efficiency. That is, it can optimize the efficiency and accuracy of reservoir inspections and solve the problems of inefficient manual inspections and delayed decision-making in the prior art.

[0114] The protection scope of the present invention includes but is not limited to the above embodiments. The protection scope of the present invention shall be based on the claims. Any replacement, deformation, and improvement of the technology that can be easily thought of by technicians in this field shall fall within the protection scope of the present invention.

Claims

1. A reservoir intelligent inspection system based on knowledge graph and digital twin, characterized in that: The system includes a perception and acquisition layer, a modeling and simulation layer, a data processing and decision support layer, a communication layer, and an interactive visualization layer; The perception and collection layer obtains image and video data, water level data, rainfall data, temperature and humidity data of the reservoir and its surrounding environment through cameras and sensors; The modeling and simulation layer maps the actual reservoir and surrounding environment into a virtual model, and imports the data obtained by the perception and collection layer into the virtual model, realizing dynamic monitoring and analysis through digital twin technology; The data processing and decision support layer processes, analyzes, mines data and provides decision support; The communication layer is responsible for the interconnection, remote control and automatic execution of tasks of the system; The interactive visualization layer provides an operation interface and data display.

2. According to claim 1, a reservoir intelligent inspection system based on knowledge graph and digital twin is characterized in that: The data processing and decision support layer integrates cross-domain knowledge graphs, combines historical data and real-time data, conducts in-depth analysis and intelligent decision-making, and generates targeted and global operation and maintenance suggestions.

3. According to claim 2, a reservoir intelligent inspection system based on knowledge graph and digital twin is characterized in that: The cross-domain knowledge graph is fused by the following method: S10: Construction of cross-domain knowledge graph: Based on the multiple fields to be integrated, a knowledge graph of the water conservancy field and other related fields is constructed, and the knowledge graph includes at least one of entities, attributes, relationships and instances; through data mining, natural language processing (NLP), entity recognition (NER) technical means, domain knowledge graphs are automatically extracted and constructed from different data sources; S20: Standardized interface design: To ensure the interoperability of cross-domain data and knowledge, design standardized data interfaces and exchange protocols. The interfaces support data transmission, access, and update between knowledge graphs in different domains. S30: Build a cross-domain graph neural network based on graph neural network and transfer learning, align and fuse graph entities in multiple fields, and build a unified graph across fields; S40: Introducing blockchain technology: Using blockchain technology to ensure data credibility, privacy protection and security of cross-domain knowledge graphs.

4. According to claim 3, a reservoir intelligent inspection system based on knowledge graph and digital twin is characterized in that: The knowledge graph fields include: (1) Water conservancy engineering: knowledge on the construction, design, operation and maintenance of reservoirs, including the design specifications, materials and construction methods of water conservancy projects. The knowledge graph in this field can provide support for the construction, maintenance and renovation of reservoirs and help analyze the performance and durability of reservoir facilities. (2) Hydrometeorological field: data related to weather, climate, and rainfall, including precipitation, flow, water level, and climate change factors. The knowledge graph in this field can help identify potential threats brought by severe weather during reservoir scheduling and inspection. (3) Environmental protection: including knowledge on water quality monitoring and ecological protection, involving water pollutants, ecological environment changes, and species diversity; domain knowledge maps can provide scientific basis for ecological restoration or pollution prevention and control to ensure the water quality and ecological balance of reservoirs; (4) Urban planning: including planning issues related to land use, infrastructure construction, urban development and reservoirs. The knowledge graph in this field can help identify the potential impact of urbanization on water quality and water resources, and assess the long-term impact of urban expansion on reservoirs. (5) Agricultural irrigation: involving water resources management, agricultural water use, and irrigation technology; including agricultural irrigation demand forecasting and effective use of water resources; the knowledge graph in this field can help to rationally dispatch water resources in reservoirs for crop growth; (6) Laws, regulations and policies: Relevant laws, regulations, policy requirements and compliance checks on reservoir management. The knowledge graph in this area provides real-time query of laws and regulations and policies related to reservoir management to ensure that the inspection and operation processes comply with government and industry standards.

5. According to claim 4, a reservoir intelligent inspection system based on knowledge graph and digital twin is characterized in that: The knowledge graph is fused based on graph neural network and transfer learning, and the fusion method includes the following steps: S31: For each field of graph neural network, the node representation of each layer is represented by the update rule of the graph convolutional network GCN, as follows: in: Indicates The node representation of the layer, That is the initial node feature; is the normalized adjacency matrix, including self-connections; It is The learned weight matrix of the layer; It is an activation function, and the linear rectification function ReLU can be used; In the cross-domain alignment task, the graphs between the two domains can be convolved in this way to obtain the node representations of the two graphs. S32: The two domains are defined as source domain A and target domain B. After constructing the graph convolutional network, the source domain A is pre-trained: a GCN model is trained on the source domain A to learn the structure and entity feature representation of the source domain. At the same time, a regularization strategy is introduced. The GCN training loss of the source domain A is set to , the regularization term of the source domain A is ,as follows: in, is the pre-training loss function, γ is a hyperparameter used to balance the regularization term Ra in the loss function and the loss La in the source domain; S33: Migrate the pre-trained model to the target domain B, introduce an adversarial training mechanism to reduce the distribution difference between the source domain and the target domain, and perform fine-tuning and adversarial training on the target domain B at the same time: During fine-tuning, the GCN network structure of the source domain A can be fixed, and the node features and relationship mapping of the target domain can be adjusted; let the fine-tuning loss of the target domain B be , the adversarial training loss is , The regularization term of the target domain B, the total loss of the training process can be expressed as: in, , It is an important hyperparameter for adjusting transfer learning and is used to balance the loss of fine-tuning and adversarial training. is the adversarial training loss; The goal of adversarial training is to achieve transfer learning by maximizing the similarity between the source domain and the target domain, which is achieved through an adversarial loss function as follows: in: is the discriminator; It is the node representation of the source domain and the target domain; It’s about fighting against loss.

6. According to claim 5, a reservoir intelligent inspection system based on knowledge graph and digital twin is characterized in that: The specific steps to introduce blockchain technology are as follows: S41: Data on-chain: When new data is added to the knowledge graph, its data hash value is written into the blockchain; S42: Verification mechanism: All updates or modifications to cross-domain knowledge graphs must be verified by consensus of multiple parties; S43: Encryption technology: Blockchain uses encryption algorithms to ensure that only authorized users can access and decrypt data; S44: Zero-knowledge proof: allows one party to prove the authenticity of a piece of information to another party without revealing the specific content of the information; S45: Distributed storage: Data is not stored in a single server or central organization, but distributed across multiple nodes; S46: Smart Contracts: Through smart contracts, data access and operations in cross-domain knowledge graphs can be automatically managed and monitored through predefined rules; S47: Permission management: Through the blockchain's permission management system, a strict access control mechanism can be set up to ensure that only authorized users can modify or access certain sensitive data.

Images