A power distribution network intelligent sensing system
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- KAIFENG POWER SUPPLY COMPANY STATE GRID HENAN ELECTRIC POWER
- Filing Date
- 2026-04-30
- Publication Date
- 2026-08-04
AI Technical Summary
虽然部分系统引入了智能终端或传感器，但往往功能单一，仅能采集电压、电流等基础电气量，缺乏对热成像、局部放电、环境状态等多维信息的融合感知能力，导致故障预警滞后、设备隐患难以及时发现
1、通过在配电网分支节点和表箱处部署集成多传感模块的边缘智能终端，实现对电流、电压、热成像、局部放电及环境参数等多维数据的同步采集与融合分析；结合内置神经网络模型的边缘计算单元，可在本地完成故障初步诊断、线损异常检测和设备状态评估，并在紧急情况下自主执行开关动作或发出预警，大幅缩短响应时间，提升配电网运行的安全性与可靠性；
Smart Images
Figure CN122512641A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power distribution network monitoring technology, specifically to a power distribution network intelligent sensing system. Background Technology
[0002] With the deepening of the construction of new power systems, the distribution network, as a key link connecting the main grid and end users, is facing multiple challenges, including the high proportion of distributed energy access, increasingly complex load patterns, and users' ever-increasing requirements for power supply reliability and power quality. Traditional distribution networks generally suffer from problems such as "many blind spots, slow response, weak coordination, and poor interaction," making it difficult to meet the development needs of high elasticity, high resilience, and intelligence.
[0003] Existing power distribution network monitoring methods largely rely on centralized data acquisition and back-end analysis, which suffers from drawbacks such as long communication links, high data processing latency, and insufficient intelligence at the edge. Although some systems have introduced intelligent terminals or sensors, these are often limited in function, only able to collect basic electrical quantities such as voltage and current, lacking the ability to fuse and perceive multi-dimensional information such as thermal imaging, partial discharge, and environmental conditions. This results in delayed fault warnings and difficulty in timely detection of equipment hazards. Furthermore, local terminals generally lack autonomous decision-making and rapid response mechanisms. In the event of leakage, overload, or equipment malfunction, they still rely on commands from the master station, making it difficult to achieve "on-site isolation and rapid power restoration." Regarding data reliability and security, current systems mostly use centralized databases to store operation logs and event records, posing a risk of data tampering and making it difficult to support future applications such as electricity market trading, carbon footprint tracking, and user authorization, which have strict requirements for data immutability. Summary of the Invention
[0004] The technical problem to be solved by this invention is to overcome the existing defects and provide an intelligent sensing system for power distribution networks, which significantly shortens the response time, improves the safety and reliability of power distribution network operation, and significantly enhances the digital operation and maintenance and intelligent scheduling capabilities of the power grid. It not only improves the observability, measurability and controllability of the power distribution network, but also builds a safe, reliable, efficient and interactive power distribution ecosystem, which can effectively solve the problems in the background technology.
[0005] To achieve the above objectives, the present invention provides the following technical solution: a smart sensing system for a power distribution network, characterized in that it comprises: Edge intelligent terminals are deployed at branch nodes and meter boxes of the power distribution network to perform multi-source fusion sensing of local electrical quantities, thermal imaging data, partial discharge signals and environmental parameters, and to execute local real-time decisions. Regional coordination nodes are used to aggregate data from the edge intelligent terminals and store key data through a blockchain network. A digital twin platform is used to build and maintain a three-dimensional digital twin of the power distribution network, enabling real-time synchronization between the physical power grid and the virtual model; The multiple edge intelligent terminals communicate with each other, as well as with the regional coordination node, through a hybrid Mesh self-organizing network.
[0006] Preferably, the edge intelligent terminal includes: The multi-sensor module integrates a current transformer, a voltage transformer, an infrared thermal imaging sensor, a temperature and humidity sensor, and a smoke detector. The edge computing unit has a built-in chip that runs a neural network model to perform preliminary fault diagnosis, abnormal line loss detection, and equipment status assessment. The secure communication unit integrates a PUF chip for device authentication and secure key generation. The local decision execution unit is used to execute local switching actions or issue warnings when a preset emergency event is detected.
[0007] Preferably, the regional coordination node performs the federated learning coordination task, and the specific steps are as follows: a. Distribute the initial parameters of the global fault prediction model to each edge intelligent terminal; b. Receive the updated model parameters from each edge intelligent terminal after training based on local data; c. Perform weighted aggregation on the received model parameters to generate a new global model, and then distribute the new global model to each edge intelligent terminal.
[0008] Preferably, the digital twin platform receives and parses multi-dimensional sensing data from regional coordination nodes; dynamically maps distribution network equipment and its operating status in three-dimensional space to form a visualized digital twin.
[0009] Preferably, it also includes a user interaction terminal, which provides users with real-time visualized information on their electricity carbon footprint and offers personalized energy-saving suggestions based on their electricity consumption habits; receives images or videos of abnormal equipment reported by users and automatically identifies and generates work orders through cloud services; and provides a virtual power plant interface, allowing users to authorize their distributed energy resources to participate in the grid ancillary services market and share revenue according to their contribution.
[0010] Preferably, the blockchain network adopts a directed acyclic graph (DAG) structure to store at least one of the following types of key data: leakage event records, power outage event records, equipment status assessment reports, and user-authorized transaction records. All on-chain data is digitally signed by the regional coordination node.
[0011] Compared with the prior art, the beneficial effects of the present invention are: 1. By deploying edge intelligent terminals with integrated multi-sensor modules at distribution network branch nodes and meter boxes, synchronous acquisition and fusion analysis of multi-dimensional data such as current, voltage, thermal imaging, partial discharge, and environmental parameters can be achieved. Combined with edge computing units with built-in neural network models, preliminary fault diagnosis, abnormal line loss detection, and equipment status assessment can be completed locally. In emergency situations, it can autonomously execute switching actions or issue early warnings, significantly shortening response time and improving the safety and reliability of distribution network operation. 2. Regional coordination nodes utilize a blockchain network based on a directed acyclic graph structure to perform tamper-proof distributed notarization of key events such as leakage current, power outages, equipment status assessment, and user authorization, and ensure data integrity through digital signatures; at the same time, a federated learning mechanism is adopted to coordinate the collaborative training of global fault prediction models by various edge terminals, so as to achieve continuous model optimization without uploading original sensitive data, thus balancing data privacy and intelligence improvement. 3. The digital twin platform, based on multi-dimensional sensing data gathered by regional coordination nodes, constructs a three-dimensional digital twin that is synchronized with the physical distribution network in real time. It dynamically maps the location, operating status and environmental information of equipment, providing high-fidelity visualization support for fault location, load forecasting, topology analysis and emergency drills, and significantly improving the digital operation and maintenance and intelligent dispatching capabilities of the power grid. In summary, this invention not only improves the observability, measurability, and controllability of the power distribution network, but also constructs a safe, reliable, efficient, and interactive power distribution ecosystem. Attached Figure Description
[0012] Figure 1 This is a system block diagram of the present invention. Detailed Implementation
[0013] The present invention can be explained in detail through the following embodiments. The purpose of disclosing the present invention is to protect all technical improvements within the scope of the present invention. In the description of the present invention, it should be understood that if there are terms such as "upper", "lower", "front", "rear", "left", "right" indicating the orientation or positional relationship, they are only corresponding to the drawings of this application for the convenience of describing the present invention, and are not intended to indicate or imply that the device or element referred to must have a specific orientation.
[0014] Please see Figure 1 This invention provides a technical solution: a smart sensing system for a power distribution network, comprising: Edge intelligent terminals are deployed at branch nodes and meter boxes of the power distribution network to perform multi-source fusion sensing of local electrical quantities, thermal imaging data, partial discharge signals and environmental parameters, and to execute local real-time decisions. The edge side is deployed nearby to realize the fusion of multi-dimensional data such as electrical quantities, thermal imaging, partial discharge, and environment, avoiding blind spots of traditional single-point monitoring. Specifically, edge intelligent terminals include: The multi-sensor module integrates a current transformer, a voltage transformer, an infrared thermal imaging sensor, a temperature and humidity sensor, and a smoke detector; it has multi-modal sensing capabilities, covering electrical, thermal, environmental, and safety dimensions, and can identify complex risks such as overheating, insulation degradation, and fire hazards at an early stage. The edge computing unit has a built-in chip that runs a neural network model to perform preliminary fault diagnosis, abnormal line loss detection, and equipment status assessment. It completes the initial fault judgment, abnormal line loss detection, and equipment health assessment on the terminal side, reducing dependence on the main station and reducing communication load. The secure communication unit integrates a PUF chip to achieve device authentication and secure key generation, providing a unique device identity and dynamic key; The local decision execution unit is used to execute local switching actions or issue warnings when a preset emergency event is detected. When an emergency event (such as severe leakage or a surge in partial discharge) is detected, the power supply can be cut off immediately or an alarm can be triggered, which significantly improves personal and equipment safety. Regional coordination nodes are used to aggregate data from the edge intelligent terminals and store key data through a blockchain network. It should be noted that the blockchain network adopts a Directed Acyclic Graph (DAG) structure to store at least one of the following types of critical data: leakage event records, power outage event records, equipment status assessment reports, and user-authorized transaction records. All data uploaded to the chain is digitally signed by regional coordination nodes. Compared to traditional chain-based blockchains, the DAG structure supports asynchronous parallel writing, making it more suitable for the high-frequency event uploading needs of a large number of terminals in the power distribution network. Critical operational events and transaction records are immutable and chronologically ordered, providing a credible chain of evidence for accident backtracking, liability determination, and carbon asset accounting. User authorization to participate in operations such as virtual power plants is digitally signed and uploaded to the chain to ensure the authenticity and verifiability of operations and prevent malicious impersonation. Furthermore, the regional coordination node performs federated learning coordination tasks, with the following specific steps: a. Distribute the initial parameters of the global fault prediction model to each edge intelligent terminal; b. Receive the updated model parameters from each edge intelligent terminal after training based on local data; c. Perform weighted aggregation on the received model parameters to generate a new global model, and then distribute the new global model to each edge intelligent terminal; Each edge terminal does not need to upload raw sensitive data (such as user electricity consumption details and equipment operation data), but only exchanges model parameters; through federated learning, the system can continuously absorb local experience from various regions and scenarios, and improve the generalization ability of models such as fault prediction and anomaly detection; the computing tasks are distributed to the edge, and regional nodes are only responsible for coordination and aggregation, which improves the scalability of the system. A digital twin platform is used to construct and maintain a three-dimensional digital twin of the distribution network, achieving real-time synchronization between the physical power grid and the virtual model. The platform receives and parses multi-dimensional sensing data from regional coordination nodes; it dynamically maps distribution network equipment and their operating status in three-dimensional space, forming a visualized digital twin. Maintenance personnel can monitor the status of all network equipment, temperature distribution, and fault locations in real time through a three-dimensional visualization interface, significantly improving maintenance efficiency. Based on the high-fidelity twin, "digital simulations" such as fault propagation simulation, load transfer scheme verification, and capacity expansion planning can be performed, reducing trial-and-error costs. It also includes a user interaction terminal, which provides users with real-time visualized information on their electricity carbon footprint and offers personalized energy-saving suggestions based on their electricity consumption habits; receives images or videos of equipment malfunctions reported by users and automatically identifies and generates work orders through cloud services; provides a virtual power plant interface, allowing users to authorize their distributed energy resources to participate in the grid ancillary services market and share revenue according to their contribution; enhances users' energy efficiency awareness and participation, guides green electricity consumption behavior through carbon footprint transparency, and promotes the coordinated development of power generation, grid, load, and storage, as well as the construction of a new power system ecosystem; Among them, the multiple edge intelligent terminals communicate with each other and with the regional coordination node through a hybrid Mesh self-organizing network. The hybrid Mesh self-organizing network supports multi-hop and self-healing communication, and is adapted to the field environment with complex power distribution network topology and poor communication conditions.
[0015] The parts of this invention not described in detail are prior art. It will be apparent to those skilled in the art that this invention is not limited to the details of the above exemplary embodiments, and that the invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be regarded as exemplary and non-limiting in all respects, and are intended to encompass all changes that fall within the meaning and scope of equivalents in the content of this invention.
Claims
1. A smart sensing system for a power distribution network, characterized in that, include: Edge intelligent terminals are deployed at branch nodes and meter boxes of the power distribution network to perform multi-source fusion sensing of local electrical quantities, thermal imaging data, partial discharge signals and environmental parameters, and to execute local real-time decisions. Regional coordination nodes are used to aggregate data from the edge intelligent terminals and store key data through a blockchain network. A digital twin platform is used to build and maintain a three-dimensional digital twin of the power distribution network, enabling real-time synchronization between the physical power grid and the virtual model; The multiple edge intelligent terminals communicate with each other, as well as with the regional coordination node, through a hybrid Mesh self-organizing network.
2. The intelligent sensing system for a power distribution network according to claim 1, characterized in that: The edge intelligent terminal includes: The multi-sensor module integrates a current transformer, a voltage transformer, an infrared thermal imaging sensor, a temperature and humidity sensor, and a smoke detector. The edge computing unit has a built-in chip that runs a neural network model to perform preliminary fault diagnosis, abnormal line loss detection, and equipment status assessment. The secure communication unit integrates a PUF chip for device authentication and secure key generation. The local decision execution unit is used to execute local switching actions or issue warnings when a preset emergency event is detected.
3. The intelligent sensing system for a power distribution network according to claim 1, characterized in that: The regional coordination node performs the federated learning coordination task, and the specific steps are as follows: a. Distribute the initial parameters of the global fault prediction model to each edge intelligent terminal; b. Receive the updated model parameters from each edge intelligent terminal after training based on local data; c. Perform weighted aggregation on the received model parameters to generate a new global model, and then distribute the new global model to each edge intelligent terminal.
4. The intelligent sensing system for a power distribution network according to claim 1, characterized in that: The digital twin platform receives and parses multi-dimensional sensing data from regional coordination nodes; it dynamically maps distribution network equipment and its operating status in three-dimensional space to form a visualized digital twin.
5. The intelligent sensing system for a power distribution network according to claim 1, characterized in that: It also includes a user interaction terminal, which provides users with real-time visualization of their electricity carbon footprint and offers personalized energy-saving suggestions based on their electricity consumption habits; Receive images or videos of abnormal devices reported by users, and automatically identify and generate work orders through cloud services; It provides a virtual power plant interface, allowing users to authorize their distributed energy resources to participate in the grid ancillary services market and share revenue based on their contribution.
6. The intelligent sensing system for a power distribution network according to claim 1, characterized in that: The blockchain network adopts a directed acyclic graph (DAG) structure to store at least one of the following types of key data: leakage event records, power outage event records, equipment status assessment reports, and user-authorized transaction records. All data uploaded to the chain is digitally signed by the regional coordination node.