Power distribution network monitoring method and system based on multi-source data

Abstract

The invention discloses a power distribution network monitoring method and system based on multi-source data. The method comprises the following steps: collecting multi-source state information, and triggering signal diffusion and data reporting when the state is judged to be abnormal; a wireless signal is sent to a signal receiving unit installed on the adjacent power distribution equipment by using a signal diffusion unit, and local multi-source data is uploaded to a cloud side collaboration platform at the same time; based on the received data, intelligent diagnosis analysis is carried out, and a diffusion diagnosis instruction set with directivity is generated and issued to specified adjacent power distribution equipment; the designated adjacent power distribution equipment receiving the wireless signal and the diffusion diagnosis instruction set carries out data acquisition and uploading according to instruction requirements, and based on new data iteration, a diagnosis and positioning result is updated until a fault point is accurately positioned; and based on the accurately positioned fault point, generating a control strategy and remotely controlling the switch equipment to realize fault isolation and non-fault area power supply recovery. According to the invention, accurate fault positioning and intelligent diagnosis are realized.

Description

Technical Field

[0001] This invention relates to the field of power system distribution network automation technology, and more specifically, to a distribution network monitoring method and system based on multi-source data. Background Technology

[0002] As the final link in the power system facing users, the reliability of the power distribution network directly affects the quality of power supply. Existing technologies include methods that detect faults by installing acquisition units and wireless communication modules on distribution equipment, sending signals to adjacent devices when an anomaly is detected, thus propagating the fault through a cascading mechanism.

[0003] However, such methods can only determine that the fault is located between the "last abnormal device" and the "first normal device," and cannot accurately locate the specific faulty device or line segment in complex topologies containing branches or ring networks. Therefore, there is an urgent need for a distribution network monitoring method that can achieve accurate location and intelligent diagnosis. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide a power distribution network monitoring method and system based on multi-source data.

[0005] To address the above problems, this invention provides a power distribution network monitoring method based on multi-source data, characterized in that the method includes the following steps: Step S1: The intelligent monitoring unit deployed on each power distribution equipment in the power distribution network collects multi-source status information and triggers signal propagation and data reporting when the status is determined to be abnormal. Step S2: Use the signal propagation unit installed on the power distribution equipment with the abnormal status to send wireless signals to the signal receiving unit installed on its adjacent power distribution equipment, and at the same time upload local multi-source data to the cloud-edge collaboration platform. Step S3: The cloud-edge collaborative platform performs intelligent diagnostic analysis based on the received data and generates a directional diffusion diagnostic instruction set, which is then sent to the designated adjacent power distribution equipment. Step S4: The designated adjacent power distribution equipment that receives the wireless signal and the diffusion diagnostic instruction set collects and uploads data according to the instruction requirements. The cloud-edge collaborative platform iteratively updates the diagnostic and positioning results based on the new data until the fault point is accurately located. Step S5: Based on the precisely located fault point, the cloud-edge collaborative platform generates control strategies and remotely controls the switching equipment to achieve fault isolation and power restoration in non-faulty areas.

[0006] Preferably, the determination of an abnormal state in step S1 specifically includes: The intelligent monitoring unit calculates the feature quantities of the collected multi-source state information in real time and inputs the feature quantities into the preset intelligent decision model; wherein, the multi-source state information includes electrical quantities, equipment state quantities and environmental quantities; the feature quantities include at least the zero-sequence component, negative-sequence component, harmonic content and phase change of electrical quantities, and the temperature change rate and vibration spectrum characteristics of equipment state quantities. The intelligent decision model outputs a comprehensive anomaly confidence level and a preliminary anomaly type identifier; When the anomaly confidence level exceeds a preset threshold and at least one feature exceeds its physical safety threshold, the power distribution equipment is determined to be in an abnormal state.

[0007] Preferably, the step of using the signal propagation unit installed on the power distribution equipment with the abnormal status to send wireless signals to the signal receiving unit installed on its adjacent power distribution equipment, while simultaneously uploading local multi-source data to the cloud-edge collaborative platform, specifically includes: The signal diffusion unit sends a data frame to the signal receiving unit of its adjacent power distribution equipment. The data frame contains at least the trigger source device ID, the trigger timestamp, and the preliminary abnormality type identifier obtained in step S1. The local multi-source data includes at least the high-sampling-rate original waveform data at the time of the fault, the status information trend data within the predetermined time period before the fault, and the abnormal confidence and preliminary abnormal type identification obtained from the local judgment.

[0008] Preferably, the intelligent diagnostic analysis in step S3 specifically includes: Step S31: Perform time alignment and fusion processing on multi-source data from one or more power distribution devices, and extract fusion feature vectors for fault diagnosis; Step S32: Input the fused feature vector into a preset multi-task processing model; the multi-task processing model simultaneously outputs the probability distribution of the fault type, the preliminary confidence distribution of the fault area, and the fault severity assessment value.

[0009] Preferably, the multi-task processing model is obtained through supervised training using a dataset containing historical fault waveform data, corresponding fault types, and exact fault location labels.

[0010] Preferably, the generation of a directional diffusion diagnostic instruction set in step S3 specifically includes: Step S33: Based on the fault type probability distribution and the preliminary confidence distribution of the fault area output by the multi-task processing model, and combined with the pre-stored distribution network topology map, dynamically generate instructions for one or more specific downstream power distribution devices; the instructions at least specify the target device ID and the targeted data acquisition requirements.

[0011] Preferably, the targeted data acquisition requirements include: increasing the sampling frequency of specific electrical quantities, initiating the measurement and reporting of specific non-electrical quantity sensors, or focusing on calculating and reporting specific fault characteristic quantities.

[0012] Preferably, the iterative update of the diagnosis and localization results in step S4 specifically includes: Based on the newly uploaded data, the cloud-edge collaboration platform repeatedly performs data fusion, feature extraction, and multi-task artificial intelligence model diagnosis to update the confidence distribution map of the fault area. Based on the updated confidence distribution map, it generates a new set of diffusion diagnosis instructions with a smaller focus until the probability of a specific device or line segment in the confidence distribution map of the fault area exceeds the predetermined convergence threshold. During the iteration process, the direction of the diffusion diagnostic command is dynamically adjusted based on the fault type: for short-circuit faults, the command is preferentially directed to downstream devices in the fault current path; for ground faults, the command requires upstream and downstream devices in the fault path to simultaneously report the phase information of zero-sequence current and voltage.

[0013] Preferably, the generation of the control strategy in step S5 specifically involves: Based on the precisely located fault point information and the pre-stored real-time power grid topology and switch status, the rule engine automatically generates a switch operation sequence with the goal of minimizing the power outage range. The operation sequence includes at least disconnecting the nearest upstream switch of the fault point and closing the tie switch to restore power supply.

[0014] The present invention also provides a power distribution network monitoring system based on multi-source data, the system comprising a field device layer, a communication network layer, and a cloud-edge collaborative platform; The field device layer includes intelligent monitoring units deployed on multiple power distribution devices in the power distribution network; The communication network layer is used to realize data communication within the field device layer and between it and the cloud-edge collaboration platform; The cloud-edge collaboration platform is used for intelligent data analysis and decision-making. The intelligent monitoring unit includes: Multi-source sensor arrays are used to collect electrical quantities, equipment status quantities, and environmental quantity data; The microprocessor (MCU) is used to process sensor data and perform local lightweight intelligent decisions to determine abnormal states. The signal propagation and data reporting module is used to send wireless signals to neighboring devices and upload multi-source data to the cloud-edge collaboration platform when the status is abnormal. The signal receiving and wake-up module is used to receive wireless signals from adjacent devices and instructions from the cloud-edge collaboration platform, and to control the working status of the intelligent monitoring unit. The cloud-edge collaboration platform includes: The data fusion and feature extraction module is used to perform time alignment and advanced feature vector extraction on the reported multi-source data. The intelligent diagnostic analysis engine has a built-in pre-trained multi-task artificial intelligence model, which is used to output fault type, fault area estimation and severity based on feature vectors; A diffusion diagnostic command scheduler is used to generate a directional diffusion diagnostic command set based on diagnostic results and power grid topology; A control strategy generator is used to generate a sequence of switching operations for isolating the fault and restoring power supply after the fault is accurately located.

[0015] As can be seen from the above technical solution, this application has at least the following beneficial effects: In terms of fault location accuracy, this method improves fault location accuracy by setting up a cloud-edge collaborative platform, performing time alignment and advanced feature vector extraction on the reported multi-source data, outputting fault type, fault area estimation and severity based on the feature vectors, and generating a directional diffusion diagnostic instruction set based on the diagnostic results and power grid topology.

[0016] From the perspective of intelligent anomaly handling, this method processes the collected multi-source data, performs local lightweight intelligent judgment to determine the state anomaly, sends wireless signals to adjacent devices and uploads multi-source data to the cloud-edge collaboration platform when the device state is abnormal, and controls the working state of the intelligent monitoring unit, thus realizing intelligent processing.

[0017] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this application do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0018] Figure 1 A flowchart of a power distribution network monitoring method based on multi-source data provided in this application embodiment; Figure 2 The framework of the power distribution network monitoring system based on multi-source data provided in the embodiments of this application; Figure 3This is a schematic diagram of a computing device provided in an embodiment of this application. Detailed Implementation

[0019] The terms "first," "second," and "third," etc., used in this application specification and accompanying drawings are used to distinguish different objects, not to limit a specific order.

[0020] In the embodiments of this application, the terms "exemplary" or "for example" are used to indicate that something is an example, illustration, or description. Any embodiment or design that is described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design. Specifically, the use of the terms "exemplary" or "for example" is intended to present the relevant concepts in a specific manner.

[0021] To ensure clarity and conciseness in the description of the following embodiments, a brief introduction to the related technologies is given first: The reliable operation of a power distribution network directly affects the power quality for end users. To overcome the latency of centralized processing at the main station, existing technologies have proposed a distributed processing method. This method installs acquisition units and wireless communication modules on the power distribution equipment. When a device detects an anomaly in its own status, it sends a wireless wake-up signal to its neighboring devices. The receiving neighboring devices then initiate self-checks; if they also find an anomaly, they continue to propagate the signal to the next level of equipment until a device detects a normal status. In this way, the fault range is narrowed down to the segment between the "last device to report an anomaly" and the "first device to report normal status." However, this method can only determine a vague fault range and cannot accurately locate the specific faulty device or line in a complex multi-branch network, making subsequent maintenance difficult.

[0022] In view of this, embodiments of this application provide a power distribution network monitoring method based on multi-source data, which can be executed by a processing device. This processing device can be a terminal or a server. Terminals include, but are not limited to, smartphones, tablets, laptops, personal digital assistants, or smart wearable devices. Servers can be cloud servers, such as central servers in a central cloud computing cluster or edge servers in an edge cloud computing cluster. Alternatively, servers can be located in a local data center. A local data center refers to a data center directly controlled by the user.

[0023] To make the technical solution of this application clearer and easier to understand, the following describes a distribution network monitoring method based on multi-source data provided by an embodiment of this application, in conjunction with the accompanying drawings. Figure 1 As shown in the figure, this is a power distribution network monitoring and self-healing method based on multi-source data and intelligent diffusion diagnosis provided in an embodiment of this application, including the following steps: Step S1: Multi-source data acquisition and anomaly triggering, specifically: Intelligent monitoring units are deployed on various power distribution equipment in the power distribution network. Each unit is an embedded system, the core of which includes: a multi-source sensor array, a microprocessor (MCU), a clock synchronization module (supporting BeiDou / GPS or IEEE 1588 protocol), and a communication module; it collects multi-source status information in real time, including electrical quantities, equipment status quantities, and environmental quantities; specifically, electrical quantity data includes three-phase voltage, current, power, frequency, and harmonic content; equipment status quantities include equipment temperature (via infrared sensors or patch temperature sensors), mechanical vibration, and operating noise; environmental quantity data includes local humidity, smoke, and images (via miniature cameras).

[0024] When any intelligent monitoring unit analyzes and finds that any one or more types of data it has collected exceed the preset threshold or meet the preset abnormal pattern, it determines that the power distribution equipment is in an abnormal state and then triggers its internal signal propagation and data reporting module.

[0025] In one embodiment, the specific process for determining abnormal status of power distribution equipment is as follows: The intelligent monitoring unit calculates the feature quantities of the collected multi-source state information in real time and inputs the feature quantities into a preset intelligent decision model. The feature quantity calculation specifically involves the MCU calculating the feature quantities of the raw data from the sensors in real time. The feature quantities include at least the zero-sequence component, negative-sequence component, harmonic content, and phase change of electrical quantities, and the temperature change rate (°C / s) and vibration spectrum characteristics (through FFT transformation) of equipment state quantities. The intelligent decision model outputs a comprehensive anomaly confidence level and a preliminary anomaly type identifier; When the anomaly confidence level exceeds a preset threshold and at least one feature exceeds its physical safety threshold, the power distribution equipment is determined to be in an abnormal state, thus avoiding malfunctions caused by momentary interference from a single sensor.

[0026] Step S2: Collaborative diffusion and data upload, specifically: The signal propagation unit installed on the power distribution equipment that is determined to be in an abnormal state sends a wireless signal to the signal receiving unit installed on its adjacent power distribution equipment, while simultaneously uploading local multi-source data to the cloud-edge collaborative platform 303. The specific process is as follows: The signal diffusion unit sends a data frame to the signal receiving unit of its adjacent power distribution equipment. The data frame contains at least the trigger source device ID, the trigger timestamp, and the preliminary abnormality type identifier obtained in step S1. The local multi-source data includes at least the high-sampling-rate original waveform data at the time of the fault, the status information trend data within the predetermined time period before the fault, and the abnormal confidence and preliminary abnormal type identification obtained from the local judgment.

[0027] The triggered signal propagation unit and data reporting module perform two parallel actions: The signal diffusion unit performs action A (wireless signal diffusion): it sends a wireless wake-up signal to the signal receiving and wake-up modules on all its directly adjacent power distribution equipment. This signal contains the unique ID of the triggering source device and a preliminary exception type code.

[0028] The data reporting unit performs action B (data reporting): it packages and uploads the recent multi-source historical data (such as data from a few seconds to a few minutes before the fault) and real-time abnormal data stored locally on the device to the cloud-edge collaboration platform 303 via power line carrier communication (PLC) and / or 4G / 5G wireless network. After receiving a wireless signal, the signal receiving unit on the adjacent power distribution equipment starts a listening timer to wait for the diffusion diagnostic instruction set from the cloud-edge collaborative platform 303; if no instruction for this device is received within the timeout period, the intelligent monitoring unit is controlled to re-enter a low-power sleep state.

[0029] Step S3: Intelligent diagnostic analysis and positioning command generation, specifically: The cloud-edge collaborative platform 303 performs intelligent diagnostic analysis based on the received data and generates a directional diffusion diagnostic instruction set, which is then sent to designated adjacent power distribution equipment. The specific process is as follows: After receiving the reported multi-source data, the cloud-edge collaboration platform 303 executes the following: Step S31: Perform time alignment and fusion processing on multi-source data from one or more power distribution devices, and extract fusion feature vectors for fault diagnosis (such as voltage sag depth, zero-sequence current amplitude, temperature rise rate, etc.).

[0030] Step S32: Input the fused feature vector into a preset multi-task processing model (e.g., a model based on a deep neural network or gradient boosting decision tree). The multi-task processing model simultaneously outputs the probability distribution of the fault type, the preliminary confidence distribution of the fault region, and the fault severity assessment value. The multi-task processing model is obtained through supervised training using a dataset containing historical fault waveform data, corresponding fault types, and exact fault location labels.

[0031] Step S33: Based on the fault type probability distribution and preliminary confidence distribution of the fault area output by the multi-task processing model, and combined with the pre-stored distribution network topology map, dynamically generate instructions for one or more specific downstream power distribution devices; the instructions at least specify the target device ID and targeted data. Targeted data acquisition requirements include: increasing the sampling frequency of specific electrical quantities, activating the measurement and reporting of specific non-electrical quantity sensors, or focusing on calculating and reporting specific fault characteristic quantities.

[0032] Step S4: Bidirectional iterative precise localization, specifically: Upon receiving the wireless signal and the diffusion diagnostic instruction set, the designated adjacent power distribution equipment collects and uploads data as required by the instructions. The cloud-edge collaborative platform 303 iteratively updates the diagnostic and location results based on the new data until the fault point is accurately located. The specific process is as follows: Neighboring devices that receive the wireless wake-up signal do not immediately report all the data, but wait and parse the diffusion diagnostic instructions from the cloud-edge collaboration platform 303; The device specified by the instruction is woken up first, and performs high-precision data acquisition and uploads the data according to the instruction requirements; Based on the newly uploaded data, the cloud-edge collaboration platform 303 repeatedly performs data fusion, feature extraction, and multi-task artificial intelligence model diagnosis to update the confidence distribution map of the fault area. Based on the updated confidence distribution map, it generates a new set of diffusion diagnosis instructions with a smaller focus until the probability of a specific device or line segment in the confidence distribution map of the fault area exceeds the predetermined convergence threshold. During the iteration process, the direction of the diffusion diagnostic command is dynamically adjusted based on the fault type: for short-circuit faults, the command is preferentially directed to downstream devices in the fault current path; for ground faults, the command requires upstream and downstream devices in the fault path to simultaneously report the phase information of zero-sequence current and voltage. This process forms a closed loop of "platform command - equipment execution - data feedback". Through multiple iterations, the platform can dynamically adjust the diagnostic focus, gradually converging from fault area estimation to ultimately pinpoint the specific faulty device or faulty line segment. This process is bidirectional, meaning that diagnostic commands can be issued simultaneously or alternately upstream and downstream along the power supply direction, effectively solving the location challenge in complex network structures.

[0033] Step S5: Fault isolation and recovery of non-faulty areas, specifically: The cloud-edge collaborative platform 303 generates control strategies and remotely controls switching equipment based on precisely located fault points, achieving fault isolation and power restoration to non-faulty areas. The specific process is as follows: After accurately locating the fault, the cloud-edge collaboration platform 303 generates a control strategy and issues commands: Based on the precisely located fault point information and the pre-stored real-time power grid topology and switch status, the rule engine automatically generates a switch operation sequence with the goal of minimizing the power outage range. The operation sequence includes at least disconnecting the nearest upstream switch of the fault point and closing the tie switch to restore power supply. Based on the precisely located fault point information and the pre-stored real-time power grid topology and switch status, the rule engine automatically generates a switch operation sequence with the goal of minimizing the power outage range. The operation sequence includes at least disconnecting the nearest upstream switch of the fault point and closing the tie switch to restore power supply.

[0034] After receiving a wireless signal, the signal receiving unit on the adjacent power distribution equipment starts a listening timer to wait for the diffusion diagnostic instruction set from the cloud-edge collaborative platform 303; if no instruction for this device is received within the timeout period, the intelligent monitoring unit is controlled to re-enter a low-power sleep state. At the same time, fault information, location results, and handling measures are automatically pushed to the mobile terminals of maintenance personnel.

[0035] Through the above methods, the present invention fully discloses a complete technical solution that combines multi-source sensing, intelligent diagnosis and collaborative control to achieve a qualitative leap in power distribution network monitoring from "passive alarm" to "active early warning, accurate diagnosis and intelligent self-healing".

[0036] like Figure 2 As shown, this embodiment of the invention also provides a power distribution network monitoring system based on multi-source data, the system including a field device layer 301, a communication network layer 302, and a cloud-edge collaborative platform 303; The field device layer 301 includes intelligent monitoring units deployed on multiple power distribution devices in the power distribution network; The communication network layer 302 is used to realize data communication within the field device layer 301 and between it and the cloud-edge collaboration platform 303; The cloud-edge collaboration platform 303 is used for intelligent data analysis and decision-making; The intelligent monitoring unit includes: Multi-source sensor arrays are used to collect electrical quantities, equipment status quantities, and environmental quantity data; The microprocessor (MCU) is used to process sensor data and perform local lightweight intelligent decisions to determine abnormal states. The signal propagation and data reporting module is used to send wireless signals to neighboring devices and upload multi-source data to the cloud-edge collaboration platform 303 when the device status is abnormal. The signal receiving and wake-up module is used to receive wireless signals from adjacent devices and instructions from the cloud-edge collaboration platform 303, and to control the working status of the intelligent monitoring unit. The cloud-edge collaboration platform 303 includes: The data fusion and feature extraction module is used to perform time alignment and advanced feature vector extraction on the reported multi-source data. The intelligent diagnostic analysis engine has a built-in pre-trained multi-task artificial intelligence model, which is used to output fault type, fault area estimation and severity based on feature vectors; A diffusion diagnostic command scheduler is used to generate a directional diffusion diagnostic command set based on diagnostic results and power grid topology; A control strategy generator is used to generate a sequence of switching operations for isolating the fault and restoring power supply after the fault is accurately located.

[0037] This application also provides a computing device. For example... Figure 3 As shown in the figure, this is a schematic diagram of a computing device provided in an embodiment of this application. The computing device 700 includes a bus 701, a processor 702, a communication interface 703, and a memory 704. The processor 702, the memory 704, and the communication interface 703 communicate with each other via the bus 701.

[0038] The 701 bus can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of representation, Figure 3 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0039] The processor 702 can be any one or more of the following processors: central processing unit (CPU), graphics processing unit (GPU), microprocessor (MP), or digital signal processor (DSP).

[0040] The communication interface 703 is used for communication with external devices.

[0041] Memory 704 may include volatile memory, such as random access memory (RAM). Memory 704 may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0042] The memory 704 stores executable code, and the processor 702 executes the executable code to perform the aforementioned power distribution network monitoring method based on multi-source data.

[0043] Specifically, in achieving Figure 2 In the case of the illustrated embodiment, and Figure 2 When the modules or units of the fault diagnosis device for the capacitive voltage transformer described in the embodiment are implemented by software, the following steps are performed: Figure 2 The software or program code required for the functions of each module / unit can be partially or entirely stored in the memory 704. The processor 702 executes the program code corresponding to each unit stored in the memory 704, and executes the aforementioned distribution network monitoring method based on multi-source data.

[0044] This application also provides a computer-readable storage medium. The computer-readable storage medium can be any available medium that a computing device can store, or a data storage device such as a data center containing one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive). The computer-readable storage medium includes instructions that instruct the computing device to execute the aforementioned power distribution network monitoring method based on multi-source data.

[0045] This application also provides a computer program product comprising one or more computer instructions. When the computer instructions are loaded and executed on a computing device, all or part of the processes or functions described in this application are generated.

[0046] The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, or data center to another website, computer, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means.

[0047] When the computer program product is executed by a computer, the computer executes any of the aforementioned methods of the multi-source data-based distribution network monitoring method. The computer program product can be a software installation package; when any of the aforementioned methods of the multi-source data-based distribution network monitoring method is required, the computer program product can be downloaded and executed on the computer.

[0048] The descriptions of the processes or structures corresponding to the above figures each have their own emphasis. For parts of a process or structure that are not described in detail, please refer to the relevant descriptions of other processes or structures.

[0049] While the present invention has been disclosed above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and all such changes and modifications will fall within the scope of protection of the present invention.

Claims

1. A method for monitoring a distribution network based on multi-source data, characterized in that, The method includes the following steps: Step S1: The intelligent monitoring unit deployed on each power distribution equipment in the power distribution network collects multi-source status information and triggers signal propagation and data reporting when the status is determined to be abnormal. Step S2: Use the signal propagation unit installed on the power distribution equipment with the abnormal status to send wireless signals to the signal receiving unit installed on its adjacent power distribution equipment, and at the same time upload local multi-source data to the cloud-edge collaboration platform. Step S3: The cloud-edge collaborative platform performs intelligent diagnostic analysis based on the received data and generates a directional diffusion diagnostic instruction set, which is then sent to the designated adjacent power distribution equipment. Step S4: The designated adjacent power distribution equipment that receives the wireless signal and the diffusion diagnostic instruction set collects and uploads data according to the instruction requirements. The cloud-edge collaborative platform iteratively updates the diagnostic and positioning results based on the new data until the fault point is accurately located. Step S5: Based on the precisely located fault point, the cloud-edge collaborative platform generates control strategies and remotely controls the switching equipment to achieve fault isolation and power restoration in non-faulty areas.

2. The distribution network monitoring method based on multi-source data according to claim 1, characterized in that, The determination of an abnormal state in step S1 specifically includes: The intelligent monitoring unit calculates the feature quantities of the collected multi-source state information in real time and inputs the feature quantities into the preset intelligent decision model; wherein, the multi-source state information includes electrical quantities, equipment state quantities and environmental quantities; the feature quantities include at least the zero-sequence component, negative-sequence component, harmonic content and phase change of electrical quantities, and the temperature change rate and vibration spectrum characteristics of equipment state quantities. The intelligent decision model outputs a comprehensive anomaly confidence level and a preliminary anomaly type identifier; When the anomaly confidence level exceeds a preset threshold and at least one feature exceeds its physical safety threshold, the power distribution equipment is determined to be in an abnormal state.

3. The distribution network monitoring method based on multi-source data according to claim 1, characterized in that, The process of using a signal propagation unit installed on a power distribution device with an abnormal status to send a wireless signal to a signal receiving unit installed on an adjacent power distribution device, while simultaneously uploading local multi-source data to the cloud-edge collaborative platform, specifically includes: The signal diffusion unit sends a data frame to the signal receiving unit of its adjacent power distribution equipment. The data frame contains at least the trigger source device ID, the trigger timestamp, and the preliminary abnormality type identifier obtained in step S1. The local multi-source data includes at least the high-sampling-rate original waveform data at the time of the fault, the status information trend data within the predetermined time period before the fault, and the abnormal confidence and preliminary abnormal type identification obtained from the local judgment.

4. The distribution network monitoring method based on multi-source data according to claim 1, characterized in that, The intelligent diagnostic analysis in step S3 specifically includes: Step S31: Perform time alignment and fusion processing on multi-source data from one or more power distribution devices, and extract fusion feature vectors for fault diagnosis; Step S32: Input the fused feature vector into a preset multi-task processing model; the multi-task processing model simultaneously outputs the probability distribution of the fault type, the preliminary confidence distribution of the fault area, and the fault severity assessment value.

5. The distribution network monitoring method based on multi-source data according to claim 4, characterized in that, The multi-task processing model is obtained through supervised training using a dataset containing historical fault waveform data, corresponding fault types, and exact fault location labels.

6. The distribution network monitoring method based on multi-source data according to claim 4, characterized in that, The specific steps in step S3 of generating a directional diffusion diagnostic instruction set include: Step S33: Based on the fault type probability distribution and the preliminary confidence distribution of the fault area output by the multi-task processing model, and combined with the pre-stored distribution network topology map, dynamically generate instructions for one or more specific downstream power distribution devices; the instructions at least specify the target device ID and the targeted data acquisition requirements.

7. The distribution network monitoring method based on multi-source data according to claim 6, characterized in that, The targeted data acquisition requirements include: increasing the sampling frequency of specific electrical quantities, activating the measurement and reporting of specific non-electrical quantity sensors, or focusing on calculating and reporting specific fault characteristic quantities.

8. The distribution network monitoring method based on multi-source data according to claim 1, characterized in that, The iterative update of the diagnosis and localization results in step S4 specifically includes: Based on the newly uploaded data, the cloud-edge collaboration platform repeatedly performs data fusion, feature extraction, and multi-task artificial intelligence model diagnosis to update the confidence distribution map of the fault area. Based on the updated confidence distribution map, it generates a new set of diffusion diagnosis instructions with a smaller focus until the probability of a specific device or line segment in the confidence distribution map of the fault area exceeds the predetermined convergence threshold. During the iteration process, the direction of the diffusion diagnostic command is dynamically adjusted based on the fault type: for short-circuit faults, the command is preferentially directed to downstream devices in the fault current path; for ground faults, the command requires upstream and downstream devices in the fault path to simultaneously report the phase information of zero-sequence current and voltage.

9. The distribution network monitoring method based on multi-source data according to claim 1, characterized in that, The specific steps for generating the control strategy in step S5 are as follows: Based on the precisely located fault point information and the pre-stored real-time power grid topology and switch status, the rule engine automatically generates a switch operation sequence with the goal of minimizing the power outage range. The operation sequence includes at least disconnecting the nearest upstream switch of the fault point and closing the tie switch to restore power supply.

10. A power distribution network monitoring system based on multi-source data, characterized in that, The system includes a field device layer, a communication network layer, and a cloud-edge collaboration platform; The field device layer includes intelligent monitoring units deployed on multiple power distribution devices in the power distribution network; The communication network layer is used to realize data communication within the field device layer and between it and the cloud-edge collaboration platform; The cloud-edge collaboration platform is used for intelligent data analysis and decision-making. The intelligent monitoring unit includes: Multi-source sensor arrays are used to collect electrical quantities, equipment status quantities, and environmental quantity data; The microprocessor (MCU) is used to process sensor data and perform local lightweight intelligent decisions to determine abnormal states. The signal propagation and data reporting module is used to send wireless signals to neighboring devices and upload multi-source data to the cloud-edge collaboration platform when the status is abnormal. The signal receiving and wake-up module is used to receive wireless signals from adjacent devices and instructions from the cloud-edge collaboration platform, and to control the working status of the intelligent monitoring unit. The cloud-edge collaboration platform includes: The data fusion and feature extraction module is used to perform time alignment and advanced feature vector extraction on the reported multi-source data. The intelligent diagnostic analysis engine has a built-in pre-trained multi-task artificial intelligence model, which is used to output fault type, fault area estimation and severity based on feature vectors; A diffusion diagnostic command scheduler is used to generate a directional diffusion diagnostic command set based on diagnostic results and power grid topology; A control strategy generator is used to generate a sequence of switching operations for isolating the fault and restoring power supply after the fault is accurately located.

Images