Fault detection method and device for power distribution network, electronic equipment and storage medium

Abstract

The invention provides a fault detection method and device for a power distribution network, electronic equipment and a computer readable storage medium, and relates to the technical field of power system monitoring. The fault detection method comprises the steps that power distribution information is collected, and the power distribution information comprises equipment operation information and voltage, current and power real-time data of a connected line; based on a binary decision tree fault prediction model, performing fault prediction on the power distribution information to obtain fault prediction information; and based on the fault prediction information and the classification and marking power distribution information, obtaining power distribution marking information, and sending the power distribution marking information to a power distribution master station, so that the power distribution master station performs fault detection on the power distribution marking information to obtain a fault detection result. The problems that fault information research and judgment are lacked, the workload of data preprocessing and analysis is huge, and the construction and maintenance cost of the digital twinborn body is high in the prior art are at least solved. The method adapts to power detection scenes.

Description

Technical Field

[0001] The present invention relates to the technical field of power system monitoring, and in particular to a method, device, electronic equipment and computer-readable storage medium for fault detection in a distribution network. Background Art

[0002] Currently, distribution network fault detection primarily relies on distribution automation technology. Based on the primary grid and equipment of the distribution network, distribution automation integrates information from these networks and related application systems to enable monitoring, control, and rapid fault isolation, enabling rapid fault resolution and improving power supply reliability. Therefore, existing technologies can utilize distribution automation terminals to capture raw data and create a digital twin of the distribution network using digital twin technology. This allows for real-time monitoring of the distribution network's operating status and analysis of fault information based on operational data.

[0003] However, existing fault detection technologies primarily rely on distribution automation (DA) and feeder automation (FA), lacking the ability to analyze fault information and provide early warnings before a fault occurs. Furthermore, distribution networks contain a large number of primary and secondary power devices, involving environmental data, equipment parameters, and distribution network operating status data. This means that distribution networks are characterized by a wide variety of data types, massive amounts of data, random multidimensionality, time-varying data, and nonlinearity. Consequently, existing data preprocessing and analysis techniques require significant workload, and the construction and maintenance costs of digital twins are high. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to address the above-mentioned deficiencies in the prior art and provide a distribution network fault detection method, device, electronic device and computer-readable storage medium. The method can realize convenient and effective processing and analysis of distribution network operation status data and fault prediction on the distribution terminal, thereby realizing collaborative fault detection of the distribution terminal and the distribution master station, improving the accuracy and efficiency of fault detection, reducing the occurrence rate of faults, and improving the stability of the distribution network.

[0005] In a first aspect, the present invention provides a fault detection method for a distribution network, which is applied to a distribution terminal. The distribution network includes a distribution terminal and a distribution master station. The fault detection method for the distribution network includes: collecting distribution information, wherein the distribution information includes equipment operation information and real-time data of voltage, current and power of the connected lines; based on a binary decision tree fault prediction model, performing fault prediction on the distribution information to obtain fault prediction information; based on the fault prediction information, classifying and marking the distribution information to obtain distribution labeling information, and sending the distribution labeling information to the distribution master station, so that the distribution master station performs fault detection on the distribution labeling information to obtain a fault detection result.

[0006] Preferably, the binary decision tree fault prediction model includes formula (1) and formula (2):

[0007]

[0008] Wherein, Gini(D) represents the Gini coefficient of the power distribution information with faults, D represents the sample data with faults in the power distribution information, and p k represents the proportion of sample data with fault k in the power distribution information in D, k = 1, 2, ... K, K represents the total number of fault types, Gini (D, A) represents the Gini index of fault A in the power distribution information, D1 represents the sample data with fault A in the power distribution information, D2 represents the sample data with fault Sample data of the fault represents all faults except fault A, Gini(D1) represents the Gini coefficient of fault A in the power distribution information, and Gini(D2) represents the Gini coefficient of fault A in the power distribution information. The Gini coefficient.

[0009] Preferably, sending the distribution marking information to the distribution master station specifically includes: encrypting the distribution marking information based on an encryption algorithm, wherein the encryption algorithm includes at least one of the following: commercial secret SM1 algorithm, commercial secret SM2 algorithm; and sending the encrypted distribution marking information to the distribution master station.

[0010] Preferably, after classifying and marking the distribution information based on the fault prediction information, obtaining the distribution labeling information, and sending the distribution labeling information to the distribution master station so that the distribution master station performs fault detection on the distribution labeling information and obtains the fault detection result, the fault detection method of the distribution network also includes: obtaining the fault detection result, wherein the fault detection result is obtained by the distribution master station performing fault detection on the distribution labeling information; and optimizing the binary decision tree fault prediction model based on the fault detection result, the fault prediction information and the pruning strategy.

[0011] In a second aspect, the present invention also provides a distribution network fault detection method, which is applied to a distribution master station. The distribution network fault detection method includes: receiving distribution labeling information sent by a distribution terminal, wherein the distribution labeling information is collected by the distribution terminal, and based on a binary decision tree fault prediction model, fault prediction is performed on the distribution information to obtain fault prediction information, and based on the fault prediction information, the distribution information is classified and labeled; fault detection is performed on the distribution labeling information to obtain a fault detection result.

[0012] Preferably, fault detection is performed on the power distribution marking information to obtain a fault detection result, which specifically includes: performing digital twinning on the power distribution marking information to obtain power distribution status data; and performing fault detection on the power distribution status data to obtain a fault detection result.

[0013] In the third aspect, the present invention also provides a fault detection device for a distribution network, which is applied to a distribution terminal and includes an acquisition module, a fault prediction module and a sending module. The acquisition module is used to collect distribution information, wherein the distribution information includes equipment operation information and real-time data of voltage, current and power of the connected lines. The fault prediction module is connected to the acquisition module and is used to perform fault prediction on the distribution information based on a binary decision tree fault prediction model to obtain fault prediction information. The sending module is connected to the fault prediction module and the acquisition module respectively and is used to generate distribution labeling information based on the fault prediction information and the distribution information, and send the distribution labeling information to the distribution master station so that the distribution master station performs fault detection on the distribution labeling information to obtain a fault detection result.

[0014] In a fourth aspect, the present invention also provides a fault detection device for a distribution network, which is applied to a distribution master station and includes a receiving module and a fault detection module. The receiving module is used to receive distribution labeling information sent by a distribution terminal, wherein the distribution labeling information is collected by the distribution terminal, and the distribution information is predicted to be faulty based on a binary decision tree fault prediction model to obtain fault prediction information, and the distribution information is classified and labeled based on the fault prediction information. The fault detection module is connected to the receiving module and is used to perform fault detection on the distribution labeling information to obtain a fault detection result.

[0015] In a fifth aspect, the present invention further provides an electronic device comprising a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to implement the distribution network fault detection method provided in the first or second aspect above.

[0016] In a sixth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the distribution network fault detection method provided in the first or second aspect above is implemented.

[0017] The present invention provides a fault detection method, device, electronic device, and computer-readable storage medium for a distribution network. The method utilizes a distribution terminal to perform fault prediction on distribution information, and based on the fault prediction information, the distribution information is labeled and classified. The method enables convenient and effective processing, analysis, and fault prediction of distribution network operating status data on the distribution terminal, thereby reducing the workload of subsequent distribution master stations in processing and analyzing distribution information. The method then sends the labeled and classified distribution information to the distribution master station for fault detection, thereby achieving efficient collaborative fault detection between the distribution terminal and the distribution master station. Therefore, the present invention enables convenient and effective processing, analysis, and fault prediction of distribution network operating status data on the distribution terminal, thereby achieving collaborative fault detection between the distribution terminal and the distribution master station, improving the accuracy and efficiency of fault detection, reducing the occurrence rate of faults, and improving the stability of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of a method for detecting a fault in a distribution network according to embodiment 1 of the present invention;

[0019] Figure 2 This is an example diagram of a power distribution terminal in Embodiment 1 of the present invention;

[0020] Figure 3 This is an example diagram of a power distribution master station in Example 1 of the present invention;

[0021] Figure 4 This is a flow chart of a method for detecting a fault in a distribution network according to embodiment 2 of the present invention;

[0022] Figure 5 This is a flow chart of a method for detecting a fault in a distribution network according to embodiment 3 of the present invention;

[0023] Figure 6 This is a schematic structural diagram of a fault detection device for a distribution network according to Embodiment 4 of the present invention;

[0024] Figure 7 This is a structural diagram of a fault detection device for a distribution network according to embodiment 5 of the present invention. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the technical solutions of the present invention, the embodiments of the present invention will be described in further detail below with reference to the accompanying drawings.

[0026] It should be understood that the specific embodiments and drawings described herein are only used to explain the present invention rather than to limit the present invention.

[0027] It is understood that, in the absence of conflict, the various embodiments of the present invention and the various features in the embodiments may be combined with each other.

[0028] It can be understood that, for the convenience of description, the drawings of the present invention only show parts related to the present invention, while parts unrelated to the present invention are not shown in the drawings.

[0029] It can be understood that each unit and module involved in the embodiments of the present invention may correspond to only one physical structure, or may be composed of multiple physical structures, or multiple units and modules may be integrated into one physical structure.

[0030] It will be understood that, without conflict, the functions and steps marked in the flowcharts and block diagrams of the present invention may occur in an order different from that marked in the drawings.

[0031] It is understood that the flowcharts and block diagrams of the present invention illustrate the possible architectures, functions, and operations of the systems, devices, equipment, and methods according to various embodiments of the present invention. Each box in the flowchart or block diagram may represent a unit, module, program segment, or code, which contains executable instructions for implementing the specified functions. Moreover, each box or combination of boxes in the block diagram and flowchart may be implemented using a hardware-based system that implements the specified functions, or may be implemented using a combination of hardware and computer instructions.

[0032] It can be understood that the units and modules involved in the embodiments of the present invention can be implemented by software or hardware. For example, the units and modules can be located in a processor.

[0033] Example 1:

[0034] like Figure 1 As shown, this embodiment provides a distribution network fault detection method, which is applied to a distribution terminal. The distribution network includes a distribution terminal and a distribution master station. The distribution network fault detection method includes:

[0035] S101, collecting power distribution information, where the power distribution information includes equipment operation information and real-time data of voltage, current and power of connected lines.

[0036] In this embodiment, Figure 2 As shown, the power distribution terminal includes a pre-processing module and an AD (Analog to Digital) conversion module, which collects equipment operation information and real-time data of voltage, current and power of the connected lines. Figure 2 Segment 1 and Segment 2 in the device are Figure 2 The middle section switch, equipment includes but is not limited to: secondary equipment of voltage sensor and current sensor; real-time data of voltage, current and power include but is not limited to: Figure 2Voltage PT (Potential Transformer, voltage transformer) 1, voltage PT2, current CT (current transformer, current transformer) 1 and current CT2.

[0037] S102 , performing fault prediction on the power distribution information based on a binary decision tree fault prediction model to obtain fault prediction information.

[0038] In this embodiment, the power distribution terminal also includes a fault detection algorithm module. Through the fault detection algorithm module, historical power distribution information and power distribution information in an insufficient or missing state, a binary decision tree fault prediction model is constructed and trained, and based on the binary decision tree fault prediction model, fault prediction is performed on the power distribution information to obtain fault prediction information.

[0039] Specifically, the binary decision tree fault prediction model includes formula (1) and formula (2):

[0040]

[0041] Among them, Gini(D) represents the Gini coefficient of the power distribution information with faults, D represents the sample data of the power distribution information with faults, and p k It represents the proportion of sample data with fault k in the power distribution information in D, k = 1, 2, ... K, K represents the total number of fault types, Gini (D, A) represents the Gini index of fault A in the power distribution information, D1 represents the sample data with fault A in the power distribution information, D2 represents the sample data with fault Sample data of the fault represents all faults except fault A, Gini(D1) represents the Gini coefficient of fault A in the power distribution information, and Gini(D2) represents the Gini coefficient of fault A in the power distribution information. The Gini coefficient.

[0042] In this embodiment, the fault types include but are not limited to: equipment (such as transformer, switch) failure, line fault (such as overcurrent, quick disconnection), overload fault, voltage abnormality, power factor abnormality, ground fault (such as zero sequence), and communication failure. Taking K=3 and the fault types including equipment fault, line fault and overload fault as an example, the fault prediction of the power distribution information is performed based on the binary decision tree fault prediction model, specifically including: according to formula (1), Gini(D1) and Gini(D2) are calculated respectively, wherein Gini(D1) includes the Gini coefficient of the power distribution information with equipment fault, the Gini coefficient of the power distribution information with line fault and the Gini coefficient of the power distribution information with overload fault, and Gini(D2) includes the Gini coefficient of the power distribution information with line fault or overload fault, the Gini coefficient of the power distribution information with equipment fault or overload fault and the Gini coefficient of the power distribution information with line fault or equipment fault; according to formula (2), the Gini index of the power distribution information with equipment fault, the Gini index of the power distribution information with line fault and the Gini index of the power distribution information with overload fault are calculated; the fault type corresponding to the smallest Gini index is taken to generate fault prediction information. This embodiment uses a binary decision tree fault prediction model to analyze the changing trend of power distribution information, and then perform predictive evaluation and reporting on the operating status of the power distribution information; by calculating the Gini coefficient and Gini index of various types of faults in the power distribution information, and selecting the Gini coefficient and Gini index of various types of faults in the power distribution information as the basis for prediction, the accuracy of the prediction results can be improved, misjudgment can be reduced, and accurate and effective fault prediction can be achieved.

[0043] It should be noted that line faults such as overcurrent, quick trip, and no pressure on the load side can be addressed locally by driving line protection measures (opening and closing the circuit breaker), improving the efficiency and timeliness of fault detection and resolution. Furthermore, fault prediction information can also serve as an alarm to facilitate early equipment maintenance.

[0044] Equipment failures include, but are not limited to, equipment failures and component aging. Equipment failures include insulation failures and mechanical wear. Component aging refers to the aging and failure of equipment components due to long-term operation. Line failures include, but are not limited to, short circuits and open circuits. A short circuit occurs between lines or between a line and the ground, potentially causing equipment damage or power outages. An open circuit occurs when a line is partially or completely disconnected, preventing normal power transmission. Overload failures include, but are not limited to, equipment overload and line overload. Equipment overload occurs when excessive load exceeds the design tolerance of the equipment, potentially causing overheating or burning. Line overload occurs when excessive current flows through the line, potentially causing overheating and damage. Voltage anomalies include, but are not limited to, overvoltage and undervoltage. Overvoltage occurs when the grid voltage exceeds the rated voltage of the equipment, causing damage. Undervoltage occurs when the voltage falls below the normal operating voltage of the equipment, potentially preventing normal operation. Power factor anomalies include, but are not limited to, low power factor. Low power factor may be caused by load imbalance or equipment problems, impacting the efficiency and stability of the power system. Grounding faults include, but are not limited to, ground faults. A ground fault refers to a failure of ground protection due to insulation failure or equipment aging, posing a threat to the safe operation of the entire system. Communication failures include, but are not limited to, data loss or delays. These problems occur when data is transmitted between the power distribution device and the main control system, potentially impacting monitoring and control capabilities.

[0045] S103, based on the fault prediction information, classify and mark the power distribution information to obtain the power distribution marking information, and send the power distribution marking information to the power distribution master station so that the power distribution master station performs fault detection on the power distribution marking information to obtain the fault detection result.

[0046] In this embodiment, the power distribution terminal also includes an MCU (Micro Controller Unit) / MPU (Micro Processor Unit) computing control module and a simulation data cache module. The data pre-processing "demand distribution" template for digital simulation issued by the power distribution master station is received through the wireless communication module. The data pre-processing "demand distribution" template is called up through the MCU / MPU computing control module, and based on the data pre-processing "demand distribution" template and fault prediction information, the power distribution information is classified and marked to obtain the power distribution labeling information, thereby improving the efficiency of the subsequent construction of the digital twin of the regional distribution network. The power distribution labeling information is cached through the simulation data cache module. This embodiment reduces the workload of the subsequent power distribution master station in processing and analyzing the power distribution information by classifying and marking the power distribution information on the power distribution terminal, and then sends the marked and classified distribution information to the power distribution master station for digital twin and fault detection. Compared with traditional digital twin modeling, the efficiency is improved by 15% and it is easy to expand and maintain.

[0047] Specifically, the power distribution marking information is sent to the power distribution master station, including steps S1031 and S1032:

[0048] S1031: Encrypt the power distribution labeling information based on an encryption algorithm, where the encryption algorithm includes at least one of the following: a commercial secret SM1 algorithm and a commercial secret SM2 algorithm.

[0049] S1032: Send the encrypted power distribution marking information to the power distribution master station.

[0050] In this embodiment, the power distribution terminal also includes a security encryption module and a wireless communication module. Through these modules, the power distribution annotation information is encrypted using the SM1 and SM2 algorithms and then transmitted back to the power distribution master station, completing the model data input of the digital twin in the power distribution master station.

[0051] like Figure 3 As shown in the figure, the power distribution master station includes the application interaction layer, digital twin, and data management layer, among which the power distribution terminal is Figure 3 The intelligent FTU (Feeder Terminal Unit) terminal for digital simulation in the power distribution is encrypted by SM1 and SM2 algorithms and then transmitted back to the power distribution master station (i.e. Figure 3 After the real-time perception in the network), the distribution master station stores and manages the distribution annotation information through the data management layer, receives the data requirements of the digital twin, generates and issues the data preprocessing "demand distribution" template for digital simulation; calls the distribution annotation information from the data management layer through the digital twin, performs digital twin on the distribution annotation information, obtains the distribution status data (i.e., digital twin), and performs fault detection on the distribution status data to obtain the fault detection results; generates distribution network early warning information through the application interaction layer and fault detection results to carry out timely maintenance and complete the regional distribution network management and maintenance.

[0052] The distribution master station stores and manages the distribution annotation information, specifically including: using a time series database to store the distribution annotation information; and performing data classification management on the distribution annotation information to provide more convenient data services for the subsequent rapid construction of digital twins.

[0053] The distribution master station creates a digital twin of the distribution annotation information. Specifically, this includes the dynamic construction, parameter adjustment, and verification and evaluation of the digital twin model based on historical distribution information, historical insufficient or missing distribution information, historical fault event records (SOE) data, and historical distribution annotation information. Based on the digital twin model, a digital twin of the regional distribution network corresponding to the distribution annotation information is established to achieve virtual simulation of the distribution network line environment and assist in fault identification, analysis, and early warning.

[0054] Optionally, in S103: based on the fault prediction information, classifying and marking the power distribution information to obtain power distribution labeling information, and sending the power distribution labeling information to the power distribution master station so that the power distribution master station performs fault detection on the power distribution labeling information. After obtaining the fault detection result, the fault detection method for the distribution network further includes:

[0055] S104: Obtain fault detection results, where the fault detection results are obtained by the distribution master station performing fault detection on the distribution labeling information. S105: Optimize the binary decision tree fault prediction model based on the fault detection results, fault prediction information, and pruning strategies. In this embodiment, the binary decision tree fault prediction model is optimized based on the fault detection results, fault prediction information, and pruning strategies. Specifically, the optimization includes: integrating the fault detection results, fault prediction information, and corresponding distribution information to form a more comprehensive training dataset. Based on the more comprehensive training dataset and pruning strategies, the model parameters are adjusted and updated in real time. Prediction information of different fault types is observed, and resampling or weighting is performed to address the imbalance of fault types, thereby preventing the model from being biased towards a certain type of fault, maintaining the model's effectiveness, and ensuring that it always adapts to the latest fault patterns and environmental changes. Pruning strategies include, but are not limited to, pre-pruning and post-pruning. In this embodiment, the binary decision tree fault prediction model is optimized through fault detection results, fault prediction information, and pruning strategies, improving the stability and generalization capability of the binary decision tree fault prediction model.

[0056] Based on a more comprehensive training data set and pruning strategy, the model parameters are adjusted and updated in real time, including: ① According to the formula Calculate the pruning coefficient α, where C(t) represents the error of the t node of the binary decision tree fault prediction model after pruning, C(T i ) The error of the binary decision tree fault prediction model before pruning i times, i = 0, 1, 2,,…, N, N represents the number of pruning times; ② Select the node with the smallest pruning coefficient α for pruning, and obtain the binary decision tree fault prediction model T i+1 ; ③ Repeat ① and ② until only the root node remains in the binary decision tree fault prediction model; ④ Independently verify the binary decision tree fault prediction model after each pruning, calculate the error of the binary decision tree fault prediction model after each pruning, and select the binary decision tree fault prediction model with the smallest error as the optimal binary decision tree fault prediction model.

[0057] It should be noted that pre-pruning refers to setting a threshold during the decision tree construction process. When the Gini coefficient falls below the threshold, node splitting is stopped. This prevents the model from becoming too complex and leading to overfitting. Post-pruning involves first building a complete decision tree and then pruning the tree through cross-validation or setting a low violation rate threshold. This removes leaf nodes that have little predictive effect, helping the model generalize better to new data.

[0058] This embodiment provides a fault detection method for a distribution network. The method uses a distribution terminal to perform fault prediction on distribution information, and marks and classifies the distribution information based on the fault prediction information, so as to realize convenient and effective processing, analysis and fault prediction of distribution network operation status data on the distribution terminal, thereby reducing the workload of subsequent distribution master stations in processing and analyzing distribution information, and then sending the marked and classified distribution information to the distribution master station for fault detection, thereby realizing efficient collaborative fault detection of the distribution terminal and the distribution master station, realizing convenient and effective processing, analysis and fault prediction of distribution network operation status data on the distribution terminal, thereby realizing collaborative fault detection of the distribution terminal and the distribution master station, improving the accuracy and efficiency of fault detection, reducing the occurrence rate of faults, and improving the stability of the distribution network.

[0059] Example 2:

[0060] like Figure 4 As shown, this embodiment provides a distribution network fault detection method, which is applied to a distribution master station. The distribution network fault detection method includes:

[0061] S201, receiving distribution labeling information sent by the distribution terminal, wherein the distribution labeling information is obtained by collecting distribution information by the distribution terminal, performing fault prediction on the distribution information based on a binary decision tree fault prediction model to obtain fault prediction information, and classifying and labeling the distribution information based on the fault prediction information.

[0062] In this embodiment, Figure 2 As shown, the power distribution terminal includes a pre-processing module and an AD conversion module, which collect equipment operation information and real-time data of voltage, current and power of the connected lines. Figure 2 Segment 1 and Segment 2 in the device are Figure 2 The middle section switch, equipment includes but is not limited to: secondary equipment of voltage sensor and current sensor; real-time data of voltage, current and power include but is not limited to: Figure 2 Voltage PT1, voltage PT2, current CT1 and current CT2 in.

[0063] The power distribution terminal also includes a fault detection algorithm module. Through the fault detection algorithm module, historical power distribution information and power distribution information in an insufficient or missing state, a binary decision tree fault prediction model is constructed and trained. Based on the binary decision tree fault prediction model, fault prediction is performed on the power distribution information to obtain fault prediction information.

[0064] Fault types include but are not limited to: equipment (such as transformer, switch) failure, line failure, overload failure, voltage abnormality, power factor abnormality, ground fault, and communication failure. Taking K=3 and the fault types including equipment fault, line fault and overload fault as an example, the distribution terminal performs fault prediction on the distribution information based on the binary decision tree fault prediction model, specifically including: according to formula (1), respectively calculating Gini(D1) and Gini(D2), wherein Gini(D1) includes the Gini coefficient of the distribution information with equipment fault, the Gini coefficient of the distribution information with line fault and the Gini coefficient of the distribution information with overload fault, and Gini(D2) includes the Gini coefficient of the distribution information with line fault or overload fault, the Gini coefficient of the distribution information with equipment fault or overload fault and the Gini coefficient of the distribution information with line fault or equipment fault; according to formula (2), calculating the Gini index of the distribution information with equipment fault, the Gini index of the distribution information with line fault and the Gini index of the distribution information with overload fault; taking the fault type corresponding to the smallest Gini index to generate fault prediction information.

[0065] The distribution terminal also includes an MCU / MPU computing control module and a simulation data cache module. The wireless communication module receives the "demand distribution" data preprocessing template for digital simulation from the distribution master station. The MCU / MPU computing control module retrieves this data preprocessing template and, based on this template and fault prediction information, classifies and labels distribution information to generate distribution labeling information, improving the efficiency of subsequent construction of a regional distribution network digital twin. The simulation data cache module caches this distribution labeling information.

[0066] The distribution terminal also includes a security encryption module and a wireless communication module. Through these modules, distribution marking information is encrypted using the SM1 and SM2 algorithms and then transmitted back to the distribution master station, completing the model data input of the digital twin in the distribution master station.

[0067] S202: Perform fault detection on the power distribution marking information to obtain a fault detection result.

[0068] Specifically, S202: performing fault detection on the power distribution marking information to obtain a fault detection result, including steps S2021 and S2022:

[0069] S2021, perform digital twinning on the power distribution marking information to obtain power distribution status data.

[0070] S2022: Perform fault detection on the power distribution status data to obtain a fault detection result.

[0071] In this embodiment, Figure 3 As shown in the figure, the power distribution master station includes the application interaction layer, digital twin, and data management layer, among which the power distribution terminal is Figure 3 The intelligent FTU terminal for digital simulation in the power distribution is encrypted by SM1 and SM2 algorithms and then transmitted back to the power distribution master station (i.e. Figure 3 After the real-time perception in the network), the distribution master station stores and manages the distribution annotation information through the data management layer, receives the data requirements of the digital twin, generates and issues the data preprocessing "demand distribution" template for digital simulation; calls the distribution annotation information from the data management layer through the digital twin, performs digital twin on the distribution annotation information, obtains the distribution status data (i.e., digital twin), and performs fault detection on the distribution status data to obtain the fault detection results; generates distribution network early warning information through the application interaction layer and fault detection results to carry out timely maintenance and complete the regional distribution network management and maintenance.

[0072] The distribution master station stores and manages the distribution annotation information, specifically including: using a time series database to store the distribution annotation information; and performing data classification management on the distribution annotation information to provide more convenient data services for the subsequent rapid construction of digital twins.

[0073] The distribution master station creates a digital twin of the distribution annotation information. This includes dynamic construction, parameter adjustment, and verification and evaluation of the digital twin model based on historical distribution information, historical insufficient or missing distribution information, historical fault event records (SOE) data, and historical distribution annotation information. Based on the digital twin model, a digital twin of the regional distribution network corresponding to the distribution annotation information is established to achieve virtual simulation of the distribution network line environment and assist in fault identification, analysis, and early warning.

[0074] This embodiment provides a fault detection method for a distribution network. The method uses a distribution terminal to perform fault prediction on distribution information, and marks and classifies the distribution information based on the fault prediction information, so as to realize convenient and effective processing, analysis and fault prediction of distribution network operation status data on the distribution terminal, thereby reducing the workload of subsequent distribution master stations in processing and analyzing distribution information, and then sending the marked and classified distribution information to the distribution master station for fault detection, thereby realizing efficient collaborative fault detection of the distribution terminal and the distribution master station, realizing convenient and effective processing, analysis and fault prediction of distribution network operation status data on the distribution terminal, thereby realizing collaborative fault detection of the distribution terminal and the distribution master station, improving the accuracy and efficiency of fault detection, reducing the occurrence rate of faults, and improving the stability of the distribution network.

[0075] Example 3:

[0076] like Figure 5 As shown, this embodiment also provides a distribution network fault detection method, including:

[0077] S11. The power distribution terminal collects power distribution information, wherein the power distribution information includes equipment operation information and real-time data of voltage, current and power of the connected lines.

[0078] In this embodiment, the power distribution terminal is Figure 5 Intelligent power distribution terminal in.

[0079] S12. The power distribution terminal performs fault prediction on the power distribution information based on a binary decision tree fault prediction model to obtain fault prediction information, and classifies and labels the power distribution information based on the fault prediction information to obtain power distribution labeling information.

[0080] In this embodiment, the binary decision tree fault prediction model is Figure 5 Cart classifier in .

[0081] S13. The power distribution terminal encrypts the power distribution labeling information based on an encryption algorithm, and sends the encrypted power distribution labeling information to the power distribution master station, wherein the encryption algorithm includes at least one of the following: a commercial secret SM1 algorithm and a commercial secret SM2 algorithm.

[0082] In this embodiment, the power distribution master station is Figure 5 The master station system in.

[0083] S14. The distribution master station receives the distribution labeling information sent by the distribution terminal, wherein the distribution labeling information is collected by the distribution terminal, and the distribution information is subjected to fault prediction based on a binary decision tree fault prediction model to obtain fault prediction information, and the distribution information is classified and labeled based on the fault prediction information.

[0084] In this embodiment, the power distribution master station is Figure 5In the master station digital twin system, the distribution marking information is Figure 5 Terminal assembly data in .

[0085] S15. The power distribution master station performs digital twinning on the power distribution marking information to obtain the power distribution status data.

[0086] In this embodiment, the power distribution marking information is Figure 5 The distribution network operation status in the digital twin is Figure 5 Real-time simulation in .

[0087] S16. The power distribution master station performs fault detection on the power distribution status data to obtain a fault detection result.

[0088] In this embodiment, the fault detection result is Figure 5 Prediction results of the distribution network digital twin system in .

[0089] This embodiment provides a fault detection method for a distribution network. The method uses a distribution terminal to perform fault prediction on distribution information, and marks and classifies the distribution information based on the fault prediction information, so as to realize convenient and effective processing, analysis and fault prediction of distribution network operation status data on the distribution terminal, thereby reducing the workload of subsequent distribution master stations in processing and analyzing distribution information, and then sending the marked and classified distribution information to the distribution master station for fault detection, thereby realizing efficient collaborative fault detection of the distribution terminal and the distribution master station, realizing convenient and effective processing, analysis and fault prediction of distribution network operation status data on the distribution terminal, thereby realizing collaborative fault detection of the distribution terminal and the distribution master station, improving the accuracy and efficiency of fault detection, reducing the occurrence rate of faults, and improving the stability of the distribution network.

[0090] Example 4:

[0091] like Figure 6 As shown, this embodiment also provides a fault detection device for a distribution network, which is applied to a distribution terminal, and includes an acquisition module 41, a fault prediction module 42 and a sending module 43. The acquisition module 41 is used to collect distribution information, wherein the distribution information includes equipment operation information and real-time data of voltage, current and power of the connected lines. The fault prediction module 42 is connected to the acquisition module 41 and is used to perform fault prediction on the distribution information based on a binary decision tree fault prediction model to obtain fault prediction information. The sending module 43 is connected to the fault prediction module 42 and the acquisition module 41 respectively and is used to generate distribution labeling information based on the fault prediction information and the distribution information, and send the distribution labeling information to the distribution master station so that the distribution master station performs fault detection on the distribution labeling information to obtain a fault detection result.

[0092] Specifically, the sending module 43 includes: an encryption unit 431 and a sending unit 432, the encryption unit 431 is used to encrypt the distribution labeling information based on an encryption algorithm, wherein the encryption algorithm includes at least one of the following: a commercial secret SM1 algorithm, a commercial secret SM2 algorithm, and the sending unit 432 is used to send the encrypted distribution labeling information to the distribution master station.

[0093] Optionally, the power distribution terminal further includes: an acquisition module 44 and an optimization module 45, the acquisition module 44 is used to obtain the fault detection result, wherein the fault detection result is obtained by the power distribution master station performing fault detection on the power distribution labeling information, and the optimization module 45 is used to optimize the binary decision tree fault prediction model based on the fault detection result, fault prediction information and pruning strategy.

[0094] It can be understood that the above-mentioned distribution network fault detection device executes the distribution network fault detection method corresponding to the embodiment 1 provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects of the scheme corresponding to the distribution network fault detection method of the embodiment 1 above, and will not be repeated here.

[0095] Example 5:

[0096] like Figure 7 As shown, this embodiment also provides a fault detection device for a distribution network, which is applied to a distribution master station and includes a receiving module 51 and a fault detection module 52. The receiving module 51 is used to receive distribution labeling information sent by a distribution terminal, wherein the distribution labeling information is collected by the distribution terminal, and the distribution information is predicted based on a binary decision tree fault prediction model to obtain fault prediction information, and the distribution information is classified and marked based on the fault prediction information. The fault detection module 52 is connected to the receiving module 51 and is used to perform fault detection on the distribution labeling information to obtain a fault detection result.

[0097] Specifically, the fault detection module 52 includes: a digital twin unit 521 and a fault detection unit 522. The digital twin unit 521 is used to perform digital twinning on the distribution labeling information to obtain distribution status data. The fault detection unit 522 is used to perform fault detection on the distribution status data to obtain a fault detection result.

[0098] It can be understood that the above-mentioned distribution network fault detection device executes the distribution network fault detection method corresponding to the embodiment 2 provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects of the scheme corresponding to the distribution network fault detection method of the embodiment 2 above, and will not be repeated here.

[0099] Example 6:

[0100] This embodiment also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to implement the distribution network fault detection method in the above-mentioned embodiment 1, embodiment 2, or embodiment 3.

[0101] Example 7:

[0102] This embodiment further provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method for detecting faults in the distribution network in the above-mentioned embodiment 1, embodiment 2, or embodiment 3 is implemented.

[0103] It will be understood that the above embodiments are merely exemplary embodiments for illustrating the principles of the present invention, and the present invention is not limited thereto. Those skilled in the art will appreciate that various modifications and improvements can be made without departing from the spirit and substance of the present invention, and such modifications and improvements are also considered to be within the scope of protection of the present invention.

Claims

1. A fault detection method for a distribution network, applied to a distribution terminal, wherein the distribution network includes a distribution terminal and a distribution master station, characterized in that: Fault detection methods for distribution networks include: Collect power distribution information, including equipment operation information and real-time data on voltage, current, and power of connected lines; Based on the binary decision tree fault prediction model, the power distribution information is predicted to obtain fault prediction information; Based on the fault prediction information, the power distribution information is classified and marked to obtain the power distribution marking information, and the power distribution marking information is sent to the power distribution master station so that the power distribution master station performs fault detection on the power distribution marking information to obtain a fault detection result.

2. The method for detecting a fault in a distribution network according to claim 1, wherein: The binary decision tree fault prediction model includes formula (1) and formula (2): Wherein, Gini(D) represents the Gini coefficient of the power distribution information with faults, D represents the sample data with faults in the power distribution information, and p k represents the proportion of sample data with fault k in the power distribution information in D, k = 1, 2, ... K, K represents the total number of fault types, Gini (D, A) represents the Gini index of fault A in the power distribution information, D1 represents the sample data with fault A in the power distribution information, D2 represents the sample data with fault Sample data of the fault represents all faults except fault A, Gini(D1) represents the Gini coefficient of fault A in the power distribution information, and Gini(D2) represents the Gini coefficient of fault A in the power distribution information. The Gini coefficient.

3. The method for detecting a fault in a distribution network according to claim 1, wherein: Sending the power distribution marking information to the power distribution master station specifically includes: Encrypting the power distribution labeling information based on an encryption algorithm, wherein the encryption algorithm includes at least one of the following: a commercial secret SM1 algorithm and a commercial secret SM2 algorithm; The encrypted power distribution marking information is sent to the power distribution master station.

4. The method for detecting a fault in a distribution network according to claim 1, wherein: After classifying and marking the power distribution information based on the fault prediction information to obtain the power distribution labeling information, and sending the power distribution labeling information to the power distribution master station so that the power distribution master station performs fault detection on the power distribution labeling information and obtains the fault detection result, the method further includes: Obtaining a fault detection result, wherein the fault detection result is obtained by the power distribution master station performing fault detection on the power distribution marking information; Based on fault detection results, fault prediction information and pruning strategy, the binary decision tree fault prediction model is optimized.

5. A method for fault detection of a distribution network, applied to a distribution master station, characterized in that: Fault detection methods for distribution networks include: receiving power distribution labeling information sent by a power distribution terminal, wherein the power distribution labeling information is obtained by collecting power distribution information from the power distribution terminal, performing fault prediction on the power distribution information based on a binary decision tree fault prediction model to obtain fault prediction information, and classifying and labeling the power distribution information based on the fault prediction information; Fault detection is performed on the power distribution marking information to obtain a fault detection result.

6. The method for detecting a fault in a distribution network according to claim 5, wherein: Performing fault detection on the power distribution marking information to obtain a fault detection result specifically includes: Performing digital twinning on the power distribution annotation information to obtain power distribution status data; Perform fault detection on the power distribution status data to obtain a fault detection result.

7. A fault detection device for a distribution network, applied to a distribution terminal, characterized in that: Including acquisition module, fault prediction module and sending module, The acquisition module is used to collect power distribution information, including equipment operation information and real-time data of voltage, current and power of connected lines. The fault prediction module is connected to the acquisition module and is used to predict the fault of the power distribution information based on the binary decision tree fault prediction model to obtain fault prediction information. The sending module is connected to the fault prediction module and the acquisition module respectively, and is used to generate distribution marking information based on the fault prediction information and the distribution information, and send the distribution marking information to the distribution master station so that the distribution master station can perform fault detection on the distribution marking information and obtain the fault detection result.

8. A fault detection device for a distribution network, applied to a main distribution station, characterized in that: Including receiving module and fault detection module, A receiving module is used to receive the distribution labeling information sent by the distribution terminal, wherein the distribution labeling information is collected by the distribution terminal, and the distribution information is subjected to fault prediction based on a binary decision tree fault prediction model to obtain fault prediction information, and the distribution information is classified and labeled based on the fault prediction information. The fault detection module is connected to the receiving module and is used to perform fault detection on the power distribution marking information to obtain a fault detection result.

9. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to implement a fault detection method for a distribution network as claimed in any one of claims 1 to 4 or claims 5 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, a fault detection method for a distribution network as claimed in any one of claims 1 to 4 or claims 5 to 6 is implemented.

Images