Distribution network unmanned aerial vehicle inspection intelligent management and control system based on multi-source data fusion

Abstract

The invention provides a distribution network unmanned aerial vehicle inspection intelligent management and control system based on multi-source data fusion. The system comprises a data acquisition module, a fusion processing module, an analysis decision module, a scheduling management and control module and a user management module. The data acquisition module is used for acquiring state information of the power distribution network in real time through various sensors carried by the unmanned aerial vehicle; the fusion processing module is used for performing standardization processing and fusion on the information data acquired by the data acquisition module; the analysis decision module analyzes the fault risk of each part of the distribution network line based on the data output by the fusion processing module and optimizes the subsequent inspection route of the unmanned aerial vehicle; the scheduling management and control module is used for managing and controlling the execution process of the inspection task of the unmanned aerial vehicle; the user management module is used for providing a user interface to complete interaction between a user and the system; according to the invention, through multi-source data fusion and intelligent analysis optimization, accurate sensing of distribution network line faults and efficient dynamic management of unmanned aerial vehicle inspection tasks are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicle (UAV) inspection systems, and particularly to an intelligent control system for UAV inspection of distribution networks based on multi-source data fusion. Background Art

[0002] With the rapid development of smart grids and UAV technologies, the inspection of power distribution network lines is gradually shifting from traditional manual inspection to automation and intelligence; distribution network lines are widely distributed and the environment is complex. The traditional manual inspection method has low efficiency, high cost, and there are significant safety hazards under harsh weather or complex terrain conditions; UAV inspection can quickly cover a large area with its flexible and mobile advantages, enabling efficient inspection of lines; however, relying solely on simple image acquisition or single-sensor data is difficult to comprehensively perceive the equipment status, resulting in inaccurate inspection results.

[0003] Referring to relevant publicly disclosed technical solutions, the technology with the publication number CN117519220A proposes a method and system for intelligent UAV inspection of distribution network lines. The method includes the following steps: S1: Conduct laser point cloud scanning on the distribution network lines and establish a three-dimensional point cloud model of the distribution network; S2: Based on the three-dimensional point cloud model, plan and simulate the flight route of the distribution network on a three-dimensional platform, and export or send the planned waypoint file to the UAV flight control system; S3: Conduct autonomous inspection of the distribution network UAV according to the route planned and simulated in S2. During the inspection process, the UAV automatically transmits the inspection data to the vehicle-mounted server and synchronizes it to the central end; S4: Automatically identify the defects and potential hazards in the distribution network during the UAV inspection process; this solution uses high-precision UAV positioning inspection technology and combines three-dimensional map information display technology to achieve autonomous UAV inspection and intelligent defect identification, promoting the process of refined management of the distribution network; however, this solution lacks multi-source data fusion analysis of the distribution network line status and cannot comprehensively perceive the operating status of the equipment; and it does not dynamically optimize the inspection tasks and cannot achieve efficient allocation and real-time adjustment of inspection resources. Summary of the Invention

[0004] The purpose of the present invention is to propose an intelligent control system for UAV inspection of distribution networks based on multi-source data fusion to address the current deficiencies.

[0005] The present invention adopts the following technical solutions:

[0006] A smart management and control system for distribution network UAV inspection based on multi-source data fusion. The system includes a data acquisition module, a fusion processing module, an analysis and decision-making module, a scheduling and management module, and a user management module. The data acquisition module is used to obtain the status information of the distribution network in real time through various sensors carried by the UAV. The fusion processing module is used to perform standardization processing and fusion on the information data obtained by the data acquisition module. The analysis and decision-making module analyzes the fault risks of various parts of the distribution network line based on the data output by the fusion processing module and optimizes the subsequent inspection route of the UAV. The scheduling and management module is used to manage and control the execution process of the UAV inspection task. The user management module is used to provide a user interface to complete the interaction between the user and the system.

[0007] The data acquisition module includes a multi-source sensor unit and an external data interface unit. The multi-source sensor unit is used to obtain the status information of the distribution network in real time through various sensors carried by the UAV during the UAV inspection process. The status information of the distribution network includes optical data and infrared imaging data of various parts of the distribution network line. The external data interface unit is used to complete the information transmission between the status information of the distribution network and other modules.

[0008] The fusion processing module includes a data preprocessing unit and a multi-source data fusion unit. The data preprocessing unit is used to perform standardization processing on the status information of the distribution network. The content of the standardization processing includes noise filtering, data calibration, missing value processing, and format conversion to ensure the consistency and reliability of data from different sensors. The multi-source data fusion unit is used to integrate the standardized data to generate a unified and highly accurate status information of the distribution network in time and space.

[0009] Further, the analysis and decision-making module includes a fault diagnosis unit and an inspection optimization unit. The fault diagnosis unit is used to analyze and diagnose the status information of the distribution network to obtain the fault risk situation of various parts of the distribution network line. The inspection optimization unit is used to optimize the subsequent inspection route of the UAV according to the diagnosis result of the fault diagnosis unit.

[0010] Further, the fault diagnosis unit includes a data entry sub-unit, a data storage sub-unit, a model establishment sub-unit, and a fault detection sub-unit. The data entry sub-unit is used to enter the status information of the distribution network obtained from the multi-source data fusion unit. The data storage sub-unit is used to store the status information of the distribution network with time stamps. The model establishment sub-unit is used to establish a distribution network status evaluation model, which is used to evaluate the fault risks of various parts of the distribution network line. The fault detection sub-unit is used to input the currently entered real-time status information of the distribution network into the distribution network status evaluation model to evaluate the fault risks of the distribution network equipment during the current UAV inspection process.

[0011] Further, the model establishment subunit completes the establishment of the distribution network status evaluation model in the following manner:

[0012] S11: Obtain historical distribution network status information; the historical distribution network status information includes optical information and infrared imaging information of various parts of the distribution network line; the distribution network status information is obtained through the historical inspection records of drones;

[0013] S12: Extract key features for evaluating the fault risk of the distribution network line from the historical distribution network status information; the key features include physical state features extracted from the optical information and temperature distribution features extracted from the infrared imaging information;

[0014] S13: Establish a distribution network status evaluation model using the key features obtained in the previous step; for a certain part of the distribution network line, the expression form of the distribution network status evaluation model is as follows:

[0015]

[0016] where F(X) is the distribution network health evaluation coefficient; X is the set of all key features corresponding to this part of the distribution network line; n is the number of key features in the key feature set; x i (t) is the specific value of the i-th key feature input into the distribution network status evaluation model at time t when the model performs the evaluation; f(x i (t)) is the normal range distribution probability density of the i-th key feature input into the distribution network status evaluation model at time t; for f(x i (t)) satisfies:

[0017]

[0018] where, represents the fluctuation variance of the i-th key feature, reflecting the fluctuation range of this key feature during the normal operation of the equipment, and is obtained by statistically analyzing the historical key feature data during normal operation in advance; u i (t) is the dynamic mean of the i-th key feature at time t, and its meaning is the typical value of this key feature during the recent normal operation;

[0019] Further, the specific working process of the fault detection subunit is as follows:

[0020] S21: Obtain real-time distribution network status information;

[0021] S22: Extract key features for evaluating the fault risk of the distribution network line from the real-time distribution network status information;

[0022] S23: Input the key features obtained in the previous step into the distribution network status evaluation model to obtain the distribution network health evaluation coefficients corresponding to each part of the distribution network line;

[0023] S24: Compare each distribution network health evaluation coefficient with a preset health threshold. If the distribution network health evaluation coefficient is lower than the health threshold, it indicates that there is a fault risk in the corresponding part of the distribution network line; if the distribution network health evaluation coefficient is higher than the health threshold, it indicates that the corresponding part of the distribution network line is in a normal state;

[0024] Further, the inspection optimization unit includes a historical risk acquisition subunit, a priority ranking subunit, and an inspection route optimization subunit; the historical risk acquisition subunit is used to acquire and store the historical fault risk information of each part of the distribution network line; the priority ranking subunit is used to determine the priority of each part of the subsequent inspection of the distribution network line by the unmanned aerial vehicle, and the inspection route optimization subunit is used to optimize the subsequent inspection route of the unmanned aerial vehicle in combination with the priority of each part of the distribution network line.

[0025] The beneficial effects achieved by the present invention are:

[0026] The present invention comprehensively acquires the optical data and infrared imaging data of the distribution network line through multi-source data fusion, improving the perception accuracy and diagnostic ability of the equipment status; by establishing a distribution network status evaluation model, quantitatively evaluating the fault risks of each part of the distribution network line, ensuring that the evaluation results can timely reflect the recent operation status of the equipment, reducing the influence of historical data deviation, and thus accurately identifying potential fault risks; by determining the inspection priorities of each part of the distribution network line and optimizing the subsequent inspection route of the unmanned aerial vehicle according to the inspection priorities, the inspection time and resource consumption are reduced, the inspection efficiency is improved, and the safe and stable operation of the distribution network system is ensured. Description of the Drawings

[0027] The present invention can be further understood from the following description in conjunction with the drawings. The components in the drawings are not necessarily drawn to scale, but the focus is on showing the principles of the embodiments. In different views, the same reference numerals designate corresponding parts.

[0028] Figure 1 It is a schematic diagram of the overall module of the present invention.

[0029] Figure 2 It is a schematic diagram of the establishment process of the distribution network status evaluation model of the present invention.

[0030] Figure 3 It is a schematic diagram of the working process of the fault detection subunit of the present invention. Detailed Embodiments

[0031] In order to make the objectives, technical solutions and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below in conjunction with its embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention; for those skilled in the art, after referring to the following detailed description, other systems, methods and / or features of this embodiment will become obvious; it is intended that all such additional systems, methods, features and advantages be included in this specification; be included within the scope of the present invention and be protected by the appended claims; additional features of the disclosed embodiments are described in the following detailed description and will be obvious in accordance with the following detailed description.

[0032] In the drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components; in the description of the present invention, it should be understood that if there are terms such as "upper", "lower", "left", "right", etc. indicating the orientation or positional relationship, they are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or component referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, the terms describing the positional relationship in the drawings are only for illustrative purposes and cannot be construed as a limitation of this patent. For those of ordinary skill in the art, the specific meanings of the above terms can be understood according to specific circumstances.

[0033] Embodiment 1:

[0034] As Figure 1 shown, this embodiment provides an intelligent control system for distribution network UAV inspection based on multi-source data fusion. The system includes a data acquisition module, a fusion processing module, an analysis and decision-making module, a scheduling and control module, and a user management module; the data acquisition module is used to obtain the distribution network status information in real time through a variety of sensors carried by the UAV; the fusion processing module is used to perform standardization processing and fusion on the information data obtained by the data acquisition module; the analysis and decision-making module analyzes the fault risks of various parts of the distribution network line based on the data output by the fusion processing module and optimizes the subsequent inspection route of the UAV; the scheduling and control module is used to manage and control the execution process of the UAV inspection task; the user management module is used to provide a user interface to complete the interaction between the user and the system;

[0035] The data acquisition module includes a multi-source sensor unit and an external data interface unit; the multi-source sensor unit is used to obtain the distribution network status information in real time through a variety of sensors carried by the UAV during the UAV inspection process. The distribution network status information includes optical data and infrared imaging data of various parts of the distribution network line; the external data interface unit is used to complete the information transmission between the distribution network status information and other modules;

[0036] The fusion processing module includes a data preprocessing unit and a multi-source data fusion unit; the data preprocessing unit is used to perform standardization processing on the distribution network state information, and the standardization processing content includes noise filtering, data calibration, missing value processing, and format conversion to ensure the consistency and reliability of different sensor data; the multi-source data fusion unit is used to integrate the data after standardization processing to generate distribution network state information that is unified and highly accurate in time and space;

[0037] The analysis and decision-making module includes a fault diagnosis unit and an inspection optimization unit; the fault diagnosis unit is used to analyze and diagnose the distribution network state information to obtain the fault risk situation of each part of the distribution network line; the inspection optimization unit is used to optimize the subsequent inspection route of the UAV according to the diagnosis result of the fault diagnosis unit;

[0038] Further, the fault diagnosis unit includes a data entry sub-unit, a data storage sub-unit, a model establishment sub-unit, and a fault detection sub-unit; the data entry sub-unit is used to enter the distribution network state information obtained from the multi-source data fusion unit; the data storage sub-unit is used to store the distribution network state information with time stamps; the model establishment sub-unit is used to establish a distribution network state evaluation model, and the distribution network state evaluation model is used to evaluate the fault risk of each part of the distribution network line; the fault detection sub-unit is used to input the currently entered real-time distribution network state information into the distribution network state evaluation model to evaluate the fault risk of the distribution network equipment during the current UAV inspection process;

[0039] Further, as Figure 2 shown, the model establishment sub-unit completes the establishment of the distribution network state evaluation model in the following manner:

[0040] S11: Obtain historical distribution network state information; the historical distribution network state information includes optical information and infrared imaging information of each part of the distribution network line; the distribution network state information is obtained through the historical inspection records of the UAV;

[0041] S12: Extract key features for evaluating the fault risk of the distribution network line from the historical distribution network state information; the key features include physical state features extracted from the optical information and temperature distribution features extracted from the infrared imaging information;

[0042] S13: Use the key features obtained in the previous step to establish a distribution network state evaluation model; for a certain part of the distribution network line, the expression form of the distribution network state evaluation model is as follows:

[0043]

[0044] Among them, F(X) is the distribution network health assessment coefficient; X is the set of all key features corresponding to this part in the distribution network line; n is the number of key features in the key feature set; x i (t) is the specific value of the i-th key feature input into the distribution network state assessment model at time t when the model performs the assessment; f(x i (t)) is the normal range distribution probability density of the i-th key feature input into the distribution network state assessment model at time t; for f(x i (t)) satisfies:

[0045]

[0046] Among them, represents the fluctuation variance of the i-th key feature, reflecting the fluctuation range of this key feature during the normal operation of the device, and is obtained by statistically analyzing the historical key feature data during normal operation in advance; u i (t) is the dynamic mean of the i-th key feature at time t, and its meaning is the typical value of this key feature during recent normal operation;

[0047] Furthermore, the dynamic mean u i (t) is updated as the distribution network state information stored in the data storage subunit is updated;

[0048] Furthermore, for the dynamic mean u i (t) of a certain key feature i, the acquisition process is as follows:

[0049] S131: Set the sliding time window T, and each time the distribution network state information stored in the data storage subunit is updated, obtain the distribution network state information within the sliding time window T with the time point where the latest distribution network state information is located as the time cut-off point;

[0050] S132: Extract the key features for evaluating the distribution network line fault risk from the distribution network state information within the time window T;

[0051] S133: Screen out multiple key feature values for calculating the dynamic mean; the specific screening process is as follows:

[0052]

[0053] Among them, x i (k) is the specific value of the i-th key feature at a certain time point k in the sliding time window T, L 1i and L 2i are respectively the upper and lower limits of the normal operation range of the key feature i; True indicates that the current x i (k) can be used for dynamic mean calculation; False indicates that the current x i(k) cannot be used for dynamic mean calculation; assume there are N key feature values available for dynamic mean calculation in the sliding time window T;

[0054] S134: Calculate the dynamic mean based on the selected key feature values:

[0055]

[0056] where, w k is the time decay weight at time point k, satisfying:

[0057] w k = exp[-γ·(t 1 -k)];

[0058] where, γ is the weight decay coefficient, used to control the decay rate of the time decay weight, set through prior experiments; t 1 is the time point where the latest distribution network state information is located within the sliding time window T; t 1 -k represents the time interval between time point t 1 and time point k;

[0059] Furthermore, as Figure 3 shown, the specific working process of the fault detection sub-unit is as follows:

[0060] S21: Obtain the real-time distribution network state information;

[0061] S22: Extract the key features for evaluating the fault risk of the distribution network line from the real-time distribution network state information;

[0062] S23: Input the key features obtained in the previous step into the distribution network state evaluation model to obtain the distribution network health evaluation coefficients corresponding to each part of the distribution network line;

[0063] S24: Compare each distribution network health evaluation coefficient with the preset health threshold. If the distribution network health evaluation coefficient is lower than the health threshold, it indicates that there is a fault risk in the distribution network line part corresponding to this distribution network health evaluation coefficient; if the distribution network health evaluation coefficient is higher than the health threshold, it indicates that the distribution network line part corresponding to this distribution network health evaluation coefficient is in a normal state;

[0064] This solution collects optical data and infrared imaging data through various sensors carried by the UAV, and fuses them to generate unified and high-precision distribution network status information, thereby improving the comprehensive perception ability of equipment status, helping to more accurately identify the physical status and temperature anomalies of equipment; by establishing a distribution network status evaluation model, the fault risks of various parts of the distribution network line are quantitatively evaluated; and dynamic mean and time decay weights are combined in the model to ensure that the evaluation results can timely reflect the recent operation status of the equipment, reduce the influence of historical data deviation, and thus accurately identify potential fault risks; furthermore, the inspection tasks of the UAV are optimized to prioritize the inspection of high-risk parts, improving the inspection efficiency and quality.

[0065] Embodiment 2:

[0066] This embodiment should be understood as including at least all the features of any one of the foregoing embodiments and being further improved on this basis;

[0067] This embodiment provides an intelligent control system for distribution network UAV inspection based on multi-source data fusion. The system includes a data acquisition module, a fusion processing module, an analysis and decision-making module, a scheduling and control module, and a user management module; the data acquisition module is used to obtain the distribution network status information in real time through various sensors carried by the UAV; the fusion processing module is used to perform standardization processing and fusion on the information data obtained by the data acquisition module; the analysis and decision-making module analyzes the fault risks of various parts of the distribution network line based on the data output by the fusion processing module and optimizes the subsequent inspection routes of the UAV; the scheduling and control module is used to manage and control the execution process of the UAV inspection task; the user management module is used to provide a user interface to complete the interaction between the user and the system;

[0068] The analysis and decision-making module includes a fault diagnosis unit and an inspection optimization unit; the fault diagnosis unit is used to analyze and diagnose the distribution network status information to obtain the fault risk situation of various parts in the distribution network line; the inspection optimization unit is used to optimize the subsequent inspection routes of the UAV according to the diagnosis results of the fault diagnosis unit;

[0069] Furthermore, the inspection optimization unit includes a historical risk acquisition subunit, a priority ranking subunit, and an inspection route optimization subunit; the historical risk acquisition subunit is used to obtain and store the historical fault risk information of various parts in the distribution network line; the priority ranking subunit is used to determine the priority of the UAV to subsequently inspect various parts of the distribution network line, and the inspection route optimization subunit is used to optimize the subsequent inspection routes of the UAV in combination with the priority of various parts in the distribution network line;

[0070] Furthermore, the priority ranking subunit ranks the inspection priorities of various parts in the distribution network line in the following manner:

[0071] Calculate the inspection priority weights for each part of the distribution network line; for a certain part j in the distribution network line:

[0072] W j = α·R j + β·I j ;

[0073] Among them, W j is the inspection priority weight of a certain part j in the distribution network line, R j is the equipment importance index of part j, which is determined according to the importance of this part in the distribution network system; α is the equipment importance weight; I j is the failure risk index of part j, and β is the failure risk weight coefficient;

[0074] Furthermore, for I j it satisfies:

[0075]

[0076] Among them, M is the total number of historical time points within a set time period with the current time as the cut-off time, l is the number of the time point; when the value of l takes M, it means this time point is the closest to the current time, and when the value of l takes 1, it represents this time point is the farthest from the current time; E(l) is the failure risk judgment function corresponding to the l-th time point, and it satisfies:

[0077]

[0078] Sort all parts in the distribution network line according to the inspection priority weights to determine the inspection priorities of each part in the distribution network line, and provide a priority ranking list for the user to select and adjust;

[0079] Furthermore, when the user selects the parts to be inspected according to the priority ranking list, the inspection route optimization sub-unit will plan the shortest path covering all the selected inspection parts, and use this path as the subsequent inspection route of the drone;

[0080] Furthermore, the scheduling and control module includes a task execution monitoring unit and a task adjustment unit; the task execution monitoring unit is used to monitor the flight status, inspection progress of the drone and the working status of the sensors carried by the drone in real time to ensure that the drone executes tasks along the set inspection route; the task adjustment unit is used to receive the drone inspection route instruction optimized and output by the inspection optimization unit, and dynamically adjust the flight path and inspection task plan of the drone according to the instruction;

[0081] Furthermore, the specific functions provided by the user management module include:

[0082] Data viewing and management: Users can view all the distribution network status information collected during the inspection through the user interface, including optical image data, infrared imaging data, and historical inspection records.

[0083] Analysis result display: The fault diagnosis results, inspection priorities, and inspection routes obtained during the inspection are graphically displayed to the users to help them understand the inspection process data and make decisions.

[0084] Task plan configuration: Users can select the line parts that need to be inspected first according to the priority ranking list provided by the system and configure the inspection tasks of the UAV.

[0085] This solution comprehensively obtains the historical risk situation of each part of the distribution network line by obtaining historical fault risk information; for each part of the distribution network line, the corresponding inspection priority is calculated by combining its importance and recent fault risk status to ensure that important parts and high-risk areas are inspected and maintained first; furthermore, the system reasonably arranges inspection tasks through priority ranking, realizes the optimal configuration of the UAV inspection route, thereby improving the inspection efficiency, reducing resource consumption, and ensuring the stable operation and timely maintenance of distribution network equipment.

[0086] The content disclosed above is only the preferred feasible embodiment of the present invention, and does not limit the protection scope of the present invention. Therefore, all equivalent technical changes made by using the content of the specification and drawings of the present invention are included in the protection scope of the present invention. In addition, the elements therein can be updated with the development of technology.

Claims

1. An intelligent management and control system for distribution network drone inspection based on multi-source data fusion, characterized in that: The system includes a data acquisition module, a fusion processing module, an analysis and decision module, a dispatching control module and a user management module; the data acquisition module is used to obtain the distribution network status information in real time through various sensors carried by the drone; the fusion processing module is used to standardize and fuse the information data obtained by the data acquisition module; the analysis and decision module analyzes the fault risks of various parts of the distribution network line based on the data output by the fusion processing module and optimizes the subsequent inspection route of the drone; The scheduling and control module is used to manage and control the execution process of the drone inspection task; the user management module is used to provide a user interface to complete the interaction between the user and the system; The data acquisition module includes a multi-source sensor unit and an external data interface unit; the multi-source sensor unit is used to obtain the distribution network status information in real time during the inspection process of the drone through various sensors carried by the drone, and the distribution network status information includes optical data and infrared imaging data of various parts of the distribution network line; the external data interface unit is used to complete the information transmission between the distribution network status information and other modules; The fusion processing module includes a data preprocessing unit and a multi-source data fusion unit; the data preprocessing unit is used to perform standardization processing on the distribution network status information, and the standardization processing content includes noise filtering, data calibration, missing value processing and format conversion to ensure the consistency and reliability of different sensor data; the multi-source data fusion unit is used to integrate the standardized data to generate unified and high-precision distribution network status information in time and space.

2. According to claim 1, a distribution network drone inspection intelligent management and control system based on multi-source data fusion is characterized in that: The analysis and decision-making module includes a fault diagnosis unit and an inspection optimization unit; The fault diagnosis unit is used to analyze and diagnose the distribution network status information to obtain the fault risk status of each part of the distribution network line; The inspection optimization unit is used to optimize the subsequent inspection routes of the UAV according to the diagnosis results of the fault diagnosis unit.

3. According to claim 1, a distribution network drone inspection intelligent management and control system based on multi-source data fusion is characterized in that: The fault diagnosis unit includes a data entry subunit, a data storage subunit, a model building subunit and a fault detection subunit; the data entry subunit is used to enter the distribution network status information obtained from the multi-source data fusion unit; the data storage subunit is used to store the distribution network status information with a timestamp; the model building subunit is used to establish a distribution network status assessment model, and the distribution network status assessment model is used to assess the fault risks of various parts of the distribution network line; the fault detection subunit is used to input the currently entered real-time distribution network status information into the distribution network status assessment model to assess the fault risk of the distribution network equipment during the current drone inspection process.

4. According to claim 1, a distribution network drone inspection intelligent management and control system based on multi-source data fusion is characterized in that: The model building subunit completes the building of the distribution network status assessment model in the following ways: S11: Acquire historical distribution network status information; the historical distribution network status information includes optical information and infrared imaging information of various parts of the distribution network line; the distribution network status information is acquired through historical inspection records of drones; S12: extracting key features for assessing the risk of distribution network line failure from historical distribution network status information; the key features include physical state features extracted from optical information and temperature distribution features extracted from infrared imaging information; S13: Use the key features obtained in the previous step to establish a distribution network status assessment model; for a certain part of the distribution network line, the distribution network status assessment model is expressed as follows: Among them, F(X) is the distribution network health assessment coefficient; X is the set of all key features corresponding to this part in the distribution network line; n is the number of key features in the key feature set; x i (t) is the specific value of the i-th key feature input into the distribution network status assessment model at the time t when the model performs the assessment; f(x i (t)) is the normal range distribution probability density of the i-th key feature input into the distribution network status assessment model at time t; for f(x i (t))Satisfy: in, u represents the fluctuation variance of the i-th key feature, reflecting the fluctuation range of the key feature when the equipment is operating normally, and is obtained by statistically analyzing the historical key feature data during normal operation in advance; i (t) is the dynamic mean of the i-th key feature at time t, which means the typical value of the key feature in normal operation in the near future.

5. According to claim 1, a distribution network drone inspection intelligent management and control system based on multi-source data fusion is characterized in that: The specific working process of the fault detection subunit is as follows: S21: Obtain real-time distribution network status information; S22: Extract key features for assessing distribution network line failure risks from real-time distribution network status information; S23: input the key features obtained in the previous step into the distribution network status assessment model to obtain the distribution network health assessment coefficient corresponding to each part of the distribution network line; S24: Compare each distribution network health assessment coefficient with a preset health threshold. If the distribution network health assessment coefficient is lower than the health threshold, it indicates that the distribution network line portion corresponding to the distribution network health assessment coefficient has a fault risk; if the distribution network health assessment coefficient is higher than the health threshold, it indicates that the distribution network line portion corresponding to the distribution network health assessment coefficient is in a normal state.

6. According to claim 1, a distribution network drone inspection intelligent management and control system based on multi-source data fusion is characterized in that: The inspection optimization unit includes a historical risk acquisition subunit, a priority sorting subunit and an inspection route optimization subunit; The historical risk acquisition subunit is used to acquire and store historical fault risk information of each part of the distribution network line; The priority sorting subunit is used to determine the priority of each part in the distribution network line for subsequent inspection by the drone, and the inspection route optimization subunit is used to optimize the subsequent inspection route of the drone in combination with the priority of each part in the distribution network line.

Images