A power distribution visualization system and method

Abstract

The present invention discloses a power distribution visualization system and method, including: an image acquisition module configured to acquire power distribution network line image information; an image intelligent recognition module configured to use a convolutional neural network to perform hidden danger recognition and / or defect recognition on the acquired image; an alarm module configured to alarm a remote operation and maintenance terminal for the recognized hidden dangers and / or defects. The power distribution visualization system of the present invention can realize the automatic inspection and diagnosis of power distribution network lines. With the precise classification of neural network technology, it can realize the automatic recognition and alarm of hidden danger sources of power distribution network lines and defects of power distribution equipment, and can accurately judge the types of hidden dangers and defects.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution visualization, and particularly to a distribution visualization system and method. Background Art

[0002] The statements in this part merely provide background technical information related to the present invention and do not necessarily constitute prior art.

[0003] The environment of distribution network lines is complex and diverse. The number of distribution lines is large, the mileage is long, and the environment around the corridors is complex and changeable. Affected by road construction, building construction, large operating vehicles, and natural disasters near the corridors, the operation safety of distribution network lines is severely threatened.

[0004] For the monitoring of distribution network lines, the traditional manual inspection mode has low inspection efficiency and cannot achieve comprehensive consideration; traditional video monitoring means are difficult to achieve full coverage. Coupled with large volume, heavy weight, complex installation, and high cost, they are not suitable for distribution line towers; high-tech means such as unmanned aerial vehicles are not suitable for use in most environments of distribution network lines. Summary of the Invention

[0005] To solve the above problems, the present invention proposes a distribution visualization system and method, which can realize remote visualization intelligent monitoring of distribution network lines.

[0006] In some embodiments, the following technical solutions are adopted:

[0007] A distribution visualization system includes:

[0008] An image acquisition module configured to acquire image information of distribution network lines;

[0009] An image intelligent recognition module configured to use a convolutional neural network to perform hidden danger recognition and / or defect recognition on the acquired images;

[0010] An alarm module configured to alarm the remote operation and maintenance terminal for the identified hidden dangers and / or defects.

[0011] As a further solution, the image intelligent recognition module uses a convolutional neural network to perform hidden danger recognition on the acquired images, specifically including:

[0012] Constructing a hidden danger sample set; the sample set includes a positive sample set and a negative sample set;

[0013] Labeling the hidden danger categories to which each sample in the sample set belongs;

[0014] Inputting the labeled positive samples and negative samples into a convolutional neural network model for training according to a set ratio;

[0015] Input the real-time obtained distribution network line image information into the trained neural network model to output the hidden danger recognition result.

[0016] As a further solution, the hidden danger recognition includes: special vehicle recognition, wildfire and smoke recognition, ice coating recognition, and foreign object suspension recognition.

[0017] As a further solution, for the incorrect hidden danger recognition result, label the defect category of the distribution network line image corresponding to this result, add the labeled image to the sample set, and retrain the neural network model.

[0018] As a further solution, the defect categories include: tower collapse and wire breakage.

[0019] As a further solution, the image intelligent recognition module can automatically extract the virtual fence of the monitored area from the obtained image information and recognize the hidden dangers within the virtual fence.

[0020] As a further solution, it further includes: an automatic work order dispatching module, configured to automatically generate an operation and maintenance work order and dispatch it to the relevant operation and maintenance terminals after identifying equipment hidden dangers or defects.

[0021] As a further solution, it further includes: a voice playback module, configured to be able to play the remote voice information of the operation and maintenance personnel; or, when the image intelligent recognition module identifies the hidden danger source and determines that the distance between the hidden danger source and the distribution network equipment is less than the safe distance, automatically play an alarm prompt sound.

[0022] As a further solution, the image intelligent recognition module transmits the obtained image data and recognition results to the front-end server through the enterprise gateway and firewall; through the port mapping method, allows external network clients to access a specific port of a specific IP of the front-end server through the firewall for real-time monitoring by the monitoring center.

[0023] In some other embodiments, the following technical solution is adopted:

[0024] A distribution visualization method includes:

[0025] Obtain distribution network line image information;

[0026] Input the obtained image information into the trained convolutional neural network to output the hidden danger recognition result and / or defect recognition result;

[0027] When the distance between the identified hidden danger source and the distribution network equipment is less than the set safe distance, or when a device defect is identified, give an alarm to the remote operation and maintenance terminal;

[0028] At the same time, automatically generate an operation and maintenance work order and dispatch it to the relevant operation and maintenance terminals.

[0029] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0030] (1) The power distribution visualization system of the present invention can realize the automatic inspection and diagnosis of distribution network lines. With the precise classification of neural network technology, it can realize the automatic identification and alarm of potential hazard sources of distribution network lines and defects of distribution network equipment, and can accurately judge the types of potential hazards and defects.

[0031] (2) Before image recognition, the present invention extracts the virtual fence of the monitoring area in advance, and only recognizes the images within the virtual fence. And through the retraining of misrecognition results, the neural network recognition model is made to have learning ability, improving the recognition accuracy.

[0032] (3) After identifying defects or potential hazards, the present invention can automatically dispatch operation and maintenance work orders, assign operation and maintenance personnel to conduct on-site inspections and defect elimination feedback, and complete the closed-loop. It can efficiently protect the distribution cable channels and prevent the occurrence of construction external damage accidents.

[0033] Other features and advantages of the additional aspects of the present invention will be partly given in the following description, partly will become obvious from the following description, or will be understood through the practice of this aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a schematic structural diagram of the power distribution visualization system in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0035] It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used in the present invention have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present application belongs.

[0036] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0037] Embodiment 1

[0038] In one or more embodiments, a power distribution visualization system is disclosed, as Figure 1 shown, including:

[0039] An image acquisition module configured to acquire distribution network line image information;

[0040] A high-definition laser camera is installed on the distribution network line poles at a set distance. Image information of the distribution network line is obtained through the high-definition laser camera. The camera takes pictures of the scene within the monitoring range at a set time interval to obtain the image information of the distribution network line within the monitoring range.

[0041] The high-definition laser camera is small in size and convenient to install. It is installed by using a hoop bundling method and supports installation methods such as self-standing poles, borrowing lamp poles, utility poles, and customizing brackets to borrow buildings, etc.

[0042] The image intelligent recognition module is configured to use a convolutional neural network to perform hidden danger recognition and / or defect recognition on the obtained images;

[0043] Specifically, for hidden danger recognition, it mainly includes: special vehicle recognition, wildfire and smoke recognition, icing recognition, and foreign object hanging recognition.

[0044] The defect categories include: recognition of tower collapse, wire breakage, etc.

[0045] In this embodiment, the specific method for using a convolutional neural network to perform hidden danger recognition on the obtained images includes:

[0046] Construct a hidden danger sample set; the sample set includes a number of positive sample sets and negative sample sets; among them, the positive samples are sample images containing hidden dangers, and the negative samples are sample images without hidden dangers; the ratio of the positive sample set to the negative sample set can be maintained at about 7:3.

[0047] The sample set is obtained from the historical data collected by the visual monitoring device installed on the pole.

[0048] Label the hidden danger category to which each sample in the positive sample set belongs;

[0049] Input the labeled positive and negative samples into the convolutional neural network model according to a set ratio for training; obtain a hidden danger recognition model.

[0050] Input the image information of the distribution network line obtained in real time into the trained hidden danger recognition model, and output the hidden danger recognition result.

[0051] For the recognition of defect categories, it is also implemented using the same principle as above and will not be elaborated here.

[0052] As an optional implementation method, after the image intelligent recognition module obtains an image, it first extracts the virtual fence of the monitored area within the monitoring range, and only performs intelligent recognition on the image information within the virtual fence, which can improve the recognition efficiency and accuracy and reduce the occurrence of false alarms.

[0053] The alarm module is configured to alarm the remote operation and maintenance terminal for the identified hidden dangers and / or defects.

[0054] An automatic work order dispatching module, configured to automatically generate an operation and maintenance work order and dispatch it to the relevant operation and maintenance terminals after identifying equipment hidden dangers or defects.

[0055] A voice playback module, configured to be able to play the remote voice information of the operation and maintenance personnel; or, when the image intelligent recognition module identifies a hidden danger source and determines that the distance between the hidden danger source and the power distribution equipment is less than the safe distance, automatically play an alarm prompt sound.

[0056] For the identified hidden danger information, if the distance between the hidden danger and the power distribution equipment is less than the set safe distance, or a power distribution equipment defect is identified, an alarm is sent to the remote operation and maintenance terminal, and at the same time, the automatic work order dispatching module automatically generates an operation and maintenance work order and dispatches it to the relevant operation and maintenance terminals.

[0057] After receiving the alarm prompt, if it is a hidden danger alarm, the operation and maintenance terminal remotely shouts through the voice playback module to stop and prompt; it is beneficial to handle the hidden danger in the first time, efficiently protect the power distribution cable channel, and prevent the occurrence of construction external damage accidents.

[0058] As an optional implementation method, the defect or hidden danger identification data can be transmitted to the deployed system front-end server through the enterprise gateway and firewall; through the port mapping method, the external network client is allowed to access the specific port of the specific IP of the system front-end server through the firewall for the monitoring center to monitor in real time.

[0059] A WeChat gateway and interface service is deployed in the system. In the form of a business system, it interacts with the WeChat background by means of secure encryption to realize the system pushing alarm information to the WeChat enterprise and previewing and viewing the detailed channel images in the enterprise WeChat, providing convenience for the patrol personnel to view the real-time status through the mobile terminal (WeChat enterprise number).

[0060] Embodiment 2

[0061] In one or more embodiments, a power distribution visualization method is disclosed, including the following processes:

[0062] Obtain the power distribution line image information;

[0063] Input the obtained image information into the trained convolutional neural network, and output the hidden danger identification result and / or defect identification result;

[0064] When the distance between the identified hidden danger source and the power distribution equipment is less than the set safe distance, or a device defect is identified, an alarm is sent to the remote operation and maintenance terminal;

[0065] At the same time, an operation and maintenance work order is automatically generated and dispatched to the relevant operation and maintenance terminals.

[0066] The specific implementation manner of the above process has been described in detail in Embodiment 1 and will not be elaborated here.

[0067] Although the specific implementation of the present invention has been described with reference to the accompanying drawings above, it is not a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications or deformations that can be made without creative efforts on the basis of the technical solution of the present invention are still within the protection scope of the present invention.

Claims

1. A power distribution visualization system, characterized in that, it includes: An image acquisition module, configured to acquire power distribution line image information; An image intelligent recognition module, configured to use a convolutional neural network to perform hidden danger recognition and / or defect recognition on the acquired images; An alarm module, configured to alarm the remote operation and maintenance terminal for the recognized hidden dangers and / or defects; The image intelligent recognition module can automatically extract the virtual fence of the monitoring area from the acquired image information and recognize the image information within the virtual fence; A voice playback module, configured to be able to play the remote voice information of the operation and maintenance personnel; or, when the image intelligent recognition module recognizes a hidden danger source and determines that the distance between the hidden danger source and the power distribution equipment is less than the safe distance, automatically play an alarm prompt sound; The hidden danger recognition includes: special vehicle recognition, wildfire and smoke recognition, icing recognition, and foreign object hanging recognition; The categories of defects include: tower collapse and wire breakage.

2. A power distribution visualization system according to claim 1, characterized in that, The image intelligent recognition module uses a convolutional neural network to perform hidden danger recognition on the acquired images, specifically including: Constructing a hidden danger sample set; the sample set includes a positive sample set and a negative sample set; Labeling the hidden danger categories to which each sample in the sample set belongs; Inputting the labeled positive and negative samples into the convolutional neural network model according to a set ratio for training; Inputting the power distribution line image information obtained in real time into the trained neural network model and outputting the hidden danger recognition result.

3. A power distribution visualization system according to claim 2, characterized in that, For the incorrect hidden danger recognition result, label the defect category of the power distribution line image of this result, add the labeled image to the sample set, and retrain the neural network model.

4. A power distribution visualization system according to claim 1, characterized in that, It further includes: An automatic work order dispatching module, configured to automatically generate an operation and maintenance work order and dispatch it to the relevant operation and maintenance terminal after identifying equipment hidden dangers or defects.

5. A power distribution visualization system according to claim 1, characterized in that, The image intelligent recognition module transmits the acquired image data and recognition results to the front-end server through the enterprise gateway and firewall; through the port mapping method, allows the external network client to access the specific port of the specific IP of the front-end server through the firewall for the monitoring center to monitor in real time.

6. A method based on the power distribution visualization system according to any one of claims 1-5, characterized in that, it includes: Acquiring power distribution line image information; Inputting the acquired image information into the trained convolutional neural network and outputting the hidden danger recognition result and / or defect recognition result; When the distance between the recognized hidden danger source and the power distribution equipment is less than the set safe distance, or when a device defect is recognized, alarm the remote operation and maintenance terminal; At the same time, automatically generate an operation and maintenance work order and dispatch it to the relevant operation and maintenance terminal.

Images