Power distribution network monitoring information monitoring equipment and method based on machine learning
Through machine learning-based monitoring information monitoring equipment and methods, the problem of low accuracy in fault judgment of distribution networks is solved, efficient and accurate fault analysis and monitoring is achieved, manual intervention is reduced, and the operation reliability of distribution networks is improved.
Patent Information
- Application Number
- CN202411831225.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-12
- Publication Date
- 2025-08-08
AI Technical Summary
The existing distribution network fault judgment system is insufficient in training effectiveness with fewer training samples and lacks weight division for different stages, resulting in low accuracy of fault judgment.
Using machine learning-based monitoring information monitoring equipment, including cameras, microphones and control devices, combined with machine learning algorithms for data collection, preprocessing and training, to build a monitoring information analysis rule base to realize intelligent analysis and decision-making.
It improves the accuracy and efficiency of fault judgment, reduces false alarms and missed reports, reduces the work burden of maintenance personnel, and improves the monitoring efficiency and accuracy of the distribution network.
Smart Images
Figure CN120454298A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of monitoring equipment, and in particular to a distribution network monitoring information monitoring device and method based on machine learning. Background Art
[0002] A distribution network is a power grid that receives electricity from the transmission grid or regional power plants and distributes it locally or in stages according to voltage to various users through distribution facilities. It consists of overhead lines, cables, towers, distribution transformers, disconnectors, VAR compensators, and other ancillary facilities, playing a crucial role in distributing electricity within the power grid. Distribution networks generally employ a closed-loop design and open-loop operation, with a radial structure. The closed-loop structure improves operational flexibility and power supply reliability; open-loop operation serves to limit short-circuit fault currents, preventing circuit breakers from exceeding their interrupting capacity and exploding, and to control the scope of faults and prevent widespread power outages.
[0003] Existing technologies have certain fault diagnosis functions, but because they rely too much on the collected data itself and lack data expansion, they cannot guarantee the effectiveness of training when there are fewer training samples. Once data expansion is performed by modifying the data, it will cause data distortion or invalidity, thereby affecting the training results. At the same time, existing technologies usually treat each piece of fault data as a whole, lack of weight division and differentiated analysis of different stages, so the fault diagnosis accuracy is low. Summary of the Invention
[0004] Based on the technical problems existing in the background technology, the present invention proposes a distribution network monitoring information monitoring device and method based on machine learning.
[0005] The present invention proposes a machine learning-based distribution network monitoring information monitoring device, comprising a monitoring device, the monitoring device including a front-end device, a communication unit configured to receive an original image acquired by the front-end device and a GPS positioning unit configured to locate a camera position, a screen unit configured to display the original image and a corrected image formed by dedistorting the original image; a sound collection unit composed of multiple microphones, the microphones converting sound waves emitted from the target area into sound pressure signals, a machine learning module for constructing a distribution network monitoring information analysis rule base; and a distribution network monitoring information analysis machine module for performing intelligent analysis and decision-making on the machine learning module.
[0006] A control device is configured to store map information related to the location of the target area relative to the sound collection device, and to complete functions such as digital collection, compression, monitoring data recording and retrieval, and hard disk recording for information analysis.
[0007] The front-end equipment preferably includes a camera, a motorized zoom lens, an infrared detector, a temperature and humidity sensor, a pan / tilt head, and a protective cover. The camera uses a built-in CCD and auxiliary circuits to capture the scene and convert it into an analog video signal, which is transmitted via a coaxial cable. The motorized zoom lens is used to adjust the shooting scene, and the pan / tilt head and protective cover provide a suitable working environment.
[0008] Preferably, the screen unit displays an icon in a dewarping area where the original image is dewarped into the corrected image, and when an icon is selected, the dewarping area where the selected icon is located is selected and the screen unit displays a polygon indicating the selected dewarping area.
[0009] Preferably, a storage unit is included, and the storage unit stores the original video images.
[0010] Preferably, the communication unit receives the original image and the corrected image from a camera.
[0011] A method for monitoring equipment for distribution network monitoring information based on machine learning, comprising the following steps:
[0012] Step 1: Collect real-time and historical data of the distribution network, including the acquisition and storage of grid big data;
[0013] Step 2: Preprocess the collected data, including data cleaning, formatting, and outlier processing;
[0014] Step 3: Use machine learning algorithms to train and learn the processed data to build a distribution network monitoring information analysis rule base;
[0015] Step 4: Perform intelligent analysis and decision-making based on the rule base trained by the machine learning module to achieve real-time monitoring and anomaly detection of the distribution network;
[0016] Step 5: The distribution network monitoring information analysis machine performs real-time monitoring based on the rule base, and issues alarms and handles abnormal situations when they are found;
[0017] Step 6: Through machine learning algorithms, the system can automatically analyze and process large amounts of data, reduce manual intervention, and improve monitoring efficiency.
[0018] The beneficial effects of the present invention are:
[0019] 1. Through training with machine learning algorithms, the system can more accurately identify abnormal situations and reduce false alarms and missed alarms. The system can automatically analyze and process large amounts of data, reduce manual intervention, improve monitoring efficiency, and handle abnormal situations when they are discovered. This not only improves the monitoring efficiency and accuracy of the distribution network, but also uses machine learning models to implement fault analysis, which facilitates improved fault diagnosis accuracy and efficiency, reduces the workload of maintenance personnel, and improves maintenance convenience. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 Flowchart proposed for the present invention;
[0021] Figure 2 This is the system connection diagram proposed by the present invention. DETAILED DESCRIPTION
[0022] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.
[0023] Reference Figure 1-2 , a distribution network monitoring information monitoring device based on machine learning, comprising a monitoring device, the monitoring device comprising a front-end device, a communication unit configured to receive an original image acquired by the front-end device and a GPS positioning unit for locating a camera position, a screen unit configured to display the original image and a corrected image formed by dedistorting the original image; a sound collection unit composed of a plurality of microphones, the microphones converting sound waves emitted from the target area into sound pressure signals, a machine learning module for constructing a distribution network monitoring information analysis rule base; a distribution network monitoring information analysis machine module for performing intelligent analysis and decision-making on the machine learning module;
[0024] A control device is configured to store map information related to the location of the target area relative to the sound collection device, and to complete functions such as digital collection, compression, monitoring data recording and retrieval, and hard disk recording for information analysis.
[0025] Specifically, the front-end equipment includes a camera, a motorized zoom lens, an infrared detector, a temperature and humidity sensor, a pan / tilt head (PTZ), and a protective cover. The camera uses a built-in CCD and auxiliary circuits to capture the scene, converting it into an analog video signal that is transmitted via a coaxial cable. The motorized zoom lens adjusts the shooting scene, while the PTZ and protective cover provide a suitable working environment.
[0026] Specifically, the screen unit displays an icon in a dewarping area where an original image is dewarped into a corrected image, and when an icon is selected, the dewarping area where the selected icon is located is selected and the screen unit displays a polygon indicating the selected dewarping area.
[0027] Specifically, it includes a storage unit, which stores the original image of the video.
[0028] Specifically, the communication unit receives an original image and a corrected image from a camera.
[0029] A method for monitoring equipment for distribution network monitoring information based on machine learning, comprising the following steps:
[0030] Step 1: Collect real-time and historical data of the distribution network, including the acquisition and storage of grid big data;
[0031] Step 2: Preprocess the collected data, including data cleaning, formatting, and outlier processing;
[0032] Step 3: Use machine learning algorithms to train and learn the processed data to build a distribution network monitoring information analysis rule base;
[0033] Step 4: Perform intelligent analysis and decision-making based on the rule base trained by the machine learning module to achieve real-time monitoring and anomaly detection of the distribution network;
[0034] Step 5: The distribution network monitoring information analysis machine performs real-time monitoring based on the rule base, and issues alarms and handles abnormal situations when they are found;
[0035] Step 6: Through machine learning algorithms, the system can automatically analyze and process large amounts of data, reduce manual intervention, and improve monitoring efficiency.
[0036] In the present invention, through the training of machine learning algorithms, the system can more accurately identify abnormal situations and reduce false alarms and missed alarms. The system can automatically analyze and process large amounts of data, reduce manual intervention, improve monitoring efficiency, and handle abnormal situations when they are found. This not only improves the monitoring efficiency and accuracy of the distribution network, but also uses machine learning models to implement fault analysis, which facilitates improving the accuracy and efficiency of fault judgment, reduces the workload of maintenance personnel, and improves the convenience of maintenance.
[0037] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.
Claims
1. A distribution network monitoring information monitoring device based on machine learning, comprising a monitoring device, characterized in that: The monitoring device includes a front-end device, a communication unit configured to receive an original image acquired by the front-end device and a GPS positioning unit configured to locate the position of a camera, a screen unit configured to display the original image and a corrected image formed by dedistorting the original image; a sound collection unit composed of multiple microphones, each of which converts sound waves emitted from the target area into sound pressure signals, and a machine learning module that constructs a distribution network monitoring information analysis rule base; A distribution network monitoring information analysis machine module that performs intelligent analysis and decision-making on the machine learning module; A control device is configured to store map information related to the position of the target area relative to the sound collection device, and to complete functions such as digital collection, compression, monitoring data recording and retrieval, and hard disk recording for information analysis.
2. The distribution network monitoring information monitoring device based on machine learning according to claim 1, characterized in that: Front-end equipment includes a camera, a motorized zoom lens, an infrared detector, a temperature and humidity sensor, a pan / tilt head (PTZ), and a protective cover. The camera uses a built-in CCD and auxiliary circuits to capture on-site conditions, converting them into analog video signals that are transmitted via a coaxial cable. The motorized zoom lens adjusts the shooting scene, while the PTZ head and protective cover provide a suitable working environment.
3. The distribution network monitoring information monitoring device based on machine learning according to claim 2 is characterized in that: The screen unit displays an icon in a dewarping area where an original image is dewarped into a corrected image, and when an icon is selected, the dewarping area where the selected icon is located is selected and the screen unit displays a polygon indicating the selected dewarping area.
4. The distribution network monitoring information monitoring device based on machine learning according to claim 3 is characterized in that: It includes a storage unit, which stores the original image of the video.
5. The distribution network monitoring information monitoring device based on machine learning according to claim 4 is characterized in that: The communication unit receives an original image and a corrected image from a camera.
6. The method of a distribution network monitoring information monitoring device based on machine learning according to claim 1, characterized in that: The steps include: Step 1: Collect real-time and historical data of the distribution network, including the acquisition and storage of grid big data; Step 2: Preprocess the collected data, including data cleaning, formatting, and outlier processing; Step 3: Use machine learning algorithms to train and learn the processed data to build a distribution network monitoring information analysis rule base; Step 4: Perform intelligent analysis and decision-making based on the rule base trained by the machine learning module to achieve real-time monitoring and anomaly detection of the distribution network; Step 5: The distribution network monitoring information analysis machine performs real-time monitoring based on the rule base, and issues alarms and handles abnormal situations when they are found; Step 6: Through machine learning algorithms, the system can automatically analyze and process large amounts of data, reduce manual intervention, and improve monitoring efficiency.