CN117113283B - State identification method and system of isolating switch - Google Patents
State identification method and system of isolating switch Download PDFInfo
- Publication number
- CN117113283B CN117113283B CN202311389065.5A CN202311389065A CN117113283B CN 117113283 B CN117113283 B CN 117113283B CN 202311389065 A CN202311389065 A CN 202311389065A CN 117113283 B CN117113283 B CN 117113283B
- Authority
- CN
- China
- Prior art keywords
- data
- data set
- point cloud
- target
- isolating switch
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 239000013598 vector Substances 0.000 claims abstract description 80
- 238000012545 processing Methods 0.000 claims abstract description 57
- 238000004458 analytical method Methods 0.000 claims abstract description 51
- 230000004927 fusion Effects 0.000 claims abstract description 31
- 238000000605 extraction Methods 0.000 claims abstract description 29
- 238000013506 data mapping Methods 0.000 claims abstract description 15
- 238000007405 data analysis Methods 0.000 claims abstract description 12
- 238000003066 decision tree Methods 0.000 claims description 15
- 238000005070 sampling Methods 0.000 claims description 15
- 238000001914 filtration Methods 0.000 claims description 12
- 230000001360 synchronised effect Effects 0.000 claims description 10
- 238000006243 chemical reaction Methods 0.000 claims description 8
- 238000004364 calculation method Methods 0.000 claims description 4
- 238000004422 calculation algorithm Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000007637 random forest analysis Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/254—Fusion techniques of classification results, e.g. of results related to same input data
- G06F18/256—Fusion techniques of classification results, e.g. of results related to same input data of results relating to different input data, e.g. multimodal recognition
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/327—Testing of circuit interrupters, switches or circuit-breakers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2123/00—Data types
- G06F2123/02—Data types in the time domain, e.g. time-series data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
- G06F2218/10—Feature extraction by analysing the shape of a waveform, e.g. extracting parameters relating to peaks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Processing (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention relates to the technical field of image processing, and discloses a method and a system for identifying the state of an isolating switch, which are used for improving the accuracy rate during the state identification of the isolating switch. Comprising the following steps: carrying out data analysis on the electric signal data to obtain an analysis data set, collecting an image data set of the isolating switch to be detected, and carrying out point cloud data mapping on the image data set to obtain a point cloud data set; collecting an infrared light data set of the isolating switch to be detected, and performing data time synchronization processing on the infrared light data set to obtain a target infrared light data set; performing data time synchronization processing on the point cloud data set to obtain a target point cloud data set; performing multi-mode data fusion on the target infrared light data set and the target point cloud data set to obtain multi-mode data, and performing feature extraction on the multi-mode data to obtain a target feature vector; and inputting the target feature vector into a state recognition model to recognize the state of the isolating switch, and obtaining target state data of the isolating switch to be detected.
Description
Technical Field
The present invention relates to the field of image processing technologies, and in particular, to a method and a system for identifying a state of an isolating switch.
Background
The isolating switch is a key device in the power system and is used for isolating a circuit, ensuring safety and carrying out maintenance and overhaul work. The isolating switch is used for opening and closing the no-load current circuit to obviously isolate the electrical equipment from the power supply. The TAU series disconnector is designed for mounting on a switchgear cabinet. The series of products are mainly applied to the direct current traction system or the power engineering industry.
With the development of technology, the isolating switch state monitoring and identifying technology based on multi-sensor data fusion is gradually in the brand-new angle. However, in the prior art, electrical signal sensors and other sensors may be affected by noise, interference or calibration problems, resulting in unstable data quality. The time resolution of the data collected by the electrical signal sensor and the image data, infrared data, etc. may be different, resulting in a problem of time synchronization between the data. Inaccurate time synchronization can complicate multi-modal data fusion and state recognition. The disconnector may be deployed in different working environments, including indoors and outdoors, each with different lighting, temperature and humidity conditions. These environmental changes may have an unstable effect on sensor data, making state recognition more complex. Therefore, the accuracy rate of the state identification of the isolating switch in the existing scheme is insufficient.
Disclosure of Invention
In view of the above, the embodiments of the present invention provide a method and a system for identifying a state of an isolating switch, which are used for improving the accuracy in identifying the state of the isolating switch.
The invention provides a state identification method of an isolating switch, which comprises the following steps: collecting electric signal data of the isolating switch to be detected through an electric signal sensor arranged on the isolating switch to be detected, and carrying out data analysis on the electric signal data to obtain an analysis data set;
acquiring an image data set of the isolating switch to be detected through a preset image acquisition device, and simultaneously, performing point cloud data mapping on the image data set to obtain a point cloud data set;
collecting an infrared data set of the isolating switch to be detected through a preset infrared collecting device, and simultaneously, carrying out data time synchronization processing on the infrared data set through the analysis data set to obtain a target infrared data set;
performing data time synchronization processing on the point cloud data set through the analysis data set to obtain a target point cloud data set;
performing multi-mode data fusion on the target infrared light data set and the target point cloud data set to obtain multi-mode data, and performing feature extraction on the multi-mode data to obtain a target feature vector;
And inputting the target feature vector into a preset state recognition model to recognize the state of the isolating switch, and obtaining the target state data of the isolating switch to be detected.
In the invention, the step of collecting the electrical signal data of the isolating switch to be detected through the electrical signal sensor arranged on the isolating switch to be detected and carrying out data analysis on the electrical signal data to obtain analysis data set comprises the following steps:
collecting electric signal data of the isolating switch to be detected through the electric signal sensor, wherein the electric signal data comprise current signal data and voltage signal data;
extracting the current signal data in a time sequence to obtain a first time sequence corresponding to the current signal data;
extracting the time sequence of the voltage signal data to obtain a second time sequence corresponding to the voltage signal data;
performing signal filtering processing on the current signal data based on the first time sequence to obtain filtered current signal data;
performing signal filtering processing on the voltage signal data based on the second time sequence to obtain filtered voltage signal data;
performing current waveform analysis on the filtered current signal data to obtain current waveform data, and performing voltage waveform analysis on the filtered voltage signal data to obtain voltage waveform data;
And carrying out data format conversion on the current waveform data and the voltage waveform data to obtain the analysis data set.
In the invention, the step of acquiring the image data set of the isolating switch to be detected through a preset image acquisition device and simultaneously performing point cloud data mapping on the image data set to obtain the point cloud data set comprises the following steps:
collecting an image data set of the isolating switch to be detected through a preset image collecting device;
calibrating the position of the isolating switch to be detected to obtain position data corresponding to the isolating switch to be detected;
constructing a space coordinate system through the position data, and extracting image depth data from the image data set based on the space coordinate system to obtain an image depth data set;
and carrying out point cloud data mapping on the image data set in the space coordinate system based on the image depth data set to obtain the point cloud data set.
In the invention, the step of acquiring the infrared light data set of the isolating switch to be detected through a preset infrared acquisition device, and simultaneously, carrying out data time synchronization processing on the infrared light data set through the analysis data set to obtain a target infrared light data set comprises the following steps:
Collecting an infrared light data set of the isolating switch to be detected through the infrared collecting device;
performing time stamp extraction on the infrared light data set to obtain first time stamp data corresponding to the infrared light data set;
performing timestamp synchronous analysis on the first time sequence and the second time sequence to obtain corresponding second timestamp data;
and carrying out data time synchronization processing on the infrared light data set based on the first time stamp data and the second time stamp data to obtain the target infrared light data set.
In the invention, the steps of performing multi-mode data fusion on the target infrared light data set and the target point cloud data set to obtain multi-mode data, and performing feature extraction on the multi-mode data to obtain a target feature vector comprise the following steps:
extracting first characteristics of the target infrared light data set to obtain an infrared characteristic vector corresponding to the target infrared light data;
extracting a point cloud projection contour from the target point cloud data to obtain corresponding point cloud projection contour data;
performing second feature extraction on the point cloud projection profile data through preset template profile data to obtain a point cloud feature vector corresponding to the target point cloud data;
Vector connection processing is carried out on the infrared characteristic vector and the point cloud characteristic vector to obtain the multi-mode data;
and extracting the characteristics of the multi-mode data to obtain the target characteristic vector.
In the present invention, the step of extracting the second feature of the point cloud projection profile data by using preset template profile data to obtain a point cloud feature vector corresponding to the target point cloud data includes:
carrying out data alignment processing on the template contour data and the point cloud projection contour data to obtain alignment contour data;
performing feature descriptor calculation on the aligned profile data to obtain a corresponding feature descriptor set, wherein the feature descriptor set comprises; profile curvature data and profile normal data;
and performing feature vector conversion on the feature descriptor set to obtain a point cloud feature vector corresponding to the target point cloud data.
In the invention, the step of inputting the target feature vector into a preset state recognition model to recognize the state of the isolating switch and obtaining the target state data of the isolating switch to be detected comprises the following steps:
inputting the target feature vector into the state recognition model to disassemble the feature values, so as to obtain a disassembled feature value data set;
Randomly sampling the disassembled characteristic value data sets to obtain a plurality of sampled data sets;
respectively constructing decision trees of each sampling data set to obtain a target decision tree corresponding to each sampling data set;
respectively inputting each sampling data set into a target decision tree corresponding to each sampling data set to perform switch state identification, so as to obtain a plurality of state identification results;
and carrying out state weighted fusion on the plurality of state identification results to obtain the target state data of the isolating switch to be detected.
The invention also provides a state identification system of the isolating switch, which comprises:
the analysis module is used for collecting the electrical signal data of the isolating switch to be detected through an electrical signal sensor arranged on the isolating switch to be detected, and carrying out data analysis on the electrical signal data to obtain an analysis data set;
the acquisition module is used for acquiring an image data set of the isolating switch to be detected through a preset image acquisition device, and simultaneously, performing point cloud data mapping on the image data set to obtain a point cloud data set;
the synchronous module is used for acquiring an infrared light data set of the isolating switch to be detected through a preset infrared acquisition device, and simultaneously, carrying out data time synchronous processing on the infrared light data set through the analysis data set to obtain a target infrared light data set;
The processing module is used for carrying out data time synchronization processing on the point cloud data set through the analysis data set to obtain a target point cloud data set;
the fusion module is used for carrying out multi-mode data fusion on the target infrared light data set and the target point cloud data set to obtain multi-mode data, and carrying out feature extraction on the multi-mode data to obtain a target feature vector;
and the identification module is used for inputting the target feature vector into a preset state identification model to identify the state of the isolating switch, so as to obtain the target state data of the isolating switch to be detected.
In the technical scheme provided by the invention, the electric signal data of the isolating switch to be detected is collected through the electric signal sensor arranged on the isolating switch to be detected, and the electric signal data is subjected to data analysis to obtain an analysis data set; acquiring an image data set of an isolating switch to be detected through a preset image acquisition device, and simultaneously, performing point cloud data mapping on the image data set to obtain a point cloud data set; collecting an infrared data set of an isolating switch to be detected through a preset infrared collecting device, and simultaneously, carrying out data time synchronization processing on the infrared data set through analyzing the data set and carrying out data time synchronization processing on a point cloud data set through analyzing the data set to obtain a point cloud data set; performing multi-mode data fusion on the target infrared light data set and the target point cloud data set to obtain multi-mode data, and performing feature extraction on the multi-mode data to obtain a target feature vector; and inputting the target feature vector into a preset state recognition model to recognize the state of the isolating switch, and obtaining target state data of the isolating switch to be detected. In the scheme, through collecting various data sources (electric signals, images, infrared light, point clouds and the like), multi-mode data fusion and feature extraction, more comprehensive and multi-angle information can be provided. The accuracy of state identification can be improved, and false alarm and missing report are reduced. Combining data of different sensor types, such as electrical signals, images and infrared light, enables more comprehensive information to be obtained. Multimodal data fusion may provide more features to better capture the complexity of the disconnector state. Through real-time data acquisition and processing, the system can monitor the disconnecting switch to be detected in real time, quickly detect state changes, and even detect potential faults and abnormal conditions, so that the safety and reliability of the system are improved, and the accuracy rate of the disconnecting switch during state identification is further improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for identifying a state of an isolating switch according to an embodiment of the present invention.
Fig. 2 is a flowchart of mapping point cloud data on an image data set according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of a state recognition system of an isolating switch according to an embodiment of the invention.
Reference numerals:
301. an analysis module; 302. an acquisition module; 303. a synchronization module; 304. a processing module; 305. a fusion module; 306. and an identification module.
Detailed Description
The following description of the embodiments of the present invention will be made apparent and fully in view of the accompanying drawings, in which some, but not all embodiments of the invention are shown. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc. are based on the directions or positional relationships shown in the drawings, are merely for convenience of describing the present invention and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and thus should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In addition, the technical features of the different embodiments of the present invention described below may be combined with each other as long as they do not collide with each other.
For easy understanding, the following describes a specific flow of an embodiment of the present invention, referring to fig. 1, fig. 1 is a flowchart of a method for identifying a state of a disconnecting switch according to an embodiment of the present invention, as shown in fig. 1, including the following steps:
s101, acquiring electric signal data of an isolating switch to be detected through an electric signal sensor arranged on the isolating switch to be detected, and carrying out data analysis on the electric signal data to obtain an analysis data set;
S102, acquiring an image data set of an isolating switch to be detected through a preset image acquisition device, and simultaneously, performing point cloud data mapping on the image data set to obtain a point cloud data set;
s103, acquiring an infrared light data set of the isolating switch to be detected through a preset infrared acquisition device, and simultaneously, carrying out data time synchronization processing on the infrared light data set through analyzing the data set to obtain a target infrared light data set;
s104, carrying out data time synchronization processing on the point cloud data set through analyzing the data set to obtain a target point cloud data set;
s105, carrying out multi-mode data fusion on the target infrared light data set and the target point cloud data set to obtain multi-mode data, and carrying out feature extraction on the multi-mode data to obtain a target feature vector;
s106, inputting the target feature vector into a preset state recognition model to recognize the state of the isolating switch, and obtaining target state data of the isolating switch to be detected.
It should be noted that, first, a current sensor and a voltage sensor are installed on a circuit of an isolating switch to be detected to measure current and voltage signals. These sensors collect electrical signal data in real time, which is typically in the form of analog signals.
For example: it is assumed that a current sensor and a voltage sensor are connected to the circuit of the disconnector to be detected. These sensors collect current and voltage data once per second, yielding the following:
current data: [5.2A,5.3A,5.1A ];
voltage data: [220V,221V,219V ].
Then, the analog signal is converted into a digital signal by data analysis.
A preset image acquisition device (e.g., a camera) is used to capture a set of image data of the disconnector to be detected. At the same time, these image data are mapped into a point cloud data set. This process may be implemented using deep learning techniques, such as structured light or stereo vision. For example: the camera captures a series of images, which are then mapped into point cloud data, including three-dimensional information of the disconnector, by image processing and computer vision techniques.
And acquiring an infrared light data set of the isolating switch to be detected by using a preset infrared acquisition device. Meanwhile, each infrared light data sample is ensured to have time information related to the infrared light data sample by means of time stamping or time recording so as to perform data time synchronization processing. For example: an infrared camera captures a series of infrared light images and a time stamp of each image is recorded. These time stamps will be used for subsequent time synchronization processing.
The parsed electrical signal data, image data set and infrared light data set are time synchronized to ensure that they correspond on the same time axis. For example: the electrical signal data, the image data and the infrared light data are all time stamped. These data are aligned by time stamps such that they correspond at the same point in time.
And carrying out multi-mode data fusion on the target infrared light data set and the target point cloud data set. And then, carrying out feature extraction on the fused multi-mode data to obtain a target feature vector. Feature extraction methods may include statistical features, deep learning feature extraction, or other suitable techniques. For example: and calculating curvature characteristics of the point cloud and temperature distribution characteristics of the infrared image by using the point cloud data and the infrared light data. These features will be combined into the target feature vector.
And inputting the target feature vector into a preset state recognition model to recognize the state of the isolating switch. The model outputs target state data of the isolating switch to be detected, such as a state label of "on" or "off".
For example: the random forest model is used, the target feature vector is input into the model, the probability of the model output 'on' state is 0.85, and the probability of the model output 'off' state is 0.15. And determining the final state according to the probability threshold.
In the scheme, the state identification of the isolating switch is realized through multi-sensor data fusion, data processing and machine learning technologies. The time synchronization processing of the data ensures the consistency of the multi-modal data. Feature extraction and model training can then optimize the accuracy of state recognition through a large number of data examples.
Through executing the steps, acquiring the electrical signal data of the isolating switch to be detected through the electrical signal sensor arranged on the isolating switch to be detected, and carrying out data analysis on the electrical signal data to obtain an analysis data set; acquiring an image data set of an isolating switch to be detected through a preset image acquisition device, and simultaneously, performing point cloud data mapping on the image data set to obtain a point cloud data set; collecting an infrared data set of an isolating switch to be detected through a preset infrared collecting device, and simultaneously, carrying out data time synchronization processing on the infrared data set through analyzing the data set and carrying out data time synchronization processing on a point cloud data set through analyzing the data set to obtain a point cloud data set; performing multi-mode data fusion on the target infrared light data set and the target point cloud data set to obtain multi-mode data, and performing feature extraction on the multi-mode data to obtain a target feature vector; and inputting the target feature vector into a preset state recognition model to recognize the state of the isolating switch, and obtaining target state data of the isolating switch to be detected. In the scheme, through collecting various data sources (electric signals, images, infrared light, point clouds and the like), multi-mode data fusion and feature extraction, more comprehensive and multi-angle information can be provided. The accuracy of state identification can be improved, and false alarm and missing report are reduced. Combining data of different sensor types, such as electrical signals, images and infrared light, enables more comprehensive information to be obtained. Multimodal data fusion may provide more features to better capture the complexity of the disconnector state. Through real-time data acquisition and processing, the system can monitor the disconnecting switch to be detected in real time, quickly detect state changes, and even detect potential faults and abnormal conditions, so that the safety and reliability of the system are improved, and the accuracy rate of the disconnecting switch during state identification is further improved.
In a specific embodiment, the process of executing step S101 may specifically include the following steps:
(1) Collecting electric signal data of an isolating switch to be detected through an electric signal sensor, wherein the electric signal data comprise current signal data and voltage signal data;
(2) Extracting the time sequence of the current signal data to obtain a first time sequence corresponding to the current signal data;
(3) Extracting the time sequence of the voltage signal data to obtain a second time sequence corresponding to the voltage signal data;
(4) Performing signal filtering processing on the current signal data based on the first time sequence to obtain filtered current signal data;
(5) Performing signal filtering processing on the voltage signal data based on the second time sequence to obtain filtered voltage signal data;
(6) Performing current waveform analysis on the filtered current signal data to obtain current waveform data, and performing voltage waveform analysis on the filtered voltage signal data to obtain voltage waveform data;
(7) And performing data format conversion on the current waveform data and the voltage waveform data to obtain an analysis data set.
Specifically, it is assumed that an isolating switch to be detected is provided, and a current sensor and a voltage sensor are installed to collect electrical signal data in real time. The sensor collects data at a frequency of once per second.
For example: at a specific point in time, sampled values of the following current and voltage signal data are obtained:
current signal data: [5.2A,5.3A,5.1A,5.4A,5.0A ];
voltage signal data: [220V,221V,219V,222V,218V ].
These data represent the current and voltage signals over time.
The acquired data is arranged in a time sequence to form a current signal time sequence and a voltage signal time sequence.
For example: the following time series was obtained:
current signal time sequence: [ t1, t2, t3, t4, t5];
voltage signal time sequence: [ t6, t7, t8, t9, t10].
These time series represent measurements of the current and voltage signals at different points in time.
And digitally filtering the current signal and the voltage signal to remove noise and interference and obtain a filtered signal.
For example: by applying a digital filter, the following filtered signal data is obtained:
filtering the current signal data: [5.2A,5.3A,5.2A,5.4A,5.3A ]
Filtering the voltage signal data: [220V,221V,220V,222V,221V ]
Now, a current waveform analysis is performed on the filtered current signal data to extract characteristics of the current waveform. For example: the following current waveform analysis was performed:
Peak detection: peaks and troughs in the current waveform are identified, for example, a peak of 5.4A is detected and a trough of 5.1A.
Frequency analysis: the frequency of the current waveform is calculated, for example, to give a frequency of 50Hz.
Waveform classification: the current waveform is compared with the waveform of the normal operation mode to determine whether an abnormality exists, for example, the waveform is found to conform to the normal operation mode.
And finally, carrying out data format conversion on the current waveform analysis result and the voltage signal data to obtain an analysis data set. For example: converting the characteristic values (peak value and frequency) of the current waveform and the voltage signal data according to a specified format to obtain an analysis data set, wherein the analysis data set comprises the following steps:
current waveform characteristics: { peak: 5.4A, frequency: 50Hz };
voltage signal data: [220V,221V,220V,222V,221V ].
In a specific embodiment, as shown in fig. 2, the process of executing step S102 may specifically include the following steps:
s201, acquiring an image data set of a disconnecting switch to be detected through a preset image acquisition device;
s202, calibrating the position of the isolating switch to be detected to obtain position data corresponding to the isolating switch to be detected;
s203, constructing a space coordinate system through the position data, and extracting image depth data from the image data set based on the space coordinate system to obtain the image depth data set;
S204, performing point cloud data mapping on the image data set in a space coordinate system based on the image depth data set to obtain the point cloud data set.
It should be noted that, a preset intelligent camera is used to continuously collect the image data set of the isolating switch to be detected at a certain frame rate. These images may include the switch itself and its surrounding environment.
For example: the camera acquires one minute of image data at a rate of 30 frames per second.
For each acquired image, a computer vision technique is used for position calibration. This means that the position of the disconnector to be detected in the image is found and converted into pixel coordinates. For example: for an image, the computer vision algorithm successfully detects the position of the switch button and marks it on the image at pixel coordinates (200, 300).
Using the known position calibration information, a three-dimensional space coordinate system is constructed to map the image into three-dimensional space. And according to the internal parameters and the external parameters of the camera and the position of the isolating switch to be detected. For example: according to the focal length and the optical center of the camera and the position of the isolating switch to be detected in the world coordinate system, the relation between the camera coordinate system and the world coordinate system can be established.
Depth information of pixel points in each image is acquired so that subsequent point cloud data can be generated. This may be achieved by using depth sensors, stereo vision or structured light, etc. For example: the depth information acquisition is performed by the structured light sensor, which can provide a depth value for each pixel point.
Mapping the image data into a three-dimensional space to generate point cloud data. This is a critical step involving combining the color information of each pixel with the corresponding depth information to obtain a three-dimensional point cloud. For example: for each pixel in the image, it may be mapped into a three-dimensional coordinate system using its color information and depth values, resulting in a set of point cloud data.
In a word, if one intelligent camera is used to collect one minute of image data, the position of the isolating switch to be detected can be calibrated through a computer vision technology, a space coordinate system is constructed, depth information is extracted, the image is mapped to a three-dimensional space, and finally a point cloud data set is generated. This point cloud data may play an important role in multimodal data fusion and state recognition.
In a specific embodiment, the process of executing step S103 may specifically include the following steps:
(1) Collecting an infrared data set of the isolating switch to be detected through an infrared collecting device;
(2) Performing time stamp extraction on the infrared light data set to obtain first time stamp data corresponding to the infrared light data set;
(3) Performing time stamp synchronous analysis on the first time sequence and the second time sequence to obtain corresponding second time stamp data;
(4) And carrying out data time synchronization processing on the infrared light data set based on the first time stamp data and the second time stamp data to obtain a target infrared light data set.
Specifically, an infrared acquisition device is used for acquiring infrared light data of the isolating switch to be detected. These data contain the infrared radiation signals emitted by the switch at different points in time. For example: the infrared acquisition device continuously acquires infrared light data within one minute.
Timestamp information is extracted from the collected infrared light data. The time stamp may be time information for each data point, typically in milliseconds or microseconds. For example: for each infrared light data point, a corresponding timestamp is extracted, for example: 0ms, 10ms, 20ms, etc.
The extracted time stamp data is divided into two sequences: a first time sequence and a second time sequence. The two sequences may come from different sources or devices and require time stamp synchronization analysis to ensure that they correspond on the same time axis. For example: it is assumed that there are two time stamp sequences, one from the infrared acquisition device and the other from the other sensor. It is necessary to find the time offset between them and synchronize them together.
And carrying out data time synchronization processing on the infrared light data set based on the time synchronization information of the first time stamp data and the second time stamp data. The infrared light data points are matched with their correct time stamps to ensure that they correspond on the time axis. For example: from the time synchronization analysis, it is known that the time offset of the first time series is 10ms. Thus, the time stamp of the second time series is subtracted by 10ms to synchronize the two time series. Next, the infrared light data points are matched with the corresponding time stamps.
In summary, it is assumed that infrared light data is collected within one minute, while a time stamp for each data point is extracted. If the infrared light data is from two different devices, the time offset between them can be determined and synchronized together through time stamp synchronization analysis. And then, matching the time stamp with the infrared light data points to obtain a target infrared light data set, and ensuring that each data point corresponds on a correct time axis.
In a specific embodiment, the process of executing the step S105 may specifically include the following steps:
(1) Extracting first characteristics of the target infrared light data set to obtain an infrared characteristic vector corresponding to the target infrared light data;
(2) Extracting a point cloud projection contour of the target point cloud data to obtain corresponding point cloud projection contour data;
(3) Carrying out second feature extraction on the point cloud projection contour data through preset template contour data to obtain a point cloud feature vector corresponding to the target point cloud data;
(4) Carrying out vector connection processing on the infrared characteristic vector and the point cloud characteristic vector to obtain multi-mode data;
(5) And extracting the characteristics of the multi-mode data to obtain a target characteristic vector.
Specifically, each infrared image in the target infrared data set is converted into a digitized data representation, typically a matrix. Important features of the infrared image, such as infrared brightness, edges, shape, etc., are extracted by, for example, filtering, edge detection, feature extraction algorithms, etc. These extracted features may constitute an infrared feature vector, where each feature corresponds to a dimension in the vector. And carrying out point cloud projection on the target point cloud data, and projecting the three-dimensional point cloud onto a plane. By processing the projected plane data, contour information of the point cloud is extracted, which may include coordinates, shape, boundary points, and the like of the contour. The preset template profile data is typically a known, labeled profile shape, which may be the coordinates of a set of profile points. Comparing the projected profile data of the target point cloud with the preset template profile data, some profile matching or alignment algorithm, such as least squares or feature point matching, may be used. The information obtained from the comparison is used to extract point cloud feature vectors describing the shape of the target point cloud data and the degree of correspondence to the template contours. And connecting the extracted infrared characteristic vector and the point cloud characteristic vector extracted in the third step into a larger vector. This can join two vectors together to form a new multi-modal data vector. For example, assume that there is a set of infrared images, each image being 128x128 pixels, representing a different scene. The average brightness of each image may be extracted as an infrared feature using image processing techniques. This will result in a set of infrared feature vectors. Each point cloud represents a three-dimensional object in the scene. These point clouds may be projected onto the XY plane and contour information extracted. For each point cloud, the coordinates of a set of contour points are obtained, which points describe the shape of the object on a plane. A set of template profile data, such as a human profile, is preset. The contours of the target point cloud are compared to these templates. The matching degree between the target point cloud and the human body template is calculated through a contour matching algorithm, and the matching degree can be expressed as a score. And (3) connecting the obtained infrared characteristic vector and the matching degree obtained in the step (3) together as a point cloud characteristic vector to form a multi-mode data vector.
And further extracting the characteristics of the connected multi-mode data vector. By training the model, high-level features in the multimodal data are extracted to better represent key information of the target. Finally, the feature vector obtained from the feature extraction step of the multi-modal data is the target feature vector.
In a specific embodiment, the process of performing the second feature extraction step on the point cloud projection profile data through the preset template profile data may specifically include the following steps:
(1) Carrying out data alignment processing on the template contour data and the point cloud projection contour data to obtain alignment contour data;
(2) Performing feature descriptor calculation on the aligned profile data to obtain a corresponding feature descriptor set, wherein the feature descriptor set comprises; profile curvature data and profile normal data;
(3) And performing feature vector conversion on the feature descriptor set to obtain a point cloud feature vector corresponding to the target point cloud data.
Specifically, first, the template profile data and the point cloud projection profile data are subjected to data alignment processing to ensure that they have the same coordinate system and scale. This may involve translation, rotation and scaling operations.
After alignment, the template profile data and the point cloud projection profile data should be aligned in the same coordinate space for comparison and feature calculation.
The aligned profile data is used to calculate feature descriptors. In this example, the feature descriptor includes contour curvature data and contour normal data.
a. Profile curvature data:
for each point, its curvature is calculated to quantify the degree of curvature of the profile. Common methods of calculating curvature include the use of sliding window methods or local fitting methods.
The curvature value of each point is taken as one of the features, forming a curvature feature set.
b. Contour normal data:
the normal is a vector perpendicular to the curved surface for describing directional information of the contour point. The normals are typically calculated by fitting a surface using a least squares method or other surface fitting technique.
For each contour point, its normal vector is calculated and the normal is taken as one of the features, forming a normal feature set.
The contour curvature data and the contour normal data are combined into a point cloud feature vector. The curvature and normal data are connected together to form a point cloud feature vector, wherein the features of each point include curvature and normal. For the target point cloud data, curvature and normal data are also calculated and combined into a point cloud feature vector to describe the shape and structure of the target point cloud.
In a specific embodiment, the process of executing step S106 may specifically include the following steps:
(1) Inputting the target feature vector into a state recognition model to disassemble the feature values, so as to obtain a disassembled feature value data set;
(2) Randomly sampling the disassembled characteristic value data sets to obtain a plurality of sampled data sets;
(3) Respectively constructing a decision tree of each sampling data set to obtain a target decision tree corresponding to each sampling data set;
(4) Respectively inputting each sampling data set into a target decision tree corresponding to each sampling data set to perform switch state identification, so as to obtain a plurality of state identification results;
(5) And carrying out state weighted fusion on the plurality of state recognition results to obtain target state data of the isolating switch to be detected.
First, a target feature vector is input into a state recognition model. This model may be a machine learning model, such as a deep neural network, for identifying the state of the disconnector.
The model output is typically a vector containing a plurality of eigenvalues, representing the probabilities or confidence of the different states. This vector is broken down into separate sets of eigenvalue data.
The feature value data sets are randomly sampled to create a plurality of different sampled data sets. Each sampled data set contains feature values randomly selected from the original set of feature values. Doing so may generate multiple samples to verify the robustness and robustness of the model.
For each sampled data set, an independent decision tree model is constructed. Decision trees are a supervised learning algorithm used to classify or predict feature values. Each sampled data set is used to train a corresponding decision tree model so that each model can independently make a state recognition decision. The feature values in each sampled dataset are used to input them into the corresponding target decision tree model, respectively. Each target decision tree model will perform state recognition based on the eigenvalues, determining the state (e.g., on or off) of the disconnector. For each sampled data set, an independent state recognition result will be obtained, representing state predictions across different data samples. A plurality of state recognition results will be obtained, each generated by an independent decision tree model.
And carrying out weighted fusion on the plurality of state recognition results to obtain final target state data. This may be done by voting or weighted averaging, etc. For example, the state recognition results may be weighted according to the performance of each model and then averaged.
The embodiment of the invention also provides a state identification system of the isolating switch, as shown in fig. 3, which specifically comprises:
The analysis module 301 is configured to collect electrical signal data of the to-be-detected isolating switch through an electrical signal sensor installed on the to-be-detected isolating switch, and perform data analysis on the electrical signal data to obtain an analysis data set;
the acquisition module 302 is configured to acquire, by using a preset image acquisition device, an image data set of the to-be-detected disconnecting switch, and simultaneously, perform point cloud data mapping on the image data set to obtain a point cloud data set;
the synchronization module 303 is configured to collect an infrared data set of the to-be-detected isolating switch through a preset infrared acquisition device, and perform data time synchronization processing on the infrared data set through the analysis data set to obtain a target infrared data set;
the processing module 304 is configured to perform data time synchronization processing on the point cloud data set through the resolved data set to obtain a target point cloud data set;
the fusion module 305 is configured to perform multi-modal data fusion on the target infrared light data set and the target point cloud data set to obtain multi-modal data, and perform feature extraction on the multi-modal data to obtain a target feature vector;
And the identification module 306 is configured to input the target feature vector into a preset state identification model to identify the state of the isolating switch, so as to obtain target state data of the isolating switch to be detected.
Through the cooperative work of the modules, the electric signal data of the isolating switch to be detected is collected through an electric signal sensor arranged on the isolating switch to be detected, and the electric signal data is subjected to data analysis to obtain an analysis data set; acquiring an image data set of an isolating switch to be detected through a preset image acquisition device, and simultaneously, performing point cloud data mapping on the image data set to obtain a point cloud data set; collecting an infrared data set of an isolating switch to be detected through a preset infrared collecting device, and simultaneously, carrying out data time synchronization processing on the infrared data set through analyzing the data set and carrying out data time synchronization processing on a point cloud data set through analyzing the data set to obtain a point cloud data set; performing multi-mode data fusion on the target infrared light data set and the target point cloud data set to obtain multi-mode data, and performing feature extraction on the multi-mode data to obtain a target feature vector; and inputting the target feature vector into a preset state recognition model to recognize the state of the isolating switch, and obtaining target state data of the isolating switch to be detected. In the scheme, through collecting various data sources (electric signals, images, infrared light, point clouds and the like), multi-mode data fusion and feature extraction, more comprehensive and multi-angle information can be provided. The accuracy of state identification can be improved, and false alarm and missing report are reduced. Combining data of different sensor types, such as electrical signals, images and infrared light, enables more comprehensive information to be obtained. Multimodal data fusion may provide more features to better capture the complexity of the disconnector state. Through real-time data acquisition and processing, the system can monitor the disconnecting switch to be detected in real time, quickly detect state changes, and even detect potential faults and abnormal conditions, so that the safety and reliability of the system are improved, and the accuracy rate of the disconnecting switch during state identification is further improved.
The above embodiments are only for illustrating the technical aspects of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the embodiments, it should be understood by those of ordinary skill in the art that: modifications and equivalents may be made to the specific embodiments of the invention without departing from the spirit and scope of the invention, which is intended to be covered by the scope of the claims.
Claims (6)
1. The method for identifying the state of the isolating switch is characterized by comprising the following steps of:
the method comprises the steps of collecting electric signal data of the isolating switch to be detected through an electric signal sensor arranged on the isolating switch to be detected, and carrying out data analysis on the electric signal data to obtain an analysis data set, wherein the method specifically comprises the following steps of: collecting electric signal data of the isolating switch to be detected through the electric signal sensor, wherein the electric signal data comprise current signal data and voltage signal data; extracting the current signal data in a time sequence to obtain a first time sequence corresponding to the current signal data; extracting the time sequence of the voltage signal data to obtain a second time sequence corresponding to the voltage signal data; performing signal filtering processing on the current signal data based on the first time sequence to obtain filtered current signal data; performing signal filtering processing on the voltage signal data based on the second time sequence to obtain filtered voltage signal data; performing current waveform analysis on the filtered current signal data to obtain current waveform data, and performing voltage waveform analysis on the filtered voltage signal data to obtain voltage waveform data; performing data format conversion on the current waveform data and the voltage waveform data to obtain the analysis data set;
Acquiring an image data set of the isolating switch to be detected through a preset image acquisition device, and simultaneously, performing point cloud data mapping on the image data set to obtain a point cloud data set;
the infrared data set of the isolating switch to be detected is acquired through a preset infrared acquisition device, and meanwhile, the infrared data set is subjected to data time synchronization processing through the analysis data set to obtain a target infrared data set, and the method specifically comprises the following steps: collecting an infrared light data set of the isolating switch to be detected through the infrared collecting device; performing time stamp extraction on the infrared light data set to obtain first time stamp data corresponding to the infrared light data set; performing timestamp synchronous analysis on the first time sequence and the second time sequence to obtain corresponding second timestamp data; based on the first timestamp data and the second timestamp data, performing data time synchronization processing on the infrared light data set to obtain the target infrared light data set;
performing data time synchronization processing on the point cloud data set through the analysis data set to obtain a target point cloud data set;
Performing multi-mode data fusion on the target infrared light data set and the target point cloud data set to obtain multi-mode data, and performing feature extraction on the multi-mode data to obtain a target feature vector;
and inputting the target feature vector into a preset state recognition model to recognize the state of the isolating switch, and obtaining the target state data of the isolating switch to be detected.
2. The method for identifying the state of the disconnecting switch according to claim 1, wherein the step of acquiring the image data set of the disconnecting switch to be detected by a preset image acquisition device, and simultaneously performing point cloud data mapping on the image data set to obtain the point cloud data set comprises the following steps:
collecting an image data set of the isolating switch to be detected through a preset image collecting device;
calibrating the position of the isolating switch to be detected to obtain position data corresponding to the isolating switch to be detected;
constructing a space coordinate system through the position data, and extracting image depth data from the image data set based on the space coordinate system to obtain an image depth data set;
and carrying out point cloud data mapping on the image data set in the space coordinate system based on the image depth data set to obtain the point cloud data set.
3. The method for identifying the state of the isolating switch according to claim 1, wherein the step of performing multi-modal data fusion on the target infrared light data set and the target point cloud data set to obtain multi-modal data, and performing feature extraction on the multi-modal data to obtain a target feature vector includes:
extracting first characteristics of the target infrared light data set to obtain an infrared characteristic vector corresponding to the target infrared light data;
extracting a point cloud projection contour from the target point cloud data to obtain corresponding point cloud projection contour data;
performing second feature extraction on the point cloud projection profile data through preset template profile data to obtain a point cloud feature vector corresponding to the target point cloud data;
vector connection processing is carried out on the infrared characteristic vector and the point cloud characteristic vector to obtain the multi-mode data;
and extracting the characteristics of the multi-mode data to obtain the target characteristic vector.
4. The method for identifying a state of an isolating switch according to claim 3, wherein the step of extracting the second feature of the point cloud projection profile data by using preset template profile data to obtain the point cloud feature vector corresponding to the target point cloud data comprises the following steps:
Carrying out data alignment processing on the template contour data and the point cloud projection contour data to obtain alignment contour data;
performing feature descriptor calculation on the aligned profile data to obtain a corresponding feature descriptor set, wherein the feature descriptor set comprises; profile curvature data and profile normal data;
and performing feature vector conversion on the feature descriptor set to obtain a point cloud feature vector corresponding to the target point cloud data.
5. The method for identifying the state of the isolating switch according to claim 1, wherein the step of inputting the target feature vector into a preset state identification model to identify the state of the isolating switch and obtaining the target state data of the isolating switch to be detected comprises the steps of:
inputting the target feature vector into the state recognition model to disassemble the feature values, so as to obtain a disassembled feature value data set;
randomly sampling the disassembled characteristic value data sets to obtain a plurality of sampled data sets;
respectively constructing decision trees of each sampling data set to obtain a target decision tree corresponding to each sampling data set;
respectively inputting each sampling data set into a target decision tree corresponding to each sampling data set to perform switch state identification, so as to obtain a plurality of state identification results;
And carrying out state weighted fusion on the plurality of state identification results to obtain the target state data of the isolating switch to be detected.
6. A state recognition system of an isolating switch for performing the state recognition method of an isolating switch according to any one of claims 1 to 5, comprising:
the analysis module is used for collecting the electrical signal data of the isolating switch to be detected through the electrical signal sensor arranged on the isolating switch to be detected, and carrying out data analysis on the electrical signal data to obtain an analysis data set, and specifically comprises the following steps: collecting electric signal data of the isolating switch to be detected through the electric signal sensor, wherein the electric signal data comprise current signal data and voltage signal data; extracting the current signal data in a time sequence to obtain a first time sequence corresponding to the current signal data; extracting the time sequence of the voltage signal data to obtain a second time sequence corresponding to the voltage signal data; performing signal filtering processing on the current signal data based on the first time sequence to obtain filtered current signal data; performing signal filtering processing on the voltage signal data based on the second time sequence to obtain filtered voltage signal data; performing current waveform analysis on the filtered current signal data to obtain current waveform data, and performing voltage waveform analysis on the filtered voltage signal data to obtain voltage waveform data; performing data format conversion on the current waveform data and the voltage waveform data to obtain the analysis data set;
The acquisition module is used for acquiring an image data set of the isolating switch to be detected through a preset image acquisition device, and simultaneously, performing point cloud data mapping on the image data set to obtain a point cloud data set;
the synchronous module is used for acquiring the infrared light data set of the isolating switch to be detected through a preset infrared acquisition device, and simultaneously, carrying out data time synchronous processing on the infrared light data set through the analysis data set to obtain a target infrared light data set, and specifically comprises the following steps: collecting an infrared light data set of the isolating switch to be detected through the infrared collecting device; performing time stamp extraction on the infrared light data set to obtain first time stamp data corresponding to the infrared light data set; performing timestamp synchronous analysis on the first time sequence and the second time sequence to obtain corresponding second timestamp data; based on the first timestamp data and the second timestamp data, performing data time synchronization processing on the infrared light data set to obtain the target infrared light data set;
the processing module is used for carrying out data time synchronization processing on the point cloud data set through the analysis data set to obtain a target point cloud data set;
The fusion module is used for carrying out multi-mode data fusion on the target infrared light data set and the target point cloud data set to obtain multi-mode data, and carrying out feature extraction on the multi-mode data to obtain a target feature vector;
and the identification module is used for inputting the target feature vector into a preset state identification model to identify the state of the isolating switch, so as to obtain the target state data of the isolating switch to be detected.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311389065.5A CN117113283B (en) | 2023-10-25 | 2023-10-25 | State identification method and system of isolating switch |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311389065.5A CN117113283B (en) | 2023-10-25 | 2023-10-25 | State identification method and system of isolating switch |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117113283A CN117113283A (en) | 2023-11-24 |
| CN117113283B true CN117113283B (en) | 2024-01-26 |
Family
ID=88800678
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311389065.5A Active CN117113283B (en) | 2023-10-25 | 2023-10-25 | State identification method and system of isolating switch |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117113283B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108389137A (en) * | 2018-02-06 | 2018-08-10 | 国网山西省电力公司电力科学研究院 | Power fault detection early warning system based on infared spectrum technology |
| CN110008877A (en) * | 2019-03-27 | 2019-07-12 | 国网内蒙古东部电力有限公司 | A kind of substation isolating-switch detection recognition method based on Faster RCNN |
| CN111563335A (en) * | 2020-07-01 | 2020-08-21 | 国网江西省电力有限公司电力科学研究院 | Laser radar-based high-voltage isolating switch closing state detection method |
| CN115761427A (en) * | 2022-11-23 | 2023-03-07 | 华南理工大学 | Device and method for accurately identifying state of high-voltage isolating switch |
| CN116108397A (en) * | 2022-12-22 | 2023-05-12 | 福建亿榕信息技术有限公司 | Electric power field operation violation identification method integrating multi-mode data analysis |
| CN116563272A (en) * | 2023-06-29 | 2023-08-08 | 深圳优立全息科技有限公司 | Isolating switch on-off state identification method based on high-precision point cloud and related device |
| CN116778288A (en) * | 2023-06-19 | 2023-09-19 | 燕山大学 | Multi-mode fusion target detection system and method |
-
2023
- 2023-10-25 CN CN202311389065.5A patent/CN117113283B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108389137A (en) * | 2018-02-06 | 2018-08-10 | 国网山西省电力公司电力科学研究院 | Power fault detection early warning system based on infared spectrum technology |
| CN110008877A (en) * | 2019-03-27 | 2019-07-12 | 国网内蒙古东部电力有限公司 | A kind of substation isolating-switch detection recognition method based on Faster RCNN |
| CN111563335A (en) * | 2020-07-01 | 2020-08-21 | 国网江西省电力有限公司电力科学研究院 | Laser radar-based high-voltage isolating switch closing state detection method |
| CN115761427A (en) * | 2022-11-23 | 2023-03-07 | 华南理工大学 | Device and method for accurately identifying state of high-voltage isolating switch |
| CN116108397A (en) * | 2022-12-22 | 2023-05-12 | 福建亿榕信息技术有限公司 | Electric power field operation violation identification method integrating multi-mode data analysis |
| CN116778288A (en) * | 2023-06-19 | 2023-09-19 | 燕山大学 | Multi-mode fusion target detection system and method |
| CN116563272A (en) * | 2023-06-29 | 2023-08-08 | 深圳优立全息科技有限公司 | Isolating switch on-off state identification method based on high-precision point cloud and related device |
Non-Patent Citations (2)
| Title |
|---|
| 基于ReliefF特征量优化及BP神经网络识别的高压隔离开关故障类型与位置诊断方法;张一茗，等.;高压电器(第02期);全文 * |
| 面向深度学习的多模态融合技术研究综述;何俊，等.;计算机工程;第46卷(第5期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117113283A (en) | 2023-11-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110826538B (en) | Abnormal off-duty identification system for electric power business hall | |
| CN109409272B (en) | Cable acceptance system and method based on machine vision | |
| CN109188148B (en) | Transmission line of electricity applied to smart grid reliably monitors system | |
| CN105447471A (en) | Infrared detection based device gas leakage identification method and apparatus | |
| CN111047568A (en) | Steam leakage defect detection and identification method and system | |
| CN112146761A (en) | Human body temperature measurement compensation method based on machine learning | |
| CN113688817A (en) | Instrument identification method and system for automatic inspection | |
| CN107369162B (en) | Method and system for generating insulator candidate target area | |
| CN117498225B (en) | Unmanned aerial vehicle intelligent power line inspection system | |
| CN116821409B (en) | Production safety management method and system based on image data processing | |
| CN117113283B (en) | State identification method and system of isolating switch | |
| WO2024119983A1 (en) | Anomality detection method and apparatus for cable terminal of rail vehicle | |
| KR101743595B1 (en) | Parital discharge diagnosis method and system, and mold transfomer deterioration monitoring system using the same. | |
| CN205228747U (en) | SF6 gas leakage discernment monitored control system based on infrared detection | |
| CN112468721A (en) | Visual acquisition method and device with automatic focusing function | |
| CN111951161A (en) | Target identification method and system and inspection robot | |
| CN114662594B (en) | Target feature recognition analysis system | |
| CN115797280A (en) | Transformer surface defect detection system based on image AI recognition algorithm | |
| CN114594102B (en) | Machine vision-based data line interface automatic detection method | |
| CN114243925B (en) | Intelligent fusion terminal-based distribution substation allergy sensing method and system | |
| CN115147591A (en) | Transformer equipment infrared image voltage heating type defect diagnosis method and system | |
| CN113702786A (en) | K-means-based multi-parameter suspension insulator insulation state evaluation method | |
| CN114529493A (en) | Cable appearance defect detection and positioning method based on binocular vision | |
| CN112200784A (en) | Intelligent defect diagnosis method for electrical equipment | |
| CN112051493A (en) | Hidden danger type identification method and device for power distribution network equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |