CN113657621A - Hidden danger monitoring method and system - Google Patents
Hidden danger monitoring method and system Download PDFInfo
- Publication number
- CN113657621A CN113657621A CN202110790050.4A CN202110790050A CN113657621A CN 113657621 A CN113657621 A CN 113657621A CN 202110790050 A CN202110790050 A CN 202110790050A CN 113657621 A CN113657621 A CN 113657621A
- Authority
- CN
- China
- Prior art keywords
- image
- distribution transformer
- transformer area
- temperature
- equipment
- 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.)
- Pending
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 29
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000009826 distribution Methods 0.000 claims abstract description 81
- 238000001514 detection method Methods 0.000 claims abstract description 23
- 238000010586 diagram Methods 0.000 claims abstract description 20
- 238000001914 filtration Methods 0.000 claims description 15
- 230000007613 environmental effect Effects 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 11
- 238000010438 heat treatment Methods 0.000 claims description 10
- 238000005259 measurement Methods 0.000 claims description 10
- 230000005540 biological transmission Effects 0.000 claims description 8
- 238000010295 mobile communication Methods 0.000 claims description 8
- 235000002566 Capsicum Nutrition 0.000 claims description 7
- 239000006002 Pepper Substances 0.000 claims description 7
- 241000722363 Piper Species 0.000 claims description 7
- 235000016761 Piper aduncum Nutrition 0.000 claims description 7
- 235000017804 Piper guineense Nutrition 0.000 claims description 7
- 235000008184 Piper nigrum Nutrition 0.000 claims description 7
- 238000011176 pooling Methods 0.000 claims description 7
- 150000003839 salts Chemical class 0.000 claims description 7
- 238000004891 communication Methods 0.000 claims description 5
- 238000005516 engineering process Methods 0.000 description 8
- 238000009529 body temperature measurement Methods 0.000 description 6
- 238000012549 training Methods 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 230000005855 radiation Effects 0.000 description 5
- 230000006872 improvement Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000013135 deep learning Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 230000008447 perception Effects 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 230000007704 transition Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000005856 abnormality Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000005562 fading Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000009545 invasion Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000005304 joining Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- G06T5/73—
Abstract
The embodiment of the application discloses a method and a system for monitoring hidden dangers, wherein the method comprises the following steps: acquiring an image of a distribution transformer area, inputting the image into a convolution layer, and extracting to obtain a characteristic diagram of the image, wherein the image is provided with a detection frame; inputting the characteristic diagram into an RPN layer, and acquiring a region candidate corresponding to the characteristic diagram; inputting the feature map and the region candidate into a RoI posing layer, and extracting to obtain a candidate feature map; inputting the candidate feature map into a full-connection layer, calculating the type of the region candidate according to the candidate feature map, determining the position of a detection frame in the image, judging whether the distribution transformer area has potential safety hazards or not according to the position, and determining the type of the potential safety hazards under the condition that the distribution transformer area has the potential safety hazards.
Description
Technical Field
The application belongs to the technical field of electric power engineering, and particularly relates to a hidden danger monitoring method and system.
Background
The development of the power industry is closely related to the whole society, and the guarantee of the safe and stable operation of power facilities is the core of all work. However, most distribution transformer stations in remote mountainous areas still stay in the traditional modes of regular inspection, periodic maintenance and fault first-aid repair due to weak communication signals, remote terrain and the like, and cannot monitor the operation conditions of equipment in a distribution room in real time, so that the accident influence is enlarged, and serious economic loss and power failure influence are caused. In addition, the remote power distribution room is unattended, and faults or accidents are caused by theft and illegal invasion; meanwhile, seasonal environment changes and adverse weather influences can cause environmental temperature/humidity changes of the power distribution room, and great influences are generated on equipment. Therefore, the current management mode of regular patrol cannot meet the maintenance requirement of the distribution transformer area.
Disclosure of Invention
The embodiment of the application aims to provide a hidden danger monitoring method and system, and the defect that a management mode of regular inspection in the prior art cannot meet the maintenance requirement of a distribution transformer area can be overcome.
In order to solve the technical problem, the present application is implemented as follows:
in a first aspect, an embodiment of the present application provides a method for monitoring hidden dangers, including:
acquiring an image of a distribution transformer area, inputting the image into a convolution layer, and extracting to obtain a characteristic diagram of the image, wherein the image is provided with a detection frame;
inputting the characteristic diagram into a regional candidate network (RPN) layer to obtain regional candidates corresponding to the characteristic diagram;
inputting the feature map and the region candidates into a region-of-interest pooling RoI posing layer, and extracting to obtain candidate feature maps;
inputting the candidate feature map into a full-connection layer, calculating the type of the region candidate according to the candidate feature map, determining the position of a detection frame in the image, judging whether the distribution transformer area has potential safety hazards or not according to the position, and determining the type of the potential safety hazards under the condition that the distribution transformer area has the potential safety hazards.
Further, before the acquiring the image of the distribution transformer area, the method further includes:
determining the signal coverage condition of the distribution transformer area through a measurement report;
aiming at the weak signal coverage area in the distribution transformer area, the mobile communication signal of the weak signal coverage area is enhanced by adjusting the azimuth angle and the downtilt angle of an antenna.
Further, after acquiring the image of the distribution transformer area, the method further includes:
removing Gaussian noise and salt and pepper noise in the image acquisition and transmission process through Gaussian filtering and median filtering;
and carrying out defogging treatment on the noise-reduced image.
Further, the defogging processing on the noise-reduced image specifically includes:
carrying out image enhancement on the denoised image;
and defogging the enhanced image by using a defogging algorithm for image restoration.
Further, the method further comprises:
setting an environment temperature reference object, a temperature rise threshold, a temperature difference threshold and a relative temperature difference threshold aiming at the equipment of the distribution transformer area;
collecting heat emitted by the surface of the equipment in the distribution transformer area to obtain the distribution condition of a thermal field on the surface of the equipment, and analyzing the distribution condition of the thermal field to obtain the surface temperature of the equipment;
and determining the heating state of the equipment according to the surface temperature, the environmental temperature reference object, the temperature rise threshold, the temperature difference threshold and the relative temperature difference threshold.
In a second aspect, an embodiment of the present application provides a hidden danger monitoring system, including:
the first acquisition module is used for acquiring an image of a distribution transformer area, inputting the image into a convolutional layer, and extracting to obtain a characteristic diagram of the image, wherein the image is provided with a detection frame;
a second obtaining module, configured to input the feature map into a RPN layer of a regional candidate network, and obtain a regional candidate corresponding to the feature map;
a third obtaining module, configured to input the feature map and the region candidates to a region-of-interest pooling RoI posing layer, and extract to obtain candidate feature maps;
the determining module is used for inputting the candidate feature map into a full-connection layer, calculating the type of the region candidate according to the candidate feature map, determining the position of the detection frame in the image, judging whether the distribution transformer area has potential safety hazards or not according to the position, and determining the type of the potential safety hazards under the condition that the distribution transformer area has the potential safety hazards.
Further, the system further comprises:
the communication module is used for determining the signal coverage condition of the distribution transformer area through a measurement report; aiming at the weak signal coverage area in the distribution transformer area, the mobile communication signal of the weak signal coverage area is enhanced by adjusting the azimuth angle and the downtilt angle of an antenna.
Further, the system further comprises:
the processing module is used for removing Gaussian noise and salt and pepper noise in the image acquisition and transmission process through Gaussian filtering and median filtering; and carrying out defogging treatment on the noise-reduced image.
Further, the processing module is specifically configured to perform image enhancement on the noise-reduced image; and defogging the enhanced image by using a defogging algorithm for image restoration.
Further, the system further comprises:
the temperature monitoring module is used for setting an environment temperature reference object, a temperature rise threshold value, a temperature difference threshold value and a relative temperature difference threshold value aiming at the equipment of the distribution transformer area; collecting heat emitted by the surface of the equipment in the distribution transformer area to obtain the distribution condition of a thermal field on the surface of the equipment, and analyzing the distribution condition of the thermal field to obtain the surface temperature of the equipment; and determining the heating state of the equipment according to the surface temperature, the environmental temperature reference object, the temperature rise threshold, the temperature difference threshold and the relative temperature difference threshold.
The embodiment of the application is through adopting hidden danger monitoring technology based on degree of depth study, the hidden danger in real time identification distribution transformer district, the degree of depth and the width of each equipment operation condition and each environmental information perception in the all-round improvement station, establish the multidimension early warning system, improve the safe operation guarantee, replace the manpower and patrol and examine, realize few and unmanned on duty, reduce the human cost, improve comprehensive benefits, realize that the management mode deals with to early warning after the fact, from extensive control to state aassessment, from the transition of regularly patrolling to state monitoring, thereby promote the intellectuality and the management lean in the distribution transformer district comprehensively.
Drawings
Fig. 1 is a schematic flowchart of a hidden danger monitoring method provided in an embodiment of the present application;
fig. 2 is a specific implementation architecture diagram of a hidden danger monitoring system according to an embodiment of the present application;
fig. 3 is a schematic structural diagram of a hidden danger monitoring system according to an embodiment of the present application.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings in combination with embodiments.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The terms first, second and the like in the description and in the claims of the present application are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It will be appreciated that the data so used may be interchanged under appropriate circumstances such that embodiments of the application may be practiced in sequences other than those illustrated or described herein, and that the terms "first," "second," and the like are generally used herein in a generic sense and do not limit the number of terms, e.g., the first term can be one or more than one.
The hidden danger monitoring method provided by the embodiment of the present application is described in detail below with reference to the accompanying drawings through specific embodiments and application scenarios thereof.
Referring to fig. 1, a schematic flow chart of a hidden danger monitoring method provided in an embodiment of the present application is shown, where the method includes the following steps:
and 104, inputting the candidate feature map into a full-connection layer, calculating the type of the region candidate according to the candidate feature map, determining the position of a detection frame in the image, judging whether the distribution transformer area has potential safety hazards or not according to the position, and determining the type of the potential safety hazards under the condition that the distribution transformer area has the potential safety hazards.
In the embodiment of the application, in order to improve the quality of the training samples, increase the diversity of the training samples, avoid overfitting of the model and improve the robustness of the model, a data enhancement technology is adopted. Sample data is enhanced from three aspects of morphology, color and unbalanced data, and an existing fast-RCNN neural network model is improved, and pictures firstly pass through Conv layers. As a CNN network target detection method, fast RCNN firstly uses a group of basic conv + relu + posing layers to extract feature maps of images. The feature maps are shared for subsequent RPN layers and full connection layers. The obtained feature maps are transferred to Region Proposal Networks (RPN Networks). In an improved RPN, firstly, a feature pyramid is utilized to extract enhanced semantic information from feature maps and increase the sensitivity of a detection system to small-scale targets, then, anchors are judged to be either positive or negative through softmax, and then, a multi-threshold accurate regression classification structure is utilized to regress and correct the anchors to obtain accurate prosals. The loss function of the RPN network is as follows:
where i denotes the ith anchor, P when the anchor is a positive samplei1, is a negative sample, then Pi*=0。tiDenotes the offset, t, of the predict box relative to the anchor boxiDenotes the offset of the ground true box relative to the anchor box, the learning objective is naturally to bring the former close to the latter value. L isclsAnd the network training is used for classifying anchors as positive and negative. L isclsTo log the softmax value. L isregFor bounding box regression network training, the formula is as follows:
wherein x, y, w, h respectively represent the center coordinate and the width and height of the box.
Then, the input feature maps and prosages are collected through a ROI Pooling (Region of interest Pooling) layer, the prossal feature maps are extracted after the information is integrated, and the feature maps are sent to a subsequent full-connection layer to judge the target category. And calculating the type of the proxy by using the proxy feature maps, and simultaneously carrying out secondary bounding box regression to obtain the final accurate position of the detection frame so as to judge whether the picture has potential safety hazards and the type of the potential safety hazards.
The method and the device are improved and optimized based on the existing target recognition Faster-RCNN algorithm framework, the image model data set is used for training iteration, the construction and the training of the image recognition model are completed, the hidden danger of the distribution and transformation platform area is recognized in real time by adopting the hidden danger monitoring technology based on deep learning, and the detection of the hidden danger of the distribution and transformation platform area is high in accuracy rate and high in detection speed. The depth and the breadth of each equipment operation condition and each environmental information perception in can all-round improvement station, establish multidimension early warning system, improve the safe operation guarantee, replace the manpower and patrol and examine, realize few and unmanned on duty, reduce the human cost, improve comprehensive benefit, realize the management mode from handling after the fact to advance the early warning, from extensive control to state evaluation, from regularly patrolling the transition of state monitoring to promote the intellectuality and the management of joining in marriage transformer substation area comprehensively.
Referring to fig. 2, a specific implementation architecture diagram of the hidden danger monitoring system provided in the embodiment of the present application is shown, and includes a data acquisition module, an intelligent analysis module, and a visual monitoring module.
Correspondingly, based on the above-mentioned architecture diagram, before the image of the distribution transformer area is acquired, normal communication with a remote area can be realized by using a signal enhancement technology, that is, the signal coverage condition of the distribution transformer area is determined through a measurement report; aiming at the weak signal coverage area in the distribution transformer area, the mobile communication signal of the weak signal coverage area is enhanced by adjusting the azimuth angle and the downtilt angle of an antenna.
Specifically, the signal coverage condition of a remote area is known through a measurement report, the state of an antenna feeder system in the area is checked for the weak coverage problem in the measurement report, and if an internal system has a problem in the operation process of the system, the signal transmission efficiency and quality are directly influenced. For a weak signal coverage area, the coverage rate and the signal strength of the signal are further improved based on the antenna in the antenna feed system. The mobile communication signals can be enhanced by considering, comparing and analyzing the aspects of operation parameters, transmitting angles, antenna heights and the like in a mode of adjusting the azimuth angle and the downtilt angle of the antenna, and the mode is most direct and simple.
Meanwhile, a multi-antenna linkage and delay combined signal transmitting mode is adopted, a plurality of signal transmitting points are arranged through field investigation of the most reasonable layout, and a plurality of repeated transmitting signals are received by utilizing a diversity technology. The method is characterized in that the high-order characteristics are learned by utilizing the low correlation of signal fading in a deep learning mode, and the modulation mode is accurately and robustly identified by constructing a classification network model, so that the strength of the acquired signal is enhanced, and the network model is optimized by utilizing an increasing signal library, so that the original signal is restored to the maximum extent, and the radio frequency spectrum analysis capability is improved. In the test process, signal emission in some areas is interfered by strong natural conditions, in the areas, the electromagnetic wave angle can be further considered, the electromagnetic wave emission position is further adjusted according to the characteristics of the electromagnetic wave and the actual shielding position of an obstacle, and the mobile communication signal is enhanced.
In addition, according to the actual situation of the distribution transformer area, the complementation of solar power generation and local transformer distribution can be adopted, the energy supply to system equipment is realized, and the stable operation of the system is ensured. Specifically, a method of forming a solar photovoltaic cell array by connecting a plurality of solar photovoltaic cell modules in series and parallel according to requirements is adopted. The solar photovoltaic cell array is a collective whole of a plurality of photovoltaic cells, is fixed on the bracket and is connected together through a lead, and is equivalent to a power supply, output voltage and power; the controller mainly controls charging and discharging of the storage battery, distributes and adjusts input and output of power, and the like. The photovoltaic cell converts the acquired solar energy into electric energy, the output power of the photovoltaic cell is directly determined by the radiation intensity and the temperature, and the output power of the photovoltaic cell can be expressed as:
wherein, PPVRThe rated capacity of the battery panel; f. ofPVThe derating factor of the cell panel is usually 0.8-0.95, and represents the influence coefficient of factors such as attenuation of the photovoltaic cell and dust on the photovoltaic cell array;represents a power temperature coefficient (%/C); t isCThe temperature of the surface of the battery at any time; t isC.STCThe temperature of the photovoltaic cell under the standard test condition is 25 ℃ internationally; gTFor the actual light radiation (kW/m) of the photovoltaic panel at the time2);GT.STCIs the light radiation quantity (kW/m) under the standard test condition2);TαThe ambient temperature at which the photovoltaic cell is located; NOCT (normal Operating Cell temperature) is a photovoltaic Cell with the radiation intensity of 800W/m2The temperature of the solar cell under the rated operation condition where the ambient temperature is 20C is generally 42C to 46C. In order to fully utilize resources to supply power continuously and avoid the phenomenon of power failure, a power supply provided by local power distribution can be used for supplying power. According to the actual situation of a project site, the local transformer station is provided with 220V and 180V power supplies, and according to the principle of mainly using solar clean energy to supply power, when solar energy cannot be used under special conditions, the local power supply is used to supply power to the composite energy power supply system, so that the reliability is greatly improved.
In addition, the defined polygonal warning area can be monitored, and whether irrelevant people enter the warning area or not can be detected. Specifically, coordinates of each point of a polygonal warning area defined by a user are obtained, a video frame image is divided, and image information of the polygonal warning area is obtained. And inputting the polygonal image into a face detection model, performing multi-face detection, and acquiring the central point coordinate of each detected face and a face image. And establishing a face tracker according to the coordinates of the central point of each face, comparing the face tracker with the previous frame of image, canceling the tracker when the number of the faces is reduced, establishing a new tracker when the number of the faces is increased, calculating the position of the centroid of the face in the current frame when the number of the faces is unchanged, determining the position change relationship of the face with the previous frame, and obtaining the face list of the current frame without face recognition. Inputting the detected face image into a face recognition model, extracting face features, traversing and comparing all faces in the current frame with a known face database, if the Euclidean distance between the comparison result and a feature descriptor of a known face Y is smaller than a set threshold value, considering the face X in the current frame as Y which is recognized by people, and otherwise, considering the face X as an illegal person. And when detecting that the operating personnel in the non-warning area enters the warning area, alarming, reminding monitoring personnel to take corresponding measures, and recording and storing illegal personnel.
In addition, after the image of the distribution transformer area is acquired, preprocessing such as defogging, image noise reduction and the like can be performed on the collected image of the distribution transformer area, namely, Gaussian noise and salt and pepper noise in the acquisition and transmission processes of the image are removed through Gaussian filtering and median filtering; and carrying out defogging treatment on the noise-reduced image.
Wherein, carry out defogging to the image after making an uproar falls, specifically include: carrying out image enhancement on the denoised image; and defogging the enhanced image by using a defogging algorithm for image restoration. Specifically, the frame extraction is carried out on the acquired video, and Gaussian filtering and median filtering are applied to remove Gaussian noise and salt and pepper noise in the acquisition and transmission processes of the digital image. And carrying out defogging treatment on the noise-reduced picture. And the defogging processing of the image is realized by the defogging algorithm in two directions of image enhancement and image restoration. Image enhanced defogging processing is firstly adopted for the image, and certain information is highlighted or weakened by improving the contrast of the image with the fog so as to reduce the influence of the fog on the image. Then, defogging is carried out on the image enhanced picture according to an atmosphere degradation model by using a defogging algorithm for image restoration, which is equivalent to inverse transformation during image forming.
In addition, the thermal infrared imager can be used for measuring the temperature of the key area of the transformer station, and the defects of the equipment in the equipment station can be judged by comprehensively considering the external heating and the internal heating of the equipment. Setting an environment temperature reference object, a temperature rise threshold, a temperature difference threshold and a relative temperature difference threshold for equipment in the distribution transformer area; collecting heat emitted by the surface of the equipment in the distribution transformer area to obtain the distribution condition of a thermal field on the surface of the equipment, and analyzing the distribution condition of the thermal field to obtain the surface temperature of the equipment; and determining the heating state of the equipment according to the surface temperature, the environmental temperature reference object, the temperature rise threshold, the temperature difference threshold and the relative temperature difference threshold.
Specifically, different heat emitted by the surfaces of the devices is collected, so that the distribution condition of the thermal field on the surfaces of the devices is obtained, and the temperature of the surfaces of the devices is presented through analysis and processing, so that the thermodynamic state of the surfaces of the devices can be determined by workers. The average temperature of the monitoring equipment is the final result obtained by infrared temperature measurement. In addition, the environmental temperature reference, the temperature rise, the temperature difference and the relative temperature difference are preset to eliminate the interference of external factors, so that the heating state of the temperature-measured equipment can be better obtained, and a correct measurement conclusion can be obtained. Wherein, the environmental temperature reference T0The temperature measuring device is an object used for reflecting the temperature of the surrounding environment, does not represent the temperature of the surrounding environment at the time, but is in the environment with the same attribute as the measured object, and can play a role in comparing the temperature measurement of the object. The temperature rise refers to the temperature of the measured object and the T in the environment with the same attribute0The difference in temperature is as follows:
Ts=Tk1-Tk2
wherein, TsFor a temperature rise, Tk1Surface temperature, T, of the object to be measuredk2Reference is made to the ambient temperature of the object.
The temperature difference refers to the temperature difference between different measured objects or different parts of the same measured object, as follows:
Tc=T1-T2
wherein, TcIs a temperature difference, T1Is a high temperature point, T2Is the low temperature point.
Relative temperature difference refers to the percentage of the temperature difference between different parts of the associated symmetric object divided by the ratio of the maximum temperature rise, as follows:
wherein, t1And T1Is the temperature difference and temperature value at the temperature measurement position, t2And T2Temperature difference and temperature value at symmetrical temperature measurement positions, T0Is a reference to the ambient temperature value of the object.
In order to better obtain the heating state of the tested equipment, a comparison body is arranged for measuring the temperature of the environment, determining the influence of the peripheral conditions on the temperature measurement result, eliminating the interference of external factors, and comparing and judging the temperature value by referring to the temperature of the equipment in normal operation and a temperature value library accumulated in abnormal operation, thereby being beneficial to obtaining a correct measurement conclusion.
In addition, in order to make up the defects that the visible light image lacks layering sense and has three-dimensional sense, auxiliary measures such as image brightness, actual correction, false color drawing contour lines and histograms for mathematical operation and the like can be adopted, and the problem that thermal image distribution diagrams of infrared radiation of a temperature-measured object are weak in signals can be solved to a certain extent.
According to the embodiment of the application, signal enhancement is carried out by adopting two modes of a multi-antenna linkage and delay combination signal transmitting mode and a deep learning mode to process the acquired signals, so that normal transmission of data is ensured; by using a local polygonal area warning technique based on mobile human body recognition, when an irrelevant person enters a warning area, immediately recognizing and generating early warning information; the collected images of the transformer area are preprocessed by adopting the technologies of defogging, Gaussian filtering noise reduction, image optimization and the like, so that the interference caused by environmental factors is avoided as much as possible, and the robustness of the model is ensured; the improved fast-RCNN algorithm is adopted according to the requirements of high detection accuracy and high speed of the hidden danger of the transformer area, the high accuracy is ensured, meanwhile, the real-time identification of the hidden danger can be realized, and compared with the common fast-RCNN algorithm, the accuracy of the detection of the hidden danger is improved from 91.2 percent to 97.6 percent; the infrared thermal imager is used for measuring the temperature of the key area of the transformer station, so that the error of temperature measurement is less than 2 ℃, and the equipment abnormality is effectively avoided, and the accident scale is further enlarged; in addition, in consideration of the problem of difficult power supply in remote areas, the power supply technology is adopted to supply power to the equipment used by the invention.
Referring to fig. 3, a schematic structural diagram of a hidden danger monitoring system provided in an embodiment of the present application is shown, where the system includes:
a first obtaining module 310, configured to obtain an image of a distribution transformer area, input the image into a convolutional layer, and extract a feature map of the image, where the image has a detection frame;
a second obtaining module 320, configured to input the feature map into a RPN layer of a regional candidate network, and obtain a regional candidate corresponding to the feature map;
a third obtaining module 330, configured to input the feature map and the region candidates into a region-of-interest pooling RoI posing layer, and extract to obtain candidate feature maps;
the determining module 340 is configured to input the candidate feature map into a full connection layer, calculate a category of the region candidate according to the candidate feature map, determine a position of the detection frame in the image, determine whether a potential safety hazard exists in the distribution transformer area according to the position, and determine the category of the potential safety hazard when the potential safety hazard exists in the distribution transformer area.
In addition, the above system further includes:
the communication module is used for determining the signal coverage condition of the distribution transformer area through a measurement report; aiming at the weak signal coverage area in the distribution transformer area, the mobile communication signal of the weak signal coverage area is enhanced by adjusting the azimuth angle and the downtilt angle of an antenna.
In addition, the above system further includes:
the processing module is used for removing Gaussian noise and salt and pepper noise in the image acquisition and transmission process through Gaussian filtering and median filtering; and carrying out defogging treatment on the noise-reduced image.
Specifically, the processing module is specifically configured to perform image enhancement on the noise-reduced image; and defogging the enhanced image by using a defogging algorithm for image restoration.
In addition, the above system further includes:
the temperature monitoring module is used for setting an environment temperature reference object, a temperature rise threshold value, a temperature difference threshold value and a relative temperature difference threshold value aiming at the equipment of the distribution transformer area; collecting heat emitted by the surface of the equipment in the distribution transformer area to obtain the distribution condition of a thermal field on the surface of the equipment, and analyzing the distribution condition of the thermal field to obtain the surface temperature of the equipment; and determining the heating state of the equipment according to the surface temperature, the environmental temperature reference object, the temperature rise threshold, the temperature difference threshold and the relative temperature difference threshold.
The hidden danger monitoring system provided by the embodiment of the application can realize each process realized in the method embodiments, and is not repeated here to avoid repetition.
The embodiment of the application is through adopting hidden danger monitoring technology based on degree of depth study, the hidden danger in real time identification distribution transformer district, the degree of depth and the width of each equipment operation condition and each environmental information perception in the all-round improvement station, establish the multidimension early warning system, improve the safe operation guarantee, replace the manpower and patrol and examine, realize few and unmanned on duty, reduce the human cost, improve comprehensive benefits, realize that the management mode deals with to early warning after the fact, from extensive control to state aassessment, from the transition of regularly patrolling to state monitoring, thereby promote the intellectuality and the management lean in the distribution transformer district comprehensively.
The virtual device in the embodiment of the present application may be a device, or may be a component, an integrated circuit, or a chip in a terminal.
The above description is only an example of the present invention, and is not intended to limit the present invention. Various modifications and alterations to this invention will become apparent to those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of the claims of the present invention.
Claims (10)
1. A hidden danger monitoring method is characterized by comprising the following steps:
acquiring an image of a distribution transformer area, inputting the image into a convolution layer, and extracting to obtain a characteristic diagram of the image, wherein the image is provided with a detection frame;
inputting the characteristic diagram into a regional candidate network (RPN) layer to obtain regional candidates corresponding to the characteristic diagram;
inputting the feature map and the region candidates into a region-of-interest pooling RoIploling layer, and extracting to obtain candidate feature maps;
inputting the candidate feature map into a full-connection layer, calculating the type of the region candidate according to the candidate feature map, determining the position of a detection frame in the image, judging whether the distribution transformer area has potential safety hazards or not according to the position, and determining the type of the potential safety hazards under the condition that the distribution transformer area has the potential safety hazards.
2. The method of claim 1, wherein before acquiring the image of the distribution transformer area, further comprising:
determining the signal coverage condition of the distribution transformer area through a measurement report;
aiming at the weak signal coverage area in the distribution transformer area, the mobile communication signal of the weak signal coverage area is enhanced by adjusting the azimuth angle and the downtilt angle of an antenna.
3. The method of claim 1, wherein after acquiring the image of the distribution transformer area, further comprising:
removing Gaussian noise and salt and pepper noise in the image acquisition and transmission process through Gaussian filtering and median filtering;
and carrying out defogging treatment on the noise-reduced image.
4. The method according to claim 3, wherein the defogging processing on the noise-reduced image specifically comprises:
carrying out image enhancement on the denoised image;
and defogging the enhanced image by using a defogging algorithm for image restoration.
5. The method of claim 1, further comprising:
setting an environment temperature reference object, a temperature rise threshold, a temperature difference threshold and a relative temperature difference threshold aiming at the equipment of the distribution transformer area;
collecting heat emitted by the surface of the equipment in the distribution transformer area to obtain the distribution condition of a thermal field on the surface of the equipment, and analyzing the distribution condition of the thermal field to obtain the surface temperature of the equipment;
and determining the heating state of the equipment according to the surface temperature, the environmental temperature reference object, the temperature rise threshold, the temperature difference threshold and the relative temperature difference threshold.
6. A hidden danger monitoring system, comprising:
the first acquisition module is used for acquiring an image of a distribution transformer area, inputting the image into a convolutional layer, and extracting to obtain a characteristic diagram of the image, wherein the image is provided with a detection frame;
a second obtaining module, configured to input the feature map into a RPN layer of a regional candidate network, and obtain a regional candidate corresponding to the feature map;
the third acquisition module is used for inputting the feature map and the region candidates into an interested region pooling RoIploling layer and extracting to obtain candidate feature maps;
the determining module is used for inputting the candidate feature map into a full-connection layer, calculating the type of the region candidate according to the candidate feature map, determining the position of the detection frame in the image, judging whether the distribution transformer area has potential safety hazards or not according to the position, and determining the type of the potential safety hazards under the condition that the distribution transformer area has the potential safety hazards.
7. The system of claim 6, further comprising:
the communication module is used for determining the signal coverage condition of the distribution transformer area through a measurement report; aiming at the weak signal coverage area in the distribution transformer area, the mobile communication signal of the weak signal coverage area is enhanced by adjusting the azimuth angle and the downtilt angle of an antenna.
8. The system of claim 6, further comprising:
the processing module is used for removing Gaussian noise and salt and pepper noise in the image acquisition and transmission process through Gaussian filtering and median filtering; and carrying out defogging treatment on the noise-reduced image.
9. The system of claim 8,
the processing module is specifically used for performing image enhancement on the denoised image; and defogging the enhanced image by using a defogging algorithm for image restoration.
10. The system of claim 6, further comprising:
the temperature monitoring module is used for setting an environment temperature reference object, a temperature rise threshold value, a temperature difference threshold value and a relative temperature difference threshold value aiming at the equipment of the distribution transformer area; collecting heat emitted by the surface of the equipment in the distribution transformer area to obtain the distribution condition of a thermal field on the surface of the equipment, and analyzing the distribution condition of the thermal field to obtain the surface temperature of the equipment; and determining the heating state of the equipment according to the surface temperature, the environmental temperature reference object, the temperature rise threshold, the temperature difference threshold and the relative temperature difference threshold.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110790050.4A CN113657621A (en) | 2021-07-13 | 2021-07-13 | Hidden danger monitoring method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110790050.4A CN113657621A (en) | 2021-07-13 | 2021-07-13 | Hidden danger monitoring method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113657621A true CN113657621A (en) | 2021-11-16 |
Family
ID=78489347
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110790050.4A Pending CN113657621A (en) | 2021-07-13 | 2021-07-13 | Hidden danger monitoring method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113657621A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115047814A (en) * | 2022-06-13 | 2022-09-13 | 国网山东省电力公司胶州市供电公司 | Pole climbing operation safety supervision system and method based on 5G |
-
2021
- 2021-07-13 CN CN202110790050.4A patent/CN113657621A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115047814A (en) * | 2022-06-13 | 2022-09-13 | 国网山东省电力公司胶州市供电公司 | Pole climbing operation safety supervision system and method based on 5G |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112734692B (en) | Defect identification method and device for power transformation equipment | |
US10269138B2 (en) | UAV inspection method for power line based on human visual system | |
WO2020199538A1 (en) | Bridge key component disease early-warning system and method based on image monitoring data | |
CN103487729B (en) | Based on the power equipments defect detection method that ultraviolet video and infrared video merge | |
CN105447471A (en) | Infrared detection based device gas leakage identification method and apparatus | |
CN103413150A (en) | Power line defect diagnosis method based on visible light image | |
CN112466000A (en) | Inspection system based on power inspection robot and inspection control method | |
CN103954897A (en) | Intelligent power grid high voltage insulation damage monitoring system and method based on ultraviolet imagery | |
CN108680833B (en) | Composite insulator defect detection system based on unmanned aerial vehicle | |
CN113284124A (en) | Photovoltaic panel defect detection method based on unmanned aerial vehicle vision | |
CN109345586A (en) | Electrical equipment discharge characteristic extracting method based on ultraviolet imagery technology | |
CN115184726B (en) | Smart power grid fault real-time monitoring and positioning system and method | |
Liao et al. | Using Matlab real-time image analysis for solar panel fault detection with UAV | |
CN115909092A (en) | Light-weight power transmission channel hidden danger distance measuring method and hidden danger early warning device | |
CN115642877A (en) | Photovoltaic module abnormal shielding detection method and system based on deep learning | |
Shou et al. | Defect detection with generative adversarial networks for electroluminescence images of solar cells | |
CN113485432A (en) | Photovoltaic power station electroluminescence intelligent diagnosis system and method based on unmanned aerial vehicle | |
CN113657621A (en) | Hidden danger monitoring method and system | |
CN117036825A (en) | Solar cell panel detection method, medium and system | |
CN107194923B (en) | Ultraviolet image diagnosis method for defect inspection of contact network power equipment | |
CN115113093A (en) | Visual insulator leakage current inspection method and equipment | |
Hwang et al. | Soiling detection for photovoltaic modules based on an intelligent method with image processing | |
CN113989209A (en) | Power line foreign matter detection method based on fast R-CNN | |
Shen et al. | Transmission line safety early warning technology based on multi-source data perception | |
Liu et al. | An efficient hot spot detection method with small sample learning for photovoltaic panels |
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 |