CN111738148B - Fault identification method using infrared inspection shooting - Google Patents
Fault identification method using infrared inspection shooting Download PDFInfo
- Publication number
- CN111738148B CN111738148B CN202010573632.2A CN202010573632A CN111738148B CN 111738148 B CN111738148 B CN 111738148B CN 202010573632 A CN202010573632 A CN 202010573632A CN 111738148 B CN111738148 B CN 111738148B
- Authority
- CN
- China
- Prior art keywords
- equipment
- detected
- camera
- inspection
- inspected
- 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 77
- 238000007689 inspection Methods 0.000 title claims abstract description 73
- 238000001514 detection method Methods 0.000 claims abstract description 58
- 230000011218 segmentation Effects 0.000 claims abstract description 22
- 238000003709 image segmentation Methods 0.000 claims abstract description 16
- 230000007547 defect Effects 0.000 claims description 35
- 238000009529 body temperature measurement Methods 0.000 claims description 6
- 238000010586 diagram Methods 0.000 claims description 4
- 230000000007 visual effect Effects 0.000 claims description 4
- 230000003287 optical effect Effects 0.000 claims description 2
- 239000003990 capacitor Substances 0.000 description 7
- 238000013528 artificial neural network Methods 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 239000012212 insulator Substances 0.000 description 4
- 238000012015 optical character recognition Methods 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000000523 sample Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/20—Scenes; Scene-specific elements in augmented reality scenes
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J19/00—Accessories fitted to manipulators, e.g. for monitoring, for viewing; Safety devices combined with or specially adapted for use in connection with manipulators
- B25J19/02—Sensing devices
- B25J19/04—Viewing devices
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2413—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on distances to training or reference patterns
- G06F18/24133—Distances to prototypes
- G06F18/24137—Distances to cluster centroïds
- G06F18/2414—Smoothing the distance, e.g. radial basis function networks [RBFN]
-
- 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
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/143—Sensing or illuminating at different wavelengths
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mechanical Engineering (AREA)
- Robotics (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Radiation Pyrometers (AREA)
Abstract
The invention provides a fault identification method utilizing infrared inspection shooting, and belongs to the field of equipment management. The invention comprises the following steps: step 1, shooting a panoramic image after a camera is set to a most wide-angle state, and storing the panoramic image; step 2, detecting and image segmentation is carried out on the panoramic image to obtain a panoramic segmentation image, and the equipment with faults in the equipment to be detected is primarily determined; step 3, determining the number of the equipment to be detected in the panoramic segmentation map, and setting the detection sequence of the inspection of the equipment to be detected; step 4, shooting the equipment to be detected to obtain a single picture of the equipment to be detected; step 5, identifying the single picture of the equipment to be detected; step 6, image segmentation is carried out on the single to-be-detected equipment picture, and whether the to-be-detected equipment corresponding to the single to-be-detected equipment picture has faults or not is determined; and 7, generating a detection report of the equipment to be detected. The invention can realize twice detection and judge whether the fault exists for each device in a short time.
Description
Technical Field
The invention belongs to the field of equipment management, and particularly relates to a fault identification method utilizing infrared inspection shooting.
Background
The fault detection of the traditional transformer substation is mainly carried out through manual detection. Because of the large number of transformer substation equipment, a large number of professionals are needed, and manpower and material resources are very consumed.
In the prior art, although the intelligent inspection robot is widely applied to equipment fault detection, whether equipment has faults or not is detected by measuring the temperature of the equipment through the inspection robot provided with the infrared probe, because the equipment in a transformer substation is numerous, the equipment cannot be positioned on specific equipment at any time, and therefore, only reference can be made.
Disclosure of Invention
The present invention has been made to solve the above problems, and an object of the present invention is to provide a fault recognition method using infrared inspection shooting, which reduces the labor consumption and cost of a transformer substation.
The invention provides a fault identification method using infrared inspection shooting, which is used for carrying out fault identification on equipment to be inspected in an area to be inspected by using an inspection robot with a camera of an infrared lens, wherein the area to be inspected is provided with a robot operation points, the inspection robot stores an area diagram to be inspected, an information list of all the equipment to be inspected and position information of all the robot operation points, and the information list is used for recording the type and the number of the equipment to be inspected and has the characteristics that: step 1, the inspection robot moves to an nth robot operation point, a camera is set to be in a most wide-angle state, then a panoramic image is shot, the panoramic image is stored, and step 2 is performed; step 2, detecting a panoramic image, generating a detection frame outside each device to be detected in the panoramic image, identifying the type of each device to be detected, then carrying out pixel level image segmentation on images in all detection frames to obtain a panoramic segmentation image, carrying out infrared pixel level temperature measurement on a device area in the panoramic segmentation image by using an infrared lens, preliminarily determining the device with a fault in the device to be detected by using a preset fault determination method, and entering step 3; step 3, determining the number of the equipment to be detected in the panoramic segmentation map as b, setting the detection sequence of the inspection of the b equipment to be detected, and entering step 4; step 4, shooting the mth equipment to be detected by the inspection robot according to the detection sequence to obtain a single picture of the equipment to be detected, and entering step 5; step 5, identifying a single to-be-detected equipment picture, confirming the type and number of to-be-detected equipment corresponding to the single to-be-detected equipment picture, and entering step 6; step 6, image segmentation is carried out on a single to-be-detected equipment picture to obtain a single equipment segmentation picture, infrared pixel level temperature measurement is carried out on an equipment area of the single equipment segmentation picture, whether to-be-detected equipment corresponding to the single to-be-detected equipment picture has faults or not is determined by using a preset fault determination method, and step 7 is carried out; step 7, generating a detection report of the equipment to be detected, deleting the information of the equipment to be detected from the information list, and entering step 8; step 8, judging whether m is equal to b, if yes, entering step 9, otherwise, enabling m=m+1, and returning to step 4; and 9, judging whether n is equal to a, ending when judging that n is equal to a, otherwise, enabling n=n+1, and returning to the step 1, wherein a, b, n and m are positive integers which are more than or equal to 1, and the initial values of n and m are 1.
The fault identification method using infrared inspection shooting provided by the invention can also have the following characteristics: in step 2, the algorithm for detecting the panorama is a fastercnn algorithm.
The fault identification method using infrared inspection shooting provided by the invention can also have the following characteristics: the algorithm of performing pixel-level image segmentation on the images in all the detection frames in the step 2 and performing image segmentation on the single to-be-detected equipment picture in the step 6 is a maskrnn algorithm.
The fault identification method using infrared inspection shooting provided by the invention can also have the following characteristics: in step 3, it is further determined whether the detection frame generated in step 2 exceeds the boundary of the panorama, if yes, the device to be detected in the detection frame is not added to the detection sequence, otherwise no operation is performed.
The fault identification method using infrared inspection shooting provided by the invention can also have the following characteristics: in step 5, the method for identifying the single to-be-detected equipment picture is optical character identification.
The fault identification method using infrared inspection shooting provided by the invention can also have the following characteristics: the most wide-angle state of the camera is a pitch angle of 0 degrees, a yaw angle of 0 degrees, and the focal length is adjusted to maximize the visual field range of the camera.
The fault identification method using infrared inspection shooting provided by the invention can also have the following characteristics: in step 4, the method for shooting the mth equipment to be inspected by the inspection robot includes the following steps: step 4-1, adjusting the camera to a most wide-angle state; step 4-2, the inspection robot adjusts the pitch angle of a camera, in the process of adjusting the pitch angle, the camera shoots a video stream, pictures in the video stream are intercepted at equal intervals, the intercepted pictures are detected, a video stream detection frame is generated outside equipment to be inspected in the intercepted pictures, the pitch angle of the camera is alpha when the video stream detection frame is closest to the center of the vertical direction of a camera screen, and the pitch angle of the camera is adjusted to alpha; step 4-3, the inspection robot adjusts the yaw angle of the camera, the camera shoots video streams in the process of adjusting the yaw angle, pictures in the video streams are intercepted at equal intervals, the intercepted pictures are detected, a video stream detection frame is generated outside equipment to be inspected in the intercepted pictures, the yaw angle of the camera is beta when the video stream detection frame is closest to the center of the horizontal direction of a screen of the camera, and the yaw angle of the camera is adjusted to be beta; and 4-4, the inspection robot adjusts the focal length of the camera, in the focal length adjusting process, the camera shoots a video stream, the pictures in the video stream are intercepted at equal intervals, the intercepted pictures are detected, a video stream detection frame is generated outside the equipment to be inspected in the intercepted pictures, and the picture when the video stream detection frame is closest to 80% of the screen area of the camera is taken as the picture of the m single equipment to be inspected.
The fault identification method using infrared inspection shooting provided by the invention can also have the following characteristics: when the type of the equipment to be detected is a capacitor, the predetermined fault judging method comprises the following steps: step 1, determining the highest temperature in the equipment area as T 1 Average temperature T 2 The method comprises the steps of carrying out a first treatment on the surface of the Step 2, when T 1 -T 2 < 2, determining that there is no fault in the capacitor, when T 1 -T 2 And (3) not less than 2, and determining that the capacitor has an emergency defect.
The fault identification method using infrared inspection shooting provided by the invention can also have the following characteristics: when the type of the equipment to be detected is a casing, the predetermined fault judging method comprises the following steps: step 1, determining the equipmentMaximum temperature in the region is T 1 Average temperature T 2 The ambient temperature is T 0 And orderStep 2, when T 1 -T 2 At 10 or less, determining that the casing is free of failure, when T 1 -T 2 When > 10 or 0.8 > r is more than or equal to 0.35, determining that the sleeve has general defects, when T 1 When more than 55 or 0.95 > r is more than or equal to 0.8, determining that the sleeve has serious defects, and when T 1 And when more than 80 or r is more than or equal to 0.95, determining that the sleeve has an emergency defect.
The fault identification method using infrared inspection shooting provided by the invention can also have the following characteristics: in the step 3, the priority sequence of the inspection sequence of the b to-be-inspected devices is set to be the to-be-inspected device with the emergency defect, the to-be-inspected device with the serious defect, the to-be-inspected device with the general defect and the to-be-inspected device without the fault in sequence.
Effects and effects of the invention
According to the fault identification method using infrared inspection shooting, when the inspection robot reaches an operation point, equipment which possibly has faults in equipment to be inspected is primarily judged by shooting a panoramic image of the operation point, and then single equipment is shot for judging again, so that the fault identification method using infrared inspection shooting can realize twice detection on each equipment in a short time.
Drawings
Fig. 1 is a route diagram of a patrol robot for patrol of a certain transformer substation using a fault recognition method using infrared patrol shooting in an embodiment of the present invention; and
fig. 2 is a flowchart of a fault identification method using infrared inspection photography in an embodiment of the present invention.
Detailed Description
In order to make the technical means, the creation features, the achievement of the purpose and the effect of the present invention easy to understand, the present invention is specifically described below with reference to the embodiments and the drawings.
In the following embodiments, image segmentation refers to a technique or process of dividing an image into a number of specific, distinct regions and presenting objects of interest.
In the following examples, the neural network framework of the fastercn algorithm is a classical neural network framework, and the trained dataset is an image containing electrical equipment. And (3) an expert in the electric power field analyzes the image, and manually marks an identification frame of the electric power equipment in the image to form a training set. During training, the neural network has the best detection result on the verification set by adjusting the training super-parameters. The fastercnn algorithm outputs power devices and their locations in the form of a detection box.
In the following examples, the neural network framework of the maskrnn algorithm is a classical neural network framework and the trained dataset is an image containing electrical equipment. And (3) an expert in the electric power field analyzes the image, and manually marks the image segmentation result of the pixel level of the electric power equipment in the image to form a training set. During training, the neural network has the best image segmentation result on the verification set by adjusting the training super-parameters. The maskrnn algorithm identifies a device region and a background region at the pixel level in the detection frame.
In the following embodiments, the inspection robot has a pan-tilt and a dual-lens camera, where the dual-lens is a visible lens and an infrared lens, respectively, and each of the two lenses can realize automatic alignment of pixel levels, the visible lens is mounted on the pan-tilt, the pan-tilt can control a pitch angle and a yaw angle of the camera, and the camera can zoom and execute a shooting command.
< example >
Fig. 1 is a route diagram of a patrol robot for patrol of a certain substation using a fault recognition method using infrared patrol shooting in an embodiment of the present invention.
As shown in fig. 1, the five-pointed star in the figure represents the initial position of the inspection robot. The black solid line represents the passageway, all passageway both sides all are provided with the equipment that awaits measuring, and all the equipment outside of awaiting measuring all has a nameplate of writing this equipment type and serial number. The black boxes represent a total of a robot work points, and in this embodiment a total of 12 robot work points, i.e., a=12. The lines with arrows represent the travel paths of the inspection robot between different work points. The area within the dashed box is the area to be inspected.
In the present embodiment, the inspection robot stores a map of an area to be inspected, an information list of all the devices to be inspected (including the types and numbers of all the devices to be inspected), position information of all the robot work points, and a moving route between the work points.
Fig. 2 is a flowchart of a fault identification method using infrared inspection photography in an embodiment of the present invention.
As shown in fig. 2, the fault identification method using infrared inspection shooting provided in this embodiment includes the following steps:
step 1, the inspection robot moves to an nth robot working point, the camera is set to be in a most wide-angle state (namely a pitch angle is 0 degrees, a yaw angle is 0 degrees, a focal length is adjusted to enable the visual field of the camera to be maximum), then a panoramic image is shot, the panoramic image is stored, a starting value of n is 1, and step 2 is entered;
step 2, detecting a panoramic image by using a trained fasternn algorithm, generating a detection frame outside each device to be detected in the panoramic image, identifying the type of each device to be detected, then performing pixel-level image segmentation on images in all detection frames by using a maskrnn algorithm to obtain a panoramic segmentation image, wherein the panoramic segmentation image comprises a device area and a background area, performing infrared pixel-level temperature measurement on the device area in the panoramic segmentation image by using an infrared lens, and initially determining the device with a fault in the device to be detected by using a preset fault determination method, and entering step 3;
in this embodiment, the fault flow of the temperature identification device is performed with reference to the power industry standard of the people's republic of China (DLT-664-2016, the specification of infrared diagnostic application of electrified devices), specifically, the predetermined fault determination method is different according to the type of the device to be detected, and specifically, the method is as follows:
when the type of the equipment to be detected is a transformer, the predetermined fault judging method comprises the following steps:
step 1, determining the highest temperature in the equipment areaIs T 1 Average temperature T 2 The ambient temperature is T 0 And order
Step 2, when T 1 -T 2 When the temperature is less than or equal to 15 or r is less than 0.35, determining that the transformer has no fault, when T 1 -T 2 When more than 15 or 0.8 > r is more than or equal to 0.35, determining that the transformer has general defects, and when T is 1 When more than 80 or 0.95 > r is more than or equal to 0.8, determining that the transformer has serious defects, and when T is 1 And when more than 110 or r is more than or equal to 0.95, determining that the transformer has emergency defects.
When the type of the equipment to be detected is a capacitor, the predetermined fault judging method comprises the following steps:
step 1, determining the highest temperature in the equipment area as T 1 Average temperature T 2 ；
Step 2, when T 1 -T 2 < 2, determining that there is no fault in the capacitor, when T 1 -T 2 And (2) determining that the capacitor has an emergency defect.
When the type of the equipment to be detected is a casing, the predetermined fault judging method comprises the following steps:
step 1, determining the highest temperature in the equipment area as T 1 Average temperature T 2 The ambient temperature is T 0 And order
Step 2, when T 1 -T 2 At 10 or less, determining that the casing is free of failure, when T 1 -T 2 When > 10 or 0.8 > r is more than or equal to 0.35, determining that the sleeve has general defects, when T 1 When more than 55 or 0.95 > r is more than or equal to 0.8, determining that the sleeve has serious defects, and when T 1 And when more than 80 or r is more than or equal to 0.95, determining that the sleeve has an emergency defect.
When the type of the equipment to be detected is a lightning arrester, the predetermined fault judging method comprises the following steps:
step 1, determining the highest temperature in the equipment area as T 1 Average temperature T 2 ；
Step 2, when T 1 -T 2 < 0.5, determining that the arrester is not faulty, when T 1 -T 2 And (5) not less than 0.5, and determining that the lightning arrester has an emergency defect.
When the type of the equipment to be detected is an insulator, the predetermined fault judging method comprises the following steps:
step 1, determining the highest temperature in the equipment area as T 1 Average temperature T 2 ；
Step 2, when T 1 -T 2 < 0.5, determining that said insulator is free of faults, when T 1 -T 2 And (5) not less than 0.5, and determining that the insulator has an emergency defect.
When the type of the equipment to be detected is inductance, the predetermined fault judging method comprises the following steps:
step 1, determining the highest temperature in the equipment area as T 1 Average temperature T 2 The ambient temperature is T 0 And order
Step 2, when T 1 -T 2 When less than or equal to 10, determining that the inductor has no fault, when T 1 -T 2 When more than 10 or 0.8 > r is more than or equal to 0.35, determining that the inductance has general defects, and when T 1 When more than 55 or 0.95 > r is more than or equal to 0.8, determining that the inductance has serious defects, and when T is 1 And when more than 80 or r is more than or equal to 0.95, determining that the inductance has an emergency defect.
When the equipment to be detected is a breaker, the predetermined fault judging method comprises the following steps:
step 1, determining the highest temperature in the equipment area as T 1 Average temperature T 2 The ambient temperature is T 0 And order
Step 2, when T 1 -T 2 When less than or equal to 10, determining that the breaker has no fault, when T 1 -T 2 When more than 10 or 0.8 > r is more than or equal to 0.35, determining that the breaker has common defects, and when T is 1 When more than 55 or 0.95 > r is more than or equal to 0.8, determining that the breaker has serious defects, and when T is 1 And when more than 80 or r is more than or equal to 0.95, determining that the circuit breaker has an emergency defect.
When the type of the equipment to be detected is an isolating switch, the predetermined fault judging method comprises the following steps:
step 1, determining the highest temperature in the equipment area as T 1 Average temperature T 2 The ambient temperature is T 0 And order
Step 2, when T 1 -T 2 When the temperature is less than or equal to 10 or r is less than 0.35, determining that the isolating switch has no fault, and when T 1 -T 2 When more than 10 or 0.8 > r is more than or equal to 0.35, determining that the isolating switch has common defects, and when T is 1 When more than 55 or 0.95 > r is more than or equal to 0.8, determining that the isolating switch has serious defects, and when T is 1 When more than 80 or r is more than or equal to 0.95, determining that the isolating switch has emergency defects.
When the type of the equipment to be detected is a current transformer, the predetermined fault judging method comprises the following steps:
step 1, determining the highest temperature in the equipment area as T 1 Average temperature T 2 The ambient temperature is T 0 And order
Step 2, when T 1 -T 2 When the current transformer is less than or equal to 10, determining that the current transformer has no fault, when T 1 -T 2 When more than 10 or 0.8 > r is more than or equal to 0.35, determining that the current transformer has common defects, and when T is 1 When more than 55 or 0.95 > r is more than or equal to 0.8, determining that the current transformer has serious defects, and when T is more than or equal to 1 And when more than 80 or r is more than or equal to 0.95, determining that the current transformer has emergency defects.
When the type of the equipment to be detected is a voltage transformer, the predetermined fault judging method comprises the following steps:
step 1, determining the highest temperature in the equipment area as T 1 Average temperature T 2 ；
Step 2, when T 1 -T 2 < 2, determining that the voltage transformer has no fault, when T 1 -T 2 And (3) not less than 2, and determining that the voltage transformer has an emergency defect.
When the type of the equipment to be inspected is a radiator, the predetermined fault judging method comprises the following steps:
step 1, determining the highest temperature in the equipment area as T 1 Average temperature T 2 The ambient temperature is T 0 And order
Step 2, when T 1 -T 2 When the temperature is less than or equal to 15 or r is less than 0.35, determining that the radiator has no fault, when T 1 -T 2 When more than 15 or 0.8 > r is more than or equal to 0.35, determining that the radiator has general defects, when T 1 When more than 80 or 0.95 > r is more than or equal to 0.8, determining that the radiator has serious defects, and when T is 1 And when more than 110 or r is more than or equal to 0.95, determining that the radiator has an emergency defect.
Step 3, judging whether each equipment to be detected detection frame in the panoramic segmentation map reaches the boundary of the panoramic segmentation map or not respectively, if yes, not adding the equipment to be detected in the detection frame into a detection sequence, otherwise, not performing any operation, determining the number of the equipment to be detected in the panoramic segmentation map as b, setting the detection sequence of the inspection of the b pieces of equipment to be detected, and entering step 4;
in this embodiment, the priority sequence is sequentially a device to be inspected with an emergency fault, a device to be inspected with a serious fault, a device to be inspected with a general fault, and a device to be inspected without a fault, if a plurality of devices to be inspected with the same fault level exist, the priority sequence is determined according to the device type, the priority sequence of the device type is sequentially a transformer, a capacitor, a sleeve, a lightning arrester, an insulator, an inductor, a circuit breaker, a disconnecting switch, a current transformer, a voltage transformer, and a radiator, and in other embodiments, the detection sequence can be adjusted according to actual requirements;
step 4, shooting the mth equipment to be detected by the inspection robot according to the detection sequence, obtaining a single picture of the equipment to be detected, and entering step 5, wherein the initial value of m is 1;
in this embodiment, the method for photographing the mth device to be inspected by the inspection robot in step 4 includes the following steps:
step 4-1, adjusting the camera to the most wide-angle state (namely, a pitch angle of 0 degrees and a yaw angle of 0 degrees), and adjusting a focal length to maximize the visual field range of the camera;
step 4-2, the inspection robot adjusts the pitch angle of a camera, in the process of adjusting the pitch angle, the camera shoots a video stream, pictures in the video stream are intercepted at equal intervals, the intercepted pictures are detected, a video stream detection frame is generated outside equipment to be inspected in the intercepted pictures, the pitch angle of the camera is alpha when the video stream detection frame is closest to the center of the vertical direction of a camera screen, and the pitch angle of the camera is adjusted to alpha;
step 4-3, the inspection robot adjusts the yaw angle of the camera, the camera shoots video streams in the process of adjusting the yaw angle, pictures in the video streams are intercepted at equal intervals, the intercepted pictures are detected, a video stream detection frame is generated outside equipment to be inspected in the intercepted pictures, the yaw angle of the camera is beta when the video stream detection frame is closest to the center of the horizontal direction of a screen of the camera, and the yaw angle of the camera is adjusted to be beta;
and 4-4, the inspection robot adjusts the focal length of the camera, in the focal length adjusting process, the camera shoots a video stream, the pictures in the video stream are intercepted at equal intervals, the intercepted pictures are detected, a video stream detection frame is generated outside the equipment to be inspected in the intercepted pictures, and the picture when the video stream detection frame is closest to 80% of the screen area of the camera is taken as the picture of the m single equipment to be inspected.
Step 5, performing Optical Character Recognition (OCR) on the single to-be-detected equipment picture, confirming the type and number of the to-be-detected equipment corresponding to the single to-be-detected equipment picture, and entering step 6; in some special cases, the name number of the nameplate of the device is recognized by OCR and is not in the device list, and at this time, the recognized name number of the device can be compared with the name number of the device on the map, and if the recognized name number of the device is mostly the same, the device can be confirmed.
Step 6, image segmentation is carried out on a single to-be-detected equipment picture to obtain a single equipment segmentation picture, infrared pixel level temperature measurement is carried out on an equipment area of the single equipment segmentation picture, whether to-be-detected equipment corresponding to the single to-be-detected equipment picture has faults or not is determined by using a preset fault determination method, and step 7 is carried out; the predetermined failure determination method used in step 6 in this embodiment is the same as the predetermined failure determination method used in step 2, and will not be described here, but in other embodiments, the predetermined failure determination methods used in step 2 and step 6 may be different;
step 7, generating a detection report of the equipment to be detected, deleting the information of the equipment to be detected from the information list, indicating that the equipment is already detected, and not detecting the equipment at a later operation point, and entering step 8;
step 8, judging whether m is equal to b, if yes, entering step 9, otherwise, enabling m=m+1, and returning to step 4;
step 9, judging whether n is equal to a, ending when judging yes, otherwise, enabling n=n+1, and returning to the step 1.
Effects and effects of the examples
According to the fault identification method using infrared inspection shooting according to the embodiment, when the inspection robot reaches an operation point, the equipment which possibly has faults in the equipment to be inspected is primarily judged by shooting the panorama of the operation point, and then the single equipment is shot for judging again, so that the fault detection can be realized twice for each equipment in a short time.
According to the fault identification method utilizing infrared inspection shooting, which is related to the embodiment, because a cradle head and a camera are adopted to control when shooting a certain specific device, the shooting pitch angle and the yaw angle of the camera are changed, and the camera is zoomed, so that the camera only shoots the specific device and occupies a fixed proportion of a screen, the shooting of each device can be realized, and the shot image is very clear and complete, thereby being beneficial to the subsequent detection and segmentation of the image.
According to the fault identification method using infrared inspection shooting according to the embodiment, since the temperature of each pixel point of the equipment area after image segmentation is calculated by using the infrared lens, whether faults exist is judged through further calculation, and the calculation method and the judgment standard are different for different types of equipment, the equipment fault judgment is more accurate.
According to the fault identification method utilizing infrared inspection shooting, which is related to the embodiment, the image is detected by adopting the fastermann algorithm trained in advance, and the image is segmented by adopting the maskrnn algorithm trained in advance, so that the method has high detection speed and accuracy on equipment types, and the segmentation of pictures is also quick and accurate, thereby laying a solid foundation for judging whether the equipment has faults or not.
The above embodiments are preferred examples of the present invention, and are not intended to limit the scope of the present invention.
Claims (8)
1. The fault identification method by utilizing infrared inspection shooting is used for carrying out fault identification on equipment to be inspected in an area to be inspected by using an inspection robot with a camera with an infrared lens, wherein the area to be inspected is provided with a robot operation points, the inspection robot stores an area diagram to be inspected, an information list of all the equipment to be inspected and position information of all the robot operation points, and the information list is used for recording the type and the number of the equipment to be inspected, and is characterized by comprising the following steps:
step 1, the inspection robot moves to an nth robot operation point, the camera is set to be in a most wide-angle state, then a panoramic image is shot, the panoramic image is stored, and step 2 is performed;
step 2, detecting the panoramic image, generating a detection frame outside each device to be detected in the panoramic image, identifying the type of each device to be detected, then carrying out pixel level image segmentation on images in all the detection frames to obtain a panoramic segmentation image, carrying out infrared pixel level temperature measurement on a device area in the panoramic segmentation image by using the infrared lens, preliminarily determining the device with a fault in the device to be detected by using a preset fault determination method, and entering step 3;
step 3, determining the number of the equipment to be detected in the panoramic segmentation map as b, setting the detection sequence of the inspection of the b equipment to be detected, and entering step 4;
step 4, according to the detection sequence, the inspection robot shoots the mth equipment to be inspected to obtain a single picture of the equipment to be inspected, and the step 5 is carried out;
step 5, identifying the single to-be-detected equipment picture, confirming the type and number of to-be-detected equipment corresponding to the single to-be-detected equipment picture, and entering step 6;
step 6, image segmentation is carried out on the single to-be-detected equipment picture to obtain a single equipment segmentation picture, infrared pixel level temperature measurement is carried out on an equipment area of the single equipment segmentation picture, a predetermined fault determination method is used for determining whether faults exist in to-be-detected equipment corresponding to the single to-be-detected equipment picture, and step 7 is carried out;
step 7, generating a detection report of the equipment to be detected, deleting the information of the equipment to be detected from the information list, and entering step 8;
step 8, judging whether m is equal to b, if yes, entering step 9, otherwise, enabling m=m+1, and returning to step 4;
step 9, judging whether n is equal to a, ending when judging yes, otherwise, making n=n+1, returning to step 1,
wherein a, b, n and m are positive integers greater than or equal to 1, the initial values of n and m are 1,
when the type of the equipment to be detected is a casing, the predetermined fault judging method comprises the following steps:
step 1, determining the highest temperature in the equipment area as T 1 Average temperature T 2 The ambient temperature is T 0 And order
Step 2, when T 1 -T 2 At 10 or less, determining that the casing is free of failure, when T 1 -T 2 When > 10 or 0.8 > r is more than or equal to 0.35, determining that the sleeve has general defects, when T 1 When more than 55 or 0.95 > r is more than or equal to 0.8, determining that the sleeve has serious defects, and when T 1 And when more than 80 or r is more than or equal to 0.95, determining that the sleeve has an emergency defect.
2. The fault identification method using infrared inspection shooting according to claim 1, wherein:
in step 2, the algorithm for detecting the panorama is a fastercnn algorithm.
3. The fault identification method using infrared inspection shooting according to claim 1, wherein:
the algorithm of performing pixel-level image segmentation on the images in all the detection frames in the step 2 and performing image segmentation on the single to-be-detected equipment picture in the step 6 is a maskrnn algorithm.
4. The fault identification method using infrared inspection shooting according to claim 1, wherein:
in step 3, it is further determined whether the detection frame generated in step 2 exceeds the boundary of the panorama, if yes, the device to be detected in the detection frame is not added to the detection sequence, otherwise no operation is performed.
5. The fault identification method using infrared inspection shooting according to claim 1, wherein:
in step 5, the method for identifying the single to-be-detected equipment picture is optical character identification.
6. The method for recognizing a fault by infrared inspection photographing according to claim 1, wherein,
the most wide-angle state of the camera is a pitch angle of 0 degrees, a yaw angle of 0 degrees, and a focal length is adjusted to maximize the visual field range of the camera.
7. The method for recognizing a fault by infrared inspection photographing according to claim 1, wherein,
in step 4, the method for shooting the mth equipment to be detected by the inspection robot includes the following steps:
step 4-1, adjusting the camera to a most wide-angle state;
step 4-2, the inspection robot adjusts the pitch angle of the camera, in the process of adjusting the pitch angle, the camera shoots video streams, images in the video streams are intercepted at equal intervals, the intercepted images are detected, a video stream detection frame is generated outside the equipment to be inspected in the intercepted images, the pitch angle of the camera is alpha when the video stream detection frame is closest to the center of the vertical direction of a camera screen, and the pitch angle of the camera is adjusted to alpha;
step 4-3, the inspection robot adjusts the yaw angle of the camera, in the process of adjusting the yaw angle, the camera shoots video streams, pictures in the video streams are intercepted at equal intervals, the intercepted pictures are detected, a video stream detection frame is generated outside the equipment to be inspected in the intercepted pictures, the yaw angle of the camera is determined to be beta when the video stream detection frame is closest to the center of the horizontal direction of a camera screen, and the yaw angle of the camera is adjusted to be beta;
and 4-4, the inspection robot adjusts the focal length of the camera, in the process of adjusting the focal length, the camera shoots a video stream, the pictures in the video stream are intercepted at equal intervals, the intercepted pictures are detected, a video stream detection frame is generated outside the equipment to be inspected in the intercepted pictures, and the picture when the video stream detection frame is closest to 80% of the screen area of the camera is taken as the picture of the m-th single equipment to be inspected.
8. The method for recognizing a fault by infrared inspection photographing according to claim 1, wherein,
in the step 3, the priority sequence of the inspection sequence of the b inspection equipment is set to be the inspection equipment with emergency faults, the inspection equipment with serious faults, the inspection equipment with general faults and the inspection equipment without faults in sequence.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010573632.2A CN111738148B (en) | 2020-06-22 | 2020-06-22 | Fault identification method using infrared inspection shooting |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010573632.2A CN111738148B (en) | 2020-06-22 | 2020-06-22 | Fault identification method using infrared inspection shooting |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111738148A CN111738148A (en) | 2020-10-02 |
| CN111738148B true CN111738148B (en) | 2024-02-20 |
Family
ID=72650353
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010573632.2A Active CN111738148B (en) | 2020-06-22 | 2020-06-22 | Fault identification method using infrared inspection shooting |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111738148B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115278016A (en) * | 2022-07-25 | 2022-11-01 | 烟台艾睿光电科技有限公司 | Infrared intelligent shooting method and device, infrared thermal imaging equipment and medium |
| CN117893933B (en) * | 2024-03-14 | 2024-05-24 | 国网上海市电力公司 | Unmanned inspection fault detection method and system for power transmission and transformation equipment |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003102706A1 (en) * | 2002-05-31 | 2003-12-11 | Fujitsu Limited | Remotely-operated robot, and robot self position identifying method |
| CN103001328A (en) * | 2012-11-19 | 2013-03-27 | 山东大学 | Fault diagnosis and assessment method of intelligent substation |
| WO2013067745A1 (en) * | 2011-11-09 | 2013-05-16 | 南方电网科学研究院有限公司 | Substation control center technology |
| CN103427366A (en) * | 2013-08-09 | 2013-12-04 | 江西理工大学 | Helicopter airborne system for controlling and recognizing power transmission line |
| CN106597185A (en) * | 2015-10-14 | 2017-04-26 | 山东鲁能智能技术有限公司 | Infrared analysis system and method of primary equipment of transformer station |
| CN110246175A (en) * | 2019-05-24 | 2019-09-17 | 国网安徽省电力有限公司检修分公司 | Intelligent Mobile Robot image detecting system and method for the panorama camera in conjunction with holder camera |
| CN110363878A (en) * | 2019-07-31 | 2019-10-22 | 国网天津市电力公司 | A kind of computer room method for inspecting based on image processing techniques |
| CN110796754A (en) * | 2019-09-17 | 2020-02-14 | 国网天津市电力公司 | Machine room inspection method based on image processing technology |
-
2020
- 2020-06-22 CN CN202010573632.2A patent/CN111738148B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2003102706A1 (en) * | 2002-05-31 | 2003-12-11 | Fujitsu Limited | Remotely-operated robot, and robot self position identifying method |
| WO2013067745A1 (en) * | 2011-11-09 | 2013-05-16 | 南方电网科学研究院有限公司 | Substation control center technology |
| CN103001328A (en) * | 2012-11-19 | 2013-03-27 | 山东大学 | Fault diagnosis and assessment method of intelligent substation |
| CN103427366A (en) * | 2013-08-09 | 2013-12-04 | 江西理工大学 | Helicopter airborne system for controlling and recognizing power transmission line |
| CN106597185A (en) * | 2015-10-14 | 2017-04-26 | 山东鲁能智能技术有限公司 | Infrared analysis system and method of primary equipment of transformer station |
| CN110246175A (en) * | 2019-05-24 | 2019-09-17 | 国网安徽省电力有限公司检修分公司 | Intelligent Mobile Robot image detecting system and method for the panorama camera in conjunction with holder camera |
| CN110363878A (en) * | 2019-07-31 | 2019-10-22 | 国网天津市电力公司 | A kind of computer room method for inspecting based on image processing techniques |
| CN110796754A (en) * | 2019-09-17 | 2020-02-14 | 国网天津市电力公司 | Machine room inspection method based on image processing technology |
Non-Patent Citations (2)
| Title |
|---|
| 基于Matlab的牵引变压器故障智能诊断方法研究;崔建强;;电气化铁道(第01期);全文 * |
| 基于图像的高压输电线间隔棒故障诊断方法;金立军;胡娟;闫书佳;高电压技术;第39卷(第5期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111738148A (en) | 2020-10-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2019111976A1 (en) | Object detection device, prediction model creation device, object detection method, and program | |
| CN110703800A (en) | Unmanned aerial vehicle-based intelligent identification method and system for electric power facilities | |
| CN111738148B (en) | Fault identification method using infrared inspection shooting | |
| WO2014058248A1 (en) | Image monitoring apparatus for estimating gradient of singleton, and method therefor | |
| CN103517041A (en) | Real-time full-view monitoring method and device based on multi-camera rotating scanning | |
| CN111738056B (en) | Heavy truck blind area target detection method based on improved YOLO v3 | |
| CN115359239A (en) | Wind power blade defect detection and positioning method and device, storage medium and electronic equipment | |
| CN113763484A (en) | Ship target positioning and speed estimation method based on video image analysis technology | |
| CN109447946B (en) | Overhead communication optical cable abnormality detection method | |
| CN114581760B (en) | Equipment fault detection method and system for machine room inspection | |
| CN112183311A (en) | Transformer substation equipment fault detection method based on bimodal data fusion | |
| CN109977850B (en) | Classroom name prompting method based on face recognition | |
| CN113160209A (en) | Target marking method and target identification method for building facade damage detection | |
| CN114037895A (en) | Unmanned aerial vehicle pole tower inspection image identification method | |
| CN113269744A (en) | Infrared intelligent identification method and system applied to defects of power transmission line equipment | |
| CN112260402A (en) | Method for monitoring state of intelligent substation inspection robot based on video monitoring | |
| CN112257514B (en) | Infrared vision intelligent detection shooting method for equipment fault inspection | |
| CN108615057B (en) | CNN-based abnormity identification method for cable tunnel lighting equipment | |
| CN113610858B (en) | Power line image real-time segmentation method based on self-supervision learning | |
| CN115147591A (en) | Transformer equipment infrared image voltage heating type defect diagnosis method and system | |
| CN114355083A (en) | Inspection robot fault identification method and system based on artificial intelligence algorithm | |
| CN115410154A (en) | Method for identifying thermal fault of electrical equipment of wind power engine room | |
| CN115049623A (en) | Device for analyzing pantograph contour through visual segmentation | |
| CN111583341B (en) | Cloud deck camera shift detection method | |
| CN110956640B (en) | Heterogeneous image edge point detection and registration method |
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 |