Identification system for welding points in power transmission line

Abstract

The invention relates to an identification system for welding points in a power transmission line, and the system comprises an intelligent identification mechanism which is used for inputting the total number of pixel points occupied by a foreground image region and various color component values of each pixel point in the foreground image region into a feedforward neural network which is trained for multiple times, obtaining an existence identifier representing whether a welding point exists in the foreground image area or not; and the command triggering mechanism is used for sending out a welding point non-detection command when the welding points do not exist in all foreground image areas in the continuous mapping picture. The identification system for the welding points in the power transmission line is compact in structure and easy and convenient to operate. The feedforward neural network can be adopted to intelligently judge whether the existence identification of the welding point exists in the foreground image area based on various visual data, and when the welding point does not exist in each foreground image area in the continuous mapping picture, the welding point non-detection command is sent out; therefore, the intelligent identification of whether the welding point exists at the wire connecting part of the power transmission line is realized.

Description

Technical Field

[0001] The present invention relates to the field of power transmission lines, and more specifically, to a system for identifying welding points in power transmission lines. Background Art

[0002] The basic process of power transmission is to create conditions for electromagnetic energy to be transmitted along the direction of the transmission line. The power transmission capacity of the line is governed by various laws of electromagnetic fields and circuits. Taking the earth potential as the reference point (zero potential), the line conductors must be under a high voltage applied by the power supply, which is called the transmission voltage. The maximum transmission power determined by the transmission line after comprehensive consideration of technical, economic and other factors is called the transmission capacity of the line. The transmission capacity is roughly proportional to the square of the transmission voltage. Therefore, increasing the transmission voltage is the main technical means to achieve large-capacity or long-distance power transmission, and it is also the main indicator of the development level of transmission technology.

[0003] However, for reasons such as convenience or irregular operation, the conductor connections of the transmission line may be welded at welding points, which poses a safety hazard to the use of the transmission line. Obviously, the prior art lacks a targeted technical solution for high-precision existence analysis to identify whether there are welding points at the conductor connections of the transmission line. Summary of the invention

[0004] In order to solve the technical problems in the prior art, the present invention provides a system for identifying welding points in a power transmission line, which can input the total number of pixels occupied by a foreground image area and the cyan component value, magenta component value, yellow component value and black component value of each pixel in the foreground image area in the CMYK space into a feedforward neural network that has completed multiple trainings to run the feedforward neural network that has completed multiple trainings, obtain an existence mark indicating whether there is a welding point in the foreground image area output by the feedforward neural network that has completed multiple trainings, and issue a welding point non-detection command when there is no welding point in each foreground image area in a continuous mapping picture, and issue a welding point detection command when there is a welding point in a certain foreground image area in the continuous mapping picture, thereby realizing intelligent identification of whether there is a welding point at the wire connection of the power transmission line, and reducing the probability of annealing caused by the presence of welding points at the wire connection of the power transmission line and the decrease in mechanical strength of the connection.

[0005] According to the present invention, a system for identifying welding points in a power transmission line is provided, the system comprising: A controlled acquisition device is arranged on a tower on which a transmission line is erected and has a built-in wireless control interface, and is used to perform directional image data acquisition of various components of the transmission line under the control of a remote control signal, wherein a collection picture obtained by directional image data acquisition performed at a conductor connection of the transmission line is output as a connection scene picture; A continuous mapping device, connected to the controlled acquisition device and comprising a dynamic processing device, a geometric correction device and a band-stop filtering device, for continuously performing white balance processing based on a dynamic threshold, geometric correction processing and band-stop filtering processing on the received connected scene picture to obtain and output a corresponding continuous mapping picture; A directional detection mechanism is connected to the continuous mapping device and is used to detect the mean square error of each brightness value corresponding to each pixel point of each image area in the received continuous mapping picture, and when the mean square error is less than a set mean square error limit, the image area is judged to be a background image area, otherwise, the image area is judged to be a foreground image area; an intelligent identification mechanism, connected to the direction detection mechanism, for inputting the total number of pixels occupied by the foreground image area and the cyan component value, magenta component value, yellow component value and black component value of each pixel in the foreground image area in the CMYK space into a feedforward neural network that has completed multiple trainings to run the feedforward neural network that has completed multiple trainings, and obtaining an existence mark indicating whether there is a welding point in the foreground image area output by the feedforward neural network that has completed multiple trainings; A command trigger mechanism connected to the intelligent identification mechanism, used to issue a welding point non-detection command when no welding points exist in each foreground image area in the continuous mapping picture, and also used to issue a welding point detection command when a welding point exists in a certain foreground image area in the continuous mapping picture; Wherein, in the feedforward neural network after completing multiple trainings, the number of trainings is monotonically positively correlated with the number of noise types in the continuous mapping picture; Among them, the controlled acquisition device is set on the iron tower where the transmission line is erected and has a built-in wireless control interface, which is used to perform directional image data acquisition of various components of the transmission line under the control of a remote control signal, including: the wireless control interface is wirelessly connected to the power equipment monitoring server, and is used to obtain an image data acquisition request from the power equipment monitoring server including the positioning information of the current monitored component.

[0006] Therefore, the present invention has at least the following beneficial technical effects: Firstly, a customized optimization process is performed on an image obtained by photographing the conductor connection of the transmission line to obtain a continuous mapping picture with higher image quality, and the mean square error of each brightness value corresponding to each pixel point in each image area in the continuous mapping picture is detected, and when the mean square error is less than a set mean square error limit, the image area is judged to be a background image area, otherwise, the image area is judged to be a foreground image area; Secondly: the total number of pixels occupied by the foreground image area and the cyan component value, magenta component value, yellow component value and black component value of each pixel in the foreground image area in the CMYK space are input into a feedforward neural network that has completed multiple trainings to run the feedforward neural network that has completed multiple trainings, and an existence mark indicating whether there is a welding point in the foreground image area output by the feedforward neural network that has completed multiple trainings is obtained, and when there is no welding point in each foreground image area in the continuous mapping picture, a welding point non-detection command is issued, and when there is a welding point in a certain foreground image area in the continuous mapping picture, a welding point detection command is issued, so as to realize intelligent identification of whether there is a welding point at the wire connection of the transmission line, and reduce the probability of annealing caused by the presence of welding points at the wire connection of the transmission line and the decrease in mechanical strength of the connection; Again: In the feedforward neural network after completing multiple trainings, the number of trainings is monotonically positively correlated with the number of noise types in the continuous mapping images, so as to achieve targeted construction of the feedforward neural network and improve the reliability and stability of the intelligent recognition results of the feedforward neural network.

[0007] The identification system of welding points in power transmission lines of the present invention has a compact structure and is easy to operate. Since it can intelligently determine whether there is a welding point in the foreground image area based on various visual data by using a feedforward neural network, and issue a welding point non-detection command when there is no welding point in each foreground image area in the continuous mapping screen, it can realize intelligent identification of whether there is a welding point at the wire connection of the power transmission line, and reduce the probability of annealing caused by the presence of welding points at the wire connection of the power transmission line and the decrease in mechanical strength of the connection. DETAILED DESCRIPTION

[0008] The implementation scheme of the system for identifying welding points in power transmission lines of the present invention will be described in detail below.

[0009] First embodiment

[0010] According to a first embodiment of the present invention, a system for identifying welding points in a power transmission line comprises: A controlled acquisition device is arranged on a tower on which a transmission line is erected and has a built-in wireless control interface, and is used to perform directional image data acquisition of various components of the transmission line under the control of a remote control signal, wherein a collection picture obtained by directional image data acquisition performed at a conductor connection of the transmission line is output as a connection scene picture; A continuous mapping device, connected to the controlled acquisition device and comprising a dynamic processing device, a geometric correction device and a band-stop filtering device, for continuously performing white balance processing based on a dynamic threshold, geometric correction processing and band-stop filtering processing on the received connected scene picture to obtain and output a corresponding continuous mapping picture; Specifically, the continuous mapping device is connected to the controlled acquisition device and includes a dynamic processing device, a geometric correction device and a band-stop filtering device, and is used to continuously perform white balance processing based on a dynamic threshold, geometric correction processing and band-stop filtering processing on the received connected scene picture to obtain and output the corresponding continuous mapping picture, including: the dynamic processing device, the geometric correction device and the band-stop filtering device are respectively implemented by using different types of ASIC chips; A directional detection mechanism is connected to the continuous mapping device and is used to detect the mean square error of each brightness value corresponding to each pixel point of each image area in the received continuous mapping picture, and when the mean square error is less than a set mean square error limit, the image area is judged to be a background image area, otherwise, the image area is judged to be a foreground image area; an intelligent identification mechanism, connected to the direction detection mechanism, for inputting the total number of pixels occupied by the foreground image area and the cyan component value, magenta component value, yellow component value and black component value of each pixel in the foreground image area in the CMYK space into a feedforward neural network that has completed multiple trainings to run the feedforward neural network that has completed multiple trainings, and obtaining an existence mark indicating whether there is a welding point in the foreground image area output by the feedforward neural network that has completed multiple trainings; A command trigger mechanism connected to the intelligent identification mechanism, used to issue a welding point non-detection command when no welding points exist in each foreground image area in the continuous mapping picture, and also used to issue a welding point detection command when a welding point exists in a certain foreground image area in the continuous mapping picture; Wherein, in the feedforward neural network after completing multiple trainings, the number of trainings is monotonically positively correlated with the number of noise types in the continuous mapping images; Among them, the controlled acquisition device is arranged on the iron tower on which the transmission line is erected and has a built-in wireless control interface, and is used to perform directional image data acquisition of various components of the transmission line under the control of a remote control signal, including: the wireless control interface is wirelessly connected to the power equipment monitoring server, and is used to obtain an image data acquisition request including positioning information of the current monitoring component from the power equipment monitoring server; The wireless control interface is wirelessly connected to the power equipment monitoring server, and is used to obtain an image data acquisition request including positioning information of a current monitoring component from the power equipment monitoring server, including: the current monitoring component includes a conductor connection of a transmission line.

[0011] Second embodiment

[0012] A second embodiment of the present invention is a system for identifying welding points in a power transmission line.

[0013] Different from the first embodiment, the identification system of welding points in the power transmission line in the second embodiment may further include the following components: A leakage protection mechanism is arranged near the directional detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filter device and is respectively connected to the directional detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filter device; Among them, the leakage protection mechanism is arranged near the directional detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device and is respectively connected to the directional detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device, including: the leakage protection mechanism is used to respectively realize leakage protection for the directional detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device.

[0014] Third embodiment

[0015] A third embodiment of the present invention is a system for identifying welding points in a power transmission line.

[0016] Different from the first embodiment, the system for identifying welding points in a power transmission line in the third embodiment may further include the following components: A flexible circuit board body is arranged near the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filter device and is respectively connected to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filter device; Among them, the flexible circuit board body is arranged near the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device and is respectively connected to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device, including: the flexible circuit board body is used to provide power supply and clock supply to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively.

[0017] Next, the specific structure of the identification system for welding points in a power transmission line of the present invention will be further described.

[0018] In the system for identifying welding points in a power transmission line according to any one of the embodiments of the present invention: A picture correction device is used to perform image data processing on the output data of the directional detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain output processing data corresponding to the directional detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively.

[0019] In the system for identifying welding points in a power transmission line according to any one of the embodiments of the present invention: The image correction device is used to perform image data processing on the output data of the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain the output processing data corresponding to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively, including: performing exponential image enhancement on the output data of the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain the output processing data corresponding to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively.

[0020] In the system for identifying welding points in a power transmission line according to any one of the embodiments of the present invention: The image correction device is used to perform image data processing on the output data of the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain the output processing data corresponding to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively, including: performing edge sharpening on the output data of the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain the output processing data corresponding to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively.

[0021] In the system for identifying welding points in a power transmission line according to any one of the embodiments of the present invention: The image correction device is used to perform image data processing on the output data of the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain the output processing data corresponding to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively, including: performing Sobel operator sharpening on the output data of the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain the output processing data corresponding to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively.

[0022] And in the system for identifying welding points in a power transmission line according to any one of the embodiments of the present invention: The image correction device is used to perform image data processing on the output data of the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain the output processing data corresponding to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively, including: performing geometric correction operation on the output data of the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain the output processing data corresponding to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively.

[0023] In addition, in the welding point identification system in the power transmission line, the total number of pixels occupied by the foreground image area and the cyan component value, magenta component value, yellow component value and black component value of each pixel in the foreground image area in the CMYK space are input into a feedforward neural network that has completed multiple trainings to run the feedforward neural network that has completed multiple trainings, and obtaining the existence identifier output by the feedforward neural network that has completed multiple trainings indicating whether there is a welding point in the foreground image area includes: using a MATLAB toolbox to simulate and test the network construction of the feedforward neural network that has completed multiple trainings.

[0024] The present invention is not limited to the above embodiments. According to the description of the above embodiments, a person skilled in the art may make some obvious changes, but these changes should fall within the protection scope of the claims of the present invention.

Claims

1. A system for identifying welding points in a power transmission line, characterized in that: The system comprises: A controlled acquisition device is arranged on a tower on which a transmission line is erected and has a built-in wireless control interface, and is used to perform directional image data acquisition of various components of the transmission line under the control of a remote control signal, wherein a collection picture obtained by directional image data acquisition performed at a conductor connection of the transmission line is output as a connection scene picture; A continuous mapping device, connected to the controlled acquisition device and comprising a dynamic processing device, a geometric correction device and a band-stop filtering device, for continuously performing white balance processing based on a dynamic threshold, geometric correction processing and band-stop filtering processing on the received connected scene picture to obtain and output a corresponding continuous mapping picture; A directional detection mechanism is connected to the continuous mapping device and is used to detect the mean square error of each brightness value corresponding to each pixel point of each image area in the received continuous mapping picture, and when the mean square error is less than a set mean square error limit, the image area is judged to be a background image area, otherwise, the image area is judged to be a foreground image area; an intelligent identification mechanism, connected to the direction detection mechanism, for inputting the total number of pixels occupied by the foreground image area and the cyan component value, magenta component value, yellow component value and black component value of each pixel in the foreground image area in the CMYK space into a feedforward neural network that has completed multiple trainings to run the feedforward neural network that has completed multiple trainings, and obtaining an existence mark indicating whether there is a welding point in the foreground image area output by the feedforward neural network that has completed multiple trainings; A command trigger mechanism connected to the intelligent identification mechanism, used to issue a welding point non-detection command when no welding points exist in each foreground image area in the continuous mapping picture, and also used to issue a welding point detection command when a welding point exists in a certain foreground image area in the continuous mapping picture; Wherein, in the feedforward neural network after completing multiple trainings, the number of trainings is monotonically positively correlated with the number of noise types in the continuous mapping images; Among them, the controlled acquisition device is set on the iron tower where the transmission line is erected and has a built-in wireless control interface, which is used to perform directional image data acquisition of various components of the transmission line under the control of a remote control signal, including: the wireless control interface is wirelessly connected to the power equipment monitoring server, and is used to obtain an image data acquisition request from the power equipment monitoring server including the positioning information of the current monitored component.

2. The system for identifying welding points in a power transmission line according to claim 1, characterized in that: The wireless control interface is wirelessly connected to the power equipment monitoring server and is used to obtain an image data acquisition request including positioning information of a currently monitored component from the power equipment monitoring server, including: the currently monitored component includes a conductor connection of a transmission line.

3. The system for identifying welding points in a power transmission line according to claim 2, characterized in that: The system further comprises: A leakage protection mechanism is arranged near the directional detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filter device and is respectively connected to the directional detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filter device; Among them, the leakage protection mechanism is arranged near the directional detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device and is respectively connected to the directional detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device, including: the leakage protection mechanism is used to respectively realize leakage protection for the directional detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device.

4. The system for identifying welding points in a power transmission line according to claim 2, characterized in that: The system further comprises: A flexible circuit board body is arranged near the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filter device and is respectively connected to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filter device; Among them, the flexible circuit board body is arranged near the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device and is respectively connected to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device, including: the flexible circuit board body is used to provide power supply and clock supply to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively.

5. The system for identifying welding points in a power transmission line according to any one of claims 2 to 4, characterized in that: A picture correction device is used to perform image data processing on the output data of the directional detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain output processing data corresponding to the directional detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively.

6. The system for identifying welding points in a power transmission line according to claim 5, characterized in that: The image correction device is used to perform image data processing on the output data of the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain the output processing data corresponding to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively, including: performing exponential image enhancement on the output data of the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain the output processing data corresponding to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively.

7. The system for identifying welding points in a power transmission line according to claim 5, characterized in that: The image correction device is used to perform image data processing on the output data of the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain the output processing data corresponding to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively, including: performing edge sharpening on the output data of the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain the output processing data corresponding to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively.

8. The system for identifying welding points in a power transmission line according to claim 5, characterized in that: The image correction device is used to perform image data processing on the output data of the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain the output processing data corresponding to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively, including: performing Sobel operator sharpening on the output data of the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain the output processing data corresponding to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively.

9. The system for identifying welding points in a power transmission line according to claim 5, characterized in that: The image correction device is used to perform image data processing on the output data of the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain the output processing data corresponding to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively, including: performing geometric correction operation on the output data of the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device to obtain the output processing data corresponding to the direction detection mechanism, the dynamic processing device, the geometric correction device and the band-stop filtering device respectively.