Unmanned aerial vehicle direct current insulator voltage measuring system and method based on optical voltage sensor

Abstract

The invention belongs to the technical field of voltage detection, and discloses an unmanned aerial vehicle direct current insulator voltage measurement system and method based on an optical voltage sensor, the system comprises an airborne module carried by an unmanned aerial vehicle, the airborne module comprises a detection unit, and the detection unit comprises an ultrasonic ranging subunit, a camera and an improved optical voltage sensor; the ultrasonic ranging subunit controls the target distance position of the unmanned aerial vehicle in the target direct-current insulator string; the camera shoots the target direct current insulator string to obtain a potential to-be-detected image; the improved optical voltage sensor comprises a target detection model and an electric field sensing element. And the target detection model uses a YOLO algorithm to obtain the center point coordinate and the number of each direct current insulator so as to sequentially obtain the voltage of each target direct current insulator. The system can measure the voltage of the direct current insulator on site, and considers the flexibility of potential distribution of the direct current insulator string, so as to improve the accuracy of measuring the voltage of the direct current insulator string.

Description

Technical Field

[0001] The present invention belongs to the technical field of voltage detection, and in particular relates to a voltage measurement system and method for a DC insulator of an unmanned aerial vehicle based on an optical voltage sensor. Background Art

[0002] On UHV DC transmission lines, degraded DC insulators may crack or break, causing their insulation resistance to decrease, increasing the risk of leakage and flashover. This affects the normal operation of the transmission line, causing local electric field distortion and uneven electric field distribution. Under the action of a strong electric field, corona discharge is easily induced, further reducing the insulation performance, while generating electromagnetic interference, affecting nearby communications and electronic equipment. Due to the degradation of insulation performance, some electric energy may leak, thereby reducing transmission efficiency. UHV DC transmission lines usually transmit a large amount of electric energy, and even a small energy loss will accumulate into a huge loss in long-term operation. Degraded DC insulators may also cause changes in line impedance, affecting the transmission of current and increasing the power loss of the transmission line. Therefore, potential detection of DC insulators on UHV DC transmission lines is a very important task.

[0003] There is currently no reliable system for non-contact detection of DC insulator potential. In the prior art, it is very difficult to measure the potential of insulator strings in DC systems. The ball gap method for measuring the potential of DC insulators is a huge workload and is only used in the laboratory, and cannot be used for on-site measurements. The variable gap and variable area electrostatic induction capacitance sensor structures can better measure the potential distribution of DC insulator strings, but the sensors need to be installed on the steel legs of the insulator strings, and can only be completed in the laboratory. In addition, it is unrealistic to use traditional contact voltage sensors mounted on drones to measure the potential distribution of DC insulator strings, because installing sensors on DC insulator strings to conduct on-site measurements of DC insulator potential on DC transmission lines does not take into account the flexibility of the potential distribution of DC insulator strings, which can easily lead to a decrease in the accuracy of measuring the voltage of DC insulator strings. Summary of the invention

[0004] Based on this, the present application proposes a DC insulator pressure measurement system and method based on an optical voltage sensor drone, which aims to perform on-site measurement of the DC insulator voltage without installing sensors on the DC insulator string, while taking into account the flexibility of the potential distribution of the DC insulator string to improve the accuracy of measuring the voltage of the DC insulator string.

[0005] The first aspect of the present application provides a DC insulator pressure measurement system for drones based on an optical voltage sensor, the system comprising: UAV, used to carry airborne modules; The airborne module comprises a detection unit, wherein the detection unit comprises an ultrasonic ranging subunit, a camera and an improved optical voltage sensor, wherein: The ultrasonic distance measuring subunit is used to control the target distance position of the UAV at the target DC insulator string; The camera is used to photograph the target DC insulator string to obtain a potential image to be measured of the target DC insulator string; The improved optical voltage sensor comprises a target detection model and an electric field sensing element, wherein: The target detection model is used to perform the YOLO algorithm on the potential image to be measured to obtain the center point coordinates and number of each DC insulator. The target detection model includes: A data preparation layer, used for collecting historical images of DC insulator strings, and annotating the historical images of the DC insulator strings with bounding boxes to form an insulator dataset; A model training layer, used for optimizing the target detection model according to the insulator dataset; The model inference layer is used to obtain the center point coordinates and number of each DC insulator based on the optimized target detection model; The electric field sensing element is used to sequentially obtain the voltage of each target DC insulator according to the center point coordinates and numbers of each DC insulator, and the electric field sensing element includes: An electric field sensing probe, used for sensing first potential information around a target DC insulator, and converting the first potential information into a first electrical signal; A light source, for emitting a light signal; A Pockels effect crystal, used for modulating the optical signal according to the first electrical signal, so that the polarization state or phase of the optical signal changes; An optical transmission sub-component, used for transmitting the modulated optical signal to the optical analysis sub-component; An optical analysis subcomponent, used for detecting polarization or phase changes of the optical signal and converting the polarization or phase changes into light intensity changes; A photoelectric detection sub-component, used for extracting second potential information from the light intensity change, and obtaining a second electrical signal according to the second potential information; The signal processing sub-component is used to amplify and filter the second electrical signal, and to obtain voltage information of each target DC insulator by using a differential measurement method.

[0006] As an optional implementation manner of the first aspect, the ultrasonic ranging subunit includes: Ultrasonic sensor, used to measure the distance between the drone and the target DC insulator string in real time according to the ranging formula; The PID controller is used to control the distance between the UAV and the target DC insulator string in real time according to the control formula.

[0007] As an optional implementation manner of the first aspect, the control formula is: , in, is the control signal output at time t, that is, the control signal sent by the PID controller, , , are the proportional, integral and differential gain coefficients respectively, is the error at time t, is the time between the initial time and time t time, is the target distance set, is the actual distance at time t.

[0008] As an optional implementation of the first aspect, the calculation formula of the center point coordinates is: , Among them, x c ,y c Represents the x-coordinate and y-coordinate of the center point of the bounding box, respectively. min 、x max Represents the minimum and maximum x-coordinates of the left and right sides of the bounding box, respectively; y min ,y max Represents the minimum and maximum y-coordinates of the upper and lower boundaries of the bounding box, respectively.

[0009] As an optional implementation manner of the first aspect, the formula of the differential measurement method is: , in, , and are the corresponding electric field strengths sensed by the electric field sensing probe when the drone moves to the n-1th DC insulator, the nth DC insulator, and the n+1th DC insulator, respectively. To calculate the electric field strength around the nth DC insulator after filtering interference.

[0010] As an optional implementation manner of the first aspect, the system further includes a ground analysis module, and the ground analysis module further includes: A ground communication unit, used for receiving voltage information of each target DC insulator sent from the airborne module; A ground processing unit is used to fuse the voltage information of each target DC insulator to obtain the potential information of the entire string of target DC insulators; The display unit is used to display the potential information of the entire string of target DC insulators.

[0011] As an optional implementation manner of the first aspect, the airborne module further includes an airborne communication unit, an airborne processing unit, and an airborne GPU computing unit, and the airborne processing unit is respectively connected to the airborne communication unit, the airborne GPU computing unit, and the detection unit, wherein: The airborne communication unit is used to transmit potential information to the ground analysis module in real time; The airborne processing unit is used to process the data from the detection unit and coordinate the work of the airborne communication unit and the airborne GPU computing unit; The onboard GPU computing unit is used to perform image processing on the potential image to be measured using a graphics processing unit (GPU).

[0012] A second aspect of the present application provides a method for measuring voltage of a DC insulator of a drone based on an optical voltage sensor, which is applied to an airborne module carried by the drone. The method comprises: controlling the drone to a target distance position of a target DC insulator string; photographing the target DC insulator string to obtain a potential-to-be-measured image of the target DC insulator string; The YOLO algorithm is performed on the potential image to be measured to obtain the coordinates and number of the center point of each DC insulator, specifically including: collecting historical images of DC insulator strings, and annotating the historical images of the DC insulator strings with bounding boxes to form an insulator data set; optimizing the target detection model according to the insulator data set; and obtaining the coordinates and number of the center point of each DC insulator based on the optimized target detection model; According to the coordinates and number of the center point of each DC insulator, the voltage of each target DC insulator is obtained in turn, which specifically includes: sensing the first potential information around the target DC insulator and converting the first potential information into a first electrical signal; emitting an optical signal; modulating the optical signal according to the first electrical signal to change the polarization state or phase of the optical signal; transmitting the modulated optical signal to an optical analysis sub-component; detecting the polarization or phase change of the optical signal, and converting the polarization or phase change into a light intensity change; extracting the second potential information from the light intensity change, and obtaining the second electrical signal according to the second potential information; amplifying and filtering the second electrical signal, and using a differential measurement method to obtain the voltage information of each target DC insulator.

[0013] The third aspect of the present application provides an electronic device, comprising a processor and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the executable instructions to implement the above-mentioned method for measuring DC insulator voltage of an unmanned aerial vehicle based on an optical voltage sensor.

[0014] The fourth aspect of the present application provides a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to perform the above-mentioned method for measuring DC insulator voltage of an unmanned aerial vehicle based on an optical voltage sensor.

[0015] Compared with the prior art, the optical voltage sensor-based unmanned aerial vehicle DC insulator pressure measurement system proposed in the present invention can more accurately detect the center point coordinates and number of each DC insulator by improving the optical voltage sensor and the target detection model, thereby improving the measurement accuracy of the potential distribution; at the same time, the measurement can be performed using the unmanned aerial vehicle equipped with an airborne module, thereby improving the efficiency and flexibility of the measurement; the ultrasonic ranging subunit is used to control the unmanned aerial vehicle at the target distance position, thereby avoiding direct contact with the DC insulator string, reducing damage to the insulator, and also improving the safety of the measurement; the target DC insulator string is photographed by a camera, thereby obtaining an overall image of the insulator string, and combined with the target detection model and the electric field sensing element, the potential distribution of each DC insulator can be measured, thereby improving the comprehensiveness of the insulator detection; the entire system integrates technologies such as unmanned aerial vehicle platform, ultrasonic ranging, image acquisition and improved optical voltage sensor, thereby realizing remote, automated and high-precision pressure measurement of the target DC insulator string, thereby meeting the needs in practical applications. Therefore, the device proposed in the present application can perform on-site measurement of DC insulator voltage without installing sensors on the DC insulator string, while taking into account the flexibility of the potential distribution of the DC insulator string to improve the accuracy of measuring the voltage of the DC insulator string.

[0016] Additional aspects and advantages of the present application will be given in part in the following description, and in part will become apparent from the following description, or will be understood through the embodiments of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A structural distribution diagram of a DC insulator pressure measurement system for drones based on optical voltage sensors proposed in an embodiment of the present application; Figure 2 This is a structural block diagram of the ultrasonic ranging subunit in the embodiment of the present application; Figure 3 This is a structural block diagram of the target detection model in the embodiment of the present application; Figure 4 This is a workflow diagram of the target detection model in the embodiment of the present application; Figure 5 This is a structural block diagram of an electric field sensing element in an embodiment of the present application; Figure 6 This is a flowchart of improving the optical voltage sensor in the embodiment of the present application; Figure 7 This is a structural block diagram of the ground analysis module in the embodiment of the present application; Figure 8 This is a structural block diagram of the airborne module in an embodiment of the present application; Fig. 9 A flow chart of a method for measuring voltage of a DC insulator of a drone based on an optical voltage sensor proposed in an embodiment of the present application; Fig.10 This is a flowchart of the pre-trained YOLO v8 model proposed in an embodiment of the present application.

[0018] The following specific implementation methods will further illustrate the present application in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0020] The terms "first", "second", etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here. In addition, the "and / or" in the specification and claims represents at least one of the connected objects, and the character " / " generally represents that the objects associated with each other are in an "or" relationship.

[0021] In order to illustrate the technical solution described in this application, a specific embodiment is provided below for illustration.

[0022] See also Figure 1 , is a structural distribution diagram of a DC insulator pressure measurement system for an optical voltage sensor drone proposed in an embodiment of the present application, the system comprising: The drone 10 is used to carry the airborne module 20; The airborne module 20 includes a detection unit 201, wherein the detection unit 201 includes an ultrasonic ranging subunit 2011, a camera 2012, and an improved optical voltage sensor 2013, wherein: The ultrasonic distance measuring subunit 2011 is used to control the drone 10 at a target distance position of a target DC insulator string; The camera 2012 is used to photograph the target DC insulator string to obtain a potential image to be measured of the target DC insulator string; The improved optical voltage sensor 2013 includes a target detection model 20131 and an electric field sensing element 20132, wherein: The target detection model 20131 is used to perform a YOLO algorithm on the potential image to be measured to obtain the center point coordinates and number of each DC insulator; The electric field sensing element 20132 is used to obtain the voltage of each target DC insulator in sequence according to the center point coordinates and number of each DC insulator.

[0023] In some embodiments, Figure 2 As shown, the ultrasonic distance measuring subunit 2011 includes: The ultrasonic sensor 2011 is used to measure the distance between the UAV 10 and the target DC insulator string in real time according to a distance measurement formula; The PID controller 2012 is used to control the distance between the UAV 10 and the target DC insulator string in real time according to the control formula.

[0024] It should be noted that the ground operator controls the drone 10 to fly to the vicinity of the target DC insulator string, and uses the ultrasonic ranging subunit 2011 to control the drone 10 at an appropriate distance from the target DC insulator string. The distance between the drone 10 and the target DC insulator string needs to be kept stable to ensure that the optical voltage sensor 2013 is at the optimal measurement distance. The positioning and control of the drone 10 are achieved through the ultrasonic sensor 2011 and the PID controller 2012. Through the ultrasonic sensor 2011 and the PID controller 2012, the drone 10 can be stably hovered at the set distance of the target DC insulator string, which is convenient for the next step.

[0025] Specifically, the ultrasonic sensor 2011 measures the distance between the drone 10 and the insulator string in real time, and the measured distance is recorded as Target distance It needs to be kept constant to ensure that the improved optical voltage sensor 2013 is located at a suitable electric field position. The distance measurement formula is: , in, is the actual distance between the UAV 10 and the target DC insulator string at time t, is the propagation speed of ultrasound in air, about 343m / s, It is the time from the transmission of ultrasonic wave to the reception of reflected echo.

[0026] In order to stabilize the drone 10 at the target distance, a PID controller 2012 is used to adjust its position. The control formula of the PID controller 2012 is: , in, is the control signal output at time t, i.e., the control signal sent by the PID controller 2012, , , are the proportional, integral and differential gain coefficients respectively, is the error at time t, is the time between the initial time and time t time, is the target distance set, is the actual distance at time t.

[0027] In some embodiments, Figure 3 As shown, it is the target detection model 20131, including: A data preparation layer, used for collecting historical images of DC insulator strings, and annotating the historical images of DC insulator strings with bounding boxes to form an insulator dataset; A model training layer, used for optimizing the target detection model 20131 according to the insulator dataset; The model inference layer is used to obtain the center point coordinates and number of each DC insulator based on the optimized target detection model 20131.

[0028] It should be noted that in the present invention, the optical voltage sensor 2013 detects the potential of the target DC insulator piece by piece, and each insulator in the target DC insulator string needs to be located and numbered, so that the detection can be completed in an orderly and smooth manner. Using the drone 10 equipped with a camera 2012 and combining it with the YOLO algorithm, automatic detection and numbering of the insulator string can be achieved. This solution not only reduces the complexity of deploying equipment on the ground, but also more conveniently covers a wide range of insulator inspection scenarios. The key is to train the model with enough labeled data in advance and optimize its performance to adapt to the flight environment of the drone 10. Figure 4 As shown, this is the workflow diagram of the target detection model 20131.

[0029] In the data preparation layer, it is necessary to collect enough historical images of DC insulator strings in advance and annotate them. Manually shoot multiple groups of 60 historical images of insulator strings at different angles and lighting conditions. The shooting angles include top view, side view, far view, near view, etc., to adapt to the diverse scenes actually shot by the drone 10, ensure data diversity, and improve the generalization ability of the model. It takes about thousands of pictures to shoot.

[0030] Furthermore, in the data annotation process, each insulator is labeled with a bounding box, without distinguishing the specific parts. The annotation tool uses the open source LabelImg; each insulator is labeled as a category, and its bounding box is labeled. At the same time, each insulator can be numbered when annotating the insulation dataset, but in fact the number can be automatically generated by sorting by position in the inference stage.

[0031] In the model training layer, the YOLO v8 model can be used. The model is lightweight and efficient. Its training process is: (1) Use the labeled insulator dataset to train the model so that it can identify the location of each insulator and output the bounding box; (2) The model outputs the location information of each insulator and generates a number through an inference algorithm; (3) Using transfer learning technology, the pre-trained model can be loaded and fine-tuned on the insulator dataset to speed up training and improve the accuracy of the model.

[0032] At the model inference layer, the trained YOLO v8 model is loaded into the computing platform. When the drone 10 is flying, the camera 2012 captures the image and detects the bounding box of each DC insulator in real time. According to the detected bounding box, the insulators are sorted by the vertical coordinate of the center point and numbered.

[0033] Among them, the calculation formula for the center point coordinates of the bounding box is: , Among them, x c ,y c Represents the x-coordinate and y-coordinate of the center point of the bounding box, respectively. min 、x max Represents the minimum and maximum x-coordinates of the left and right sides of the bounding box, respectively; y min ,y max Represents the minimum and maximum y-coordinates of the upper and lower boundaries of the bounding box, respectively.

[0034] In some embodiments, Figure 5 As shown, the electric field sensing element 20132 includes: An electric field sensing probe, used for sensing first potential information around a target DC insulator, and converting the first potential information into a first electrical signal; A light source, for emitting a light signal; A Pockels effect crystal, used for modulating the optical signal according to the first electrical signal, so that the polarization state or phase of the optical signal changes; An optical transmission subcomponent, used for transmitting the optical signal to the optical analysis subcomponent; An optical analysis subcomponent, used for detecting polarization or phase changes of the optical signal and converting the polarization or phase changes into light intensity changes; A photoelectric detection sub-component, used for extracting second potential information from the light intensity change, and obtaining a second electrical signal according to the second potential information; The signal processing sub-component is used to amplify and filter the second electrical signal, and to obtain voltage information of each target DC insulator by using a differential measurement method.

[0035] It should be noted that the traditional optical voltage sensor is composed of a light source, an electro-optical crystal sensing unit, a photodetector, a data acquisition card, a signal processing system, etc. In practical applications, the measured potential is applied to the metal plates on both sides of the electro-optical crystal sensing unit, which is a contact voltage sensor. Obviously, it is unrealistic to measure the potential distribution of the DC insulator string on the drone 10. The present application proposes an improved optical voltage sensor 2013, which can detect the potential in a non-contact situation by adding an electric field sensing element 20132 to the optical voltage sensor and combining the electric field sensing technology. The electric field probe is used to detect the electric field distribution of a space around a piece in the insulator string, and the potential of the insulator steel cap of the piece is derived. The electric field sensing probe usually senses the electric field within a spatial range, rather than a single point electric field. Therefore, this means that it senses the comprehensive electric field distribution of a region. Whether it is a traditional optical voltage sensor or an improved optical voltage sensor 2013, the working principle is based on the Pockels effect. The Pockels effect refers to the phenomenon that the refractive index of certain electro-optical materials (such as crystals) changes under the action of an external electric field. This change is proportional to the strength of the applied electric field and only exists in crystals with non-centrosymmetric structures. For the improved optical voltage sensor 2013, the external electric field is usually generated by the potential to be measured, and the Pockels effect is used to modulate the polarization state of the light wave.

[0036] This application adopts a differential measurement method to obtain the electric field information of the target insulator by simultaneously measuring the electric fields of multiple adjacent areas and calculating the differences between these measured values. This can eliminate the electric field interference from adjacent insulators and highlight the electric field generated by the target insulator. The formula of the differential measurement method is: , in, , and are the corresponding electric field strengths sensed by the electric field sensing probe when the drone moves to the n-1th DC insulator, the nth DC insulator, and the n+1th DC insulator, respectively. To calculate the electric field strength around the nth DC insulator after filtering interference.

[0037] For example: , E 9. E 10 , E 11 The corresponding electric field strength sensed by the electric field sensing probe when the drone moves to the 9th, 10th, and 11th DC insulators. E’ 10 To calculate the electric field strength around the 10th insulator after filtering interference. The electric field interference of insulators at other positions is ignored in this application.

[0038] like Figure 6 The figure shows the working flow chart of the improved optical voltage sensor 2013. The working flow of the improved optical voltage sensor 2013 is that the electric field sensing probe senses the electric field around the DC insulator and converts it into an electric signal according to the potential information, so that the light source emits a light signal which is modulated by the Pockels effect crystal. The modulated light signal carries the information related to the potential of the DC insulator. The light signal is transmitted to the optical analysis sub-component through the optical transmission sub-component, and the polarization or phase change of the light signal is detected and converted into a light intensity change. After the potential information is extracted, the light signal is then transferred to the photoelectric detection sub-component to realize the conversion of the light signal into an electric signal, and finally amplified and filtered by the signal processing sub-component.

[0039] In some embodiments, a DC insulator pressure measurement system based on an optical voltage sensor UAV further includes a ground analysis module 30, such as Figure 7 As shown, the ground analysis module 30 includes: The ground communication unit 301 is used to receive the voltage information of each target DC insulator sent by the airborne module 20; A ground processing unit 302 is used to fuse the voltage information of each target DC insulator to obtain the potential information of the entire string of target DC insulators; The display unit 303 is used to display the potential information of the entire string of target DC insulators.

[0040] In the above-mentioned ground analysis module 30, the ground communication unit 301 is used to receive the voltage information of each target DC insulator sent from the airborne module 20, thereby realizing data communication between the ground and the airborne module 20, so that ground personnel can obtain the insulator voltage information in real time, and provide data support for the subsequent insulator status evaluation; the ground processing unit 302 is used to fuse the voltage information of each target DC insulator to obtain the potential information of the entire string of target DC insulators. Through data fusion processing, the accuracy and reliability of the insulator potential information are improved, which provides a strong guarantee for the comprehensive evaluation of the operating status of the insulators; the display unit 303 is used to display the potential information of the entire string of target DC insulators, so that ground personnel can intuitively understand the voltage conditions of the insulators, which is convenient for timely discovery of abnormal conditions of the insulators, so as to take corresponding maintenance measures.

[0041] In some embodiments, Figure 8 As shown, it is an airborne module 20, which also includes an airborne communication unit 202, an airborne processing unit 203 and an airborne GPU computing unit 204, and the airborne processing unit 203 is connected to the airborne communication unit 202, the airborne GPU computing unit 204, and the detection unit 201 respectively, wherein: The airborne communication unit 202 is used to transmit potential information with the ground analysis module 30 in real time; The onboard processing unit 203 is used to process the data from the detection unit 201 and coordinate the work of the onboard communication unit 202 and the onboard GPU computing unit 204; The onboard GPU computing unit 204 is used to perform image processing on the potential image to be measured using a graphics processing unit (GPU).

[0042] In the above-mentioned airborne module 20, the main function of the airborne communication unit 202 is to transmit real-time potential information with the ground analysis module 30. This design enables the airborne module 20 to exchange data with the ground module in a timely manner to ensure the real-time and accuracy of the information, which is crucial for real-time monitoring and ground analysis during the flight, and helps to make timely adjustments to the flight status and warn of potential risks; the airborne processing unit 203 is responsible for processing the data from the detection unit 201 and coordinating the work of the airborne communication unit 202 and the airborne GPU computing unit 204. Since the detection unit 201 may generate a large amount of data, the airborne processing unit needs to process it efficiently. Process these data to ensure the accuracy and completeness of the data. In addition, it also needs to coordinate with other modules to ensure the efficient operation of the entire airborne module 20. This design enables the module to cope with complex flight environments and ensure the safety and stability of the flight. The airborne GPU computing unit 204 uses a graphics processor GPU to perform image processing on the potential image to be measured. Since the GPU has powerful parallel processing capabilities, using the GPU for image processing can greatly improve the processing speed and efficiency, which is crucial for real-time image processing and analysis during flight, and helps to improve the quality and accuracy of the image, thereby providing more accurate navigation and positioning information for the flight.

[0043] See also Fig. 9 , which is a flow chart of a method for measuring voltage of a DC insulator of a UAV based on an optical voltage sensor proposed in an embodiment of the present application. The method is applied to an airborne module carried by the UAV, and the proposed method includes steps S01 to S04.

[0044] S01: Controlling the UAV to a target distance position of a target DC insulator string; S02: photographing the target DC insulator string to obtain a potential image to be measured of the target DC insulator string; S03: performing a YOLO algorithm on the potential image to be measured to obtain the center point coordinates and number of each DC insulator; S04: According to the center point coordinates and serial numbers of each DC insulator, the voltage of each target DC insulator is obtained in sequence.

[0045] It should be noted that the YOLO v8 model can be pre-trained, and then deployed on the drone for performing the YOLO algorithm on the potential image to be measured in step S03. Fig.10 As shown, this is a flowchart for pre-training the YOLO v8 model.

[0046] Specifically, it is necessary to collect enough sets of 60 insulator string images at different angles and different lighting conditions in advance and annotate them. Manually take multiple sets of 60 insulator string images at different angles and different lighting conditions, and it takes about thousands of pictures to take pictures; use the open source LabelImg to annotate and number each insulator with a bounding box. Then use the YOLO v8 model training on a high-GPU performance workstation in advance: use the annotated insulator data set to train the model so that the model can identify the position of each insulator and output the bounding box; the model outputs the bounding box position of each insulator and generates a number through the inference algorithm; using transfer learning technology, the pre-trained model can be loaded and fine-tuned on the insulator data set to speed up the training speed and improve the accuracy of the model. After the training is completed, the YOLO v8 model is deployed on the drone. After that, the drone uses the onboard GPU computing device to analyze and process the DC insulator image obtained by the high-definition camera, detects the bounding box position of each DC insulator in real time, and locates and numbers the entire string of DC insulators in sequence by sorting the vertical coordinate of the center point.

[0047] Furthermore, suppose that the ground operator hovers the drone around a string of 60 UHV DC insulators and uses an ultrasonic sensor combined with a PID algorithm to keep the drone at a set target distance d of the DC insulator string. set ,Then the trained YOLO v8 model locates and numbers the DC insulator strings.

[0048] Furthermore, when the UAV obtains the positioning and numbering information of the DC insulator string, the onboard processor enables the ultrasonic ranging sensor, high-definition camera and improved optical voltage sensor to work together, and measures each DC insulator to be tested by continuously adjusting the position of the UAV. After the onboard processor obtains the insulator potential data, the DC insulator voltage is transmitted back to the ground communication module through the wireless communication of the airborne communication module. After the ground communication module gives the DC insulator string voltage data to the ground processor, the ground processor displays the voltage of the DC insulator string through the display screen, that is, the voltage at the measurement point of the improved optical voltage sensor on each insulator.

[0049] In summary, the working process of the system is: the ground operator controls the UAV to fly to the vicinity of the DC insulator string, and uses the ultrasonic ranging sensor and PID control algorithm to realize the positioning of the UAV and control the UAV at the appropriate distance of the DC insulator string; the YOLO model detects the position and number of each insulator in real time, and after locating the insulator, the improved optical voltage sensor obtains the potential of the first insulator of the DC insulator string; the onboard processor coordinates the ultrasonic ranging and YOLO model to adjust the position of the UAV to the specified position of the second piece of the DC insulator string, and then the onboard processor issues a command to the improved optical voltage sensor to measure the second DC insulator, and the voltage data of the entire string of DC insulators is obtained in this way. The voltage data is transmitted to the ground processor through the wireless transmission module, and the voltage of the DC insulator string is displayed on the display screen.

[0050] On the other hand, the present application also proposes an electronic device, which includes a processor and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the executable instructions to implement the above-mentioned method for measuring DC insulator voltage of an unmanned aerial vehicle based on an optical voltage sensor.

[0051] On the other hand, the present application also proposes a computer-readable storage medium. When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device can execute the above-mentioned method for measuring DC insulator voltage of an unmanned aerial vehicle based on an optical voltage sensor.

[0052] It should be noted that, in this article, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises one..." does not exclude the presence of other identical elements in the process, method, article or device including the element. In addition, it should be noted that the scope of the method and device in the embodiment of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved, for example, the described method may be performed in an order different from that described, and various steps may also be added, omitted, or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0053] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0054] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A UAV DC insulator pressure measurement system based on an optical voltage sensor, characterized in that: The system comprises: UAV, used to carry airborne modules; The airborne module comprises a detection unit, wherein the detection unit comprises an ultrasonic ranging subunit, a camera and an improved optical voltage sensor, wherein: The ultrasonic distance measuring subunit is used to control the target distance position of the UAV at the target DC insulator string; The camera is used to photograph the target DC insulator string to obtain a potential image to be measured of the target DC insulator string; The improved optical voltage sensor comprises a target detection model and an electric field sensing element, wherein: The target detection model is used to perform the YOLO algorithm on the potential image to be measured to obtain the center point coordinates and number of each DC insulator. The target detection model includes: A data preparation layer, used for collecting historical images of DC insulator strings, and annotating the historical images of DC insulator strings with bounding boxes to form an insulator dataset; A model training layer, used for optimizing the target detection model according to the insulator dataset; The model inference layer is used to obtain the center point coordinates and number of each DC insulator based on the optimized target detection model; The electric field sensing element is used to sequentially obtain the voltage of each target DC insulator according to the center point coordinates and numbers of each DC insulator, and the electric field sensing element includes: An electric field sensing probe, used for sensing first potential information around a target DC insulator, and converting the first potential information into a first electrical signal; A light source, for emitting a light signal; A Pockels effect crystal, used for modulating the optical signal according to the first electrical signal, so that the polarization state or phase of the optical signal changes; An optical transmission sub-component, used for transmitting the modulated optical signal to the optical analysis sub-component; An optical analysis subcomponent, used for detecting polarization or phase changes of the optical signal and converting the polarization or phase changes into light intensity changes; A photoelectric detection sub-component, used for extracting second potential information from the light intensity change, and obtaining a second electrical signal according to the second potential information; The signal processing sub-component is used to amplify and filter the second electrical signal, and to obtain voltage information of each target DC insulator by using a differential measurement method.

2. The system according to claim 1, characterized in that: The ultrasonic ranging subunit comprises: Ultrasonic sensor, used to measure the distance between the drone and the target DC insulator string in real time according to the ranging formula; The PID controller is used to control the distance between the UAV and the target DC insulator string in real time according to the control formula.

3. The system according to claim 2, characterized in that: The control formula is: ， in, is the control signal output at time t, that is, the control signal sent by the PID controller, , , are the proportional, integral and differential gain coefficients respectively, is the error at time t, is the time between the initial time and time t time, is the target distance set, is the actual distance at time t.

4. The system according to claim 1, characterized in that: The calculation formula of the center point coordinates is: ， Among them, x c ,y c Represents the x-coordinate and y-coordinate of the center point of the bounding box, respectively. min 、x max Represents the minimum and maximum x-coordinates of the left and right sides of the bounding box, respectively; y min ,y max Represents the minimum and maximum y-coordinates of the upper and lower boundaries of the bounding box, respectively.

5. The system according to claim 1, characterized in that: The formula for the differential measurement method is: ， in, , and are the corresponding electric field strengths sensed by the electric field sensing probe when the drone moves to the n-1th DC insulator, the nth DC insulator, and the n+1th DC insulator, respectively. To calculate the electric field strength around the nth DC insulator after filtering interference.

6. The system according to claim 1, characterized in that: The system further includes a ground analysis module, and the ground analysis module further includes: A ground communication unit, used for receiving voltage information of each target DC insulator sent from the airborne module; A ground processing unit is used to fuse the voltage information of each target DC insulator to obtain the potential information of the entire string of target DC insulators; The display unit is used to display the potential information of the entire string of target DC insulators.

7. The system according to claim 6, characterized in that: The airborne module further includes an airborne communication unit, an airborne processing unit and an airborne GPU computing unit, wherein the airborne processing unit is connected to the airborne communication unit, the airborne GPU computing unit and the detection unit respectively, wherein: The airborne communication unit is used to transmit potential information to the ground analysis module in real time; The airborne processing unit is used to process the data from the detection unit and coordinate the work of the airborne communication unit and the airborne GPU computing unit; The onboard GPU computing unit is used to perform image processing on the potential image to be measured using a graphics processing unit (GPU).

8. A method for measuring voltage of DC insulators of unmanned aerial vehicles based on optical voltage sensors, characterized in that: Applied to an airborne module carried by a drone, the method comprises: controlling the drone to a target distance position of a target DC insulator string; photographing the target DC insulator string to obtain a potential-to-be-measured image of the target DC insulator string; The YOLO algorithm is performed on the potential image to be measured to obtain the coordinates and number of the center point of each DC insulator, specifically including: collecting historical images of DC insulator strings, and annotating the historical images of the DC insulator strings with bounding boxes to form an insulator data set; optimizing the target detection model according to the insulator data set; and obtaining the coordinates and number of the center point of each DC insulator based on the optimized target detection model; According to the coordinates and number of the center point of each DC insulator, the voltage of each target DC insulator is obtained in turn, which specifically includes: sensing the first potential information around the target DC insulator and converting the first potential information into a first electrical signal; emitting an optical signal; modulating the optical signal according to the first electrical signal to change the polarization state or phase of the optical signal; transmitting the modulated optical signal to an optical analysis sub-component; detecting the polarization or phase change of the optical signal, and converting the polarization or phase change into a light intensity change; extracting the second potential information from the light intensity change, and obtaining the second electrical signal according to the second potential information; amplifying and filtering the second electrical signal, and using a differential measurement method to obtain the voltage information of each target DC insulator.

9. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; Wherein, the processor is configured to execute the executable instructions to implement the method for measuring voltage of DC insulators of unmanned aerial vehicles based on optical voltage sensors as described in claim 8.

10. A computer-readable storage medium, characterized in that: When the instructions in the computer-readable storage medium are executed by a processor of an electronic device, the electronic device is enabled to execute a method for measuring voltage of DC insulators of an unmanned aerial vehicle based on an optical voltage sensor as described in claim 8.

Images