Cleaning and flaw detection system and method based on unmanned aerial vehicle

Abstract

The invention discloses a cleaning and flaw detection system and method based on an unmanned aerial vehicle, and the system comprises an unmanned aerial vehicle body which carries at least the following modules: a collection module which is used for collecting infrared scanning data and image data of a target area; the control module is used for acquiring the infrared scanning data and the image data, acquiring laser parameters, flaw detection parameters, a cleaning path and a flaw detection path which are determined on the basis of the infrared scanning data and the image data, sending the laser parameters to the laser cleaning module and sending the flaw detection parameters to the flaw detection module; the laser cleaning module is used for conducting laser cleaning on the target area based on the laser parameters in the process that the unmanned aerial vehicle body moves according to the cleaning path; and the flaw detection module is used for collecting flaw detection data of the target area based on the flaw detection parameters in the process that the unmanned aerial vehicle body moves according to the flaw detection path after laser cleaning of the target area is completed, and the automation level and the maintenance operation efficiency can be improved.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle (UAV) control technology, specifically to a UAV-based cleaning and flaw detection system and method. Background Technology

[0002] Photovoltaic power generation has been widely used and developed rapidly in recent years due to its environmental and renewable advantages. Solar photovoltaic panels, as the core component of photovoltaic power generation, are generally installed outdoors at high altitudes to receive sunlight and generate electricity. Therefore, regular and efficient cleaning and maintenance of photovoltaic panels is a crucial step in ensuring the power generation efficiency of the power plant. However, existing high-altitude cleaning methods have significant shortcomings.

[0003] Currently, conventional high-altitude cleaning methods mainly include manual cleaning, which requires operators to work at heights, posing serious safety hazards and is ineffective against stubborn stains. Furthermore, existing drone cleaning devices generally suffer from poor cleaning performance and require frequent water replenishment or spraying of cleaning fluid, making it difficult to achieve efficient and precise automated operations.

[0004] Therefore, there is an urgent need for a combination of cleaning and flaw detection equipment that can be integrated into a drone platform to achieve efficient, non-contact cleaning of high-altitude areas and to detect internal defects in material structures after cleaning, thereby improving the automation level of drone high-altitude operations. Summary of the Invention

[0005] This application provides a cleaning and flaw detection system and method based on unmanned aerial vehicles (UAVs), aiming to solve the problem of achieving efficient, non-contact cleaning and automated internal defect detection of high-altitude areas.

[0006] In a first aspect, this application provides a cleaning and flaw detection system based on a drone, including a drone body, and the drone body is equipped with at least the following modules: The acquisition module is used to acquire infrared scanning data and image data of the target area; The control module is used to acquire infrared scanning data and image data, as well as laser parameters, flaw detection parameters, cleaning path and flaw detection path determined based on infrared scanning data and image data, and to send the laser parameters to the laser cleaning module and the flaw detection parameters to the flaw detection module. The laser cleaning module is used to perform laser cleaning on the target area based on laser parameters while the drone moves along the cleaning path. The flaw detection module is used to collect flaw detection data of the target area based on flaw detection parameters while the UAV moves along the flaw detection path after laser cleaning of the target area is completed.

[0007] In some embodiments, the control module is further configured to: Dirt identification is performed on image data to determine the location and type of dirt within the target area; Based on infrared scanning data, determine the dirt thickness corresponding to the location of the dirt; Based on the type and thickness of the dirt, the laser parameters corresponding to the location of the dirt are determined.

[0008] In some embodiments, the control module is further configured to: Based on the location of the dirt, its type, and its thickness, a cleaning path is generated for the target area.

[0009] In some embodiments, the control module is further configured to: Use the cleaning path as the flaw detection path corresponding to the target area; or... Based on image data, generate the flaw detection path corresponding to the target area.

[0010] In some embodiments, the control module is further configured to: Based on image data, identify the material type corresponding to the target area; Based on infrared scanning data, determine the material thickness corresponding to the target area; Based on the material type and thickness, determine the flaw detection parameters corresponding to the target area.

[0011] In some embodiments, the control module is further configured to: Acquire temperature data and cleaning images of the cleaned areas within the target region; The cleaning quality assessment is performed based on temperature data and cleaning images, and the assessment results are obtained. The laser parameters are updated based on the evaluation results, enabling the laser cleaning module to perform laser cleaning on the target area based on the updated laser parameters as the drone moves along the cleaning path.

[0012] In some embodiments, the control module is further configured to: After the laser cleaning of the target area is completed, the laser cleaning module is stopped and the flaw detection module is started.

[0013] In some embodiments, the flaw detection module is further configured to: During the movement of the UAV body along the flaw detection path, flaw detection data is collected inside the target area based on the flaw detection parameters; Flaw detection images are generated based on flaw detection data and sent to the control module, which then identifies defects in the images.

[0014] In some embodiments, the laser cleaning module includes: Laser source; An optical transmission component is used to transmit the laser beam emitted by the laser source to the laser cleaning probe; A laser cleaning probe, including a focusing lens assembly, is used to focus a laser beam to form a focused spot on the target area; Temperature sensing components are used to monitor temperature data of the target area in real time.

[0015] Secondly, this application also provides a cleaning and flaw detection method based on a drone, applied to a drone-based cleaning and flaw detection system. The system includes a drone body, and the drone body is equipped with a data acquisition module, a control module, a laser cleaning module, and a flaw detection module. The method includes: The acquisition module collects infrared scanning data and image data of the target area. Acquire laser parameters, flaw detection parameters, cleaning path, and flaw detection path determined based on infrared scanning data and image data; The laser parameters are sent to the laser cleaning module, and the flaw detection parameters are sent to the flaw detection module. As the drone moves along the cleaning path, the laser cleaning module controls the laser cleaning module to perform laser cleaning on the target area based on laser parameters. As the UAV moves along the flaw detection path, the flaw detection module controls the collection of flaw detection data on the target area based on the flaw detection parameters.

[0016] Beneficial Effects: The UAV-based cleaning and flaw detection system of this application includes a UAV body, which is also equipped with a data acquisition module, a control module, a laser cleaning module, and a flaw detection module. The data acquisition module acquires infrared scanning data and image data of the target area, and the control module acquires laser parameters, flaw detection parameters, cleaning path, and flaw detection path determined based on the infrared scanning data and image data. As the UAV moves along the cleaning path, the laser cleaning module performs laser cleaning on the target area according to the laser parameters, and as the UAV moves along the flaw detection path, the flaw detection module acquires flaw detection data on the target area according to the flaw detection parameters. This achieves adaptive acquisition of the operating parameters of the target area, ensuring the cleaning effect and accurate flaw detection of the target area.

[0017] Furthermore, by integrating a laser cleaning module and a flaw detection module onto the drone, laser cleaning is performed using a laser cleaning model, and while ensuring the cleaning effect, the flaw detection module detects flaws in the target area to identify internal defects in the cleaned target area. This achieves integrated non-contact cleaning and damage detection of the target area, thereby improving the automation level of drone high-altitude operations and the efficiency of damage detection maintenance operations. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of a drone-based cleaning and flaw detection system provided by an exemplary embodiment of this disclosure; Figure 2 This is a schematic diagram of the specific structure of the UAV-based cleaning and flaw detection system provided in the exemplary embodiments of this disclosure; Figure 3 This is a flowchart illustrating a UAV-based cleaning and flaw detection method provided by an exemplary embodiment of this disclosure.

[0020] Reference numerals: 1-UAV body; 2-Control module; 3-UAV arm connecting component; 4-UAV propeller; 5-Solar panel assembly; 6-Equipment support and connecting bracket; 7-UAV landing gear; 8-Equipment storage / battery compartment; 9-Laser cleaning equipment transmitter head; 10-Acquisition module. Detailed Implementation

[0021] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] In the description of this application, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are used only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, "a plurality of" means two or more, unless otherwise explicitly specified.

[0023] Firstly, this embodiment provides a cleaning and flaw detection system based on unmanned aerial vehicles (UAVs), such as... Figure 1 As shown, the system includes a drone body 101, and the drone body 101 is equipped with at least the following modules: Acquisition module 102 is used to acquire infrared scanning data and image data of the target area; The control module 103 is used to acquire infrared scanning data and image data, as well as to acquire laser parameters, flaw detection parameters, cleaning path and flaw detection path determined based on infrared scanning data and image data, and to send the laser parameters to the laser cleaning module and the flaw detection parameters to the flaw detection module. Laser cleaning module 104 is used to perform laser cleaning on the target area based on laser parameters during the movement of the drone body along the cleaning path; The flaw detection module 105 is used to collect flaw detection data of the target area based on flaw detection parameters during the movement of the UAV body along the flaw detection path after the laser cleaning of the target area is completed.

[0024] The target area refers to the area to be cleaned and inspected, such as areas on metal materials like photovoltaic panels or wind turbine blades. Laser parameters are laser control parameters used to instruct the laser cleaning module to emit the required laser beam; these include laser power and frequency. Inspection parameters are control parameters used to instruct the inspection module to collect inspection data for inspection purposes. Inspection data is signal data acquired by the inspection module. Generally, the inspection module may include an ultrasonic inspection component, in which case the inspection data can be signal data collected from the target area using the ultrasonic inspection component.

[0025] The cleaning path refers to the movement path of the UAV body when performing cleaning operations via the laser cleaning module. The flaw detection path refers to the movement path of the UAV body when performing flaw detection operations via the flaw detection module. Flaw detection refers to the operation of detecting defects in the material inside the target area, such as material defects like pores and cracks.

[0026] For example, a cleaning and flaw detection system based on a drone includes a drone body 101, which is further equipped with a data acquisition module 102, a control module 103, a laser cleaning module 104, and a flaw detection module 105. The data acquisition module 102 also includes an infrared acquisition component and an image acquisition component (e.g., a camera). The control module 103 is connected to a host computer via a communication module to receive instructions from the host computer and achieve real-time data transmission and remote control with the host computer.

[0027] The UAV 101 receives cleaning instructions for the target area from the host computer via the control module 103, flies to the target area according to the cleaning instructions, and collects infrared scanning data and image data of the target area through the infrared acquisition component and image acquisition component of the acquisition module 102, and sends the infrared scanning data and image data to the control module 103.

[0028] The control module 103 receives infrared scanning data and image data. It can send the infrared scanning data and image data to a host computer, and the host computer determines the laser parameters, flaw detection parameters, and the corresponding cleaning and flaw detection paths for the target area based on the infrared scanning data and image data. Alternatively, the control module 103 can use a processor to determine the laser parameters, flaw detection parameters, and the corresponding cleaning and flaw detection paths for the target area based on the infrared scanning data and image data. Then, it sends the laser parameters to the laser cleaning module 104 and the flaw detection parameters to the flaw detection module 105.

[0029] The drone 101 begins its movement from the starting point along the cleaning path and activates the laser cleaning module 104. As the drone 101 moves along the cleaning path, the laser cleaning module 104 emits a laser beam towards the target area for laser cleaning. For example, the laser source in the laser cleaning module 104 outputs 1064nm near-infrared light, which is transmitted to the laser cleaning probe via an optical transmission component. The laser cleaning probe automatically focuses based on the target distance and angle, ensuring cleaning accuracy and energy concentration.

[0030] After laser cleaning of the target area is completed, the laser cleaning module 104 stops working, and the drone 101 starts moving from the starting position according to the flaw detection path. The flaw detection module 105 then starts working, so that during the movement of the drone 101 along the flaw detection path, the flaw detection module 105 collects flaw detection data on the internal material of the target area according to the flaw detection parameters, and sends the flaw detection data to the control module 103 for defect detection, or the control module 103 sends the defect data to the host computer for defect detection. The flaw detection module 105 may also include an image processor, used to generate a flaw detection image corresponding to the target area based on the collected flaw detection data, and send the flaw detection image to the control module 103 or the host computer for defect detection.

[0031] In some embodiments, the drone 101 is also equipped with a power supply module, which may be a solar photovoltaic panel, for powering the various modules of the drone-based cleaning and flaw detection system and the drone itself.

[0032] The bottom of the drone 101 can also be equipped with a dust suction module for sucking up debris from the target area, such as dust and fumes generated during laser cleaning. Specifically, the dust suction module includes: a suction port, a conveying channel, a dust collection cup, and a suction motor. The suction port is located on the lower surface of the drone's shell. The conveying channel transports debris sucked in by the suction port to the dust collection cup. The dust collection cup is detachable for easy periodic cleaning. The suction motor provides the power for suction. A roller with a brush is located at the suction port. During suction, the roller is driven by the motor through a transmission mechanism to rotate, thus rotating the brush and assisting in cleaning the area. The roller is made of wear-resistant material and has good anti-slip properties.

[0033] The outer surface of the drone 101 is equipped with a tear-resistant film to protect the drone from damage during the cleaning process. The motors involved in the drone-based cleaning and flaw detection system are all designed to be waterproof and dustproof to ensure the reliable operation of the system under various weather conditions.

[0034] In this embodiment, the acquisition module collects infrared scanning data and image data of the target area, and the control module obtains laser parameters, flaw detection parameters, cleaning path, and flaw detection path determined based on the infrared scanning data and image data. This enables the laser cleaning module to perform laser cleaning on the target area according to the laser parameters while the UAV moves along the cleaning path, and the flaw detection module to collect flaw detection data on the target area according to the flaw detection parameters while the UAV moves along the flaw detection path. This achieves adaptive acquisition of the operation parameters of the target area, ensuring the cleaning effect and accurate flaw detection of the target area.

[0035] Furthermore, by integrating a laser cleaning module and a flaw detection module onto the drone, laser cleaning is performed using a laser cleaning model, and while ensuring the cleaning effect, the flaw detection module detects flaws in the target area to identify internal defects in the cleaned target area. This achieves integrated non-contact cleaning and damage detection of the target area, thereby improving the automation level of drone high-altitude operations and the efficiency of damage detection maintenance operations.

[0036] In some embodiments, when the control module performs the step of acquiring laser parameters determined based on infrared scanning data and image data, it is specifically used for: Dirt identification is performed on image data to determine the location and type of dirt within the target area; Based on infrared scanning data, determine the dirt thickness corresponding to the location of the dirt; Based on the type and thickness of the dirt, the laser parameters corresponding to the location of the dirt are determined.

[0037] For example, the control module is equipped with a processor. After acquiring the infrared scanning data and image data collected by the acquisition module from the target area, the processor's image recognition algorithm can be used to identify dirt in the image data, determining the type and location of the dirt in the image. The dirt type could be, for example, rust, oil, or dust. Then, a mapping model is used to transform the coordinate system of the image location to obtain the dirt location in the target coordinate system. The target coordinate system can be the world coordinate system or a coordinate system with a specified location in the target area as its origin. Specifically, it can be based on identifying positioning features within the target area, such as bolt holes or building joints, and constructing a spatial coordinate system using these features as the origin as the target coordinate system.

[0038] Then, based on the infrared scanning data, the thickness of the dirt corresponding to the location is determined, and a preset thickness threshold corresponding to the dirt type is obtained. The dirt thickness is compared with the preset thickness threshold, and the dirt level, i.e., the severity of the dirt, is determined based on the comparison result. The laser parameters corresponding to the dirt level are obtained as the laser parameters for the corresponding dirt location. For example, if the image data identifies the dirt type in the target area as rust, and the infrared scanning data determines that the dirt thickness is 1mm, which is greater than the preset thickness threshold of 0.5mm, the dirt level is determined to be a thick rust layer. Then, laser parameters specifically for cleaning thick rust layers are used, such as laser power, defocusing amount, and frequency.

[0039] In this embodiment, by identifying the type and thickness of dirt in the target area to determine the corresponding laser parameters, the laser parameters of the target area can be adaptively acquired, thereby ensuring accurate cleaning and cleaning effect of the target area.

[0040] In some embodiments, when the control module performs the step of acquiring the cleaning path determined based on infrared scan data and image data, it is specifically used for: Based on the location of the dirt, its type, and its thickness, a cleaning path is generated for the target area.

[0041] For example, the control module, through the processor, divides the target area into multiple sub-regions based on the location of dirt within the target area. Then, it determines the dirt level of the corresponding sub-region based on the dirt type and thickness corresponding to the dirt location. Since the sub-regions may include clean areas, the dirt level also includes the dirt-free level corresponding to the clean areas.

[0042] Using path planning algorithms, a cleaning path is generated for the target area based on image data and the dirt level of each sub-region. Specifically, this can be done by determining the cleaning order of each sub-region according to its dirt level, and then generating a cleaning path accordingly. For example, cleaning areas with thick dirt first, then areas with light dirt, and skipping clean areas. Alternatively, operators can adjust and optimize the path in real time using a host computer based on a reference path.

[0043] In one embodiment, the control module adjusts the laser parameters and / or flaw detection parameters in real time according to different cleaning needs and cleaning effects. First, it collects preset data on cleaning needs and real-time operating condition feedback data. Then, it completes data calculation and decision-making through mapping models, path optimization algorithms and value judgments, and finally generates specific adjustment instructions for the laser cleaning module and / or flaw detection module.

[0044] Taking laser parameter adjustment as an example, this can involve collecting preset data on cleaning requirements, such as dirt judgment thresholds and cleaning qualification thresholds, as well as collecting real-time operating condition feedback data, such as current laser parameters and flaw detection parameters. The image data is then used to identify the two-dimensional dirt location, dirt type, and dirt thickness. The thickness threshold from the dirt judgment threshold is used to determine the dirt thickness, and combined with the dirt type, the severity of dirt at each two-dimensional dirt location is determined. A mapping module is then used to convert the two-dimensional dirt locations in the image data into three-dimensional coordinates. Based on the laser parameters corresponding to the dirt severity, the laser parameters for the corresponding three-dimensional dirt location are determined.

[0045] A path planning algorithm is used to generate a cleaning path based on the severity of dirt at each three-dimensional dirt location. This allows the drone to automatically adjust its current laser parameters to the corresponding laser parameters at different dirt locations as it moves along the cleaning path.

[0046] Alternatively, images of the cleaned area can be captured, and the cleaning quality can be assessed. The assessment results can then be used to determine a threshold for acceptable cleaning. If the assessment result does not meet the acceptable standard, it indicates that the laser cleaning effect of the corresponding laser parameters in the cleaned area is insufficient. In this case, the laser parameters for the remaining uncleaned areas in the target area will be adjusted.

[0047] In this embodiment, a cleaning path is generated based on the dirt level of the dirt in the target area, which improves the efficiency and accuracy of the cleaning operation.

[0048] In some embodiments, when the control module performs the step of acquiring the flaw detection path determined based on infrared scanning data and image data, it is specifically used for: Use the cleaning path as the flaw detection path corresponding to the target area; or... Based on image data, generate the flaw detection path corresponding to the target area.

[0049] For example, the control module, after acquiring the cleaning path determined based on infrared scan data and image data, uses the cleaning path as the flaw detection path. Alternatively, the processor uses a path planning algorithm to regenerate a path based on the image data as the flaw detection path for the target area. Specifically, this could involve identifying the area of ​​the target area based on the image data and generating a path of corresponding shape as the flaw detection path, for example, generating a bow-shaped path for a large target area. Alternatively, the operator could adjust and optimize the path in real time based on a reference path via a host computer.

[0050] In some embodiments, the control module obtains the corresponding laser parameters according to the dirt level of different sub-regions in the cleaning path and sends them to the laser cleaning module, so that when the drone moves to different sub-regions according to the cleaning path, the laser cleaning module emits laser beams to perform laser cleaning on the sub-regions according to the laser parameters corresponding to the sub-regions.

[0051] In this embodiment, by using the cleaning path as the flaw detection path, or by regenerating the flaw detection path, the flaw detection accuracy of the target area can be improved.

[0052] In some embodiments, when the control module performs the step of acquiring laser parameters, flaw detection parameters, cleaning path, and flaw detection path determined based on infrared scanning data and image data, it is specifically used for: Based on image data, identify the material type corresponding to the target area; Based on infrared scanning data, determine the material thickness corresponding to the target area; Based on the material type and thickness, determine the flaw detection parameters corresponding to the target area.

[0053] For example, the processor of the control module uses a recognition algorithm to identify the image data and determine the material type corresponding to the target area, such as cover glass, frame aluminum alloy, etc. Generally, the material of the target area is consistent. Then, the material thickness corresponding to the target area is determined based on the infrared scanning data. Based on the material type and material thickness of the target area, the flaw detection parameters corresponding to the target area are determined.

[0054] In this embodiment, by determining the flaw detection parameters based on the material type and thickness of the target area, adaptive acquisition of flaw detection parameters can be achieved, thereby improving the flaw detection accuracy of the target area.

[0055] In some embodiments, the control module is further configured to: Acquire temperature data and cleaning images of the cleaned areas within the target region; The cleaning quality assessment is performed based on temperature data and cleaning images, and the assessment results are obtained. The laser parameters are updated based on the evaluation results, enabling the laser cleaning module to perform laser cleaning on the target area based on the updated laser parameters as the drone moves along the cleaning path.

[0056] For example, a temperature sensor can be configured in the laser cleaning module. During the laser cleaning of the target area, the acquisition module can acquire images of the cleaned area in the target area as cleaning images, and the temperature sensor can acquire temperature data of the cleaned area in the target area.

[0057] The control module can send the acquired cleaning images and temperature data to the host computer for cleaning quality assessment and obtain the assessment results returned by the host computer. Alternatively, the control module can use the processor to perform cleaning quality assessment on the cleaning images and temperature data and obtain the assessment results.

[0058] Cleaning quality assessment can specifically involve preprocessing image and temperature data, such as denoising image data and filtering temperature data. Evaluation metrics are then extracted from the preprocessed clean images and temperature data, including stain coverage, surface reflectivity uniformity, maximum temperature, and operational parameter characteristics.

[0059] This can be achieved by identifying areas of residual dirt in a clean image and calculating the ratio of the residual dirt area to the target area to obtain the dirt coverage rate; performing image processing on the clean image to extract the surface reflectivity uniformity; and extracting parameter features from the laser parameters, such as laser power stability.

[0060] Each evaluation indicator is compared with its corresponding preset pass / fail threshold. The comparison result for each indicator is converted into an evaluation score. The evaluation scores for each indicator are then weighted and summed to obtain the final evaluation score. The evaluation result is then determined based on the evaluation score. The evaluation result may be, for example, excellent, good, pass, or fail.

[0061] The control module updates the laser parameters based on the evaluation results, enabling the laser cleaning module to emit laser beams to clean the target area according to the updated laser parameters as the drone moves along the cleaning path. For example, if the evaluation result is excellent, the current laser parameters are maintained and the drone's movement speed is increased to improve work efficiency; if the evaluation result is good, the focal length is fine-tuned or the scan line spacing is adjusted to optimize the overlap rate, which refers to the proportion of overlap between two adjacent laser spots or two scan lines; if the evaluation result is acceptable, and there is slight dirt residue, the laser power can be increased and the working speed reduced, or if there are large temperature fluctuations, the pulse mode can be switched and the power reduced; if the evaluation result is unacceptable, and there is severe dirt residue, the maximum power is adjusted, the focal length is shortened, and the speed is slowed down, or if over-cleaning occurs, the power is reduced, the focal length is lengthened, and the pulse mode is switched.

[0062] In some embodiments, after laser cleaning of the target area is completed, the drone can return to the starting position of the cleaning path, collect images of the target area as cleaning images, evaluate the cleaning quality of the cleaning images through the control module or host computer, and determine the laser parameters corresponding to the next target area based on the evaluation results.

[0063] In this embodiment, by evaluating the cleaning quality of the cleaned image and adjusting the laser parameters based on the evaluation results, adaptive adjustment of the laser parameters can be achieved, thereby improving laser accuracy and cleaning effect.

[0064] In some embodiments, the control module is further configured to: After the laser cleaning of the target area is completed, the laser cleaning module is stopped and the flaw detection module is started.

[0065] For example, a laser cleaning module and a flaw detection module can be mounted at different locations on the drone, and after the laser cleaning of the target area is completed, the laser cleaning module can be controlled to stop working, and the flaw detection module can be controlled to start working.

[0066] Laser cleaning and flaw detection modules can also be integrated into the same location on the drone to save space. The laser cleaning module emits a laser beam towards the target area through a laser emitter, while the flaw detection module collects flaw detection data of the target area through a flaw detection probe.

[0067] In one embodiment, since the flaw detection probe needs to be close to the surface of the target area to collect flaw detection data, the flaw detection probe can be set as a telescopic probe. When the flaw detection module starts working, the telescopic component extends the flaw detection probe to be close to the surface of the target area to collect flaw detection data, and when the flaw detection module stops working, the telescopic component is controlled to reset the flaw detection probe.

[0068] In one embodiment, the drone is further provided with a working position and a standby position. The working position is used to house the laser emitter or flaw detection probe in the working state, and the standby position is used to house the laser emitter or flaw detection probe in the non-working state. Specifically, when the laser cleaning module is working, the laser emitter is located in the working position and emits a laser beam towards the target area, while the flaw detection probe is located in the standby position and does not work. When the laser cleaning module is stopped working, and when the flaw detection module is started working, the positions of the laser emitter and the flaw detection probe are swapped, so that the flaw detection probe is located in the working position to collect flaw detection data from the target area, and the laser emitter is located in the standby position and does not work.

[0069] In this embodiment, by controlling the laser cleaning module to stop working and controlling the flaw detection module to start working after the laser cleaning of the target area is completed, a seamless connection between laser cleaning and damage detection can be achieved, ensuring the cleaning effect and improving the maintenance efficiency of damage detection.

[0070] In some embodiments, the flaw detection module is further configured to: During the movement of the UAV body along the flaw detection path, flaw detection data is collected inside the target area based on the flaw detection parameters; Flaw detection images are generated based on flaw detection data and sent to the control module, which then identifies defects in the images.

[0071] For example, the flaw detection module includes an ultrasonic flaw detection component (specifically, an ultrasonic flaw detector) and an image processor (e.g., an embedded processor). During the movement of the UAV along the flaw detection path, the ultrasonic flaw detection component collects flaw detection data within the target area based on flaw detection parameters. Then, the image processor generates a flaw detection image based on the flaw detection data. Specifically, this can involve using a wavelet threshold denoising algorithm to filter high-altitude electromagnetic and vibration interference, calibrating the reflection coefficient of a standard test block to eliminate probe coupling differences, preserving the true signal of the flaw detection data, extracting defect features based on the acoustic reflection laws of the defects, such as cracks (continuous high-amplitude echoes and fixed time intervals), pores (discrete low-amplitude echoes and frequency fluctuations), inclusions (medium-to-high amplitude echoes and time abrupt changes), and stratification (double-peak echoes and amplitude symmetry). A support vector machine classification model is used to identify the defect features, calculate the vertical depth of the defects, quantify the defect size through echo amplitude integration, and generate a flaw detection image based on the identified defect features, vertical depth, and size.

[0072] The flaw detection module then sends the flaw detection images to the control module. The control module identifies defects in the flaw detection images, such as performing defect statistics and defect level assessment to generate a standardized defect report, which is then uploaded to the host computer in real time. The host computer classifies the defects according to their severity (e.g., minor, moderate, severe) and triggers corresponding warnings. If the defect affects structural safety, the control system suspends the cleaning operation, and the drone hovers or returns to base to avoid secondary damage. Subsequent actions include pushing the defect report to maintenance personnel and simultaneously displaying images of the cleaned area to provide data support for developing repair plans (e.g., welding repair, reinforcement).

[0073] In this embodiment, by collecting flaw detection data within the target area and generating flaw detection images for defect identification, material defects within the target area can be accurately located and the degree of damage quantified, thereby improving maintenance efficiency.

[0074] In some embodiments, the laser cleaning module includes: Laser source; An optical transmission component is used to transmit the laser beam emitted by the laser source to the laser cleaning probe; A laser cleaning probe, including a focusing lens assembly, is used to focus a laser beam to form a focused spot on the target area; Temperature sensing components are used to monitor temperature data of the target area in real time.

[0075] For example, a laser cleaning system includes a laser source, an optical transmission component, a laser cleaning probe (i.e., a laser emitter) and a temperature sensing component (i.e., a temperature sensor).

[0076] The laser source uses a high-power-density semiconductor laser source and outputs near-infrared light with a wavelength of 1064nm.

[0077] The optical transmission component includes optical fibers and optical lenses, which are used to shape and position the laser beam output from the laser source and transmit it to the laser cleaning probe.

[0078] The laser cleaning probe includes a focusing lens assembly, which consists of three plano-concave mirrors and one plano-convex mirror. This assembly automatically focuses the laser light according to different distances and angles, ensuring precision and energy concentration during the cleaning process. Specifically, the laser cleaning head employs a three-segment focusing lens design: a double-cemented achromatic correction mirror, a motorized zoom lens assembly, and a plano-convex focusing mirror. This design is adapted to the transmission characteristics of infrared lasers and automatically adjusts the focal length based on changes in the distance and surface tilt of the target. The principles of each component are as follows: Double-cemented correction mirror: By bonding two types of glass with different refractive indices, chromatic aberration and spherical aberration in laser transmission are offset, preventing beam divergence; Motorized zoom lens assembly: By changing the relative distance between the two zoom lenses, the equivalent focal length of the entire assembly is adjusted, achieving continuous focal length adjustment within the range of 150-300mm; Plano-convex focusing mirror: This focuses the corrected and zoomed parallel laser beam to a single point, utilizing the high light transmittance of quartz material to reduce energy loss, ultimately forming a high-power-density focused spot.

[0079] A temperature sensor can be installed on the laser cleaning probe to monitor the temperature of the cleaning area in real time and prevent damage caused by overheating.

[0080] In this embodiment, the laser cleaning module includes a laser cleaning probe, which automatically focuses based on the target distance and angle to ensure cleaning accuracy and energy concentration.

[0081] In one specific embodiment, such as Figure 2 The schematic diagram of the specific structure of the UAV-based cleaning and flaw detection system shown includes: The drone body 1 is used to integrate and carry a laser cleaning module, a flaw detection module, a data acquisition module, a control module, and a power module, providing installation support and protection for each module. The drone body 1 has four folding arms arranged along its circumference. The four folding arms are symmetrical about each other with respect to the center line. The internal and middle parts of the folding arms are connected as a whole. Each folding arm has a propeller at its outer end. The lower surface of the middle part is fixedly connected to a bracket. The entire drone body is made by 3D printing and is lightweight.

[0082] Control module 2 is responsible for receiving instructions from the host computer, coordinating the work of various modules such as the laser cleaning module, flaw detection module, data acquisition module, and power module, and processing and transmitting operation data.

[0083] The drone arm connecting component 3 is used to securely connect the arm and the fuselage, ensuring the structural stability of the drone during flight and enabling the power system (such as motors and propellers) on the arm to operate normally.

[0084] The drone's propeller 4 generates lift through rotation, enabling the drone to take off, hover, and move, providing aerial mobility for the entire operation.

[0085] Solar panel assembly 5 is used to convert solar energy into electrical energy to power the drone's power module, control module, laser cleaning module, and flaw detection module. It can also charge the backup battery, realizing the solar charging function and improving the equipment's endurance.

[0086] The equipment support and connection bracket 6 is used to support the solar panel assembly and some equipment modules, ensuring the stability of the installation position of each component on the drone, and also serves to connect different modules.

[0087] The drone landing gear 7 provides support and cushioning during takeoff and landing, protecting the fuselage and onboard equipment from ground impact damage.

[0088] The equipment storage / battery compartment 8 can be used to store spare batteries or other small equipment parts, providing additional power support or storage space for the equipment.

[0089] The laser cleaning equipment emitter head 9 is used to emit a laser beam. The laser energy acts on the surface of high-rise buildings and dangerous bridges to remove dirt and provide a clean inspection surface for subsequent flaw detection operations. At the same time, it can also integrate flaw detection probes (such as ultrasonic probes, eddy current probes, etc.) into the flaw detection module, so that flaw detection operations can be performed directly after cleaning to detect defects inside the structure or on the surface.

[0090] The acquisition module 10 is used to capture surface conditions and upload them to the backend, such as the control module or host computer, in real time.

[0091] The main parameters of the UAV-based cleaning and flaw detection equipment combination provided in this embodiment are as follows: Drone body: Drone model: DJII Inspire 2, maximum flight speed: 60km / h, maximum flight altitude: 500 meters, drone dimensions: 430mm×320mm×260mm, drone weight: 1.2 kg, battery type: lithium battery, capacity: 5000mAh; Laser cleaning module: Laser model: YLR-1064-200W, output wavelength: 1064nm, output power: 200W, beam quality number: M2<1.2, focusing lens combination: composed of 3 plano-concave mirrors and 1 plano-convex mirror, focal length: 200mm, temperature sensor model: PT100; Flaw detection module: Flaw sensor model: ET8000-Ultrasound flaw detector, flaw detection frequency: 50kHz, flaw detection depth: 100mm, data processing unit: Intel i7 processor, 16GB memory, running Windows 10 operating system; Control module: Processor: STM32F4 processor; Communication module: 5G router; The acquisition module includes an infrared image acquisition component and a visible light image acquisition component: camera models: FLIRLEPTON 3.5 and Canon EOS M50 respectively, resolution: 640×480 and 1920×1080 respectively; Vacuuming system: Vacuum motor: 400W centrifugal pump, dust collection cup capacity: 3 liters, roller material: polyoxymethylene, diameter 100mm, width 50mm.

[0092] The above parameters are only a preferred embodiment of the present invention. Without departing from the principle and spirit of the present invention, the parameters of each module and component can be adjusted and replaced.

[0093] On the other hand, this embodiment provides a cleaning and flaw detection method based on unmanned aerial vehicles (UAVs), such as... Figure 3 As shown, an unmanned aerial vehicle (UAV)-based cleaning and flaw detection method system, applicable to any of the above-mentioned methods, includes the following steps: S301, acquire the infrared scanning data and image data collected by the acquisition module from the target area; S302, acquire the laser parameters, flaw detection parameters, cleaning path and flaw detection path determined based on infrared scanning data and image data; S303 sends laser parameters to the laser cleaning module and flaw detection parameters to the flaw detection module; S304: During the movement of the drone body along the cleaning path, the laser cleaning module is controlled to perform laser cleaning on the target area based on laser parameters; S305: During the movement of the UAV body along the flaw detection path, the flaw detection module controls the flaw detection module to collect flaw detection data on the target area based on the flaw detection parameters.

[0094] For any parts of this method not described in detail in the embodiments, please refer to the corresponding tables in the above text, and they will not be repeated here.

[0095] Thirdly, embodiments of this application provide a computer-readable storage medium storing computer instructions, which are loaded by a processor to execute the arrangements in any of the methods described above. In embodiments of this application, the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc.

[0096] Fourthly, embodiments of this application provide a computer program product, including a computer program or instructions, which are executed by a processor to implement the steps of any of the methods described above.

[0097] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0098] The above provides a detailed description of a cleaning and flaw detection system and method based on unmanned aerial vehicles (UAVs) provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A cleaning and flaw detection system based on unmanned aerial vehicles (UAVs), characterized in that, Including the drone body, the drone body is equipped with at least the following modules: The acquisition module is used to acquire infrared scanning data and image data of the target area; The control module is used to acquire the infrared scanning data and the image data, and to acquire the laser parameters, flaw detection parameters, cleaning path and flaw detection path determined based on the infrared scanning data and the image data, and to send the laser parameters to the laser cleaning module and the flaw detection parameters to the flaw detection module. A laser cleaning module is used to perform laser cleaning on the target area based on the laser parameters as the drone body moves along the cleaning path. The flaw detection module is used to collect flaw detection data of the target area based on the flaw detection parameters during the movement of the UAV body along the flaw detection path after the laser cleaning of the target area is completed.

2. The system according to claim 1, characterized in that, The control module is also used for: Dirt identification is performed on the image data to determine the location and type of dirt within the target area; Based on the infrared scanning data, the thickness of the dirt corresponding to the location of the dirt is determined; Based on the type and thickness of the dirt, the laser parameters corresponding to the location of the dirt are determined.

3. The system according to claim 2, characterized in that, The control module is also used for: Based on the location of the dirt and the corresponding dirt type and thickness, a cleaning path is generated for the target area.

4. The system according to claim 1, characterized in that, The control module is also used for: Use the cleaning path as the flaw detection path corresponding to the target area; or... Based on the image data, a flaw detection path corresponding to the target area is generated.

5. The system according to claim 1, characterized in that, The control module is also used for: Based on the image data, the material type corresponding to the target area is identified; Based on the infrared scanning data, the material thickness corresponding to the target area is determined; Based on the material type and the material thickness, the flaw detection parameters corresponding to the target area are determined.

6. The system according to claim 1, characterized in that, The control module is also used for: Acquire temperature data and cleaning images of the cleaned areas within the target region; The cleaning quality assessment is performed based on the temperature data and the cleaning image, and the assessment result is obtained. Based on the evaluation results, the laser parameters are updated so that the laser cleaning module performs laser cleaning on the target area based on the updated laser parameters as the drone moves along the cleaning path.

7. The system according to claim 1, characterized in that, The control module is also used for: After the laser cleaning of the target area is completed, the laser cleaning module is controlled to stop working, and the flaw detection module is controlled to start working.

8. The system according to claim 1, characterized in that, The flaw detection module is also used for: During the movement of the UAV body along the flaw detection path, flaw detection data is collected inside the target area based on the flaw detection parameters. Based on the flaw detection data, a flaw detection image is generated and sent to the control module, which then performs defect identification on the flaw detection image.

9. The system according to claim 1, characterized in that, The laser cleaning module includes: Laser source; An optical transmission component is used to transmit the laser beam emitted by the laser source to the laser cleaning probe; A laser cleaning probe, including a focusing lens group, is used to focus the laser beam to form a focused spot on the target area; A temperature sensing component is used to monitor the temperature data of the target area in real time.

10. A cleaning and flaw detection method based on unmanned aerial vehicles (UAVs), characterized in that, An application is made to a drone-based cleaning and flaw detection system, the system comprising a drone body equipped with a data acquisition module, a control module, a laser cleaning module, and a flaw detection module, the method comprising: The infrared scanning data and image data collected by the acquisition module on the target area are obtained; Acquire the laser parameters, flaw detection parameters, cleaning path, and flaw detection path determined based on the infrared scanning data and the image data; The laser parameters are sent to the laser cleaning module, and the flaw detection parameters are sent to the flaw detection module; During the movement of the drone body along the cleaning path, the laser cleaning module is controlled to perform laser cleaning on the target area based on the laser parameters; During the movement of the UAV body along the flaw detection path, the flaw detection module is controlled to collect flaw detection data on the target area based on the flaw detection parameters.

Images