Cleaning equipment and cleaning method

Abstract

The invention discloses cleaning equipment and a cleaning method, and relates to the technical field of cleaning devices. The cleaning equipment comprises an objective table, a laser transmitter, a jet ejector, an image acquisition module and a controller. The objective table is used for bearing a to-be-cleaned part; the laser transmitter is used for transmitting laser to a to-be-cleaned part so as to clean dirt on the surface of the to-be-cleaned part; the jet flow ejector is used for ejecting liquid to the to-be-cleaned part so as to clean dirt on the surface of the to-be-cleaned part; the image acquisition module is used for acquiring image data of a to-be-cleaned part; the controller is electrically connected with the laser transmitter, the jet flow ejector and the image acquisition module; the controller is used for controlling the working state of the laser transmitter and / or the working state of the jet ejector based on smudginess parameters of the to-be-cleaned surface obtained by the image data, the smudginess parameters comprise the smudginess type, the smudginess thickness and the smudginess position, and the working state at least comprises a start-stop state.

Description

Technical Field

[0001] This application relates to the field of cleaning equipment technology, and in particular to a cleaning device and cleaning method. Background Technology

[0002] As industrial manufacturing, precision electronics, and cultural relic restoration fields increasingly demand higher precision, efficiency, and environmental friendliness in surface cleaning, traditional methods such as chemical cleaning and mechanical grinding are no longer sufficient to meet the needs of high-end applications due to their drawbacks, including environmental pollution, damage to substrates, and poor cleaning consistency. Against this backdrop, laser cleaning technology has gained widespread attention due to its advantages of being non-contact, consumable-free, and highly precise. It uses laser energy to vaporize and peel off contaminants, achieving highly efficient cleaning.

[0003] In related technologies, laser cleaning equipment is not effective at cleaning stubborn dirt and cannot meet the needs of high-end fields for efficient and stable cleaning. Summary of the Invention

[0004] Based on this, this application provides a cleaning device and a cleaning method to solve the problem that laser cleaning devices in the related art have poor cleaning effects on stubborn dirt.

[0005] In a first aspect, embodiments of this application provide a cleaning device, comprising:

[0006] A platform for holding items to be cleaned;

[0007] A laser emitter is used to emit a laser beam toward the part to be cleaned in order to clean the part;

[0008] A jet injector is used to spray liquid onto the part to be cleaned in order to clean the part;

[0009] The image acquisition module is used to acquire image data of the part to be cleaned;

[0010] The controller is electrically connected to the laser emitter, the jet injector, and the image acquisition module. Based on the dirt parameters of the surface to be cleaned obtained from the image data, the controller controls the working state of the laser emitter and / or the working state of the jet injector. The dirt parameters include dirt type, dirt thickness, and dirt location. The working state includes at least a start / stop state.

[0011] In some embodiments, the cleaning device further includes:

[0012] The image analysis module is electrically connected to both the image acquisition module and the controller, and the image analysis module obtains the dirt parameters based on the image data.

[0013] In some embodiments, the jet injector includes a plurality of nozzles, wherein the liquid jets ejected from the plurality of nozzles and the laser beam emitted from the laser emitter converge at the same location on the surface of the part to be cleaned.

[0014] In some embodiments, the cleaning device includes a plurality of image acquisition modules, which are used to acquire image data of the same area of ​​the object to be cleaned from different angles.

[0015] In some embodiments, the cleaning device further includes a drive mechanism electrically connected to the controller, wherein the laser emitter and the jet ejector are both mounted on the output end of the drive mechanism; the controller controls the drive mechanism to adjust the positions of the laser emitter and the jet ejector.

[0016] Secondly, embodiments of this application provide a cleaning method, implemented based on the cleaning equipment described in the first aspect, the cleaning method comprising:

[0017] Collect image data of the part to be cleaned;

[0018] Based on the image data, the dirt parameters of the part to be cleaned are obtained, including dirt type, dirt thickness and dirt location;

[0019] Based on the dirt parameters, the laser emitter is controlled to emit a laser towards the dirt on the part to be cleaned, and / or the jet injector is controlled to spray liquid towards the dirt on the part to be cleaned.

[0020] In some embodiments, controlling the laser emitter to emit laser light towards the dirt on the part to be cleaned based on the dirt parameters, and / or controlling the jet injector to spray liquid towards the dirt on the part to be cleaned, includes:

[0021] Control the jet injector to spray liquid onto the dirt on the part to be cleaned;

[0022] The laser emitter is controlled to emit a laser at a first emission power toward the dirt on the part to be cleaned.

[0023] The laser emitter is controlled to emit laser light at a second emission power toward the dirt on the part to be cleaned; wherein the first emission power is greater than the second emission power.

[0024] In some embodiments, while controlling the laser emitter to emit laser light at a second emission power toward the dirt on the part to be cleaned, the jet injector is also controlled to spray liquid toward the dirt on the part to be cleaned.

[0025] In some embodiments, during the cleaning process of controlling the laser emitter to emit laser and / or the jet injector to spray liquid, real-time image data of the cleaning area on the part to be cleaned is continuously acquired;

[0026] Based on the real-time image data, compare the changes in texture features and spectral features of the cleaned area during the cleaning process;

[0027] Based on the changes in texture features and spectral features, the cleanliness index of the clean area is obtained.

[0028] In some embodiments, it is determined whether the cleanliness index tends to stabilize;

[0029] If the cleanliness index tends to stabilize, determine whether the cleanliness index has reached the target cleanliness threshold;

[0030] If the cleanliness index does not reach the target cleanliness threshold, an enhanced cleaning strategy is executed; the enhanced cleaning strategy includes at least one of the following: increasing the emission power of the laser emitter, reducing the scanning speed of the laser emitter, and increasing the jet impact intensity of the jet ejector.

[0031] In some embodiments, during the cleaning process of controlling the laser emitter to emit laser and / or the jet injector to spray liquid, real-time image data of the cleaning area on the part to be cleaned is continuously acquired;

[0032] Based on the real-time image data, compare the changes in texture features and spectral features of the cleaned area before and after cleaning;

[0033] Based on the changes in texture features and spectral features, the cleanliness index of the clean area is obtained;

[0034] Determine whether the cleanliness index has reached the target cleanliness threshold;

[0035] If the cleanliness index does not reach the target cleaning threshold, then control the jet injector to spray liquid onto the dirt on the part to be cleaned, control the laser emitter to emit laser light onto the dirt on the part to be cleaned at the first emission power, and control the laser emitter to emit laser light onto the dirt on the part to be cleaned at the second emission power, until the cleanliness index reaches the target cleaning threshold or reaches the preset maximum number of repetitions.

[0036] This application has at least the following beneficial effects:

[0037] The cleaning equipment provided in this application includes a stage, a laser emitter, a jet ejector, an image acquisition module, and a controller. Through the coordinated configuration of these components, three operating modes can be flexibly implemented: laser-only cleaning, jet-only cleaning, and laser-jet combined cleaning, effectively expanding application scenarios. Laser-only cleaning is suitable for waterless environments, sensitive material surfaces, or the rapid treatment of thin, slightly soiled surfaces, avoiding damage to the parts being cleaned through non-contact energy action. Jet-only cleaning is suitable for low-damage cleaning and the removal of dust or easily soluble dirt, achieving gentle and efficient cleaning through liquid impact. In laser-jet combined cleaning, the vaporization and abrasion effects of the laser energy complement the dirt softening, debris removal, and cooling functions of the jet, improving the removal efficiency of thick, stubborn dirt (such as rust and paint layers), reducing the heat-affected zone generated by laser operation, lowering the risk of deformation or aging of the parts being cleaned, and ensuring cleaning accuracy and surface quality. Meanwhile, relying on the image acquisition module to acquire images and in conjunction with the adaptive control of the controller, the cleaning equipment provided in this application can be precisely adapted to the parts to be cleaned with different materials and different dirt characteristics, taking into account both the excellent cleaning effect and the flexibility of operation, and is suitable for the cleaning needs of multiple industries such as precision manufacturing, automotive parts processing, and electronic component production. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is a schematic diagram of the structure of the cleaning equipment in some embodiments of this application.

[0040] Figure 2 This is a flowchart of a cleaning method in some embodiments of this application.

[0041] Figure 3 This is a flowchart of a sub-step of step S300 of the cleaning method in some embodiments of this application.

[0042] Explanation of reference numerals in the attached figures:

[0043] 100-Cleaning equipment, 110-Stage, 120-Laser emitter, 130-Jet injector, 131-Nozzle, 140-Image acquisition module, 150-Image analysis module, 160-Controller, 170-Drive mechanism. Detailed Implementation

[0044] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions in the embodiments of this application will be described in more detail below with reference to the accompanying drawings. In the drawings, the same or similar reference numerals denote the same or similar components or components having the same or similar functions throughout. The described embodiments are some, but not all, of the embodiments of this application. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. The embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0045] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, an indirect connection through an intermediate medium, or the internal communication between two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0046] In the description of this application, it should be understood that the terms "upper", "lower", "front", "back", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the accompanying drawings, and are 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.

[0047] The terms “first,” “second,” and “third” (if any) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0048] Furthermore, the terms “comprising” and “having”, and any variations thereof, are intended to cover non-exclusive inclusion, such that a process, method, system, product, or display that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such process, method, product, or display.

[0049] In related technologies, laser cleaning equipment is not effective at cleaning stubborn dirt and cannot meet the needs of high-end fields for efficient and stable cleaning.

[0050] In view of this, the inventors have designed a cleaning device and a cleaning method. The cleaning device and cleaning method provided in the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0051] like Figure 1 As shown, the cleaning device 100 includes: a stage 110, a laser emitter 120, a jet sprayer 130, an image acquisition module 140, and a controller 160. The stage 110 is used to hold the object to be cleaned; the laser emitter 120 is used to emit laser light onto the object to be cleaned to clean the dirt on the surface of the object; the jet sprayer 130 is used to spray liquid onto the object to be cleaned to clean the dirt on the surface of the object; the image acquisition module 140 is used to acquire image data of the object to be cleaned; the controller 160 is electrically connected to the laser emitter 120, the jet sprayer 130, and the image acquisition module 140; the controller 160 controls the working state of the laser emitter 120 and / or the jet sprayer 130 based on the dirt parameters of the surface to be cleaned obtained from the image data. The dirt parameters include dirt type, dirt thickness, and dirt location, and the working state includes at least a start / stop state.

[0052] The stage 110 is used to stably support the parts to be cleaned, ensuring the positional stability of the parts during the cleaning process and preventing displacement from affecting the accuracy of the laser and jet. The stage 110 can adopt a flat or hollow structure, and the size of the stage 110 can be adapted to the actual specifications of the parts to be cleaned, which is not limited in this application. The stage 110 can be made of high-strength corrosion-resistant materials such as high-strength aluminum alloy or stainless steel, and the surface can be anodized or passivated to have both wear resistance and corrosion resistance. If the parts to be cleaned are fragile materials (such as glass or ceramics), a cushioning pad of a certain thickness (such as a polyurethane cushioning pad) can be laid on the surface of the stage 110 to reduce the risk of damage during contact between the parts to be cleaned and the stage 110. In addition, the stage 110 can be designed with an adjustable height and angle to adapt to different cleaning angles and cleaning heights. Specifically, it can be a manually adjustable stage 110 or an electric stage 110, which is not limited in this application.

[0053] The laser emitter 120 emits a laser beam towards dirt on the surface of the part to be cleaned. The laser energy vaporizes, peels off, or breaks down the dirt, thus achieving non-contact cleaning. The laser emitter 120 can adopt a pulsed or continuous wave structure, etc., and is not limited in this application. The output power, pulse frequency, and scanning speed of the laser emitter 120 can be adjusted as needed to adapt to the cleaning requirements of dirt of different thicknesses and types.

[0054] The jet ejector 130 is used to spray liquid onto the surface of the part to be cleaned. Through the impact of the liquid on the surface, it can directly peel off dirt adhering to the surface, break down the adhesion between dirt and the surface, and simultaneously soften the dirt, instantly remove debris generated during laser cleaning, and cool the surface. The liquid sprayed by the jet ejector 130 can be water, deionized water, neutral cleaning solution, etc., and can be mixed with compressed air, nitrogen, or other gases to form a gas-liquid mixed jet, enhancing the cavitation effect and cleaning impact force. The cavitation effect refers to the instantaneous high-pressure shock wave generated by the rapid collapse of air bubbles inside the liquid when the gas-liquid mixed jet contacts the surface of the part to be cleaned. This shock wave can help peel off tiny dirt particles from the surface or break down the adhesion between dirt and the surface, improving the cleaning effect.

[0055] The jet ejector 130 has a variety of structures, including a nozzle 131, pipelines, a water pump, and a liquid supply device. These components work together to store, pressurize, transport, and eject the liquid. The liquid supply device stores the liquid required for ejection. The pipelines form sealed connections with the nozzle 131, the water pump, and the liquid supply device. The water pump draws liquid from the liquid supply device, pressurizes it, and then transports it through the pipeline to the nozzle 131, from which the liquid is finally ejected. Pressure sensors and flow control valves can be installed on the pipelines to collect jet pressure and flow data in real time and feed them back to the controller 160 for dynamic adjustment, achieving precise adaptation of jet parameters. The orifice diameter of the nozzle 131 can be flexibly designed according to the jet pressure and cleaning accuracy, and is not limited in this application. Its material can be wear-resistant materials such as hard alloy or zirconia ceramic, combining corrosion resistance and lightweight characteristics, making it suitable for long-term high-pressure jetting scenarios.

[0056] The image acquisition module 140 is used to acquire image data of the surface of the part to be cleaned, providing data support for image analysis by the subsequent image analysis module 150. The image acquisition module 140 may include an imaging unit and an illumination unit. The imaging unit can be a general-purpose imaging unit or a multispectral imaging unit. The general-purpose imaging unit can be an industrial camera equipped with a charge-coupled device (CCD) or complementary metal-oxide-semiconductor (CMOS) sensor. The lenses of these cameras can be fixed-focus or zoom, and the field of view can be flexibly adapted according to the size of the stage 110, which is not limited in this application. The multispectral imaging unit can capture multi-band spectral information of the surface of the part to be cleaned, identify specific substances or elements in contaminants, confirm contaminants based on big data, and help to more accurately distinguish types of dirt with similar chemical compositions. The illumination unit can be a light-emitting diode (LED) light source, and the image acquisition module 140 can be equipped with multiple illumination units to illuminate various areas of the part to be cleaned, reducing the impact of shadows on image details.

[0057] The cleaning equipment 100 may also include an image analysis module 150, which is electrically connected to the image acquisition module 140 and the controller 160. The image analysis module 150 receives and processes image data transmitted by the image acquisition module 140 to detect dirt parameters on the surface of the part to be cleaned. The dirt parameters include dirt type, dirt thickness, and dirt location. The specific acquisition method for each parameter can be achieved through the hardware configuration and algorithm support of the image analysis module 150, and is not limited in this application. The image analysis module 150 can use an industrial computer or an embedded processing module as its hardware carrier. Its central processing unit frequency, memory, and storage capacity can be flexibly adapted according to data processing needs to ensure the stable operation of algorithms such as image preprocessing, dirt segmentation, and parameter recognition. At the algorithm level, deep learning (such as convolutional neural networks) or traditional image recognition algorithms can be used. Different algorithm selections can adapt to the recognition needs of different cleaning scenarios. Based on the characteristics and optimization space of the selected algorithm, the dirt recognition accuracy can be optimized and improved in combination with the actual scenario, and the dirt location positioning accuracy can be flexibly adjusted according to cleaning needs.

[0058] Of course, the cleaning device 100 can also analyze images based on the cloud, meaning that the cleaning device 100 does not have an image analysis module 150. In this case, the controller 160 can transmit the image data acquired by the image acquisition module 140 to the cloud intelligent engine or other electronic devices (which have computing capabilities) via data communication protocols (such as Bluetooth, cellular network, Wi-Fi, etc.); the cloud intelligent engine or the electronic device will then process the image data and determine the dirt parameters, and then transmit the aforementioned dirt parameters to the controller 160.

[0059] The type of dirt can be determined by extracting the texture features, grayscale distribution features, or spectral features of the dirty area in the image, comparing the extracted features with a preset dirt feature database (containing feature templates of common dirt such as rust, paint, grease, and oxide layer), and calculating the feature similarity through an algorithm to determine the type of dirt.

[0060] The thickness of dirt can be obtained based on differences in image grayscale values. A dirt thickness-grayscale value calibration model (pre-trained and fitted using dirt samples of different standard thicknesses) can be established, and the thickness can be inferred by combining the grayscale difference between the dirty area and the clean surface area of ​​the part to be cleaned. Alternatively, the transmission characteristics of multispectral images can be utilized; by analyzing the differences in light absorption and reflection in the dirty area at different wavelengths, a thickness estimation model can be constructed to achieve quantitative assessment. Dirt thickness can also be identified using a binocular camera. Based on the principle of binocular parallax, parallax images are acquired, parallax information is extracted, and combined with calibration parameters to calculate the height difference between the dirt and the clean surface, thus achieving thickness quantification.

[0061] The location of dirt can be obtained by first separating the dirty area from the image using image segmentation algorithms (such as threshold segmentation and semantic segmentation) to determine its pixel coordinates in the image coordinate system. Then, combined with the calibration parameters of the image acquisition module 140 (such as camera intrinsics, shooting distance, and lens focal length), the pixel coordinates are mapped to the actual physical coordinates of the surface of the part to be cleaned through a coordinate transformation algorithm, thereby locating the specific location of the dirt.

[0062] The controller 160 is electrically connected to the laser emitter 120, the jet injector 130, and the image analysis module 150. It receives dirt parameters output by the image analysis module 150 and automatically adjusts the operating state of the laser emitter 120 and / or the jet injector 130 based on these parameters. Simultaneously, it coordinates the timing of each module to ensure automated and precise operation of the cleaning process. The operating state may include start / stop status, operating parameters, laser irradiation position, jet injection position, etc., which are not limited in this application.

[0063] The start / stop state refers to the laser emitter 120 and / or the jet ejector 130 being in a working (or starting) state or a stopped working state. The laser emitter 120 being in a working state includes at least a change in the position of the laser emitter 120 or the emission of a laser. The jet ejector 130 being in a working state includes at least a change in the position of the jet ejector 130 or the ejection of a jet. The operating parameters include at least the laser parameters, the jet parameters, and the coordination timing between the laser emitter 120 and the jet ejector 130.

[0064] Specifically, the controller 160 controls the operating state of the laser emitter 120 and / or the jet ejector 130, including the following three control scenarios:

[0065] The first option is to simultaneously control the start / stop status of the laser emitter 120 and the jet ejector 130, adjust the operating parameters of the laser emitter 120 and the jet ejector 130, and adjust the laser irradiation position and the jet ejection position, etc.; the second option is to only control the start / stop status, operating parameters, and laser irradiation position of the laser emitter 120; and the third option is to only control the start / stop status, operating parameters, and jet ejection position of the jet ejector 130.

[0066] The controller 160 can preset a cleaning strategy library, matching the corresponding laser parameters (such as emission power, scanning speed, pulse frequency), jet parameters (such as pressure, flow rate, liquid type), and the timing of their coordination for different types and thicknesses of dirt (such as using a sequence of jet pretreatment, laser ablation, and jet cleaning for thick dirt layers, or using a sequence of synchronous action of laser and jet for precision cleaning).

[0067] Specifically, for the same cleaning cycle, the timing coordination between the laser emitter 120 and the jet ejector 130 includes at least sequential timing, synchronous timing, and staggered timing.

[0068] Specifically, the sequential timing is defined as follows: the start time stamp of laser emission from laser emitter 120 is different from the start time stamp of jet ejection from jet ejector 130, i.e., the start time stamp of laser emission from laser emitter 120 is earlier or later than the start time stamp of jet ejection from jet ejector 130; the synchronous timing is defined as follows: the start time stamp of laser emission from laser emitter 120 is the same as the start time stamp of jet ejection from jet ejector 130, or the end time stamp of laser emission from laser emitter 120 is the same as the end time stamp of jet ejection from jet ejector 130; the interleaved timing is defined as follows: the time period corresponding to laser emission from laser emitter 120 and the time period corresponding to jet ejection from jet ejector 130 do not contain the same timestamp.

[0069] The cleaning equipment 100 provided in this application includes a stage 110, a laser emitter 120, a jet ejector 130, an image acquisition module 140, an image analysis module 150, and a controller 160. Through the coordinated configuration of these components, three operating modes can be flexibly realized: laser-only cleaning, jet-only cleaning, and laser-jet combined cleaning, effectively expanding application scenarios. Laser-only cleaning is suitable for waterless environments, sensitive material surfaces, or the rapid treatment of slightly thin dirt, avoiding damage to the parts being cleaned through non-contact energy action. Jet-only cleaning is suitable for low-damage cleaning and the removal of dust or easily soluble dirt, achieving gentle and efficient cleaning through liquid impact. In laser-jet combined cleaning, the vaporization and stripping effects of the laser energy complement the dirt softening, debris removal, and cooling functions of the jet, improving the stripping efficiency of thick, stubborn dirt (such as rust and paint layers), reducing the heat-affected zone generated by laser operation, lowering the risk of deformation or aging of the parts being cleaned, and ensuring cleaning accuracy and surface quality. Meanwhile, relying on the image acquisition module 140 to acquire images and the image analysis module 150 to accurately detect dirt parameters, and with the adaptive control of the controller 160, the cleaning equipment 100 provided in this application can be accurately adapted to the parts to be cleaned with different materials and different dirt characteristics, taking into account both the excellent cleaning effect and the flexibility of operation, and is suitable for the cleaning needs of multiple industries such as precision manufacturing, automotive parts processing, and electronic component production.

[0070] In some embodiments, the jet injector 130 includes a plurality of nozzles 131, and the liquid jets ejected by the plurality of nozzles 131 and the laser beam emitted by the laser emitter 120 converge at the same position on the surface of the part to be cleaned.

[0071] In other words, the liquid jets ejected from multiple nozzles 131 intersect at the same target location on the surface of the part to be cleaned, and the laser beam emitted by the laser emitter 120 is also focused on this target location. This design enables multiple liquid jets and the laser beam to form a strong synergistic effect at the same target location, improving cleaning effectiveness and work efficiency. Specifically, the superimposed impact effect formed by the convergence of multiple jets can more efficiently soften thick layers of dirt, peel off stubbornly adhered deposits, and quickly remove debris and residues generated by laser vaporization, effectively preventing secondary adhesion of dirt or secondary contamination of the cleaned area; at the same time, the convergence of multiple jets can enhance the directional cooling effect on the laser action area, further reducing the adverse effects of laser thermal impact on non-dirty areas and ensuring the stability of the surface quality of the part to be cleaned.

[0072] In some embodiments, the cleaning device 100 includes a plurality of image acquisition modules 140, which are used to acquire image data of the same area of ​​the object to be cleaned from different angles.

[0073] The layout of multiple image acquisition modules 140 is adapted to the need for multiple image acquisition modules 140 to acquire the same area from multiple perspectives. This can reduce the loss of detail caused by a single image acquisition module 140 having a fixed shooting angle, so that the three-dimensional shape of dirt in the same area of ​​the part to be cleaned can be completely captured, providing data support for the calculation of dirt thickness.

[0074] In some embodiments, the cleaning device 100 further includes a drive mechanism 170 electrically connected to the controller 160, and the laser emitter 120 and the jet ejector 130 are both mounted on the output end of the drive mechanism 170; the controller 160 controls the drive mechanism 170 to adjust the position of the laser emitter 120 and the jet ejector 130.

[0075] The drive mechanism 170, in conjunction with the controller 160, adjusts the positions of the laser emitter 120 and the jet ejector 130, enabling them to move along a preset trajectory. This ensures that the laser emitted by the laser emitter 120 and the jet ejected by the jet ejector 130 precisely target the cleaning area on the surface of the workpiece. The drive mechanism 170 can take various structural forms, such as a three-axis linear drive module or a multi-degree-of-freedom robotic arm, and is not limited in this application. These embodiments, through the control of the drive mechanism 170 by the controller 160, can automatically move the laser emitter 120 and the jet ejector 130 to the corresponding area based on the dirt position parameters fed back by the image analysis module 150, eliminating the need for manual adjustment and thus improving the automation level of the cleaning process.

[0076] like Figure 2 As shown, based on the same inventive concept, this application also provides a cleaning method, implemented using the aforementioned cleaning device 100, the cleaning method comprising:

[0077] S100: Acquires image data of the part to be cleaned.

[0078] Before acquiring image data, the part to be cleaned must be placed stably on the stage 110. The height and angle of the stage 110 should be adjusted according to the size, shape, and material characteristics of the part to be cleaned, ensuring that the area to be cleaned is fully exposed within the field of view of the image acquisition module 140. After preparation, the illumination unit of the image acquisition module 140 is turned on, and the brightness and angle of the light source are adjusted to reduce the impact of shadows, reflections, and other interference factors on image quality, avoiding missed or false detections of dirt due to lighting issues. Subsequently, image data of the part to be cleaned is acquired through the imaging unit of the image acquisition module 140.

[0079] S200: Based on image data, obtain the dirt parameters of the part to be cleaned, including dirt type, dirt thickness, and dirt location.

[0080] The dirt parameters of the part to be cleaned can be obtained through the image analysis module 150 of the cleaning device 100. The execution process of the image analysis module 150 may include the following steps: first, receiving the raw image data transmitted by the image acquisition module 140 and performing preprocessing operations on it. The preprocessing operations may include noise reduction, image enhancement, cropping, etc., to improve the contrast between the dirty area and the clean surface of the part to be cleaned, and to remove irrelevant background interference, thereby providing support for the accurate detection of dirt parameters. After preprocessing, the dirty area and the clean area are divided from the image by an image segmentation algorithm to clarify the outline range of the dirt. Further, features can be extracted and parameters analyzed for each segmented dirty area. By comparing the texture, grayscale distribution, and other features of the dirt with a preset dirt feature database, the type of dirt is determined; the dirt thickness is quantified by using a dirt thickness-grayscale value calibration model or multispectral analysis method; and the pixel coordinates of the dirty area are mapped to the actual physical coordinates of the surface of the part to be cleaned by a coordinate transformation algorithm to determine the location of the dirt. During parameter analysis, cross-validation can be used to remove invalid data with excessive errors, ensuring the accuracy and reliability of contaminated parameters.

[0081] S300, based on the dirt parameters, controls the laser emitter 120 to emit a laser towards the dirt on the part to be cleaned, and / or controls the jet injector 130 to spray liquid towards the dirt on the part to be cleaned.

[0082] Specifically, the controller 160 first receives the dirt parameters transmitted by the image analysis module 150, and then matches the corresponding cleaning scheme with the built-in cleaning strategy library. For example, if the dirt is easy to clean, such as light dust or a small amount of grease, or if the part to be cleaned is a sensitive material such as glass or ceramic, the jet injector 130 can be started independently, and the appropriate pressure and flow rate of liquid can be selected to spray onto the dirty area according to the characteristics of the dirt. If it is in a waterless working environment, or if the dirt is a slightly dry attachment, the laser emitter 120 can be started independently, and the laser can be emitted onto the dirty area with low power and high scanning speed laser parameters. If the dirt is a thick and stubborn type, such as rust or paint layer, the laser emitter 120 and the jet injector 130 can be started simultaneously and work together according to a preset sequence, such as first softening the dirt with the jet, then vaporizing and peeling it off with the laser, and finally cleaning the residual debris with the jet.

[0083] In some embodiments, the jet ejector 130 includes a plurality of nozzles 131, and the controller 160 divides the area to be cleaned based on the location of the dirt, and dynamically selects the nozzle group in the corresponding area of ​​the jet ejector 130 to be activated according to the distribution of the divided area and the real-time power of the laser emitter 120.

[0084] Specifically, the controller 160, based on the physical coordinates of the dirt locations, divides the surface of the part to be cleaned into several independent cleaning areas. Each area corresponds to a set of nozzles 131 of the jet injector 130 (the nozzle groups can be preset in an array, with each set of nozzles 131 covering a fixed area). When the laser emitter 120 moves to the target area, the controller 160 only activates the nozzle group corresponding to that area, while the nozzle groups in other areas remain closed. This achieves precise alignment between the jet and the laser area, reducing the risk of non-clean areas being contaminated by the sprayed liquid, and also reducing liquid consumption. For example, when there are multiple discrete dirt points on the surface of the part to be cleaned, the laser operates on each point, activating only the nozzle group of the area to which that point belongs, further improving the targeted cleaning.

[0085] In some embodiments, the number of nozzles 131 in operation is positively correlated with the laser power to improve cleaning targeting and scene adaptability.

[0086] The controller 160 can have a built-in power-nozzle 131 quantity mapping table. The higher the laser power (e.g., for removing thick, stubborn dirt), the more nozzles 131 in the corresponding area can be activated. This enhances the impact intensity and cavitation effect through the superposition of multiple jets, simultaneously strengthening debris removal and cooling. Conversely, the lower the laser power (e.g., for cleaning trace residues), the fewer nozzles 131 can be activated, switching to a fine-stream jet mode to avoid damaging the surface of the workpiece with high-pressure jets. The proportional relationship between the two can be preset to a basic ratio, such as adding two working nozzles 131 for every 100W increase in laser power. This can be dynamically fine-tuned according to the dirt thickness to ensure precise adaptation to the synergistic effect and cleaning scenario.

[0087] In some embodiments, the controller 160 combines the real-time scanning trajectory and moving speed of the laser emitter 120 to control the nozzle group in the corresponding area to dynamically adjust its working state according to the laser, so as to achieve synchronous following and coverage of the jet and the laser.

[0088] Specifically, as the laser moves along a preset trajectory (such as a serpentine or spiral path), the controller 160 adjusts the direction or position of the nozzle assembly via the drive mechanism 170, ensuring that the jet always covers the area just touched by the laser. Furthermore, the moving speed of the nozzle 131 is positively correlated with the laser scanning speed; the faster the laser scanning speed, the faster the nozzle 131 moves, ensuring timely removal of debris. When the scanning speed slows down, the nozzle 131 also slows down accordingly, extending the jet's contact time and enhancing cooling and residue removal effects. This solution reduces the probability of misalignment between the jet and the laser's contact area, making it particularly suitable for cleaning irregularly shaped dirt.

[0089] In some embodiments, the controller 160 adaptively switches the nozzle 131 jetting mode according to the working mode of the laser emitter 120, so as to achieve coordinated adaptation of the working modes of the laser and the jet.

[0090] The laser emitter 120 operates in two modes: pulsed and continuous wave. When the laser emitter 120 operates in pulsed mode, the controller 160 switches the nozzle 131's jetting mode to pulsed jetting, and the frequency and phase of the jetting pulse are synchronized with the laser pulse. For example, the start time of the nozzle 131's jetting pulse is set to lag behind the start time of the laser pulse by 5-10 ms, allowing the laser to vaporize and remove contaminants first, followed by pulsed jet cleaning of debris, avoiding the absorption of laser energy by the liquid due to synchronous pulses. For high-frequency laser pulses (e.g., above 1000 Hz), the nozzle 131's jetting pulse frequency can be matched in a 1:1 or 1:2 ratio, forming a high-frequency alternating synergy between laser impact and jet cleaning, improving the efficiency of removing stubborn contaminants while reducing liquid loss during continuous jetting. If the laser operates in continuous wave mode, the nozzle 131 switches to continuous jetting mode, with the jetting pressure dynamically adjusted according to the laser power to balance cooling effect and liquid consumption.

[0091] like Figure 3 As shown, step S300 may include steps S310, S320 and S330.

[0092] S310 controls the jet injector 130 to spray liquid onto the dirt on the part to be cleaned.

[0093] In this step, liquid is sprayed onto the dirt on the part to be cleaned through the jet injector 130. The impact of the liquid directly removes some of the easily peelable dirt, while wetting and softening the remaining dirt, reducing the difficulty of subsequent laser cleaning, reducing the amount of debris flying during laser action, and improving the overall cleaning efficiency and effect.

[0094] In some embodiments, the controller 160 can send a start signal to the jet injector 130 based on the dirt parameters transmitted by the image analysis module 150, and simultaneously match the appropriate jet parameters, liquid type and jet method. For example, for hard and sticky dirt such as cement residue and thick oil stains, a higher pressure liquid jet can be selected and a pulse jet method can be used to enhance the softening and peeling effect; for loosely attached thick layers of dust, a medium pressure water jet can be selected to avoid dirt splashing and spreading and improve cleaning targeting. The jetting time is set according to the dirt thickness.

[0095] S320 controls the laser emitter 120 to emit a laser at a first emission power toward the dirt on the part to be cleaned.

[0096] The controller 160 sends signals to the laser emitter 120 and the drive mechanism 170, activating the laser emitter 120 and switching it to the first emission power mode. The first emission power is a high-power setting, the specific value of which is matched according to the thickness of the dirt and the material of the part to be cleaned, and is not limited in this application. At the same time, the drive mechanism 170 moves the laser emitter 120 to the dirty area and performs laser irradiation according to a preset scanning trajectory (such as serpentine scanning or spiral scanning). The scanning speed is adjusted according to the power and the thickness of the dirt. This step uses a high-power laser to quickly vaporize and peel off the softened thick layer of dirt.

[0097] S330, control the laser emitter 120 to emit laser light at a second emission power toward the dirt on the part to be cleaned, wherein the first emission power is greater than the second emission power.

[0098] The controller 160 sends signals to the image acquisition module 140, the laser emitter 120, and the drive mechanism 170. After the image acquisition module 140 confirms that the thick layer of dirt has been largely removed, it controls the laser emitter 120 to switch to a second emission power mode. The second emission power is a low-power setting, and the specific value is matched according to the first emission power, the amount of dirt residue, and the material of the part to be cleaned, and is not limited in this application. At the same time, the drive mechanism 170 drives the laser emitter 120 to perform a second scan on the original dirty area. The scanning speed is adjusted according to the power and the amount of residual dirt to avoid damage to the surface of the part to be cleaned due to prolonged low-power irradiation. The range of the second scan can be slightly smaller than the scanning range in step S320, focusing on the area where residual dirt is concentrated. If the image acquisition module 140 detects that there is little residual dirt, it can further reduce the scanning range or increase the scanning speed to shorten the cleaning cycle. After the image acquisition module 140 reports that there is no obvious dirt residue in the cleaned area, the controller 160 can control the laser emitter 120 to stop working.

[0099] Since the low-power laser cleaning stage requires focusing on the removal and surface trimming of residual trace dirt, and the laser action is prone to the adsorption of trace debris or slight local temperature rise, which may affect the cleaning accuracy or damage the surface material of the part to be cleaned, in some embodiments, while controlling the laser emitter 120 to emit laser at a second emission power to the dirt on the part to be cleaned, the jet ejector 130 is also simultaneously controlled to spray liquid onto the dirt on the part to be cleaned.

[0100] With this design, the jet can flush away the trace debris generated by laser cleaning in real time, preventing debris from adhering to the cleaning area again and improving the surface cleaning accuracy; the liquid can also quickly remove the local heat generated by the laser action, reducing the risk of temperature rise caused by long-term scanning of low-power lasers and avoiding unexpected degradation of the surface performance of the parts to be cleaned.

[0101] Specifically, while sending a second emission power start signal to the laser emitter 120, the controller 160 simultaneously sends a suitable jet control signal to the jet ejector 130. The jet parameters must match the low-power laser cleaning scenario, and a low-pressure, fine-flow liquid jet can be used. The liquid type can continue from the clean water or neutral detergent in S310, ensuring compatibility with the material of the part to be cleaned. The jetting method can be a continuous fine-flow jet, with the jetting range precisely aligned with the laser secondary scanning range, removing trace debris generated by laser cleaning in real time. After cleaning is completed, the controller 160 synchronously controls the laser emitter 120 and the jet ejector 130 to stop working.

[0102] In some embodiments, during the cleaning process of controlling the laser emitter 120 to emit laser and / or the jet injector 130 to spray liquid, the image acquisition module 140 continuously acquires real-time image data of the cleaning area on the part to be cleaned; and based on the real-time image data, compares the changes in texture features and spectral features of the cleaning area during the cleaning process; and obtains the cleanliness index of the cleaning area according to the changes in texture features and spectral features.

[0103] In other words, during the cleaning process of step S300, the image acquisition module 140 continuously acquires real-time image data of the area to be cleaned on the part to be cleaned and transmits this data to the image analysis module 150 in real time. After receiving the image data, the image analysis module 150 performs targeted feature comparison between the real-time image and the historical images stored during the cleaning process. The comparison dimensions can be set to two categories: texture features and spectral features. For texture features, parameters such as gray-level co-occurrence matrix (energy, entropy, contrast), edge density, and texture uniformity can be extracted. The dynamic changes in surface roughness and texture regularity of the area during the cleaning process (such as the regularization trend of surface texture after dirt is gradually peeled off) can be quantified through feature quantification algorithms. For spectral features, a preset characteristic band can be set according to the target dirt (such as the 2.3μm absorption peak corresponding to grease and the 1.5μm reflection peak corresponding to the oxide layer). The dynamic differences in reflectance and absorptivity of this band can be analyzed through spectral analysis algorithms, thereby capturing the reduction trend of trace residual dirt during the cleaning process (such as the attenuation of the absorption signal of grease in the 2.3μm band and the enhancement of the reflection signal of the oxide layer in the 1.5μm band). Based on the dynamic trends of the two types of features mentioned above, and combined with the real-time quantitative results of dirt residue, the image analysis module 150 further calculates a cleanliness index. This index can be synthesized using multi-dimensional parameter weighting (including the proportion of dirt residue area, texture similarity, and spectral matching degree), and outputs real-time results through preset quantitative standards (such as cleanliness ≥95% being considered satisfactory). Finally, the image analysis module 150 feeds back the cleanliness index to the controller 160 in real time, providing data support for the controller 160 to dynamically adjust cleaning parameters (such as laser power, jet pressure, scanning speed, operation time, etc.) or terminate the cleaning operation, ensuring that the cleaning effect accurately meets the standards.

[0104] The cleanliness index is a normalized parameter that comprehensively quantifies the dynamic changes in dirt residue, the improvement trend in texture regularity, and the enhancement effect of spectral response consistency during the cleaning process. Its value ranges from 0 to 100%, with higher values ​​indicating better cleaning results. The calculation process of the cleanliness index includes dynamic quantification of three types of features: texture, spectrum, and dirt residue; fixed weight allocation; comprehensive synthesis; and dynamic calibration. It can be calculated through the following feasible steps, the specific process of which is as follows:

[0105] First, dynamic quantification of basic features is performed, focusing on the dynamic changes during the cleaning process:

[0106] Quantification of the percentage of residual dirt area (denoted as P): The aforementioned image segmentation algorithms (such as threshold segmentation and semantic segmentation) can be used to extract the dirty areas in real-time images during the cleaning process. The ratio of the number of pixels in the dirty area to the total number of pixels in the clean area is calculated using the formula: P = (number of pixels in the dirty area / total number of pixels in the clean area) × 100%. This parameter reflects the dynamic change in the degree of visible dirt residue during the cleaning process. In the early stage of cleaning, P is close to 100%, and it gradually decreases as cleaning progresses, providing a basic reference for the dynamic analysis of texture and spectral features.

[0107] Texture similarity quantification (denoted as T): Extract texture feature parameters (such as energy, entropy, and contrast of the gray-level co-occurrence matrix, or edge density, texture uniformity, etc.) of historical images (such as the cleaning start time or the previous frame image) and the current real-time image during the cleaning process. The similarity between the two can be calculated using algorithms such as cosine similarity and Euclidean distance. The formula is T = 1 - |real-time texture parameter - historical texture parameter| / maximum value of historical texture parameter × 100%. The value of T ranges from 0 to 100%, and the value gradually increases with the cleaning process. This represents the continuous improvement of the consistency between the real-time image texture and the target texture (the target texture can be preset based on the standard sample image after the cleaning is completed, or dynamically updated through multiple historical images during the cleaning process), which intuitively reflects the improvement trend of texture features during the cleaning process.

[0108] Spectral matching quantification (denoted as S): Characteristic spectral bands (e.g., 2.3μm absorption peak for grease, 1.5μm reflection peak for oxide layers) can be preset based on the target dirt type. Spectral response values ​​(reflectance R or absorptivity A) of historical and real-time images during the cleaning process are extracted using spectral analysis algorithms. The correlation coefficient between the two is calculated using the formula: S = (covariance of historical and real-time spectral response values) / (standard deviation of historical spectral response values ​​× standard deviation of real-time spectral response values) × 100%. The value of S ranges from 0 to 100%, gradually increasing as the cleaning process progresses. This indicates a continuous improvement in the matching degree between the real-time image spectral characteristics and the target spectral characteristics (which can be preset based on the standard sample spectrum after cleaning is completed). It visually reflects the optimization trend of spectral characteristics during the cleaning process and indirectly reflects the gradual reduction of trace amounts of residual dirt.

[0109] Next, feature weights are assigned. Based on the requirement that the cleaning process must simultaneously incorporate the effects of texture feature changes and spectral feature changes, a fixed weight assignment mode can be adopted, where... The weight of the corresponding area of ​​dirt residue Weights corresponding to texture similarity The weights corresponding to the spectral matching degree, and + + =1. The weight values ​​follow the collaborative contribution logic of the three types of parameters, for example, setting... =0.3、 =0.35、 =0.35, ensuring a balanced contribution of texture and spectral features to the cleanliness index. The weight parameters can be pre-stored in the preset parameter library of the controller 160, and can be fine-tuned according to the material of the part to be cleaned and the type of dirt. The weights of texture features and spectral features must not be zero, ensuring consistency with the core logic of dual-feature collaborative monitoring.

[0110] Next, a comprehensive index synthesis is performed. The quantified basic feature parameters and their corresponding weights are weighted and summed to obtain the initial cleanliness index (denoted as ). The following formula can be used to calculate it: The role of (1-P / 100) is to convert the proportion of dirt residue into a positive indicator (the smaller P is, the larger this value is), which is consistent with the positive trend of texture similarity T and spectral matching S, ensuring that the three types of parameters contribute to the cleanliness index in a unified direction and can reflect the dynamic improvement effect in the cleaning process simultaneously.

[0111] Finally, dynamic calibration and optimization were performed to match the real-time monitoring requirements of the cleaning process. This involved incorporating iterative trends from multiple frames of historical data to refine the initial cleanliness indicators. Dynamic calibration is performed to obtain the final cleanliness index C: If Exceeding the preset reasonable range (e.g., due to image noise) >100% or If <0%), then it is corrected to 100% or 0% through truncation; if If it is within a reasonable range, then use the dynamic calibration formula. (in , The calibration coefficient for the k-th frame can be optimized by using conventional algorithms in the field, such as linear fitting and least squares fitting, to fit the historical cleanliness index data of the previous k-1 frames in real time. After calibration, the final cleanliness index C will still be in the range of 0-100%, ensuring that the index can track the dynamic changes of the cleaning process in real time, avoid the evaluation deviation caused by the fluctuation of a single frame image, and further improve the accuracy and reliability of cleanliness judgment.

[0112] In some embodiments, it is determined whether the cleanliness index tends to be stable; if the cleanliness index tends to be stable, it is determined whether the cleanliness index has reached the target cleaning threshold; if the cleanliness index has not reached the target cleaning threshold, an enhanced cleaning strategy is executed; the enhanced cleaning strategy includes at least one of the following: increasing the emission power of the laser emitter 120, reducing the scanning speed of the laser emitter 120, and increasing the jet impact intensity of the jet ejector 130.

[0113] The cleanliness index tends to stabilize when, within a preset time window (e.g., 3-5 seconds, the specific duration can be set according to the cleaning scenario), the change in the cleanliness index is within a preset fluctuation threshold range (e.g., fluctuation does not exceed ±2%). In other words, the index no longer increases significantly with the extension of cleaning time, indicating that the current cleaning parameters have reached a bottleneck in their ability to remove dirt.

[0114] The target cleaning threshold is a pre-set quantitative standard for judging whether the cleaning effect meets the standard. Its value is usually determined based on the application scenario of the part to be cleaned, industry cleaning standards, or user needs (such as the target cleaning threshold for precision electronic components can be set to cleanliness ≥98%, and the target cleaning threshold for ordinary mechanical parts can be set to cleanliness ≥90%). This threshold matches the calculation dimension of the cleanliness index and is used to clarify the minimum quality requirements that the cleaning operation must achieve.

[0115] When the cleanliness index stabilizes but does not reach the target cleaning threshold, it indicates that conventional cleaning parameters are no longer effective in removing the remaining dirt, and this dirt can be identified as stubborn dirt. At this point, a localized enhanced cleaning strategy is triggered: increasing the laser emission power enhances the laser's vaporization and removal capabilities for stubborn dirt; decreasing the laser scanning speed prolongs the interaction time between the laser and the dirt; and increasing the jet impact intensity strengthens the mechanical removal effect of the liquid on stubborn dirt. Depending on the type of stubborn dirt (such as rust, hardened adhesive layers, etc.) and the material characteristics of the part to be cleaned, a single strategy or a combination of strategies can be selected to ensure the cleaning effect meets the standards while minimizing the risk of damage to the surface of the part to be cleaned.

[0116] In some embodiments, during the cleaning process of controlling the laser emitter 120 to emit laser and / or the jet injector 130 to spray liquid, real-time image data of the cleaning area on the workpiece to be cleaned is continuously acquired; based on the real-time image data, the changes in texture features and spectral features of the cleaning area before and after cleaning are compared; based on the changes in texture features and spectral features, a cleanliness index of the cleaning area is obtained; it is determined whether the cleanliness index has reached the target cleaning threshold; if the cleanliness index has not reached the target cleaning threshold, the jet injector 130 is controlled to spray liquid onto the dirt on the workpiece to be cleaned, the laser emitter 120 is controlled to emit laser at a first emission power onto the dirt on the workpiece to be cleaned, and the laser emitter 120 is controlled to emit laser at a second emission power onto the dirt on the workpiece to be cleaned, until the cleanliness index reaches the target cleaning threshold or the preset maximum number of repetitions is reached.

[0117] In other words, during the cleaning process of step S300, the image acquisition module 140 continuously acquires real-time image data of the cleaning area on the part to be cleaned and transmits this data to the image analysis module 150 in real time. After receiving the image data, the image analysis module 150 first calls the baseline image of the cleaning area that was pre-acquired and stored before cleaning, and performs targeted feature comparison between the real-time image and the baseline image. The comparison focuses on two categories: texture features and spectral features. For texture features, core parameters such as gray-level co-occurrence matrix (energy, entropy, contrast), edge density, and texture uniformity are extracted to quantitatively analyze the changes in surface roughness and texture regularity of the cleaning area compared to before cleaning. For spectral features, the differences in reflectance and absorptivity of preset feature bands (such as the 2.3μm absorption peak corresponding to grease and the 1.5μm reflection peak corresponding to the oxide layer) are analyzed to accurately capture the reduction of trace residual dirt before and after cleaning.

[0118] Based on the comprehensive analysis results of the above-mentioned texture feature changes and spectral feature changes, the image analysis module 150 further calculates the cleanliness index of the clean area, which quantitatively reflects the actual effect of the cleaning operation.

[0119] Subsequently, the image analysis module 150 feeds back the calculated cleanliness index to the controller 160 in real time. The controller 160 then determines whether the cleanliness index has reached the preset target cleaning threshold. If the target cleaning threshold has not been reached, the controller 160 triggers the repeated cleaning control logic and repeats steps S310, S320, and S330 until the cleanliness index reaches the target cleaning threshold, or the controller 160 detects that the number of repeated operations has reached the preset maximum number of repetitions, and then terminates the current cleaning process.

[0120] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A cleaning device, characterized in that, include: A stage (110) is used to hold items to be cleaned; A laser emitter (120) is used to emit a laser beam toward the part to be cleaned in order to clean the part to be cleaned; A jet injector (130) is used to spray liquid onto the part to be cleaned in order to clean the part; Image acquisition module (140) is used to acquire image data of the part to be cleaned; The controller (160) is electrically connected to the laser emitter (120), the jet ejector (130), and the image acquisition module (140). The controller (160) controls the working state of the laser emitter (120) and / or the working state of the jet ejector (130) based on the dirt parameters of the surface to be cleaned obtained from the image data. The dirt parameters include dirt type, dirt thickness, and dirt location. The working state includes at least a start / stop state.

2. The cleaning equipment according to claim 1, characterized in that, The cleaning equipment also includes: The image analysis module (150) is electrically connected to both the image acquisition module (140) and the controller (160). The image analysis module (150) obtains the dirt parameters based on the image data.

3. The cleaning equipment according to claim 1, characterized in that, The jet injector (130) includes multiple nozzles (131), and the liquid jets ejected by the multiple nozzles (131) and the laser beam emitted by the laser emitter (120) converge at the same position on the surface of the part to be cleaned.

4. The cleaning equipment according to claim 1, characterized in that, The cleaning device (100) includes a plurality of image acquisition modules (140), which are used to acquire image data of the same area of ​​the part to be cleaned from different angles.

5. A cleaning method, characterized in that, The cleaning method is implemented based on the cleaning device (100) according to any one of claims 1-4, and includes: Collect image data of the part to be cleaned; Based on the image data, the dirt parameters of the part to be cleaned are obtained, including dirt type, dirt thickness and dirt location; Based on the dirt parameters, the laser emitter (120) is controlled to emit a laser towards the dirt on the part to be cleaned, and / or the jet injector (130) is controlled to spray liquid towards the dirt on the part to be cleaned.

6. The cleaning method according to claim 5, characterized in that, The step of controlling the laser emitter (120) to emit laser light onto the dirt on the part to be cleaned based on the dirt parameters, and / or controlling the jet injector (130) to spray liquid onto the dirt on the part to be cleaned, includes: Control the jet injector (130) to spray liquid onto the dirt on the part to be cleaned; The laser emitter (120) is controlled to emit laser light at a first emission power toward the dirt on the part to be cleaned; The laser emitter (120) is controlled to emit laser light at a second emission power toward the dirt on the part to be cleaned; wherein the first emission power is greater than the second emission power.

7. The cleaning method according to claim 6, characterized in that, While controlling the laser emitter (120) to emit laser light at a second emission power toward the dirt on the part to be cleaned, the jet injector (130) is also controlled to spray liquid toward the dirt on the part to be cleaned.

8. The cleaning method according to claim 6, characterized in that, During the cleaning process of controlling the laser emitter (120) to emit laser and / or the jet injector (130) to spray liquid, real-time image data of the cleaning area on the part to be cleaned is continuously collected; Based on the real-time image data, compare the changes in texture features and spectral features of the cleaned area during the cleaning process; Based on the changes in texture features and spectral features, the cleanliness index of the clean area is obtained.

9. The cleaning method according to claim 8, characterized in that, Determine whether the cleanliness index tends to stabilize; If the cleanliness index tends to stabilize, determine whether the cleanliness index has reached the target cleanliness threshold; If the cleanliness index does not reach the target cleanliness threshold, an enhanced cleaning strategy is executed; the enhanced cleaning strategy includes at least one of the following: increasing the emission power of the laser emitter (120), reducing the scanning speed of the laser emitter (120), and increasing the jet impact intensity of the jet ejector (130).

10. The cleaning method according to claim 6, characterized in that, During the cleaning process of controlling the laser emitter (120) to emit laser and / or the jet injector (130) to spray liquid, real-time image data of the cleaning area on the part to be cleaned is continuously collected; Based on the real-time image data, compare the changes in texture features and spectral features of the cleaned area before and after cleaning; Based on the changes in texture features and spectral features, the cleanliness index of the clean area is obtained; Determine whether the cleanliness index has reached the target cleanliness threshold; If the cleanliness index does not reach the target cleaning threshold, the jet injector (130) is controlled to spray liquid onto the dirt on the part to be cleaned, the laser emitter (120) is controlled to emit laser light onto the dirt on the part to be cleaned at the first emission power, and the laser emitter (120) is controlled to emit laser light onto the dirt on the part to be cleaned at the second emission power, until the cleanliness index reaches the target cleaning threshold or the preset maximum number of repetitions is reached.

Images