Laser cleaning system and method based on coaxial visual feedback

Abstract

The invention relates to a laser cleaning system and method based on coaxial visual feedback. The system comprises a laser emitting module used for emitting cleaning laser f1 with the wavelength of lambda1; the visual monitoring module comprises an illumination light emitting unit and an image acquisition unit; the illumination light emitting unit is used for emitting auxiliary illumination light f2 with the wavelength of lambda2, and the image acquisition unit is used for acquiring image information of a cleaning area of a target workpiece; the light path control module is used for combining the cleaning laser f1 and the auxiliary illumination light f2 to form combined light F and guiding the combined light F to irradiate the cleaning surface of the target workpiece; the light filtering assembly is used for filtering the reflected light p with the feature information of the cleaning area of the target workpiece and guiding the reflected light p to enter the image acquisition unit; and the analysis and control module is connected with the laser emission module and the visual monitoring module and is used for issuing a time sequence excitation command, receiving image information and analyzing and processing image features.

Description

Technical Field

[0001] This invention relates to the field of laser processing and surface treatment technology, and specifically to a laser cleaning system and method based on coaxial visual feedback. Background Technology

[0002] Laser cleaning is an advanced industrial cleaning technology that uses the interaction of a high-energy, short-pulse laser beam with the surface of a material to remove contaminants such as rust, paint, and oil. In current industrial applications, laser cleaning equipment often employs an open-loop control mode, where operators pre-set process parameters such as laser power, frequency, and scanning speed based on experience, and the equipment then operates according to the set program. However, because the types and distribution of contaminants on the surface of the workpiece to be cleaned are often uneven, this fixed-parameter mode can easily lead to incomplete cleaning or thermal damage to the substrate material.

[0003] To improve cleaning quality and process safety, existing technologies have introduced online monitoring and feedback mechanisms to achieve adaptive adjustments to process parameters. Currently, the main online monitoring methods include acoustic monitoring, spectral monitoring, and visual monitoring. However, acoustic and spectral monitoring are easily affected by strong ambient noise and background light in real industrial environments, limiting their practicality. Visual monitoring, on the other hand, effectively avoids such interference, using machine vision to analyze the surface and identify the type and distribution of contaminants, offering high reliability. In existing visual monitoring solutions, image sensors are typically arranged at an angle to the cleaning laser head to acquire real-time images of the cleaning area's topography. This method offers advantages such as simple installation and low cost.

[0004] However, the aforementioned visual monitoring technologies still have significant limitations in practical applications: First, the strong plasma glow generated during laser cleaning can severely interfere with imaging, leading to decreased image contrast and difficulty in feature recognition; Second, the inherent parallax and occlusion problems of the paraxial detection method make it difficult to effectively monitor contaminants in structures such as deep holes, trenches, and complex curved surfaces. Summary of the Invention

[0005] The purpose of this invention is to provide a laser cleaning system and method based on coaxial visual feedback. This system has excellent contaminant feature recognition capabilities and can effectively detect the type and morphology of contaminants in deep holes, grooves, and complex curved surfaces on the surface of the workpiece to be cleaned. It can greatly improve the cleaning quality, has high cleaning safety, and has good applicability.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] In a first aspect, the present invention provides a laser cleaning system based on coaxial visual feedback, the system comprising:

[0008] The laser emitting module is used to emit a cleaning laser f1 with a wavelength of λ1;

[0009] The visual monitoring module includes an illumination light emitting unit and an image acquisition unit; the illumination light emitting unit is used to emit auxiliary illumination light f2 with a wavelength of λ2, and the image acquisition unit is used to acquire image information of the cleaning area of ​​the target workpiece;

[0010] The optical path control module is used to combine the cleaning laser f1 and the auxiliary illumination light f2 to form a combined beam F; it is also used to guide the combined beam F to irradiate the cleaning surface of the target workpiece.

[0011] A filter assembly is used to filter the reflected light p carrying the feature information of the target workpiece cleaning area and guide the reflected light p into the image acquisition unit.

[0012] The analysis and control module, connected to the laser emission module and the visual monitoring module, is used to issue timing excitation commands, receive image information, and perform image feature analysis and processing.

[0013] As a preferred embodiment of the present invention, the wavelength λ1 of the cleaning laser f1 is not equal to the wavelength λ2 of the auxiliary illumination light f2.

[0014] As a preferred embodiment of the present invention, the wavelength of the cleaning laser f1 is 1064nm; the wavelength range of the auxiliary illumination light f2 is 450nm±10nm.

[0015] As a preferred embodiment of the present invention, the analysis and control module includes a synchronous control unit and a host computer unit connected to each other; the synchronous control unit is connected to the laser emission module and the image acquisition unit, and is used to receive timing excitation commands from the host computer unit, send trigger pulse signals to the laser emission module, and send shutter opening signals to the image acquisition unit.

[0016] As a preferred embodiment of the present invention, the filtering component includes a narrowband filter disposed in front of the image acquisition unit, wherein the center wavelength of the narrowband filter matches wavelength λ2 and cuts off light of wavelength λ1.

[0017] As a preferred embodiment of the present invention, the optical path control module includes a polarizing beam splitter, a dichroic mirror, a galvanometer, and a field mirror; the dichroic mirror refracts light with wavelength λ1 and reflects light with wavelength λ2.

[0018] Secondly, the present invention provides a laser cleaning method based on coaxial visual feedback, the method comprising:

[0019] S01, at time T0, the analysis and control module sends a laser timing excitation command to the laser emission module and starts the timer; the laser emission module emits a cleaning laser f1 with wavelength λ1 to the optical path control module;

[0020] S02, the timer executes a countdown Δt, and the analysis and control module sends a visual trigger signal to the visual monitoring module at time T0 + Δt. The visual monitoring module includes an illumination light emitting unit and an image acquisition unit; the visual trigger signal includes a strobe signal and a shutter opening signal.

[0021] S03, the illumination light emitting unit receives the strobe signal and emits auxiliary illumination light f2 to the optical path control module; the image acquisition unit receives the shutter opening signal and acquires a feature image of the target workpiece cleaning area;

[0022] S04, the image acquisition unit uploads the feature image to the analysis and control module, the analysis and control module filters and analyzes the feature image, and generates a feedback control command; the feedback control command is used to control the execution state of the laser emission module.

[0023] As a preferred embodiment of the present invention, the specific steps of the analysis and control module in filtering and analyzing the feature image are as follows:

[0024] Graphical features are extracted from the feature image, and the cleanliness measurement value Z is obtained by analyzing the graphic features based on a preset analysis algorithm. k ;

[0025] The cleanliness measurement value Z is based on the asynchronous Kalman filter backtracking correction method. k Perform asynchronous updates and generate an estimate of the current residual value X based on the updated estimate. k；

[0026] The energy U of the next pulse is calculated based on a pre-defined cleaning state-space model. k+1 Based on the next pulse energy U k+1 Generate feedback control commands to control the execution state of the laser emission module at the next moment.

[0027] As a preferred embodiment of the present invention, this laser cleaning method based on coaxial visual feedback further includes an over-cleaning prevention step:

[0028] Based on the current estimated residual value X k The estimated value X of the residue at the next moment is obtained by predicting the state space of the pre-set cleaning process. k+1 ;

[0029] Compare the estimated residual amount X at the next moment. k+1And the magnitude of the cleanliness threshold R; if X k+1 If X > R, then control the laser emitting module 1 to perform the laser emitting action at the next moment; if X k+1 If the value is less than or equal to R, then the cleaning action on the current cleaning area of ​​the target workpiece will be stopped.

[0030] As a preferred embodiment of the present invention, the cleaning state space model is specifically as follows:

[0031] X k+1 = AX k + BU k

[0032] Among them, X k For the thickness of pollutants, U k Let A be the laser energy density, and B be preset weighting coefficients.

[0033] In summary, the present invention has the following beneficial effects:

[0034] This invention employs a coaxial optical path design using dichroic mirrors, ensuring constant synchronization between visual monitoring and laser focus. This eliminates the parallax and occlusion problems inherent in traditional paraxial detection, making it particularly suitable for monitoring deep holes, trenches, and complex curved surfaces. Furthermore, by introducing an asynchronous Kalman filter algorithm, this invention effectively compensates for system lag caused by image acquisition and processing, achieving millisecond-level accurate prediction of the current residual state and adaptive adjustment of process parameters. This ensures cleaning quality while effectively preventing substrate damage. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention 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 the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a system structure block diagram of the present invention;

[0037] Figure 2 This is a flowchart of the method of the present invention;

[0038] Figure 3 This is a schematic diagram of a laser cleaning system structure according to an embodiment of the present invention;

[0039] Figure 4 This is a schematic diagram of the optical path control module in an embodiment of the present invention;

[0040] Figure 5 This is a schematic diagram of timing triggering in an embodiment of the present invention;

[0041] Figure 6 This is a flowchart of the method steps in an embodiment of the present invention.

[0042] In the diagram: Laser emission module 1, visual monitoring module 2, illumination light emission unit 21, image acquisition unit 22, optical path control module 3, polarizing beam splitter 31, dichroic mirror 32, galvanometer 33, field mirror 34, filter assembly 4, narrowband filter 41, analysis and control module 5, synchronous control unit 51, host computer unit 52, target workpiece 6. Detailed Implementation

[0043] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed merely to enable those skilled in the art to better understand and implement the subject matter described herein, and are not intended to limit the scope, applicability, or examples set forth in the claims. The function and arrangement of the elements discussed may be changed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the various examples. For example, the described methods may be performed in a different order than described, and steps may be added, omitted, or combined. Furthermore, features described in some examples may be combined in other examples.

[0044] The following is in conjunction with the appendix Figure 1 , 3 Sections 4 and 5 provide a further detailed description of the laser cleaning system of the present invention. This embodiment uses a laser cleaning system as an example, but the application of the present invention is not limited thereto.

[0045] Example 1: A laser cleaning system based on coaxial visual feedback. The system includes a laser emitting module 1, a visual monitoring module 2, an optical path control module 3, a filter assembly 4, and an analysis and control module 5. In this example, the laser emitting module 1 emits a cleaning laser f1 with a wavelength of 1064 nm; the visual monitoring module 2 includes an illumination light emitting unit 21 and an image acquisition unit 22. The illumination light emitting unit 21 emits an auxiliary illumination light f2 with a wavelength of 450 nm ± 10 nm. The image acquisition unit 22 uses a high-resolution camera to acquire image information of the cleaning area of ​​the target workpiece 6; the optical path control module 3 combines the cleaning laser f1 and the auxiliary illumination light f2 to form a combined beam F, and guides it to the cleaning surface of the target workpiece 6; the filter assembly 4 filters the reflected light p; the analysis and control module 5 is connected to the laser emitting module 1 and the visual monitoring module 2 to achieve timing control and image analysis.

[0046] The optical path control module 3 specifically consists of a polarizing beam splitter 31, a dichroic mirror 32, a galvanometer 33, and a field mirror 34. In this embodiment, the dichroic mirror 32 has high transmittance (greater than 99.5%) for the 1064nm cleaning laser f1 and high reflectance (greater than 95%) for the 450nm auxiliary illumination light f2. The cleaning laser f1 is transmitted through the dichroic mirror 32, and the auxiliary illumination light f2 is reflected by the polarizing beam splitter 31 and the dichroic mirror 32. After coupling, the two beams are focused onto the surface of the target workpiece 6 by the galvanometer 33 and the field mirror 34. The reflected light p from the surface of the target workpiece 6 carries the topographic information of the cleaning area, returns along the original optical path, passes through the field mirror 34 and the galvanometer 33, and is reflected by the dichroic mirror 32 into the visual imaging branch. This optical path design simplifies the system structure, improves the stability of the optical path and the cleaning efficiency, and is suitable for high-speed dynamic cleaning scenarios.

[0047] The filter assembly 4 includes a narrowband filter 41, which is positioned in front of the image acquisition unit 22. The center wavelength of the narrowband filter 41 is matched to 450nm, and it has high cutoff characteristics for 1064nm wavelength and broadband plasma radiation, ensuring that the image acquisition unit 22 only receives 450nm reflected light and avoids interference. This filtering design ensures that the image acquisition unit acquires only effective surface morphology information, improving the accuracy and anti-interference capability of cleanliness measurement.

[0048] The analysis and control module 5 includes a synchronization control unit 51 and a host computer unit 52. The synchronization control unit 51 receives timing commands from the host computer unit 52, sends trigger pulse signals to the laser emission module 1, and sends shutter opening signals to the image acquisition unit 22. The host computer unit 52 runs a closed-loop control algorithm, processes image data, and generates feedback control commands. This architecture reduces system latency, supports millisecond-level feedback adjustments, makes the cleaning process highly adaptive, and avoids errors caused by human intervention.

[0049] Example 2, as Figure 2 , Figure 5 and Figure 6 As shown in the figure, this embodiment provides a laser cleaning method based on coaxial visual feedback. The specific steps of the method are as follows.

[0050] To address the strong plasma interference during laser cleaning, this embodiment employs a synchronous control unit 51 to execute time-gated logic. The timing control process is as follows:

[0051] At time T0, the synchronous control unit 51 sends a trigger signal (laser timing excitation command) to the laser emission module 1, and the laser emission module 1 emits the cleaning laser f1. At this time, the contaminants on the surface of the target workpiece 6 are vaporized and ionized, generating a high-intensity plasma glow, with the intensity reaching its peak within microseconds.

[0052] Simultaneously, the synchronization control unit 51 starts an internal timer, delaying for a preset time Δt (typically a few microseconds to tens of microseconds). During this period, the laser pulse ends, the plasma decays rapidly, but the surface morphology remains.

[0053] The cleaning process begins and parameters are initialized. At time T0+Δt, the synchronous control unit 51 sends a visual trigger signal to the visual monitoring module 2, including a strobe signal (triggering the illumination light emitting unit 21 to emit auxiliary illumination light f2) and a shutter opening signal (triggering the image acquisition unit 22 to expose). The image acquisition unit 22 acquires feature images of the surface of the target workpiece 6 within a short exposure time. Because the narrow-band filter 41 filters out all stray light except for 450nm, and the exposure time avoids the period of strongest plasma emission, the image background is clean and has high contrast. This timing-gated method proposed in this invention effectively avoids plasma interference and improves the reliability of contaminant identification by separating laser emission and image acquisition.

[0054] Subsequently, the image acquisition unit 22 uploads the feature image to the host computer unit 52, which performs image processing: first, it extracts graphic features (such as the area or texture of contaminant coverage), and then calculates the cleanliness measurement value Z. k Then, based on the asynchronous Kalman filter backtracking correction method, Z is... k Perform asynchronous updates to generate an estimated value X of the current pollutant residue. k .

[0055] Then, finally, based on the cleaning state-space model X... k+1 = AX k + BU k Calculate the energy U of the next pulse k+1 ; where X k For the thickness of pollutants, U k Let A be the laser energy density, and B be preset weighting coefficients.

[0056] Simultaneously, feedback control commands are generated to adjust the parameters of laser emission module 1.

[0057] In addition, this laser cleaning method based on coaxial visual feedback also includes an over-cleaning prevention step:

[0058] Based on the current estimated residual value X k The estimated value X of the residue at the next moment is obtained by predicting the state space of the pre-set cleaning process. k+1 ;

[0059] Compare the estimated residual amount X at the next moment. k+1 And the magnitude of the cleanliness threshold R; if X k+1If R >, then the laser emitting module 1 is controlled to perform the laser emitting action at the next moment, for example, the PID controller outputs a signal to increase the laser power or decrease the scanning speed; if X k+1 If the value is less than or equal to R, the controller output signal will reduce power or perform a skip step to prevent overheating damage to the substrate.

[0060] Several embodiments of this disclosure have been described above. These descriptions are exemplary and not exhaustive, and are not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is chosen to best explain the principles, practical applications, or technological improvements to the embodiments in the market, or to enable others skilled in the art to understand the embodiments disclosed herein.

Claims

1. A laser cleaning system based on coaxial visual feedback, characterized in that the system... include: Laser emitting module (1) is used to emit a cleaning laser f1 with a wavelength of λ1; The visual monitoring module (2) includes an illumination light emitting unit (21) and an image acquisition unit (22); the illumination light emitting unit (21) is used to emit auxiliary illumination light f2 with a wavelength of λ2, and the image acquisition unit (22) is used to acquire image information of the cleaning area of ​​the target workpiece (6); The optical path control module (3) is used to combine the cleaning laser f1 and the auxiliary illumination light f2 to form a combined beam F; it is also used to guide the combined beam F to irradiate the cleaning surface of the target workpiece (6); The filter assembly (4) is used to filter the reflected light p carrying the feature information of the cleaning area of ​​the target workpiece (6) and guide the reflected light p into the image acquisition unit (22). The analysis and control module (5) is connected to the laser emission module (1) and the visual monitoring module (2) and is used to issue timing excitation commands, receive image information and perform image feature analysis and processing.

2. The laser cleaning system based on coaxial visual feedback according to claim 1, characterized in that, The wavelength λ1 of the cleaning laser f1 is not equal to the wavelength λ2 of the auxiliary illumination light f2.

3. The laser cleaning system based on coaxial visual feedback according to claim 2, characterized in that, The wavelength of the cleaning laser f1 is 1064nm; the wavelength range of the auxiliary illumination light f2 is 450nm±10nm.

4. The laser cleaning system based on coaxial visual feedback according to claim 1, characterized in that, The analysis and control module (5) includes a synchronous control unit (51) and a host computer unit (52) connected to each other; the synchronous control unit (51) is connected to the laser emission module (1) and the image acquisition unit (22), and is used to receive timing excitation commands from the host computer unit (52), send trigger pulse signals to the laser emission module (1) and send shutter opening signals to the image acquisition unit (22).

5. A laser cleaning system based on coaxial visual feedback according to claim 2, characterized in that, The filter assembly (4) includes a narrowband filter (41) disposed in front of the image acquisition unit (22). The center wavelength of the narrowband filter (41) matches the wavelength λ2 and cuts off light with wavelength λ1.

6. The laser cleaning system based on coaxial visual feedback according to claim 5, characterized in that, The optical path control module (3) includes a polarizing beam splitter (31), a dichroic mirror (32), a galvanometer (33), and a field mirror (34); the dichroic mirror (32) refracts light with wavelength λ1 and reflects light with wavelength λ2.

7. A laser cleaning method based on coaxial visual feedback, characterized in that, The methods include: S01, the analysis and control module (5) sends the first timing excitation command to the laser emission module (1) at time T0 and starts the timer; the laser emission module (1) emits a cleaning laser f1 with wavelength λ1 to the optical path control module (3); S02, the timer executes a countdown Δt, and the analysis and control module (5) sends a visual trigger signal to the visual monitoring module (2) at time T0+Δt. The visual monitoring module (2) includes an illumination light emitting unit (21) and an image acquisition unit (22). The visual trigger signal includes a strobe signal and a shutter opening signal. S03, the illumination light emitting unit (21) receives the strobe signal and emits auxiliary illumination light f2 to the optical path control module (3); the image acquisition unit (22) receives the shutter opening signal and acquires the feature image of the cleaning area of ​​the target workpiece (6); S04, the image acquisition unit (22) uploads the feature image to the analysis and control module (5), the analysis and control module (5) filters and analyzes the feature image, and generates a feedback control command; the feedback control command is used to control the execution state of the laser emission module (1).

8. The laser cleaning method based on coaxial visual feedback according to claim 7, characterized in that, The specific steps of the analysis and control module (5) in filtering and analyzing the feature image are as follows: Graphical features are extracted from the feature image, and the cleanliness measurement value Z is obtained by analyzing the graphic features based on a preset analysis algorithm. k ; The cleanliness measurement value Z is based on the asynchronous Kalman filter backtracking correction method. k Perform asynchronous updates and generate an estimate of the current residual value X based on the updated estimate. k； The energy U of the next pulse is calculated based on a pre-defined cleaning state-space model. k+1 ; Based on the next pulse energy U k+1 Generate a feedback control command for controlling the execution state of the laser emission module (1) at the next moment.

9. The laser cleaning method based on coaxial visual feedback according to claim 8, characterized in that, This laser cleaning method based on coaxial visual feedback also includes an over-cleaning defense step: Based on the current estimated residual value X k The estimated value X of the residue at the next moment is obtained by predicting the state space of the pre-set cleaning process. k+1 ; Compare the estimated residual amount X at the next moment. k+1 And the magnitude of the cleanliness threshold R; if X k+1 If R >, then control the laser emitting module (1) to perform the laser emitting action at the next moment; if X k+1 If ≤R, then stop the cleaning action on the current cleaning area of ​​the target workpiece (6).

10. The laser cleaning method based on coaxial visual feedback according to claim 9, characterized in that, The cleaning state space model is specifically as follows: X k+1 = AX k + BU k Among them, X k For the thickness of pollutants, U k Let A be the laser energy density, and B be preset weighting coefficients.

Images