Laser processing method and device for special shell

Abstract

The invention discloses a laser processing method and device for a special shell, and the method comprises the steps: S1, obtaining the structural feature data of the shell through three-dimensional scanning, planning a laser processing path according to the structural feature data, and generating a processing path file; s2, starting a dust removal assembly, emitting positioning laser at the same time, measuring the optical path offset through a photoelectric detection assembly, and generating calibration parameters; s3, the light transmittance change of the light blocking assembly is controlled based on the calibration parameters, and a stable laser beam is formed by adopting a closed-loop control strategy; s4, a stable laser beam irradiates the surface of the shell along the machining path, and the irradiation position is dynamically corrected in combination with the deformation prediction model; s5, local laser strengthening is conducted on the machined key area, and meanwhile the machining path file of the next period is optimized; a light blocking assembly with adjustable light transmittance is combined with a closed-loop control strategy to dynamically adjust laser energy, deformation prediction and spectral analysis provide real-time quality guarantee, the process is continuously optimized through iterative optimization of a digital model, and a high-quality shell finished product is obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of laser processing, and in particular to a laser processing method and system for special shells. Background Art

[0002] As the core component of the nuclear power plant safety system, the nuclear reactor shell is the second protection barrier to ensure the safe operation of the reactor (the first is the fuel cladding), and undertakes key functions such as maintaining the integrity of the pressure boundary and blocking the leakage of radioactive substances. Modern nuclear reactor shells are mostly made of high-strength alloy steel, and their structures show asymmetric and multi-curved surface characteristics. In welding processing, full-thickness penetration needs to be achieved, and the width of the heat-affected zone needs to be strictly controlled within the micron level. Due to the technical requirements of nuclear safety class I equipment, the shell welds must meet the strict standards of zero defects and full-life-cycle reliability, which puts extremely high requirements on the energy control accuracy and process stability of laser processing.

[0003] Currently, the laser processing of nuclear reactor shells mainly uses high-power fiber lasers in combination with mechanical optical path adjustment systems for continuous welding by presetting power parameters. However, during the energy ramp-up process (about 50 - 100 ms) of the laser when it starts, an unstable laser beam lower than the set threshold will be generated. At the same time, the optical path scattering caused by suspended particles in the processing environment will cause a 10% - 15% fluctuation in the actual irradiation energy. Although the existing technology intercepts the initial laser with a physical baffle, the mechanical light-blocking component has a slow response speed (>20 ms) and is difficult to accurately match the laser power change curve, resulting in defects such as incomplete penetration or excessive penetration depth in the starting section of the weld. More seriously, when processing complex curved surface areas, the traditional method cannot compensate in real time for the energy distribution distortion caused by the deflection of the optical path, and ultimately leads to a penetration uniformity deviation of the shell weld exceeding ±12%, directly affecting the reliability of the nuclear safety barrier.

[0004] In view of this, it is necessary to improve the laser processing of shells in the existing technology to solve the problems of unstable initial laser energy and quality defects caused by optical path deviation. Summary of the Invention

[0005] The purpose of the present invention is to provide a laser processing method and system for special shells to solve the above technical problems.

[0006] To achieve this purpose, the present invention adopts the following technical solutions: A laser processing method for special shells, comprising the following steps: S1, obtaining the structural feature data of the shell through three-dimensional scanning, planning the laser processing path according to the structural feature data, and generating a processing path file including coordinate positions, power parameters, and moving speeds; S2. Based on the machining path file, start the dust removal component to remove impurities in the optical path. Meanwhile, emit positioning laser and measure the optical path offset through the photoelectric detection component to generate calibration parameters for the laser beam angle. S3. Based on the calibration parameters, control the light transmittance change of the light blocking component, and gradually increase the laser power to the set threshold using a closed-loop control strategy to form a stable laser beam. S4. Irradiate the surface of the shell along the machining path with the stable laser beam, dynamically correct the irradiation position in combination with the deformation prediction model, and simultaneously detect the weld state in real time through photosensitive analysis and trigger abnormal correction. S5. Perform local laser strengthening on the key areas after machining, and simultaneously collect machining data to construct an optimized digital model to optimize the machining path file for the next machining cycle.

[0007] Optionally, the step S1 specifically includes the following steps: S11. Obtain three-dimensional point cloud data through laser scanning, and simultaneously capture the shell thickness distribution characteristics using an infrared thermal imager to generate an original three-dimensional structure data packet. S12. Perform multi-source data fusion processing, match and calibrate the three-dimensional point cloud data with the preset shell CAD model, identify the geometric parameters of the asymmetric weld track, special-shaped curved surface contour, and thickness mutation area on the shell surface, and generate a characteristic parameter matrix. S13. According to the light absorption rate and thermal conductivity of the shell material, calculate the theoretical laser energy requirement value for each region, generate a material-energy correlation mapping table, and dynamically correct the three-dimensional point cloud data. S14. Use the ant colony algorithm to optimize the path of the characteristic parameter matrix, automatically generate a spiral progressive scanning path for the thickness mutation area, and generate an interleaved compensation path for the asymmetric weld area, and output an initial machining path plan. S15. According to the material-energy correlation mapping table and the initial machining path plan, establish a mathematical relationship model between the spatial coordinate position, laser power, and moving speed, and generate a dynamic control instruction set. S16. Associate the dynamic control instruction set with the nuclear radiation protection parameters, and embed the power attenuation instruction and cooling cycle parameters of the radiation shielding area in the path file to form a machining path file with anti-radiation enhancement.

[0008] Optionally, the dynamic control instruction set specifically includes the following parameters: For the special-shaped curved surface contour with a curvature radius not exceeding the first threshold: the power increase coefficient is 1.2 - 1.5, and the speed is reduced to 70% of the reference value. For the thickness mutation area with a thickness change gradient not less than the second threshold: the power ramp slope is 0.8 kW / mm, and the speed fluctuation threshold is ±5%. The asymmetric structural angle area is used as the asymmetric weld trajectory: the path offset compensation amount is 0.05-0.3mm.

[0009] Optionally, the step S2 specifically includes the following steps: S21, identifying high dust risk areas according to the spatial coordinate parameters in the processing path file, dynamically adjusting the local wind speed and airflow direction of the dust collection device, and forming a directional dust removal strategy; S22, executing the coordinated control of graded laser positioning and dust removal, firstly emitting a low-power positioning laser to scan the processing area, and synchronously starting a pulsed dust removal airflow to remove suspended particles in the optical path by capturing the intensity distribution of the scattered signal; S23 measures multi-dimensional optical path offset parameters, uses a dual-frequency positioning laser beam to illuminate the target area in a time-sharing manner, collects spot position deviation data through a high-speed CMOS sensor array, builds an optical path offset prediction model based on the historical processing database, and outputs a primary calibration data set: S24, generating dynamic environmental compensation calibration parameters, fusing the primary calibration data set with the temperature data and humidity data collected in real time by the environmental sensor, predicting the dynamic distortion of the optical path through a deep learning algorithm, and generating calibration parameters for the laser beam angle.

[0010] Optionally, the primary calibration data set specifically includes the following parameters: Horizontal deflection angle compensation value, the range is ±0.1°~±1.2°; Vertical focus position offset, the range is -0.05~+0.3mm; Energy attenuation correction factor ranges from 1.05 to 1.25 times.

[0011] Optionally, the light blocking assembly includes a mounting plate disposed inside the dust removal assembly, the mounting plate being provided with a mounting hole, an annular photosensor and a photoelectric modulator being disposed in the mounting hole, the annular photosensor being used to detect photosensitivity data during the welding process, and the photoelectric modulator being used to control the transmittance of the light blocking assembly through an electrical signal.

[0012] Optionally, step S3 specifically includes the following steps: S31, initializing the light blocking component and establishing a transmittance control channel, setting the initial transmittance of the photoelectric modulator to 0.1%-0.5% based on the energy attenuation gradient in the calibration parameter package, and activating the annular photosensor to monitor the laser incident angle; S32, performing dynamic matching of multi-zone transmittance, and controlling the photoelectric modulator to form a gradient transmittance distribution in the XY plane according to the angle compensation matrix in the calibration parameters, wherein the transmittance increase rate in the high energy demand area is 5% / ms, and the transmittance increase rate in the low risk area is 2% / ms; S33. Implement closed-loop power ramp loading, collect laser power data in real time through an annular photosensitive sensor, and dynamically adjust the electrical signal intensity of the electro-optical modulator by combining a preset light intensity threshold and a PID control algorithm, so that the laser power rises from the initial power to 99% of the set value in an exponential curve within 50 ms. S34. When it is detected that the laser power volatility region is stable and lasts for a preset duration, lock the electrical signal parameters of the electro-optical modulator, generate a stable beam identifier, and form a stable laser beam.

[0013] Optionally, step S4 specifically includes the following steps: S41. Load the stable laser beam parameters and initialize the processing path, perform spatio-temporal synchronous matching between the stable laser beam identifier and the spatial coordinates in the processing path file, and generate a dynamic processing instruction set including the laser power-movement speed coupling relationship. S42. Predict the deformation amount of the current irradiation area based on the thermo-mechanical coupling model, and adjust the three-dimensional offset of the laser focus in real time according to the thermal expansion coefficient and residual stress distribution data of the housing material. S43. Collect the light intensity data of the welding molten pool through the annular photosensitive sensor, and obtain the real-time weld quality based on the monitored light intensity data. S44. Trigger a correction mechanism according to the obtained real-time weld quality. When it is detected that the fluctuation of the molten pool length exceeds the standard or the keyhole collapse frequency is greater than 3 times / second, automatically execute the following correction mechanism: Instantly increase the laser power to 120% of the set value and maintain it for 10 ms to eliminate the lack of penetration defect. Append a 0.1 mm offset in the path normal direction to compensate for thermal deformation. Record the abnormal coordinates and mark them as key areas.

[0014] Optionally, step S5 specifically includes the following steps: S51. According to the records of the correction mechanism and the process parameters in the processing path file, extract the coordinate data of the stress concentration area, the sealing surface, and the penetration depth fluctuation exceeding the standard, and generate a key area coordinate set and the corresponding energy density threshold. S52. Perform gradient energy laser strengthening treatment, regulate the laser power and dwell time based on the energy density threshold, and perform impact strengthening on the sealing surface area in an annular path at 1.2 - 1.5 times the reference power density. S53. Integrate the processing process characteristic data, collect the light intensity characteristic data of the strengthening area through the annular photosensitive sensor, synchronously extract the laser power, movement speed, and deformation compensation parameters in the processing path file, and generate a characteristic data set including the temperature distribution in the strengthening area, the intensity ratio of the molten pool solidification characteristic spectrum line, and the statistical distribution of deformation compensation. S54. Construct a material constitutive model according to the material-energy correlation mapping table and the deformation amount predicted by the thermal-mechanical coupling model; S54. Couple the feature dataset and the material constitutive model, and use an optimized digital model to train process parameters to output an updated parameter set: S55. Integrate and optimize the updated parameter set with the original processing path file, and output the next-generation path file after verifying the process stability.

[0015] The present invention also provides a laser processing device for a special shell to implement the laser processing method for the special shell as described above. The laser processing device specifically includes: A laser generator, which uses a fiber-coupled continuous laser and is equipped with an optical component for beam shaping; A laser adjustment component, which carries the laser generator and is used to adjust the position and angle of the laser generator; A dust removal component, which is arranged opposite to the laser generator and has a dust removal cavity inside; A light shielding component, which is arranged in the dust removal cavity and includes an annular photosensitive sensor array and an electrochromic glass modulator; A control host, which is used to coordinately control the operation of each component and provide corresponding control strategies.

[0016] Compared with the prior art, the present invention has the following beneficial effects: First, obtain the shell structure features through three-dimensional scanning and generate a processing path file. Subsequently, start the dust removal component to purify the light path and calibrate the laser beam angle parameters. Based on the calibration data, dynamically adjust the light transmittance of the light shielding component to output a stable laser beam. During the processing, combine deformation prediction to real-time correct the irradiation position and monitor the weld state. Finally, perform strengthening treatment on the key area and optimize the processing parameters of the next cycle based on data feedback to form a closed-loop intelligent processing system; This method uses a light shielding component with adjustable light transmittance combined with a closed-loop control strategy to dynamically adjust the laser energy. Deformation prediction and spectral analysis provide real-time quality assurance. The iterative optimization of the digital model continuously optimizes the process, solves the processing defects caused by energy fluctuations and light path deviation in traditional laser processing, and obtains high-quality special shells. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0018] The structures, proportions, sizes, etc. shown in the accompanying drawings of this specification are only used to cooperate with the content disclosed in the specification for those familiar with this technology to understand and read, and are not used to limit the limiting conditions under which the present invention can be implemented. Therefore, they do not have substantial technical significance. Any modification of the structure, change in the proportional relationship, or adjustment of the size, without affecting the effects that the present invention can produce and the purposes that can be achieved, should still fall within the scope covered by the technical content disclosed in the present invention.

[0019] Figure 1 It is a schematic overall layout diagram of the laser processing device for the special shell in the second embodiment; Figure 2 It is a partial sectional view schematic diagram of the laser processing device for the special shell in the second embodiment; Figure 3 It is a schematic diagram of the light shielding component of the laser processing device for the special shell in the third embodiment; Figure 4 It is one of the schematic flow diagrams of the laser processing method for the special shell in the first embodiment; Figure 5 It is the second schematic flow diagram of the laser processing method for the special shell in the first embodiment. Detailed implementation manners

[0020] To make the object, features, and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the embodiments described below are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope protected by the present invention.

[0021] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by terms such as "upper", "lower", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation to the present invention. It should be noted that when a component is considered to be "connected" to another component, it can be directly connected to the other component or there may be an intermediate component present.

[0022] The technical solutions of the present invention will be further described below in conjunction with the accompanying drawings and through specific implementation manners.

[0023] Embodiment 1: Combined with Figure 4 and Figure 5As shown in the figure, an embodiment of the present invention provides a laser processing method for a special shell, including the following steps: S1. Obtain the structural feature data of the shell through three-dimensional scanning, plan the laser processing path according to the structural feature data, and generate a processing path file including coordinate positions, power parameters, and moving speeds. Obtaining the structural feature data of the shell through three-dimensional scanning and generating a processing path file realizes the precise digital modeling of complex shells (such as curved surfaces and special-shaped structures). The path file integrates the collaborative optimization of coordinate positions, power parameters, and moving speeds, solving the problem of processing trajectory deviation caused by structural deviation in traditional preset paths.

[0024] S2. Based on the processing path file, start the dust removal component to remove impurities in the optical path, and at the same time emit positioning laser and measure the optical path offset through the photoelectric detection component to generate calibration parameters for the laser beam angle. Before starting the processing, perform dust removal and optical path offset measurement synchronously, directly intervening in the energy fluctuations caused by the scattering of suspended particles. The dust removal component removes impurities in the optical path, reducing scattering losses from the source; the photoelectric detection component measures the offset in real time through the positioning laser and generates calibration parameters, which can quickly compensate for the optical path deflection caused by mechanical vibration or assembly errors (such as distortion during complex curved surface processing). Improve the stability of the optical path and provide guarantee for the stable output of high-energy density laser in the follow-up.

[0025] S3. Based on the calibration parameters, control the change of the light transmittance of the light blocking component, and adopt a closed-loop control strategy to gradually increase the laser power to the set threshold to form a stable laser beam. Replace the traditional mechanical baffle with a light blocking component with adjustable light transmittance (such as an electro-optic modulator or a liquid crystal modulator), and combine it with a closed-loop control strategy to increase the response speed from >20ms of the mechanical baffle to the microsecond level, accurately matching the power ramp curve (50 - 100ms) at the start-up stage of the laser. By dynamically adjusting the light transmittance, unstable low-energy laser (such as stray light in the initial stage) can be filtered out, and smoothly transition to the set threshold power, completely eliminating the defects of incomplete penetration or excessive penetration depth at the starting section of the weld, while reducing the risk of energy overshoot.

[0026] S4. Irradiate the surface of the shell with the stable laser beam along the processing path, dynamically correct the irradiation position in combination with the deformation prediction model, and at the same time detect the weld state in real time through photosensitive analysis and trigger abnormal correction. During the processing, predict the heat deformation trend of the shell through the deformation prediction model (such as local offset caused by thermal expansion), and dynamically correct the laser irradiation position, solving the problem of energy distribution distortion caused by optical path deflection or material deformation in complex curved surface processing. Combining real-time weld state detection by spectroscopy or thermal imaging, abnormal conditions in the molten pool (such as pores or cracks) can be identified, triggering immediate correction of power or path to ensure uniform penetration.

[0027] S5. Locally laser strengthen the processed key area, and at the same time collect processing data to construct an optimized digital model to optimize the processing path file for the next processing cycle.

[0028] Locally laser strengthen the key area to specifically improve the structural strength and make up for possible microscopic defects during the processing. At the same time, collect processing data (such as actual power curve, deformation compensation amount) to construct an optimized digital model, and iteratively update the path file for the next cycle through machine learning to achieve self-optimization of process parameters. This closed-loop optimization mechanism continuously improves the processing accuracy and efficiency with the number of iterations, especially suitable for large-scale production of multi-batch special shells.

[0029] The working principle of the present invention is as follows: First, obtain the shell structure characteristics through three-dimensional scanning and generate a processing path file. Then, start the dust removal component to purify the optical path and calibrate the laser beam angle parameters. Based on the calibration data, dynamically adjust the light transmittance of the light blocking component to output a stable laser beam. During the processing, combine deformation prediction to real-time correct the irradiation position and monitor the weld state. Finally, strengthen the key area and optimize the processing parameters for the next cycle based on data feedback to form a closed-loop intelligent processing system; this method uses a light blocking component with adjustable light transmittance combined with a closed-loop control strategy to dynamically adjust the laser energy, deformation prediction and spectral analysis provide real-time quality assurance, and the iterative optimization of the digital model continuously optimizes the process, solves the processing defects caused by energy fluctuations and optical path offsets in traditional laser processing, and obtains high-quality special shells.

[0030] In this embodiment, specifically, step S1 specifically includes the following steps: S11. Obtain three-dimensional point cloud data through laser scanning, and at the same time use an infrared thermal imager to capture the shell thickness distribution characteristics to generate an original three-dimensional structure data packet; The three-dimensional point cloud data emits a laser beam through a laser scanner and receives the reflection signal to generate a high-density three-dimensional coordinate point set, accurately describing the geometric shape of the shell surface.

[0031] The infrared thermal imager infers the material thickness distribution by detecting the infrared radiation intensity difference on the shell surface (for example, the thick area has slow heat dissipation and strong thermal radiation signal).

[0032] Capture the geometric morphology and thickness characteristics at the same time, avoid energy misjudgment caused by ignoring material distribution differences in traditional single scanning methods, and provide multi-dimensional data support for subsequent path planning.

[0033] S12. Perform multi-source data fusion processing, match and calibrate the three-dimensional point cloud data with the preset shell CAD model, identify the geometric parameters of the asymmetric weld track, special-shaped surface contour and thickness mutation area on the shell surface, and generate a characteristic parameter matrix; Align the measured point cloud data with the preset CAD model using a point cloud registration algorithm (such as the ICP Iterative Closest Point), and extract the offset of the asymmetric weld track, the curvature parameters of the special-shaped surface, and the boundary of the thickness mutation area (such as the area where the thickness suddenly increases from 5 mm to 15 mm) through residual analysis. Preferably, the characteristic region can be classified based on a deep learning model to generate a characteristic parameter matrix containing position, curvature, and thickness.

[0034] Eliminate the systematic deviation between the machining reference coordinate system and the physical model, accurately locate the thickness mutation area, and avoid the path planning failure caused by model mismatch.

[0035] S13. Calculate the theoretical laser energy requirement value for each region according to the light absorption rate and thermal conductivity of the shell material, generate a material-energy correlation mapping table, and dynamically correct the three-dimensional point cloud data; The light absorption rate is the proportion of the energy absorbed by the material for a laser with a specific wavelength (for example, the light absorption rate of zirconium alloy for 1064 nm laser is about 35%).

[0036] The thermal conductivity is the heat conduction ability of the material (such as 80 W / m·K), which affects the laser heat accumulation and the molten pool diffusion speed.

[0037] Dynamically generate an energy requirement table (for example, an 8 kW power is required for the 10 mm thick area), solve the problem of insufficient or excessive melting depth in the thickness mutation area in the traditional fixed power mode, and improve the energy utilization rate.

[0038] S14. Use the ant colony algorithm to optimize the path of the characteristic parameter matrix, automatically generate a spiral progressive scanning path for the thickness mutation area, and generate an interleaved compensation path for the asymmetric weld area, and output the initial machining path plan; The ant colony algorithm is an intelligent optimization algorithm that simulates the foraging path of ants. It finds the shortest path through the "pheromone" weight. In this scheme, a thickness weight factor and a heat accumulation penalty term are introduced.

[0039] For the spiral progressive path, gradually increase the laser power density in a spiral trajectory in the thickness mutation area to avoid sudden thermal stress changes.

[0040] The interleaved compensation path scans alternately left and right in the asymmetric weld area to offset the unilateral thermal deformation.

[0041] The spiral path reduces the temperature gradient in the thickness mutation area, and the interleaved path suppresses the weld warping (deformation amount < 0.05 mm), improving the processing efficiency.

[0042] S15. According to the material-energy correlation mapping table and the initial machining path plan, establish a mathematical relationship model between the spatial coordinate position, laser power, and moving speed, and generate a dynamic control instruction set; Mathematical relationship model, establishing the correlation rule between the spatial coordinates (X, Y, Z) and the laser power (P) and the moving speed (V) (for example, P = K·V·T, where T is the local thickness).

[0043] Dynamic control instruction set, converting the mathematical model into a machine-executable control signal in milliseconds, realizing the real-time synchronization of laser parameters and the processing trajectory (response delay < 1ms), and eliminating the energy step defect caused by traditional segmented control.

[0044] S16, associating the dynamic control instruction set with the nuclear radiation protection parameters, embedding the power attenuation instruction and the cooling cycle parameters of the radiation shielding area in the path file to form a radiation-resistant enhanced processing path file.

[0045] Power attenuation instruction, automatically reducing the power to a safe threshold (for example, 10kW → 6kW) in the radiation-sensitive area (such as the weld intersection point) to reduce lattice damage.

[0046] Cooling cycle parameter: Inserting a forced cooling interval in the high heat accumulation area (for example, pausing for 0.5s after every 5mm scan) to prevent microcracks.

[0047] Reducing the lattice distortion rate in the radiation-sensitive area, the cooling strategy reduces the microcrack density, meeting the anti-radiation life requirement of the nuclear container.

[0048] In this embodiment, it is further illustrated that the dynamic control instruction set specifically includes the following parameters: Taking the area with a curvature radius not exceeding the first threshold as the profile of the special-shaped curved surface: the power boost coefficient is 1.2 - 1.5, and the speed is reduced to 70% of the reference value; the power boost coefficient (1.2 - 1.5 times) compensates for the surface reflection loss by increasing the energy density; the speed is reduced to 70% of the reference value to extend the heat action time and ensure the consistency of the melt depth.

[0049] Taking the area with a thickness change gradient not less than the second threshold as the thickness mutation area: the power ramp slope is 0.8kW / mm, and the speed fluctuation threshold is ±5%; avoiding the insufficient energy caused by the sudden increase in thickness, and the speed fluctuation limit of ±5% to prevent the out-of-control flow of the molten pool.

[0050] Taking the non-symmetric weld seam trajectory in the non-symmetric structure included angle area: the path offset compensation amount is 0.05 - 0.3mm, offsetting the spot offset caused by the geometric asymmetry of the laser beam to ensure the alignment of the weld center.

[0051] In this embodiment, it is specifically illustrated that step S2 specifically includes the following steps: S21, identifying the high-dust risk area according to the spatial coordinate parameters in the processing path file, dynamically adjusting the local wind speed and air flow direction of the dust suction device to form a directional dust removal strategy; In high-dust-risk areas, by analyzing the spatial coordinate parameters in the processing path file (such as dust-prone areas like weld intersections and curved surface depressions), combined with historical dust deposition data, dynamically mark the areas where the dust concentration may exceed the standard.

[0052] Directional dust removal strategy, using an adjustable dust suction device, adjusts the local wind speed (for example, from 5 m / s to 15 m / s) and air flow direction (such as blowing at an angle of 30°) in real time according to the dust distribution, forming a local strong air flow field to directionally remove suspended particles.

[0053] Precisely remove dust in high-dust areas, avoiding the waste of air flow in traditional full-area dust removal, and at the same time reducing the energy fluctuations caused by laser scattering.

[0054] S22, perform hierarchical laser positioning and dust removal collaborative control. First, emit a low-power positioning laser to scan the processing area. By capturing the intensity distribution of the scattered signal, synchronously start the pulsed dust removal air flow to remove the suspended particles in the optical path. Hierarchical laser positioning, first emit a low-power positioning laser to scan the processing area. By using a photoelectric sensor to capture the intensity distribution of the scattered signal, identify the concentration of suspended particles (such as a signal attenuation > 20% is determined as a heavily polluted area).

[0055] Pulsed dust removal air flow, synchronously trigger a high-frequency pulsed air flow (such as a 200 Hz short-time strong air flow) according to the intensity of the scattered signal, and only turn on the dust removal instantaneously in front of the laser scanning path to avoid the interference of continuous air flow on the processing stability.

[0056] The low-power positioning laser avoids preheating the material in advance, and the pulsed dust removal reduces the air flow energy consumption, ensuring the stability of the main laser beam energy transmission.

[0057] S23, measure multi-dimensional optical path offset parameters. Use a dual-frequency positioning laser beam to irradiate the target area at different times. Collect the spot position deviation data through a high-speed CMOS sensor array, and combine the historical processing database to construct an optical path offset prediction model, and output the primary calibration data set: Dual-frequency positioning laser beam: Emit two kinds of lasers with different wavelengths (such as 532 nm green light and 1064 nm infrared light) at different times. Utilize the difference in scattering characteristics of different wavelengths to separate the environmental interference signals and accurately measure the spot position deviation (with an accuracy of ±0.01 mm).

[0058] Optical path offset prediction model: Based on the real-time deviation data collected by the high-speed CMOS sensor (such as the offset in the X / Y direction and the angle deflection), combined with the historical processing database (such as vibration and assembly error records), construct a time series prediction model (such as an LSTM neural network) to output the optical path offset trend within the next 0.1 second.

[0059] Dual-frequency laser eliminates ambient light interference, and the detection error of the spot position is <0.05mm; the prediction model shortens the compensation response time for the optical path offset.

[0060] S24. Generate dynamic environment compensation calibration parameters, fuse the primary calibration data set with the temperature data and humidity data collected in real time by the environmental sensors, and predict the dynamic distortion amount of the optical path through a deep learning algorithm to generate calibration parameters for the laser beam angle.

[0061] Fuse the real-time data of the temperature sensor (measurement range: -20°C to 200°C) and the humidity sensor (accuracy: ±2%RH) with the primary calibration data set (such as spot deviation, predicted offset), and quantify the comprehensive influence of environmental factors on the optical path (for example, the beam deflects 0.001° for every 1°C increase in temperature).

[0062] Train a convolutional neural network (CNN) model for deep learning dynamic distortion prediction. Input the environmental data and historical offset records, and output the dynamic compensation value of the optical path angle (such as the compensation deflection angle is 0.05° to 0.5°).

[0063] In this embodiment, it is further explained that the primary calibration data set specifically includes the following parameters: Horizontal deflection angle compensation value, whose range is ±0.1° to ±1.2°; measure the deflection angle of the optical path through a dual-frequency laser (such as a 0.5° deflection caused by vibration or assembly error), generate a reverse compensation instruction, and correct the alignment deviation of the laser beam in the horizontal direction.

[0064] Vertical focusing position offset, whose range is -0.05 to +0.3mm; adjust the position of the focusing lens in real time according to the spot imaging analysis to compensate for the focus drift caused by the thermal lens effect or dust scattering (for example, when there is a +0.2mm offset, reverse compensate -0.2mm).

[0065] Energy attenuation correction coefficient, whose range is 1.05 to 1.25 times. Based on the light intensity attenuation data detected by the annular photosensitive sensor, dynamically increase the output power to maintain the set energy threshold.

[0066] In this embodiment, it is specifically explained that the light blocking component includes a mounting plate disposed inside the dust removal component. The mounting plate is provided with mounting holes, and an annular photosensitive sensor and an electro-optical modulator are disposed in the mounting holes. The annular photosensitive sensor is used to detect the photosensitive data during the welding process, and the electro-optical modulator is used to control the light transmittance of the light blocking component through an electrical signal.

[0067] It should be noted that the electro-optical modulator uses a lithium niobate electro-optical modulator. By applying an external voltage to change the crystal refractive index, the light transmittance is dynamically adjusted to replace the physical occlusion of the traditional mechanical baffle.

[0068] The response speed of the electro-optical modulator is <1 μs (the traditional mechanical shutter >20 ms), and the adjustment accuracy of the light transmittance is relatively high; the annular photosensitive sensor synchronously monitors the scattered light to reduce the ineffective radiation during the energy ramp-up stage.

[0069] In this embodiment, specifically, step S3 specifically includes the following steps: S31, Initialize the light-blocking component and establish a light transmittance control channel. Based on the energy attenuation gradient in the calibration parameter package, set the initial light transmittance of the electro-optical modulator to 0.1% - 0.5%, and at the same time activate the annular photosensitive sensor to monitor the laser incident angle; The light transmittance control channel controls the refractive index of the electro-optical modulator (such as lithium niobate electro-optic crystal) through an electrical signal (such as a 0 - 10V voltage) to form a light transmittance adjustment path. The initial light transmittance is set to 0.1% - 0.5% (close to the fully enclosed state) to intercept the stray light when the laser is started.

[0070] The annular photosensitive sensor monitors. The sensor annular array real-time detects the laser incident angle (such as the angle deviation between the central spot and the edge scattered light) to prevent energy leakage caused by the optical path deviation.

[0071] S32, Perform multi-region light transmittance dynamic matching. According to the angle compensation matrix in the calibration parameters, control the electro-optical modulator to form a gradient light transmittance distribution in the X - Y plane, where the light transmittance increase rate in the high-energy demand region is 5% / ms, and the light transmittance increase rate in the low-risk region is 2% / ms to achieve on-demand energy distribution.

[0072] According to the horizontal / vertical offset in the calibration parameters (such as the ±0.1° deflection output by S24), divide the electro-optical modulator into multiple control regions (such as a 10×10 grid), assign an independent light transmittance adjustment rate to each grid, and establish an angle compensation matrix.

[0073] The multi-region gradient adjustment makes the energy distribution match the processing requirements, and the response speed in the high-energy region is increased by 150%, improving the energy utilization rate.

[0074] S33, Implement closed-loop power ramp loading. Real-time collect the laser power data through the annular photosensitive sensor, and dynamically adjust the electrical signal intensity of the electro-optical modulator in combination with the preset light intensity threshold and the PID control algorithm, so that the laser power rises from the initial power to 99% of the set value within 50 ms according to an exponential curve; The PID control algorithm, through the proportional-integral-derivative algorithm, compares the actual power collected by the annular photosensitive sensor with the preset threshold (such as 10 kW) in real time, and dynamically adjusts the modulator voltage to control the light transmittance to climb according to an exponential curve (such as the time constant τ = 15 ms).

[0075] The exponential curve power increases, the initial light transmittance increases rapidly, and then increases slowly approaching the set value, avoiding the molten pool splash caused by the stepwise power impact.

[0076] S34. When it is detected that the laser power volatility region is stable and lasts for a preset duration, lock the electrical signal parameters of the electro-optical modulator, generate a stable beam identifier, and form a stable laser beam.

[0077] The power volatility is detected by statistically analyzing the data of the annular photosensitive sensor, calculating the power standard deviation to determine the stability, and locking the modulator voltage.

[0078] The stable beam identifier generates a digital signal (such as a high-level pulse) to trigger the processing system to enter the steady state mode, ensuring that the subsequent processing light intensity is constant.

[0079] In this embodiment, specifically, step S4 specifically includes the following steps: S41. Load the stable laser beam parameters and initialize the processing path, perform spatio-temporal synchronous matching between the stable laser beam identifier and the spatial coordinates in the processing path file, and generate a dynamic processing instruction set including the laser power - moving speed coupling relationship; Through the timestamp alignment technology, synchronize the power parameters of the stable laser beam and the coordinate moving speed of the processing path file at the millisecond level to ensure real-time matching of the laser energy and the moving trajectory. For example, in the curvature mutation area, synchronously increase the power to 12 kW and reduce the speed to 0.35 m / s.

[0080] The dynamic processing instruction set generates control instructions including the power - speed coupling relationship (such as a set of P - V parameters corresponding to each 0.1 mm coordinate point), and the microsecond-level instruction execution is realized through the FPGA hardware.

[0081] S42. Predict the deformation amount of the current irradiation area based on the thermal - mechanical coupling model, and adjust the three-dimensional offset of the laser focus in real time according to the thermal expansion coefficient and residual stress distribution data of the shell material (the X / Y axis compensation range is ±0.05 - 0.2 mm, and the Z axis compensation is ±0.01 - 0.05 mm).

[0082] Based on the material thermal expansion coefficient and residual stress distribution (such as the welding residual stress of 200 MPa), predict the transient deformation amount of the irradiation area (such as a 0.2 mm offset caused by thermal expansion) through finite element simulation, and establish a thermal - mechanical coupling model.

[0083] Three-dimensional focus compensation, adjust the galvanometer system in real time according to the predicted deformation amount, and dynamically correct the laser focus position.

[0084] S43. Collect the light intensity data of the welding molten pool through the annular photosensitive sensor, and obtain the real-time weld quality based on the monitored light intensity data; The annular photosensitive sensor collects the light intensity distribution in the molten pool area in real time. The weld quality algorithm judges the penetration state through the light intensity fluctuation frequency (normal <1Hz) and the spatial distribution uniformity, and identifies defects such as incomplete penetration (light intensity drops suddenly by 30%) or pores (local light intensity rises suddenly).

[0085] S44. According to the obtained real-time weld quality, trigger the correction mechanism. When it is detected that the fluctuation of the molten pool length exceeds the standard or the keyhole collapse frequency is greater than 3 times per second, automatically execute the following correction mechanism: Instantly increase the laser power to 120% of the set value and keep it for 10ms to eliminate the incomplete penetration defect; trigger the PID overshoot algorithm, adjust the laser power to 120% of the rated value and keep it for 10ms to force the molten pool to be stable.

[0086] Append an offset of 0.1mm in the path normal direction to compensate for thermal deformation and offset the weld offset caused by deformation.

[0087] Record the abnormal coordinates and mark them as key areas. Record the defect positions in the database, and perform local remelting or strengthening on this area during subsequent processing.

[0088] In this embodiment, specifically, step S5 specifically includes the following steps: S51. According to the records of the correction mechanism and the process parameters in the processing path file, extract the coordinate data of the stress concentration area, the sealing surface and the penetration depth fluctuation exceeding the standard, and generate the key area coordinate set and the corresponding energy density threshold; Stress concentration area identification. Based on the records of the correction mechanism (such as the coordinates of the penetration depth fluctuation exceeding the standard) and the process parameters (such as welding speed, power), screen out the stress concentration areas (such as weld intersections, curvature mutation areas) through machine learning classification algorithms (such as support vector machines).

[0089] Energy density threshold generation. Combine the material fatigue limit (such as 300MPa for zirconium alloy) and the penetration depth fluctuation data to calculate the required energy density for each area (for example, the sealing surface requires 1.5 times the reference energy density to improve the density).

[0090] Precisely locate the key defect areas, improve the energy threshold matching degree, and avoid problems such as overburning or insufficient strengthening caused by traditional empirical thresholds.

[0091] S52. Perform gradient energy laser strengthening treatment. Based on the energy density threshold, regulate the laser power and the dwell time. In the sealing surface area, use a circular path to perform impact strengthening at 1.2 - 1.5 times the reference power density; Gradient energy regulation. Dynamically adjust the laser power (such as increasing from 10kW to 15kW) and the dwell time (such as extending from 0.5ms to 1.2ms) according to the energy density threshold. Use a circular scanning path (diameter 0.2 - 0.5mm) in the sealing surface to form a uniform strengthening layer.

[0092] The impact strengthening mechanism induces micro-plastic deformation on the material surface with a high power density (such as 1.5 times the reference value), refines the grain size (from 50 μm to 10 μm), and improves the anti-creep performance of the sealing surface.

[0093] S53, integrating the characteristic data of the processing process, collecting the light intensity characteristic data of the strengthening area through a ring-shaped photosensitive sensor, synchronously extracting the laser power, moving speed, and deformation compensation parameters in the processing path file, and generating a characteristic data set including the temperature distribution in the strengthening area, the intensity ratio of the solidification characteristic spectral lines of the molten pool, and the statistical distribution of deformation compensation: Collect the light intensity distribution in the strengthening area through a ring-shaped photosensitive sensor (such as the central spot intensity of 12 kW / cm² and the proportion of edge scattered light < 5%), synchronously extract the deformation compensation parameters in the processing path, and construct a multi-dimensional data set including the temperature gradient (such as 200 °C / mm) and the solidification characteristic spectral line (such as the intensity ratio of the Fe ion spectral line of 1:0.8).

[0094] The data set covers the full-dimensional characteristics of physics-chemistry-geometry, provides high-confidence input for model training, and improves the accuracy of feature correlation.

[0095] S54, construct a material constitutive model according to the material-energy correlation mapping table and the deformation amount predicted by the thermal-mechanical coupling model; S54, couple the characteristic data set and the material constitutive model, and adopt an optimized digital model to train the process parameters to output an updated parameter set: The updated parameter set includes the correction value of the path curvature-power correlation coefficient, the dynamic adjustment threshold of the spot overlap rate, and the early warning sensitivity level of the abnormal area. Quantify the influence of laser processing on the microstructure (such as dislocation density) and macroscopic properties (such as tensile strength).

[0096] S55, fuse and optimize the updated parameter set with the original processing path file, and output the next-generation path file after verifying the process stability.

[0097] It should be noted that the genetic algorithm is used to perform multi-objective optimization on the characteristic data set and the constitutive model (objective function: melt depth uniformity + minimum energy consumption), output the updated parameter set, verify the stability of the updated parameters through virtual processing simulation, and generate the next-generation path file.

[0098] Embodiment 2: Combined with Figures 1 to 3 As shown, the present invention also provides a laser processing device for a special shell, which is used to implement the laser processing method for the special shell in Embodiment 1. The laser processing device specifically includes: The laser generator 10 is a fiber-coupled continuous laser and is equipped with an optical component for beam shaping. The emission end of the laser generator 10 faces the processing station 50 of the special housing.

[0099] The laser adjustment component 20 carries the laser generator 10 and is used to adjust the position and angle of the laser generator 10.

[0100] The dust removal component 30 is arranged opposite to the laser generator 10, and a dust removal cavity 31 is arranged inside it.

[0101] The light blocking component 40 is arranged in the dust removal cavity 31. The light blocking component 40 includes a mounting plate 41 arranged in the dust removal cavity 31. The mounting plate 41 is provided with mounting holes, and an annular photosensor 42 and an electro-optical modulator 43 are arranged in the mounting holes.

[0102] The control host is used to coordinate and control the operation of each component and provide corresponding control strategies.

[0103] As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A laser processing method for a special shell, characterized in that, It includes the following steps: S1. Obtain the structural feature data of the housing through 3D scanning, plan the laser processing path according to the structural feature data, and generate a processing path file including coordinate positions, power parameters, and moving speeds; S2. Based on the processing path file, start the dust removal component to remove impurities in the optical path. At the same time, emit positioning laser and measure the optical path offset through the photoelectric detection component to generate calibration parameters for the laser beam angle; S3. Control the light transmittance change of the light blocking component based on the calibration parameters, and gradually increase the laser power to the set threshold using a closed-loop control strategy to form a stable laser beam; S4. Irradiate the surface of the housing with the stable laser beam along the processing path, dynamically correct the irradiation position in combination with the deformation prediction model, and at the same time detect the weld state in real time through photosensitive analysis and trigger abnormal correction; S5. Perform local laser strengthening on the key areas after processing, and at the same time collect processing data to construct an optimized digital model to optimize the processing path file for the next processing cycle.

2. The laser processing method for the special shell according to claim 1, wherein The step S1 specifically includes the following steps: S11. Obtain 3D point cloud data through laser scanning, and at the same time use an infrared thermal imager to capture the housing thickness distribution characteristics to generate an original 3D structure data packet; S12. Perform multi-source data fusion processing, match and calibrate the 3D point cloud data with the preset housing CAD model, identify the geometric parameters of the asymmetric weld track, special-shaped curved surface contour, and thickness mutation area on the housing surface, and generate a feature parameter matrix; S13. Calculate the theoretical laser energy requirement values for each region according to the light absorption rate and thermal conductivity of the housing material, generate a material-energy correlation mapping table, and dynamically correct the 3D point cloud data; S14. Use the ant colony algorithm to optimize the path of the feature parameter matrix, automatically generate a spiral progressive scanning path for the thickness mutation area, generate an interleaved compensation path for the asymmetric weld area, and output an initial processing path plan; S15. According to the material-energy correlation mapping table and the initial processing path plan, establish a mathematical relationship model between the spatial coordinate position, laser power, and moving speed, and generate a dynamic control instruction set; S16. Associate the dynamic control instruction set with the nuclear radiation protection parameters, and embed the power attenuation instruction and cooling cycle parameters of the radiation shielding area in the path file to form a radiation-resistant enhanced processing path file.

3. The laser processing method of the special shell according to claim 2, characterized in that, The dynamic control instruction set specifically includes the following parameters: For the special-shaped curved surface contour with a curvature radius not exceeding the first threshold: the power increase coefficient is 1.2 - 1.5, and the speed is reduced to 70% of the reference value; For the thickness mutation area with a thickness change gradient not less than the second threshold: the power ramp slope is 0.8 kW / mm, and the speed fluctuation threshold is ±5%; For the asymmetric weld track in the asymmetric structure angle area: the path offset compensation amount is 0.05 - 0.3 mm.

4. The laser processing method of the special shell according to claim 1, wherein, The step S2 specifically includes the following steps: S21. Identify the high-dust risk areas according to the spatial coordinate parameters in the processing path file, dynamically adjust the local wind speed and air flow direction of the dust suction device to form a directional dust removal strategy; S22, executing the coordinated control of graded laser positioning and dust removal, firstly emitting a low-power positioning laser to scan the processing area, and synchronously starting a pulsed dust removal airflow to remove suspended particles in the optical path by capturing the intensity distribution of the scattered signal; S23 measures multi-dimensional optical path offset parameters, uses a dual-frequency positioning laser beam to illuminate the target area in a time-sharing manner, collects spot position deviation data through a high-speed CMOS sensor array, builds an optical path offset prediction model based on the historical processing database, and outputs a primary calibration data set: S24, generating dynamic environmental compensation calibration parameters, fusing the primary calibration data set with the temperature data and humidity data collected in real time by the environmental sensor, predicting the dynamic distortion of the optical path through a deep learning algorithm, and generating calibration parameters for the laser beam angle.

5. The laser processing method for the special shell according to claim 4, characterized in that, The primary calibration data set specifically includes the following parameters: Horizontal deflection angle compensation value, the range is ±0.1°~±1.2°; Vertical focus position offset, the range is -0.05~+0.3mm; Energy attenuation correction factor ranges from 1.05 to 1.25 times.

6. The laser processing method for the special shell according to claim 1, characterized in that The light blocking assembly includes a mounting plate arranged inside the dust removal assembly, the mounting plate is provided with a mounting hole, an annular photosensor and a photoelectric modulator are arranged in the mounting hole, the annular photosensor is used to detect photosensitivity data during the welding process, and the photoelectric modulator is used to control the transmittance of the light blocking assembly through an electrical signal.

7. The laser processing method for the special housing according to claim 6, characterized in that, The step S3 specifically includes the following steps: S31, initializing the light blocking component and establishing a transmittance control channel, setting the initial transmittance of the photoelectric modulator to 0.1%-0.5% based on the energy attenuation gradient in the calibration parameter package, and activating the annular photosensor to monitor the laser incident angle; S32, performing dynamic matching of multi-zone transmittance, and controlling the photoelectric modulator to form a gradient transmittance distribution in the XY plane according to the angle compensation matrix in the calibration parameters, wherein the transmittance increase rate in the high energy demand area is 5% / ms, and the transmittance increase rate in the low risk area is 2% / ms; S33, implement closed-loop power ramp loading, collect laser power data in real time through an annular photosensor, dynamically adjust the electrical signal intensity of the photoelectric modulator in combination with a preset light intensity threshold and a PID control algorithm, so that the laser power increases from the initial power to 99% of the set value within 50ms according to an exponential curve; S34, when it is detected that the laser power fluctuation rate area is stable and lasts for a preset time, the electrical signal parameters of the photoelectric modulator are locked, a stable beam mark is generated, and a stable laser beam is formed.

8. The laser processing method for the special shell according to claim 7, characterized in that The step S4 specifically includes the following steps: S41, loading stable laser beam parameters and initializing a processing path, performing spatiotemporal synchronization matching between the stable laser beam identifier and the spatial coordinates in the processing path file, and generating a dynamic processing instruction set including a laser power-moving speed coupling relationship; S42, predicting the deformation of the current irradiation area based on the thermal-mechanical coupling model, and adjusting the three-dimensional offset of the laser focus in real time according to the thermal expansion coefficient and residual stress distribution data of the shell material; S43. Collect the light intensity data of the welding molten pool through the annular photosensor, and obtain the real-time weld quality based on the monitored light intensity data. S44. Trigger the correction mechanism according to the obtained real-time weld quality. When it is detected that the fluctuation of the molten pool length exceeds the standard or the keyhole collapse frequency is greater than 3 times per second, automatically execute the following correction mechanism: Instantly increase the laser power to 120% of the set value and maintain it for 10 ms to eliminate the lack of penetration defect. Append an offset of 0.1 mm in the path normal direction to compensate for thermal deformation. Record the abnormal coordinates and mark them as key areas.

9. The laser processing method for the special shell according to claim 8, characterized in that, The specific steps of step S5 are as follows: S51. According to the records of the correction mechanism and the process parameters in the processing path file, extract the coordinate data of the stress concentration area, the sealing surface and the penetration depth fluctuation exceeding the standard, and generate a key area coordinate set and the corresponding energy density threshold. S52. Perform gradient energy laser strengthening treatment. Based on the energy density threshold, regulate the laser power and residence time, and perform impact strengthening on the sealing surface area with an annular path at 1.2 - 1.5 times the reference power density. S53. Integrate the characteristic data of the processing process. Collect the light intensity characteristic data of the strengthening area through the annular photosensor, synchronously extract the laser power, moving speed and deformation compensation parameters in the processing path file, and generate a characteristic data set including the temperature distribution in the strengthening area, the intensity ratio of the molten pool solidification characteristic spectrum line and the statistical distribution of deformation compensation. S54. According to the material - energy correlation mapping table and the deformation amount predicted by the thermal - mechanical coupling model, construct a material constitutive model. S54. Couple the characteristic data set and the material constitutive model, and use the optimized digital model to train the process parameters to output an updated parameter set. S55. Integrate and optimize the updated parameter set with the original processing path file, and output the next-generation path file after verifying the process stability.

10. A laser processing device for a special shell, characterized in that, A laser processing method for a special shell as described in any one of claims 1 to 9, wherein the laser processing device specifically includes: A laser generator, using a fiber-coupled continuous laser and equipped with an optical component for beam shaping. A laser adjustment component, carrying the laser generator, used to adjust the position and angle of the laser generator. A dust removal component, arranged opposite to the laser generator, and having a dust removal cavity inside. A light shielding component, arranged in the dust removal cavity, including an annular photosensor array and an electrochromic glass modulator. A control host, used to coordinate and control the operation of each component and provide corresponding control strategies.

Images