Large maritime work equipment welding double-beam laser strengthening device and strengthening method based on high-energy ray online monitoring

Abstract

The invention relates to a large maritime work equipment welding double-beam laser strengthening device and method based on high-energy ray online monitoring, and the device comprises a laser welding device, a double-beam laser shock strengthening device, an X-ray detection system, a workbench and an upper computer. The laser welding device is located on the upper right portion of the workbench, the double-beam laser shock device is located on the upper left portion of the workbench, and the X-ray detection system is located on the side face of the workbench. According to the device, the material can undergo stronger plastic deformation in the solidification process, so that the grain structure of a cladding layer is more effectively refined, the stress distribution in the material is better adjusted, the accumulation of residual stress is reduced, meanwhile, the coupling condition of shock waves in a melting area is monitored in real time in the welding process by virtue of a high-energy X-ray detection system, and the welding quality is improved. Waveforms are segmented by establishing a deep learning model, and a data set is established by using an image captured by X-rays to optimize process parameters.

Description

Technical Field

[0001] The present invention relates to a method and a device for optimizing process parameters by applying double laser shock to a molten area and detecting shock wave coupling using X-rays during laser additive manufacturing welding, belonging to the technical field of laser additive manufacturing. Background Art

[0002] my country's marine environment is complex, with frequent typhoons, high salt content and high humidity, which has caused the welds of large marine equipment (such as offshore wind turbines, offshore platforms, etc.) to be affected by factors such as seawater corrosion, wind and wave impact and temperature changes for a long time. The welds formed by traditional welding methods are prone to corrosion, fatigue crack initiation and expansion in such harsh environments, resulting in reduced weld strength and damage to the overall structural integrity of the equipment. For example, in the high salt fog environment, the welds of offshore wind turbines are prone to electrochemical reactions on the metal surface, causing corrosion pits, which then become stress concentration points, accelerate crack generation, and seriously threaten the safe operation and service life of the wind turbines.

[0003] During the manufacturing and use of large-scale marine equipment, the weld area is subjected to complex stress states, including welding residual stress, stress caused by working load, and thermal stress caused by environmental factors. Existing welding technology is difficult to effectively regulate the stress distribution inside the weld, and stress concentration is relatively common, which reduces the load-bearing capacity and fatigue resistance of the weld. In the offshore platform structure, due to the action of waves and wind, stress concentration at the weld may cause brittle fracture, resulting in serious safety accidents and economic losses.

[0004] Nowadays, damage to large offshore equipment mostly occurs in the weld area. Offshore wind turbines and offshore platforms have increasingly higher requirements for weld strength. Therefore, this paper proposes a method of using dual-beam laser shock strengthening to improve the weld strength of large offshore equipment.

[0005] The dual-beam laser shock peening process involves multiple key parameters, such as laser energy density, pulse width, spot diameter, pulse frequency, number of shocks, and the relative position and angle of the two laser beams. These parameters are coupled with each other and have a complex effect on the strengthening effect. At present, there is a lack of accurate theoretical models to comprehensively describe the relationship between the parameters and their quantitative connection with the evolution of the microstructure of the material and the improvement of the macroscopic performance, which makes it difficult to accurately determine the optimal parameter combination in the actual process, and the process stability and repeatability are poor. This technology uses an X-ray detection system to monitor the coupling of shock waves in the molten area, but the existing monitoring technology still has certain limitations in data acquisition speed, resolution and accuracy. At the same time, how to quickly and effectively feed back the information obtained from real-time monitoring to the control system to achieve dynamic adjustment of laser parameters is still a key problem that needs to be solved. During the laser shock peening process, changes inside the material occur instantly. If the monitoring feedback is not timely, the best time for process adjustment may be missed, affecting the consistency and reliability of the strengthening effect. Summary of the invention

[0006] The present invention is aimed at the technical problems existing in the prior art and provides a large-scale marine equipment welding dual-beam laser strengthening device and strengthening method based on high-energy ray online monitoring.

[0007] In order to achieve the above-mentioned purpose, the technical solution of the present invention is as follows: a dual-beam laser strengthening device for welding large-scale marine engineering equipment based on high-energy ray online monitoring, the device comprising a laser welding device, a dual-beam laser shock strengthening device, an X-ray detection system, a workbench and a host computer, the laser welding device is located at the upper right of the workbench, the dual-beam laser shock strengthening device is located at the upper left of the workbench, and the X-ray detection system is located on the side of the workbench.

[0008] The laser welding device includes a laser welding head 100 and a laser welding head power supply and power control module 101, wherein the laser welding head 100 is used to process materials, and the laser welding head power supply and power control module 101 is used to control power switches and parameters such as the power of laser processing. The laser welding device has a high power density after laser focusing, which can make the material absorb a large amount of energy in a very short time to reach a state of melting, evaporation or sintering, providing the necessary energy conditions for the accumulation and forming of the material; the welding speed is fast, the depth is large, and the deformation is small, and small holes can be generated in the metal material. The laser energy is transmitted to the deep part of the workpiece through the small holes to reduce lateral diffusion. Therefore, in one scanning process, the material fusion depth is large, the welding speed is fast, and the welding area per unit time is large; and welding can be performed at room temperature or under special conditions, and the welding equipment is simple.

[0009] The dual-beam laser shock hardening device includes a mechanical gripper, a mechanical gripper controller, a laser shock hardening head 1, a power supply and power control module of the laser shock hardening head 1, a laser shock hardening head 2, and a power supply and power control module of the laser shock hardening head 2, wherein the mechanical gripper includes a mechanical gripper controller, a left arm, a left gripper and a right arm, a right gripper, a support plate, a slider, and a base. The mechanical gripper controller can realize the up and down movement of the base and the opening and closing angles of the left and right arms to control the specific orientation and specific position of the dual-beam laser shock hardening. The support plate is located at the top for connecting the mechanical gripper and the slider, the bottom of the support plate is connected to the base, and the slider is located on the base and connected to the left and right arms. The dual-beam laser shock device can use the dual-beam laser to synergistically act on the molten area on the surface of the material, so that the surface of the material instantly heats up and forms a molten layer, generates a strong shock wave, and causes the material to produce plastic deformation, thereby refining the grains, improving the strength and hardness of the material, and improving the stress state inside the material; and can change the energy size and opening and closing angle of the two laser beams in real time according to different situations. An energy distribution model is established to achieve accurate energy distribution of dual-beam lasers on the surface or in different areas of the material to achieve the best strengthening effect, such as uniform grain refinement, ideal residual stress adjustment, and improvement of the overall strength and hardness of the material, improve energy utilization efficiency, and reduce energy waste. The input parameters are material characteristic parameters, laser parameters, workpiece geometry and size; the output parameters are the energy distribution ratio of the dual-beam laser in each area of ​​the material and the predicted strengthening effect index. The laser-material interaction model is divided into an energy absorption model and a heat conduction model. Regarding the energy absorption model, the Beer-Lambert law is used to calculate the material's absorption energy of the laser according to the optical properties of the material and the laser wavelength. The law is expressed as I(x) = I0exp(-αx), where I(x) is the intensity of the laser at a depth x in the material, I0 is the incident laser intensity, and α is the absorption coefficient of the material (related to the material type and laser wavelength). The total energy absorbed by the material within different depth ranges is calculated by integration, providing a basis for the subsequent calculation of the energy distribution inside the material. Regarding the heat conduction model, based on the heat conduction equation (where ρ is the material density, c is the specific heat capacity, T is the temperature, t is the time, k is the thermal conductivity, and q is the heat source term generated by the laser action), calculate the propagation and distribution of the heat generated by the laser shock in the material. The heat source term q is related to the laser absorption energy. The absorbed laser energy is converted into thermal power density and substituted into the heat conduction equation for solution. The heat conduction equation can be solved by numerical calculation methods such as the finite difference method and the finite element method to obtain the temperature distribution of the material at different times and positions. The temperature distribution reflects the accumulation and diffusion of the laser energy in the material and is an important manifestation of the energy distribution effect. Regarding the energy distribution algorithm, the surface or interior of the material is divided into multiple areas according to the geometry and strengthening requirements of the workpiece. For the weld area, stress concentration area or key parts with high strength requirements, a higher weight value can be set, indicating that these areas need to allocate more energy. In the central area of ​​the weld, the weight can be set to 1.5-2.0, while in the area far away from the weld, the weight can be set to 0.8-1.0. By establishing an energy distribution model, the energy distribution can be dynamically adjusted to adapt to the complex needs of materials in different stages and different areas.

[0010] Among them, the X-ray detection system is composed of an X-ray emitter 5, which is used to detect the coupling of two laser-enhanced shock waves inside the material, transmit the collected images to the host computer 6, and the host computer 6 displays the detected images, collects a large amount of X-ray image data, including images under different laser parameters and shock conditions, for training the deep learning model. And the image is annotated to mark the position and shape of the shock wave. The convolutional neural network (CNN) is selected as the deep learning model for the image segmentation task, which can accurately segment the contour of the target object. By adjusting the number of layers of the network, the size and number of filters, the activation function and other parameters, the model is properly adjusted and optimized. The annotated image data is divided into a training set, a validation set and a test set. The deep learning model is trained using the training set, and the parameters of the model are optimized so that it can accurately segment the waveform of the shock wave. During the training process, the stochastic gradient descent optimization algorithm can be used, and appropriate parameters such as the learning rate and batch size can be set. At the same time, data enhancement techniques such as rotation, flipping, scaling, etc. can be used to increase the diversity of data and improve the robustness of the model. Subsequently, the trained model is evaluated using the validation set, and the model's accuracy, recall, F1 value and other indicators are calculated to evaluate the model's performance. Based on the evaluation results, the model is further optimized and adjusted. The trained deep learning model is applied to actual X-ray image data to segment the shock wave waveform. The performance of the model can be finally verified using the test set and compared with the manually labeled results to evaluate the accuracy and reliability of the model.

[0011] The dual-beam laser shock strengthening method uses two laser beams to act synergistically on the molten area on the surface of the material, so that the surface of the material is instantly heated and a molten layer is formed. During the process, the two laser beams generate strong shock waves, which propagate into the material. The two shock waves interact with each other to produce an interference effect, and the interference effect can generate a complex stress field inside the material. This stress field promotes the generation, movement and interaction of dislocations. In the constructive interference area, high pressure increases the dislocation density in the material. The increase in dislocations can improve the strength and hardness of the material because dislocations can hinder the slip movement in the crystal. At the same time, the distribution of dislocations is also affected by the interference effect. Different interference patterns will cause the distribution of dislocations in the material to show a specific law, thereby affecting the mechanical properties of the material. The interference effect will also affect the formation and distribution of twins. Under high stress, the material may undergo twin deformation and form twins. The uneven stress distribution in the interference area will cause twins to form preferentially at specific locations. The presence of twins can change the crystal structure of the material and improve the toughness and plastic deformation ability of the material. By controlling the interference effect of the laser shock wave, the distribution and number of twins can be adjusted to optimize the comprehensive performance of the material. There are three different methods, namely, two lasers impacting at the same time; one laser shock wave is used as the driving force for dislocation movement, and when the first laser wave is reflected, the second laser shock wave is immediately used to strengthen the dislocation movement, and the interval between the two laser waves is 5 nanoseconds; one laser wave is used as the driving force for dislocation movement, and when it is reflected, two laser shock waves are used to strengthen the dislocation movement. Set a function H, H is a function of material thickness and laser head scanning speed, go to a point p (x, y, z) inside the material, two laser heads S1 and S2, amplitudes A1 and A2 respectively, initial phases φ1 and φ2 respectively, angular frequencies ω1 and ω2, and the vibration expression is:

[0012]

[0013] The vibration caused by laser head 1 and laser head 2 at point p is:

[0014]

[0015] The resultant vibration at point P is:

[0016]

[0017] Set a function D1, D1 is a function of dislocation density and interference ripple intensity, D1 = kI, I is the interference ripple intensity, k is a proportional coefficient, which is determined by the characteristics of the instrument and the basic properties of the experimental material itself. Set a function D2, D2 is a function of dislocation density and interference ripple spatial frequency, D2 = mf2, f is the spatial frequency of the interference ripple, m is a proportional coefficient, which comprehensively considers the effects of various factors other than dislocation density on the ripple interference spatial frequency. Combining the above two relationships, we can get the relationship between dislocation density and interference ripple:

[0018] D=D1+D2=kI+mf2.

[0019] A dual-beam laser strengthening method for welding large offshore equipment based on high-energy ray online monitoring, the method comprising the following steps:

[0020] Step 1: First, select 316 stainless steel as the experimental material and place the material on the workbench.

[0021] Step 2: Then, check whether all components of the device are working properly, including the stability of the working platform, the laser output stability of the laser welding device, whether the mechanical gripper used to control the dual-beam laser shock peening can be used normally, the detection sensitivity of the X-ray detection system, and the data processing capability of the controller.

[0022] Step 3: Set the processing parameters according to the experimental material 316 stainless steel, set the laser power to 200W, so that the material can be melted well, while avoiding excessive spatter and energy waste. Considering the thermal diffusion characteristics of stainless steel and the desire to obtain a better grain refinement effect, set the pulse frequency to 30Hz. Set the spot size to 0.3mm. Set the laser beam incident angle to an angle perpendicular or nearly perpendicular to the inclined surface to ensure that the laser energy can effectively act on the processed surface and improve the cladding quality. Set the scanning speed to 5mm / s, which can meet the processing accuracy requirements of the parts while ensuring the uniform thickness of the cladding layer.

[0023] Step 4: For the two laser beams used for laser shock strengthening, select the common Nd:YAG laser with a wavelength of 1064nm. This wavelength has a good absorption effect on the laser shock strengthening of 316 stainless steel and can effectively transfer laser energy to the material surface. The laser energy density is set to 3-5GW / cm 2. Energy density is one of the key parameters of laser shock strengthening. For 316 stainless steel, a sufficiently strong shock wave can be generated within this energy density range to cause plastic deformation on the surface of the material, thereby achieving a strengthening effect. An energy density that is too low may not achieve the expected strengthening effect, while an energy density that is too high may cause damage to the surface of the material. The pulse width is set to 10-20ns within this range, so that the laser energy can be concentratedly released in a short time, generating a high-amplitude shock wave, effectively driving the plastic deformation of the material surface. The spot diameter is set to 2-4mm. The size of the spot diameter affects the effective area and energy distribution of the laser shock. A smaller spot diameter can produce a higher energy density, but a smaller effective area; a larger spot diameter has a larger effective area, but a relatively lower energy density. Selecting a spot diameter of 2-4mm can cover a certain area of ​​the material surface while ensuring sufficient energy density to achieve a uniform strengthening effect. The pulse frequency is set to 5-10Hz. Set the number of shocks to 3-5 times. For smaller parts or areas with higher strengthening requirements, the number of shocks can be appropriately increased, but care should be taken to avoid excessive shocks that may cause material performance degradation.

[0024] Step 5: Before laser shock treatment, the workpiece is pre-deformed and then laser shock hardening is performed. First, the pre-deformation is determined to be within the strain range of 0.1%-0.5%. This range can appropriately increase the dislocation density inside the material, preparing for a better dislocation strengthening effect in subsequent laser shock hardening. Then, the 316 alloy workpiece is stretched using a tensile testing machine. Both ends of the workpiece are fixed on the fixture of the tensile testing machine, and a tensile force is applied at a rate of 0.3mm / min to cause the workpiece to elongate axially. After reaching the predetermined pre-deformation, the tension is maintained for 40 seconds to allow the internal microstructure of the material to be fully adjusted, and then the tension is unloaded.

[0025] Step 6: Then, start the laser welding laser head to perform layer-by-layer deposition molding on the 316 stainless steel specimen placed on the workbench. At the same time, the shock wave generated by the dual laser beams used for laser shock strengthening acts on the molten area, and the laser parameters are controlled to ensure that the molten area is kept at a suitable temperature and state so that the shock wave can be effectively transmitted to the inside of the material to cause plastic deformation of the material. In this process, the laser deposition part continuously provides molten material, which works in synergy with the laser shock to refine the grain structure of the cladding layer and reduce the generation of defects such as pores and cracks.

[0026] Step 7: During the laser deposition and shock process, the X-ray detection system is used to detect the coupling of shock waves in the molten area in real time. The X-ray emitter emits X-rays at a frequency of 10 frames per second to penetrate the material. The detector receives the X-ray intensity information after passing through the material and converts it into a digital image signal. These image data are transmitted to the host computer system in real time. The host computer analyzes the received image. First, the coupling of the two laser shock waves is observed. The synergy and consistency of the two laser shock effects are evaluated by the morphology, position and intensity distribution of the shock wave waveform in the image. For example, if the two laser shock waves show clear, regular and mutually superimposed and enhanced waveforms in the image, it means that the coupling effect is good; on the contrary, if the waveforms are chaotic, interfere with each other or the intensity is uneven, it may be necessary to adjust the angle, energy and other parameters of the laser shock strengthening head. Multiple experiments were conducted by changing the angles of the two laser shock strengthening heads. The above laser shock strengthening and X-ray detection processes were repeated in each experiment to obtain shock wave coupling image data at different angles. The collected interactive waveform data is deeply analyzed to extract the characteristic parameters of the waveform, including the exact position and amplitude of the peak and trough, and the duration of the waveform. These characteristic parameters can quantitatively describe key information such as the intensity, frequency and action time of the shock wave.

[0027] Step 8: Based on the image data obtained by X-ray detection, a deep learning model is established for image segmentation and data analysis. The collected interactive waveform data is analyzed to extract the characteristic parameters of the waveform, including the position and amplitude of the peak and trough, as well as the duration of the waveform. The characteristic parameters of the interactive waveform are correlated with the detection results of the strengthening effect to find the relationship between the two and establish a mathematical model. First, a large amount of X-ray image data is collected, including images under different laser parameters and shock conditions, and the images are annotated to mark the position and shape of the shock wave. Then, a convolutional neural network (CNN) is selected as the deep learning model for the image segmentation task. In order to train the model, the annotated image data needs to be divided into a training set, a validation set, and a test set. Next, the training set is used to train the deep learning model. During the training process, optimization algorithms such as stochastic gradient descent can be used, and appropriate parameters such as learning rate and batch size can be set. At the same time, data enhancement techniques such as rotation, flipping, scaling, etc. can be used to increase the diversity of the data and improve the robustness of the model. After that, the trained model is evaluated using the validation set, and the model's accuracy, recall rate, F1 value and other indicators are calculated to evaluate the performance of the model. According to the evaluation results, the model is further optimized and adjusted, and finally the trained deep learning model is applied to the actual X-ray image data to segment the shock wave waveform. The performance of the model can be finally verified using the test set and compared with the manually labeled results to evaluate the accuracy and reliability of the model. Through the above steps, a deep learning model that can accurately segment the shock wave waveform in the X-ray image can be trained. Then, the collected interactive waveform data is analyzed to extract the characteristic parameters of the waveform, such as the position and amplitude of the peak and trough, and the duration of the waveform. The characteristic parameters of these interactive waveforms are correlated with the test results of the strengthening effect (such as residual stress, dislocation density, etc.) to find the relationship between the two, so as to establish a mathematical model that can predict the strengthening effect based on the interactive waveform. Using the established model, the strengthening effect can be predicted based on the interactive waveform, and real-time feedback can be given to the laser generator control system to achieve dynamic adjustment of the laser parameters.

[0028] Finally, the study found that the laser shock peening effect is best when the two laser beams are at a 45-degree angle to the horizontal plane. The experimental effect measured by using the same parameter setting for the two laser beams is better than setting different parameters separately.

[0029] Compared with the prior art, the advantages of the present invention are as follows:

[0030] 1. Grain refinement and stress adjustment: During the laser deposition process of the material, the dual-beam laser shock peening device causes the surface of the material to instantly heat up to form a molten layer, generating a strong shock wave and causing the material to undergo plastic deformation. According to dislocation theory, plastic deformation increases dislocation density, and dislocation movement and interaction hinder crystal slip movement, thereby refining the grains and improving material strength and hardness. For example, experimental observations have shown that the grain size of 316 stainless steel specimens treated with dual-beam laser shock peening is significantly smaller than that of untreated specimens, and the material hardness is increased by about 18%. This process can also improve the internal stress state of the material and reduce residual stress accumulation. The shock wave generated by the laser shock propagates inside the material, redistributes the internal stress of the material, reduces the residual stress peak, avoids crack initiation and expansion caused by stress concentration, and improves the material's fatigue resistance and stability.

[0031] 2. Advantages of laser welding equipment: It can focus the laser beam to a very small area, so that the material absorbs a large amount of energy in a very short time, reaching the state of melting, evaporation or sintering, and providing sufficient energy for material accumulation and molding. For example, in laser additive manufacturing of metal materials, metal powder or wire can be melted quickly to achieve high-precision molding, and the molding accuracy can reach ±3mm. In addition, the welding conditions are flexible, and welding can be performed at room temperature or under special conditions. The equipment is relatively simple and does not require complex preheating or post-processing procedures, which reduces production costs and process complexity. It is suitable for on-site welding and repair operations of various large marine equipment.

[0032] 3. Characteristics of the dual-beam laser shock peening device: The dual-beam lasers act synergistically on the molten area to generate stronger shock waves, further enhancing the plastic deformation effect of the material and making the grain refinement effect more significant. By comparing the single-beam laser shock peening experiment, the grain refinement of the material after dual-beam laser shock peening increased by 20%-50%. And the parameters can be adjusted in real time. The mechanical gripper and controller can change the energy size and opening and closing angle of the two laser beams in real time to adapt to different materials and process requirements and optimize the strengthening effect. For example, for 316 stainless steel plates of different thicknesses, the laser parameters can be adjusted to ensure the formation of a uniform strengthening area inside the material and improve the overall performance of the material.

[0033] 4. X-ray detection system and process optimization: It can detect the coupling situation in real time, detect the coupling situation of shock waves in the molten area in real time at a frequency of 10 frames per second, and obtain the interaction information of the two laser shock waves in time to provide a basis for adjusting the process parameters. Subsequently, a large amount of X-ray image data is used to train the deep learning model, accurately segment the shock wave waveform, extract characteristic parameters, and analyze the correlation with the strengthening effect to establish a mathematical model. After comparing the model prediction with the actual detection, the prediction error of the dislocation density is within ±5%, which realizes the effective prediction of the strengthening effect and the optimization of the laser parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a diagram of the structure of a device that uses dual laser shock peening and X-rays to detect coupling conditions.

[0035] Figure 2 This is the coupling of the shock waves of two laser beams detected by high-energy rays.

[0036] Figure 3 The flowchart of the whole process is as follows:

[0037] Figure 4 This is a top view of the mechanical gripper.

[0038] Figure 5 A side view of the mechanical gripper.

[0039] The meanings of the grid numbers in the figure are as follows: 100, laser welding head; 101, power supply and power control module of laser welding head; 200, mechanical gripper; 201, mechanical gripper controller; 202, left arm; 203, left gripper; 204, right arm; 205, right gripper; 206, support plate; 207, slider; 208, base; 300, laser shock peening head one; 301, power supply and power control module of laser shock peening head; 302, laser shock peening head two; 303, power supply and power control module of laser shock peening head two; 4, workbench; 5, X-ray emitter; 6, host computer; 7, molten pool; 8, experimental sample. DETAILED DESCRIPTION

[0040] In order to deepen the understanding of the present invention, the present embodiment is described in detail below with reference to the accompanying drawings.

[0041] Example 1: See Figure 1-Figure 5 , a large-scale marine equipment welding dual-beam laser strengthening device based on high-energy ray online monitoring, consists of a laser welding device, a dual-beam laser shock strengthening device, an X-ray detection system, a workbench and a host computer. Figure 1As shown, the whole device is mainly composed of a laser welding device, a dual-beam laser shock peening device, an X-ray detection system, a workbench 4 and a host computer 6. The laser welding device is located at the upper right of the workbench 4, the dual-beam laser shock peening device is located at the upper left of the workbench 4, the X-ray detection system is located on the side of the workbench 4, and the host computer 6 is located on the left side of the whole device. Among them, the laser head of the laser welding part focuses the laser beam from the laser on a very small area, so that the material in this area can absorb a large amount of energy in a very short time, thereby achieving the state of melting, evaporation or sintering of the material, and providing the necessary energy conditions for the accumulation and molding of the material. For example, in the laser additive manufacturing of metal materials, the high-energy laser output by the laser head can quickly melt and solidify metal powder or metal wire to form the required part shape. The dual-beam laser shock peening part is used to generate high-amplitude shock waves on the surface of the material, causing the material to undergo plastic deformation, thereby refining the grains, improving the strength and hardness of the material, and improving the stress state inside the material. The X-ray detection system can penetrate the material and obtain the structural information inside the material, and is used to detect the coupling of the two laser shock waves. The workbench is used to carry objects, and the host computer is used to receive data.

[0042] The laser welding device is composed of a laser welding head 100 and a laser welding head power supply and power control module 101. The laser welding head 100 is used to process materials, and its function is to focus the laser beam on the surface of the material so that the material absorbs energy to achieve a state of melting, evaporation or sintering, thereby achieving the accumulation and forming of the material. The laser welding head power supply and power control module 101 is used to control the power switch of the laser welding head 100 and the power of the laser processing and other parameters to ensure the stability and accuracy of the laser output.

[0043] The dual-beam laser shock peening device comprises a mechanical gripper 200, a mechanical gripper controller 201, a laser shock peening head 1 300, a power supply and power control module 301 of the laser shock peening head 1, a laser shock peening head 2 302, and a power supply and power control module 303 of the laser shock peening head 2, wherein the mechanical gripper 200 comprises a mechanical gripper controller 201, a left arm 202, a left gripper 203 and a right arm 204, a right gripper 205, a support plate 206, a slider 207, and a base 208. The mechanical gripper controller 201 can realize the up and down movement of the base 208 and the opening and closing angle adjustment of the left arm 202 and the right arm 204, so as to accurately control the specific orientation and position of the dual-beam laser shock peening. The support plate 206 is located at the top, connecting the mechanical gripper 200 and the slider 207, and its bottom is connected to the base 208. The slider 207 is located on the base 208 and connected to the left arm 202 and the right arm 204 to ensure the stability of the mechanical gripper 200 during movement. The laser shock peening head 1 300 and the laser shock peening head 2 302 provide energy through their respective power supply and power control modules 301 and 303, respectively, to generate high-amplitude shock waves to act on the molten area on the surface of the material.

[0044] The X-ray detection system is composed of an X-ray emitter 5, whose main function is to detect the coupling of two laser-enhanced shock waves inside the material, to obtain the internal structure information of the material by emitting X-rays to penetrate the material, and to transmit the collected images to the host computer 6. The host computer 6 is responsible for displaying the detected images for the operator to analyze and process.

[0045] Example 2: See Figure 3 , using the above-mentioned dual laser shock strengthening and X-ray coupling detection device to conduct experiments, including the following steps:

[0046] Step 1: First, select 316 stainless steel as the experimental material and place the material on workbench 4.

[0047] Step 2: Then, check whether all parts of the device are working normally, including the stability of the work platform, to ensure that the workbench will not shake or move during the welding and impact strengthening process, affecting the processing accuracy; check the laser output stability of the laser welding device to ensure constant laser energy output to avoid unstable welding quality due to energy fluctuations; confirm whether the mechanical clamp used to control the dual-beam laser shock strengthening can be used normally, including whether the opening and closing action of the clamp and the position adjustment function are accurate and reliable; test the detection sensitivity of the X-ray detection system and the data processing capability of the controller to ensure that the image data of the shock wave coupling situation can be accurately acquired and processed.

[0048] Step 3: Set the processing parameters according to the experimental material 316 stainless steel, set the laser power to 200W, so that the material can be melted well, while avoiding excessive spatter and energy waste. Considering the thermal diffusion characteristics of stainless steel and the desire to obtain a better grain refinement effect, set the pulse frequency to 30Hz. Set the spot size to 0.3mm. Set the laser beam incident angle to an angle perpendicular or nearly perpendicular to the inclined surface to ensure that the laser energy can effectively act on the processed surface and improve the cladding quality. Set the scanning speed to 5mm / s, which can meet the processing accuracy requirements of the parts while ensuring the uniform thickness of the cladding layer.

[0049] Step 4: For the two laser beams used for laser shock strengthening, select the common Nd:YAG laser with a wavelength of 1064nm. This wavelength has a good absorption effect on the laser shock strengthening of 316 stainless steel and can effectively transfer laser energy to the material surface. The laser energy density is set to 3-5GW / cm 2 . Energy density is one of the key parameters of laser shock strengthening. For 316 stainless steel, a sufficiently strong shock wave can be generated within this energy density range to cause plastic deformation on the surface of the material, thereby achieving a strengthening effect. An energy density that is too low may not achieve the expected strengthening effect, while an energy density that is too high may cause damage to the surface of the material. The pulse width is set to 10-20ns within this range, so that the laser energy can be concentratedly released in a short time, generating a high-amplitude shock wave, effectively driving the plastic deformation of the material surface. The spot diameter is set to 2-4mm. The size of the spot diameter affects the effective area and energy distribution of the laser shock. A smaller spot diameter can produce a higher energy density, but a smaller effective area; a larger spot diameter has a larger effective area, but a relatively lower energy density. Selecting a spot diameter of 2-4mm can cover a certain area of ​​the material surface while ensuring sufficient energy density to achieve a uniform strengthening effect. The pulse frequency is set to 5-10Hz. Set the number of shocks to 3-5 times. For smaller parts or areas with higher strengthening requirements, the number of shocks can be appropriately increased, but care should be taken to avoid excessive shocks that may cause material performance degradation.

[0050] Step 5: Then, start the laser welding laser head to perform layer-by-layer deposition molding on the 316 stainless steel specimen placed on the workbench. In this process, the laser welding device focuses the laser beam on the surface of the material, and melts and solidifies the material point by point and layer by layer according to the preset scanning path and parameters to form the desired shape and structure. At the same time, the shock wave generated by the dual laser beams used for laser shock strengthening acts precisely on the molten area, and the laser parameters (such as energy density, pulse frequency, etc.) are controlled to ensure that the molten area remains at a suitable temperature and state. The appropriate temperature range can ensure that the material has good fluidity and cladding properties, and is conducive to the propagation and interaction of shock waves inside the material, causing effective plastic deformation of the material, refining the grain structure of the cladding layer, and reducing the generation of defects such as pores and cracks.

[0051] Step 6: During the laser deposition and shock process, the X-ray detection system is used to detect the coupling of shock waves in the molten area in real time. The X-ray detection system acquires detection data at a frequency of 10 frames per second and transmits the data to the host computer system in real time. Finally, the image received by the host computer is analyzed to observe the coupling of the two laser shock waves. Multiple experiments are performed by changing the angles of the two laser shock strengthening heads. The collected interactive waveform data is analyzed to extract the characteristic parameters of the waveform, such as the position and amplitude of the peak and trough, and the duration of the waveform. The characteristic parameters of the interactive waveform are correlated with the detection results of the strengthening effect (residual stress, dislocation density, etc.) to find the relationship between the two and establish a mathematical model. First, a large amount of X-ray image data is collected, including images under different laser parameters and shock conditions, and the images are annotated to mark the position and shape of the shock wave. Then a convolutional neural network (CNN) is selected as the deep learning model for the image segmentation task. In order to train the model, the annotated image data needs to be divided into a training set, a validation set, and a test set. Next, the deep learning model is trained using the training set. During the training process, optimization algorithms such as stochastic gradient descent can be used, and appropriate parameters such as learning rate and batch size can be set. At the same time, data enhancement techniques such as rotation, flipping, and scaling can be used to increase the diversity of data and improve the robustness of the model. After that, the trained model is evaluated using the validation set, and the model's accuracy, recall rate, F1 value and other indicators are calculated to evaluate the performance of the model. According to the evaluation results, the model is further optimized and adjusted. Finally, the trained deep learning model is applied to the actual X-ray image data to segment the shock wave waveform. The performance of the model can be finally verified using the test set and compared with the manually labeled results to evaluate the accuracy and reliability of the model. Through the above steps, a deep learning model that can accurately segment the shock wave waveform in the X-ray image can be trained. Then, the collected interactive waveform data is analyzed to extract the characteristic parameters of the waveform, such as the position and amplitude of the peak and trough, and the duration of the waveform. The characteristic parameters of these interactive waveforms are correlated with the test results of the strengthening effect (such as residual stress, dislocation density, etc.) to find the relationship between the two, so as to establish a mathematical model that can predict the strengthening effect based on the interactive waveform. Using the established model, the strengthening effect can be predicted according to the interaction waveform, or the interaction waveform can be optimized by adjusting the laser parameters to obtain a better strengthening effect.

[0052] Finally, the study found that the laser shock peening effect is best when the two laser beams are at a 45-degree angle to the horizontal plane. The experimental effect measured by using the same parameter setting for the two laser beams is better than setting different parameters separately.

[0053] It should be noted that the above embodiments are not intended to limit the protection scope of the present invention, and equivalent changes or substitutions made on the basis of the above technical solutions all fall within the protection scope of the claims of the present invention.

Claims

1. A dual-beam laser strengthening device for welding large offshore equipment based on high-energy ray online monitoring, characterized in that: The device includes a laser welding device, a dual-beam laser shock peening device, an X-ray detection system, a workbench and a host computer. The laser welding device is located at the upper right of the workbench, the dual-beam laser shock peening device is located at the upper left of the workbench, the X-ray detection system is located on the side of the workbench to capture images of the inside of the workpiece, and the host computer is located on the left side of the overall device.

2. According to claim 1, the large-scale marine equipment welding dual-beam laser strengthening device based on high-energy ray online monitoring is characterized in that: The laser welding device comprises a laser welding head, a laser welding head power supply and a power control module, wherein the laser welding head is used for processing materials, and the laser welding head power supply and the power control module are used for controlling the power switch and the power parameters of the laser processing.

3. The large-scale marine equipment welding dual-beam laser strengthening device based on high-energy ray online monitoring according to claim 1 is characterized in that The dual-beam laser shock strengthening device includes a mechanical clamp, a mechanical clamp controller, a laser shock strengthening head one, a power supply and power control module of the laser shock strengthening head one, a laser shock strengthening head two, and a power supply and power control module of the laser shock strengthening head two. The mechanical clamp includes a mechanical clamp controller, a left arm, a left clamp, a right arm, a right clamp, a support plate, a slider, and a base. The mechanical clamp controller realizes the up and down movement of the base and the opening and closing angles of the left arm and the right arm to control the specific orientation and specific position of the dual-beam laser shock strengthening. The support plate is located at the top for connecting the mechanical clamp and the slider. The bottom of the support plate is connected to the base. The slider is located on the base and connected to the left arm and the right arm.

4. The large-scale marine equipment welding dual-beam laser strengthening device based on high-energy ray online monitoring according to claim 1 is characterized in that ,The X-ray detection system consists of an X-ray emitter to detect the coupling of two ,laser-enhanced shock waves inside the material, and transmits the ,collected images to the host computer, which displays the detected images.

5. A dual-beam laser strengthening method for welding large offshore equipment based on high-energy ray online monitoring, characterized in that , the method comprises the following steps: Step 1: First, select 316 stainless steel as the experimental material and place the material on the workbench. Step 2: Then, check whether the various components of the device are working properly, including the stability of the working platform, the laser output stability of the laser welding device, whether the mechanical gripper used to control the dual-beam laser shock peening can be used normally, the detection sensitivity of the X-ray detection system, and the data processing capability of the controller. Step 3: Set the processing parameters according to the experimental material 316 stainless steel, set the laser power to 200W to melt the material well, set the pulse frequency to 30Hz, set the spot size to 0.3mm, set the laser beam incident angle to be perpendicular or nearly perpendicular to the inclined surface to ensure that the laser energy can effectively act on the processing surface and improve the cladding quality, and set the scanning speed to 5mm / s. Step 4: For the two laser beams used for laser shock peening, select the common Nd:YAG laser with a wavelength of 1064nm and a laser energy density of 3-5GW / cm 2 , the pulse width is set to 10-20ns. Within this range, the laser energy is concentratedly released in a short time to generate a high-amplitude shock wave. The spot diameter is set to 2-4mm, the pulse frequency is set to 5-10Hz, and the number of shocks is set to 3-5 times; Step 5: Before laser shock treatment, the workpiece is pre-deformed and then laser shock hardening is performed. First, the pre-deformation is determined to be within the strain range of 0.1%-0.5%. This range can appropriately increase the dislocation density inside the material, preparing for a better dislocation strengthening effect of subsequent laser shock hardening. Then, the 316 alloy workpiece is stretched using a tensile testing machine. Both ends of the workpiece are fixed on the fixture of the tensile testing machine, and a tensile force is applied at a rate of 0.3mm / min to cause the workpiece to elongate axially. After reaching the predetermined pre-deformation, the tension is maintained for 40 seconds to allow the internal microstructure of the material to be fully adjusted, and then the tension is unloaded. Step 6: Then, start the laser welding laser head to perform layer-by-layer deposition molding on the 316 stainless steel specimen placed on the workbench. In this process, the laser welding device focuses the laser beam on the material surface, and melts and solidifies the material point by point and layer by layer according to the preset scanning path and parameters to form the required shape and structure. At the same time, the shock wave generated by the dual laser beams used for laser shock strengthening acts precisely on the molten area, and the laser parameters are controlled to ensure that the molten area is kept at a suitable temperature and state. Step 7: During the laser deposition and shock process, the X-ray detection system is used to detect the coupling of shock waves in the molten area in real time. The X-ray emitter emits X-rays at a frequency of 10 frames per second to penetrate the material. The detector receives the X-ray intensity information after passing through the material and converts it into a digital image signal. These image data are transmitted to the host computer system in real time. The host computer analyzes the received images and first observes the coupling of the two laser shock waves. The synergy and consistency of the two laser shock effects are evaluated through the morphology, position and intensity distribution of the shock wave waveform in the image. Step 8: Based on the image data obtained by X-ray detection, a deep learning model is established for image segmentation and data analysis. The collected interactive waveform data is analyzed to extract the characteristic parameters of the waveform, including the position and amplitude of the peak and trough, as well as the duration of the waveform. The characteristic parameters of the interactive waveform are correlated with the detection results of the strengthening effect to find the relationship between the two and establish a mathematical model. First, a large amount of X-ray image data is collected, including images under different laser parameters and shock conditions, and the images are annotated to mark the position and shape of the shock wave. Then, a convolutional neural network (CNN) is selected as the deep learning model for the image segmentation task. Deep learning model. In order to train the model, the labeled image data needs to be divided into training set, validation set and test set. Next, the training set is used to train the deep learning model. After that, the validation set is used to evaluate the trained model. The accuracy, recall rate, F1 value and other indicators of the model are calculated to evaluate the performance of the model. According to the evaluation results, the model is further optimized and adjusted. Finally, the trained deep learning model is applied to the actual X-ray image data to segment the waveform of the shock wave. The test set is used to finally verify the performance of the model and compare it with the results of manual annotation to evaluate the accuracy and reliability of the model.

Images