Method and system for removing steel wire rod oxide scale on line based on laser technology

Abstract

The invention relates to the technical field of laser processing, and discloses a method and system for online removal of steel wire rod oxide scale based on the laser technology, and the method comprises the following steps: establishing a relationship between a release stress and an adhesive force, and determining a relationship between different process parameters and oxide scale removal effects; initial laser process parameters are determined by combining the current material attributes of the wire rod and the performance of the oxide skin; the initial laser process parameters are dynamically adjusted according to the thickness distribution condition of the oxidation layer on the surface of the wire rod; performing laser treatment on the oxide scale on the surface of the wire rod by using a rotary laser head; and detecting the removal effect of the scale on the surface of the wire rod by using an image recognition technology, and dynamically optimizing model parameters according to a detection result. The device can remove scale and stains on the surface of the wire rod in the continuous production process at a time, and the problems that in the prior art, the scale on the surface of the wire rod cannot be completely removed, and the scale on the surface of a workpiece cannot be efficiently treated online are solved.

Description

Technical Field

[0001] The present invention relates to the field of laser processing technology, and in particular to a method and system for online removal of wire rod iron oxide scale based on laser technology. Background Art

[0002] In the metal surface deep processing production process of the metallurgical industry, surface cleaning technology has always been the key link to improve product quality and production efficiency. Traditional surface cleaning methods mainly include pickling process and mechanical stripping process, but both methods have significant disadvantages and limitations, especially in terms of environmental protection, cost control and process efficiency.

[0003] The pickling process is the most common method for cleaning metal surfaces. The workpiece is immersed in a pickling tank and cleaned by chemical reaction between the acid and the oxide layer or oil on the metal surface. Although pickling can effectively remove the oxide layer on the metal surface, there are a lot of environmental pollution risks in the process. The replacement of acid, the recycling of acid and the treatment of sewage all require complex supporting facilities and high investment costs. At the same time, the acid emissions and pollutants generated during the pickling process are difficult to handle, and it is impossible to achieve completely zero emissions, which violates the increasingly stringent environmental protection requirements. What's more serious is that the acid is highly toxic, and long-term exposure may lead to occupational diseases and pose a threat to the personal safety of operators.

[0004] The mechanical stripping process mainly removes the oxide layer on the surface of the workpiece by machining. Although this method can remove surface impurities more thoroughly, it causes a large amount of metal loss and reduces the yield rate. Moreover, machining fluids are inevitably used during machining, which not only causes environmental pollution but also increases production costs. At the same time, mechanical stripping is difficult to process for precision workpieces and is difficult to adapt to high-precision and high-demand surface treatment.

[0005] As a relatively advanced surface cleaning method, laser cleaning technology has gradually been used to remove oxides from metal surfaces due to its high energy density, cleanliness and high efficiency. Laser cleaning technology irradiates the metal surface with a high-energy laser beam, instantly evaporating or exciting the surface oxides or contaminants to achieve the purpose of cleaning. Compared with pickling and mechanical stripping, laser cleaning has the advantages of low energy consumption, no chemical pollution, and no need to contact the workpiece surface. However, the current laser cleaning technology still faces certain limitations in practical applications. In particular, when multiple cleanings are required to completely remove the iron oxide scale, repeated cleaning will lead to inefficiency and increase production time and cost. In addition, how to efficiently remove the iron oxide scale on the surface of the workpiece during continuous deep processing production is still a technical problem that needs to be solved urgently.

[0006] Therefore, the present invention proposes a method and system for online removal of wire rod iron oxide scale based on laser technology. Summary of the Invention

[0007] In view of the problems in the related art, the present invention provides a method and a system for online removing scale from wire rods based on laser technology to overcome the above-mentioned technical problems existing in the existing related art.

[0008] To this end, the specific technical solutions adopted by the present invention are as follows:

[0009] According to one aspect of the present invention, there is provided a method for online removing scale from wire rods based on laser technology, including the following steps:

[0010] S1. According to the temperature distribution and displacement distribution caused by laser heating, establish the relationship between the detachment stress and the adhesion force, determine the relationship between different process parameters and the scale removal effect, and combine the current wire rod material properties and scale properties to determine the initial laser process parameters;

[0011] S2. Use the oxide layer thickness identification model combined with terahertz technology to identify the thickness distribution of the oxide layer on the surface of the wire rod, and dynamically adjust the initial laser process parameters according to the thickness distribution of the oxide layer on the surface of the wire rod to obtain the optimal laser process parameters based on the thickness distribution;

[0012] S3. Based on the optimal laser process parameters, use a rotating laser head to perform laser treatment on the scale on the surface of the wire rod, and collect metal powder through negative pressure technology; use image recognition technology to detect the scale removal effect on the surface of the wire rod, and dynamically optimize the model parameters according to the detection results.

[0013] Further, the step of establishing the relationship between the detachment stress and the adhesion force according to the temperature distribution and displacement distribution caused by laser heating, determining the relationship between different process parameters and the scale removal effect, and combining the current wire rod material properties and scale properties to determine the initial laser process parameters includes the following steps:

[0014] S11. According to the laser power, material properties and heat conduction characteristics, establish a temperature distribution model caused by laser heating; based on the temperature gradient caused by laser heating and the material thermal expansion characteristics, establish a displacement and stress distribution model caused by laser heating;

[0015] S12. Determine the detachment stress between the scale and the substrate according to the thermal stress, calculate the adhesion force between the scale and the substrate, and determine the critical value between the detachment stress and the adhesion force according to the calculation results of the adhesion force and the detachment stress;

[0016] S13. Use the finite element analysis method to determine the relationship between the laser parameters, material parameters and the scale cleaning and removal effect, and combine the current wire rod material properties and scale properties to determine the initial laser process parameters.

[0017] Furthermore, the expression of the temperature distribution model caused by laser heating is as follows:

[0018]

[0019] The expression of the displacement distribution model caused by laser is as follows:

[0020]

[0021]

[0022] In the formula, T represents temperature, t represents time, represents the Laplace operator of temperature, represents the thermal diffusivity of the material, Q(r) represents the distribution of laser power at position r, ρ is the density of the material, c represents the specific heat capacity of the material, represents thermal stress, represents the temperature change, E represents the elastic modulus, represents the thermal expansion coefficient of the material, u(x, y, z) represents the distribution of displacement in space, and L represents the length of the path.

[0023] Furthermore, the calculation formula for the adhesion force between the scale and the substrate is as follows:

[0024]

[0025] In the formula, F represents the adhesion force between the scale and the substrate, represents the adhesion strength between the scale and the substrate, and A represents the contact area between the scale and the substrate.

[0026] Furthermore, the steps to determine the relationship between laser parameters, material parameters and the cleaning and removal effect of the scale by using the finite element analysis method, and to determine the initial laser process parameters in combination with the current properties of the wire rod material and the scale performance are as follows:

[0027] S131. Based on the temperature distribution model, displacement and stress distribution models, combined with the critical value between the detachment stress and the adhesion force, simulate the influence of different laser process parameters on the scale removal effect, and determine the basic laser process parameters according to the simulation results;

[0028] S132. Obtain the property data of the wire rod material to be processed, the property data of the scale and the laser parameter data, optimize the process conditions of the laser power and scanning speed in the basic laser process parameters through finite element simulation, and determine the initial laser process parameters according to the optimization results.

[0029] Further, the method for identifying the thickness distribution of the oxide layer on the surface of the wire rod by using the oxide layer thickness identification model in combination with terahertz technology, and dynamically adjusting the initial laser process parameters according to the thickness distribution of the oxide layer on the surface of the wire rod to obtain the optimal laser process parameters based on the thickness distribution includes the following steps:

[0030] S21. Based on the pre-constructed oxide layer thickness identification model, identify the thickness distribution of the oxide layer on the surface of the wire rod to be processed, and obtain the first distribution result of the oxide layer thickness;

[0031] S22. Use terahertz technology to identify the thickness distribution of the oxide layer on the surface of the wire rod to be processed, and obtain the second distribution result of the oxide layer thickness;

[0032] S23. Based on the first distribution result and the second distribution result of the oxide layer thickness, determine the thickness distribution of the oxide layer on the surface of the wire rod to be processed;

[0033] S24. According to the thickness distribution of the oxide layer on the surface of the wire rod to be processed, dynamically adjust the initial laser process parameters at different positions on the surface of the wire rod to obtain the optimal laser process parameters based on the oxide layer thickness.

[0034] Further, the step of identifying the thickness distribution of the oxide layer on the surface of the wire rod to be processed based on the pre-constructed oxide layer thickness identification model and obtaining the first distribution result of the oxide layer thickness includes the following steps:

[0035] S211. Obtain the surface image of the wire rod to be processed and perform preprocessing, analyze the brightness difference in the surface image, extract the gray values of different regions on the surface of the oxide layer, calculate the local brightness distribution in the surface image, obtain the average brightness of the local area by means of a sliding window, and statistically analyze the global brightness distribution to generate a gray histogram;

[0036] S212. Use the local binary pattern method to extract the texture features of the surface image, and identify the boundary between the oxide layer and the substrate by means of edge detection; based on the pre-constructed oxide layer thickness identification model, output the oxide layer thickness corresponding to the texture features of the surface image, and obtain the first distribution result of the oxide layer thickness at different positions on the surface of the wire rod.

[0037] Further, the step of using terahertz technology to identify the thickness distribution of the oxide layer on the surface of the wire rod to be processed and obtaining the second distribution result of the oxide layer thickness includes the following steps:

[0038] S221. Use ultrashort pulse terahertz waves to irradiate the surface of the wire rod to be processed, measure the reflection, transmission or absorption of the terahertz waves by the oxide layer, and record the amplitude, phase and time delay information of the terahertz waves;

[0039] S222. Calculate the reflection and transmission time differences of terahertz waves at different oxide layer thicknesses through time-domain analysis. Use Fourier transform to convert the time-domain signals into frequency-domain signals, extract the spectral characteristics at different frequencies, and calculate the absorption and reflectivity of the oxide layer for terahertz waves at specific frequencies to determine its thickness information;

[0040] S223. Based on the transmission or reflection time difference of terahertz waves, calculate the thickness of the oxide layer in combination with the refractive index of the oxide layer, and use a mathematical model to invert the thickness distribution of the oxide layer to obtain the second distribution result of the oxide layer thickness at different positions on the surface of the wire rod.

[0041] Further, determining the thickness distribution of the oxide layer on the surface of the wire rod to be processed based on the first distribution result and the second distribution result of the oxide layer thickness includes the following steps:

[0042] S231. Obtain the first distribution result and the second distribution result of the oxide layer thickness at different positions on the surface of the wire rod, and perform data alignment and preprocessing; based on the grid division method on the surface of the wire rod, establish the spatial correspondence relationship between the first distribution result data and the second distribution result data, and use the feature matching method to spatially align the thickness distribution data of the first distribution result and the second distribution result;

[0043] S232. Use the weighted average method in combination with a preset weight coefficient to fuse the first distribution result data and the second distribution result data to obtain the final thickness value, and generate a distribution map of the oxide layer thickness at different positions on the surface of the wire rod according to the final thickness value.

[0044] According to another aspect of the present invention, there is provided a system for online removing scale from wire rods based on laser technology, including an initial laser process parameter determination module, a laser process parameter optimization module, and a scale online removing module;

[0045] Among them, the initial laser process parameter determination module is used to establish the relationship between the detachment stress and the adhesion force according to the temperature distribution and displacement distribution caused by laser heating, determine the relationship between different process parameters and the scale removal effect, and combine the current wire rod material properties and scale performance to determine the initial laser process parameters;

[0046] The laser process parameter optimization module is used to identify the thickness distribution of the oxide layer on the surface of the wire rod by using the oxide layer thickness identification model in combination with terahertz technology, and dynamically adjust the initial laser process parameters according to the thickness distribution of the oxide layer on the surface of the wire rod to obtain the optimal laser process parameters based on the thickness distribution;

[0047] The online scale removal module is used to perform laser treatment on the scale on the surface of the wire rod using a rotating laser head based on the optimal laser process parameters, and collect metal powder through negative pressure technology; detect the removal effect of the scale on the surface of the wire rod using image recognition technology, and dynamically optimize the model parameters according to the detection results.

[0048] The beneficial effects of the present invention are as follows:

[0049] 1) Through the high energy density and precise control of the laser, the present invention can quickly heat the scale to the falling-off temperature, achieve rapid removal of the scale, and can remove the scale and stains on the surface of the wire rod in the continuous production process at one time, solving the pain points in the prior art that the scale on the surface of the wire rod cannot be completely removed and the scale on the surface of the workpiece cannot be efficiently processed online. Compared with traditional methods, the laser removal process does not require chemical substances or a large amount of mechanical force, has a high removal efficiency, and will not cause damage to the base material.

[0050] 2) By precisely adjusting the laser power and scanning speed, the present invention can accurately process scales with different thicknesses and properties. By dynamically adjusting the process parameters, it can ensure that the scale removal process always maintains the best state, avoiding overheating or damaging the surface of the wire rod.

[0051] 3) By integrating an advanced oxide layer thickness identification model and terahertz technology, the present invention can monitor the thickness distribution of the oxide layer on the surface of the wire rod in real time, and dynamically adjust the laser process parameters according to the actual situation. This real-time feedback mechanism ensures that the scale removal effect in the laser treatment process always remains at the best level.

[0052] 4) By combining with the properties of the wire rod material, the properties of the scale, and laser technology, the present invention can adapt to wire rod materials of different types and sizes. Whether it is a thin oxide layer or a thick oxide layer, ideal removal effects can be obtained by flexibly adjusting the process parameters.

[0053] 5) By combining image recognition technology and negative pressure collection technology, the present invention can detect the removal effect in real time and perform dynamic optimization, ensuring the thoroughness and uniformity of scale removal, and further improving the automation and intelligence level of the process. Description of the Drawings

[0054] 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 to be used in the embodiments. Obviously, the drawings in the following description 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.

[0055] Figure 1It is a flowchart of a method for online removal of scale from wire rods based on laser technology according to an embodiment of the present invention. Detailed implementation manners

[0056] To further illustrate each embodiment, the present invention provides accompanying drawings, which are a part of the disclosure of the present invention. They are mainly used to illustrate the embodiments and can be used in conjunction with the relevant descriptions in the specification to explain the operating principles of the embodiments. With reference to these contents, those of ordinary skill in the art should be able to understand other possible implementation manners and the advantages of the present invention. The components in the figures are not drawn to scale, and similar component symbols are usually used to represent similar components.

[0057] According to an embodiment of the present invention, there are provided a method and a system for online removal of scale from wire rods based on laser technology.

[0058] Now, the present invention will be further described in conjunction with the accompanying drawings and the detailed implementation manners. As Figure 1 shown, according to an embodiment of the present invention, there is provided a method for online removal of scale from wire rods based on laser technology, including the following steps:

[0059] S1. Based on the temperature distribution and displacement distribution caused by laser heating, establish the relationship between the detachment stress and the adhesion force, determine the relationship between different process parameters and the scale removal effect, and combine the current wire rod material properties and scale properties to determine the initial laser process parameters;

[0060] Among them, the step of establishing the relationship between the detachment stress and the adhesion force, determining the relationship between different process parameters and the scale removal effect, and combining the current wire rod material properties and scale properties to determine the initial laser process parameters includes the following steps:

[0061] S11. Based on the laser power, material properties, and heat conduction characteristics, establish a temperature distribution model caused by laser heating; based on the temperature gradient caused by laser heating and the material thermal expansion characteristics, establish a displacement and stress distribution model caused by laser heating;

[0062] Specifically, when laser heats the metal surface, the laser energy will be absorbed by the material and converted into heat energy, resulting in an increase in the material surface temperature, which in turn affects the temperature distribution of the material. The temperature distribution model needs to comprehensively consider physical properties such as laser power, thermal conductivity, density, and specific heat capacity of the material. The following are the basic steps for establishing the temperature distribution model:

[0063] 1) Temperature distribution equation: Based on the heat conduction equation and the laser power distribution, the temperature distribution caused by laser heating can be described by a two-dimensional heat conduction equation, that is, the expression of the temperature distribution model caused by laser heating is:

[0064]

[0065] In the formula, T represents temperature, t represents time, the Laplace operator of temperature (representing heat diffusion), represents the thermal diffusivity of the material (determined by thermal conductivity, density, and specific heat capacity), Q(r) represents the distribution of laser power at position r, ρ is the density of the material, and c represents the specific heat capacity of the material;

[0066] 2) Laser power distribution Q(r): The distribution of laser power is usually determined by the shape of the laser spot. For a Gaussian laser beam, the laser power follows a Gaussian distribution in the radial direction:

[0067]

[0068] In the formula, Q0 represents the maximum value of the laser power, r represents the radial distance from the center of the laser beam, and k represents the width of the spot;

[0069] 3) Steady-state temperature distribution: For the steady state (ignoring time variation), the temperature distribution equation simplifies to:

[0070]

[0071] This equation can obtain the temperature distribution through numerical solution methods (such as the finite difference method or the finite element method), thereby obtaining the temperature distribution caused by laser heating;

[0072] Laser heating will cause thermal expansion of the material. The existence of a temperature gradient makes the expansion degree of the material in different regions different, resulting in the formation of internal stress, which in turn causes displacement. According to the relationship between thermal stress and the thermal expansion characteristics of the material, a displacement and stress distribution model can be established, including the basic steps:

[0073] 1) Thermal stress model: The generation of thermal stress stems from the temperature gradient and the thermal expansion of the material. The distribution of thermal stress can be described by the following thermal stress formula, that is, the expression of the displacement distribution model caused by the laser is:

[0074]

[0075] In the formula, represents thermal stress, represents the temperature change, E represents the elastic modulus, represents the thermal expansion coefficient of the material;

[0076] 2) Displacement model: Thermal stress will cause the material to deform, specifically manifested as displacement. The distribution of displacement can be calculated by the equations of elasticity, assuming that the material obeys linear elastic behavior during heating. The displacement model can be expressed by the following equation:

[0077]

[0078] In the formula, represents the divergence of stress, f represents the external force density (for example, the body force caused by laser heating), and the relationship between stress and displacement can be calculated through this equation;

[0079] 3) Displacement caused by thermal expansion: According to the temperature gradient and the thermal expansion characteristics of the material, the displacement can be calculated by the following formula:

[0080]

[0081] In the formula, u(x, y, z) represents the distribution of displacement in space, and L represents the length of the path;

[0082] 4) Finite element analysis of stress and displacement: Based on the temperature distribution, thermal stress, and thermal expansion characteristics, the displacement and stress can be numerically calculated by the finite element analysis method (FEA) to obtain the accurate temperature, stress, and displacement distributions;

[0083] S12. Determine the detachment stress between the scale and the substrate according to the thermal stress, calculate the adhesion force between the scale and the substrate, and determine the critical value between the detachment stress and the adhesion force according to the calculation results of the adhesion force and the detachment stress; specifically including:

[0084] 1) Determine the detachment stress between the scale and the substrate according to the thermal stress:

[0085] During the laser heating process, due to the existence of the temperature gradient, there will be a thermal expansion difference between the substrate and the scale, which will cause thermal stress and further affect the adhesion force between the scale and the substrate. The detachment stress refers to the minimum stress required for the scale to start separating from the substrate under the external action. The detachment stress between the scale and the substrate can be further estimated by calculating the magnitude of the thermal stress;

[0086] Assume that the thermal expansion coefficients of the scale and the substrate are different. When the scale is heated by laser, the temperature difference between the scale and the substrate will generate thermal stress; the detachment stress refers to the minimum stress that needs to be overcome when the scale and the substrate start to separate. Assume that the adhesion force of the scale increases with the increase of the thermal stress. When the thermal stress reaches a certain value, the scale will start to detach from the substrate surface, and it can be calculated by the difference between the stress caused by thermal expansion and the adhesion force between the scale and the substrate;

[0087] 2) Calculate according to the adhesion force between the scale and the substrate:

[0088] The adhesion force between the mill scale and the substrate refers to the adhesion strength between the two, which determines whether the mill scale can be detached during the laser cleaning process. The adhesion force can be calculated by the following formula:

[0089]

[0090] In the formula, F represents the adhesion force between the mill scale and the substrate, represents the adhesion strength between the mill scale and the substrate. The adhesion strength is part of the material properties and is usually determined through experiments. A represents the contact area between the mill scale and the substrate, which can be obtained by scanning the surface of the oxide layer or estimating through a model;

[0091] 3) Determine the critical value between the detachment stress and the adhesion force:

[0092] The critical value refers to the critical point between the adhesion force and the thermal stress between the mill scale and the substrate. When the thermal stress is greater than or equal to the adhesion force, the mill scale will start to detach from the substrate, forming the critical condition for the detachment stress;

[0093] S13. Use the finite element analysis method to determine the relationship between the laser parameters, material parameters, and the cleaning and removal effect of the mill scale, and combine the current wire rod material properties and mill scale performance to determine the initial laser process parameters;

[0094] Specifically, the use of the finite element analysis method to determine the relationship between the laser parameters, material parameters, and the cleaning and removal effect of the mill scale, and combine the current wire rod material properties and mill scale performance to determine the initial laser process parameters includes the following steps:

[0095] S131. Based on the temperature distribution model, displacement and stress distribution model, and combined with the critical value between the detachment stress and the adhesion force, simulate the influence of different laser process parameters on the removal effect of the mill scale, and determine the basic laser process parameters according to the simulation results;

[0096] In order to optimize the laser cleaning process parameters and effectively remove the mill scale, based on the temperature distribution model, displacement and stress distribution model, and combined with the critical value between the detachment stress and the adhesion force, the influence of different laser process parameters (such as laser power, scanning speed, etc.) on the removal effect of the mill scale can be simulated. The following is an overview of the steps:

[0097] 1) Overview of the simulation steps

[0098] 1.1) Determine the range of laser process parameters: First, set a range of laser process parameters. These parameters usually include:

[0099] Laser power: Affects the energy transfer efficiency, thus affecting the heating rate and thermal stress;

[0100] Scanning speed: Determines the speed at which the laser beam moves on the workpiece surface, thereby affecting the heating time;

[0101] Laser beam diameter (spot size): Affects the heat distribution and directly influences the temperature distribution and stress distribution;

[0102] Pulse frequency (for pulsed lasers): Affects the repetition frequency of laser pulses, thus influencing the energy input per unit time;

[0103] These process parameters will affect the temperature distribution on the metal surface, thermal expansion, stress distribution, and the detachment of scale. Therefore, it is necessary to simulate the removal effects under different process conditions;

[0104] 1.2) Calculate the temperature distribution under different laser process parameters using the temperature distribution model: According to parameters such as laser power, scanning speed, and spot size, calculate the temperature distribution on the metal surface after laser heating through the temperature distribution equation (heat conduction equation). By changing different laser parameters, simulate different temperature distributions, and thus obtain the temperature field under different laser process conditions;

[0105] 1.3) Calculate the displacement and stress distributions based on the temperature distribution: Through the thermal stress equation, calculate the thermal stress and displacement generated under different laser process conditions. Different temperature distributions will result in different thermal expansion effects, thereby generating different displacement and stress fields. By simulating these stresses and displacements, the adhesion force and detachment stress between the scale and the substrate under different laser parameters can be estimated;

[0106] 1.4) Calculate the detachment stress and adhesion force: According to the relationship between thermal stress and material thermal expansion, combined with the adhesion force between the scale and the substrate, calculate the magnitude of the detachment stress under different temperature distribution conditions. If the thermal stress is greater than the adhesion force, the scale will start to detach;

[0107] 1.5) Compare the removal effects under different laser process conditions: Based on the magnitude of the detachment stress and the change in adhesion force, determine the removal effect of the scale under each laser process condition. If the detachment stress reaches or exceeds the adhesion force, the scale can be removed. Through this simulation process, the influence of different laser process parameters on the scale removal effect can be evaluated;

[0108] 2) Determine the basic laser process parameters according to the simulation results

[0109] 2.1) Analyze the simulation results: The simulation results will show the scale removal effects under different laser process parameters, including:

[0110] Removal efficiency: That is, under different laser process parameters, the speed and degree of scale removal;

[0111] Degree of thermal damage: Excessive laser power or too low scanning speed may cause overheating and thermal damage of the metal substrate, thus affecting the quality of the final product;

[0112] Surface quality: It is necessary to ensure that there is no obvious damage or deformation on the surface of the workpiece after laser cleaning;

[0113] According to these simulation results, analyze the influence of different process parameters on the removal effect of mill scale, find the most suitable laser process parameters, and avoid substrate damage caused by excessive thermal stress;

[0114] 2.2) Optimize the basic laser process parameters: By comparing the simulation results, optimize the best laser power, scanning speed and other process parameters. Generally speaking, the basic laser process parameters should meet the following conditions:

[0115] A reasonable match between laser power and scanning speed to ensure that the temperature distribution can reach the conditions required to relieve stress, while avoiding excessive heat causing substrate damage;

[0116] The laser beam diameter and spot size should ensure uniform heat distribution to improve the removal efficiency and avoid surface overheating;

[0117] Finally, the basic laser process parameters can be further optimized through the finite element analysis method (FEA) or other numerical methods to ensure that the mill scale can be removed efficiently and uniformly without damaging the workpiece;

[0118] S132. Obtain the property data of the wire rod material to be processed, the property data of the mill scale, and the laser parameter data. Optimize the process conditions of the laser power and scanning speed in the basic laser process parameters through finite element simulation (such as ABAQUS, ANSYS, etc.), and determine the initial laser process parameters according to the optimization results (that is, according to the simulation results, select the laser power and scanning speed that can efficiently remove the mill scale, avoid overheating damage and ensure the surface quality);

[0119] The physical properties of the wire rod substrate include: density, specific heat capacity, thermal conductivity, elastic modulus, wire rod substrate thickness and coefficient of thermal expansion;

[0120] The physical properties of the mill scale include: density, specific heat capacity, thermal conductivity, elastic modulus, mill scale thickness and coefficient of thermal expansion;

[0121] S2. Use the oxide layer thickness identification model combined with terahertz technology to identify the thickness distribution of the oxide layer on the surface of the wire rod, and dynamically adjust the initial laser process parameters according to the thickness distribution of the oxide layer on the surface of the wire rod to obtain the best laser process parameters based on the thickness distribution;

[0122] The method of using the oxide layer thickness recognition model in combination with terahertz technology to recognize the thickness distribution of the oxide layer on the surface of the wire rod, and dynamically adjusting the initial laser process parameters according to the thickness distribution of the oxide layer on the surface of the wire rod to obtain the optimal laser process parameters based on the thickness distribution includes the following steps:

[0123] S21, based on a pre-constructed oxide layer thickness identification model, identifying the thickness distribution of the oxide layer on the surface of the wire rod to be processed, and obtaining a first distribution result of the oxide layer thickness;

[0124] Specifically, the method of identifying the thickness distribution of the oxide layer on the surface of the wire rod to be processed based on the pre-constructed oxide layer thickness identification model and obtaining the first distribution result of the oxide layer thickness includes the following steps:

[0125] S211, obtaining a surface image of the wire rod to be processed and performing preprocessing, analyzing the brightness difference in the surface image, extracting the grayscale values ​​of different areas on the surface of the oxide layer, calculating the local brightness distribution in the surface image, obtaining the average brightness of the local area by a sliding window method, and calculating the global brightness distribution to generate a grayscale histogram; specifically including:

[0126] 1) Obtain the surface image of the wire rod to be processed: Use a high-resolution image acquisition device (such as a digital camera, microscope or scanning electron microscope, etc.) to take the surface image of the wire rod. The image should be as clear as possible and be able to show the details of the iron oxide scale. Ensure that the image is clear and can show the obvious difference between the iron oxide scale and the base material;

[0127] 2) Image preprocessing:

[0128] Denoising: After image acquisition, it is usually necessary to remove the noise in the image to improve the accuracy of subsequent analysis. Common noise removal methods include Gaussian blur, median filtering, etc. These methods can effectively remove subtle noise or background noise;

[0129] Grayscale processing: Convert the image from color mode to grayscale mode to simplify the subsequent brightness analysis. Grayscale images only consider brightness information, not color information, which is very helpful for analyzing the distribution of oxide layers and brightness differences;

[0130] Contrast enhancement: Use histogram equalization to enhance the contrast of the image. This step can make the brightness difference more obvious, especially at the boundary between the oxidized iron area and the substrate, improving the image's recognizability;

[0131] 3) Brightness difference analysis:

[0132] Local brightness distribution calculation: To analyze the brightness differences in an image, a sliding window method can be adopted. The image is divided into multiple small blocks, and local brightness analysis is performed on each small block through the sliding window technique. The brightness of each window area can be obtained by calculating the average gray value of all pixels within that area. In this way, the local brightness distribution of different regions of the image can be obtained, and further, the thickness of the oxide layer, reflection characteristics, etc. of different regions on the surface can be understood;

[0133] Global brightness distribution statistics: The global brightness distribution refers to the brightness distribution of the entire image. The histogram of the global brightness distribution can be obtained by calculating the gray values of all pixels in the image and counting the frequency of occurrence of each gray value;

[0134] 4) Generate a gray histogram: A gray histogram is a commonly used image analysis tool used to represent the distribution of different gray values in an image. By counting the number of pixels with each gray value in the image, a graph of the gray value and its corresponding number of pixels can be plotted;

[0135] Brightness distribution: From the gray histogram, the distribution of regions with higher and lower brightness in the image can be intuitively seen. For example, if the brightness of the scale area is higher, there will be more pixel points on the right side (close to the gray value of 255) of the histogram; if the substrate area is darker, there will be more pixel points on the left side (close to the gray value of 0) of the histogram;

[0136] Analyze the brightness differences in the image: By comparing the gray values of different regions, the brightness differences between the scale and the substrate can be analyzed. High-brightness regions usually correspond to the scale with a higher reflectivity, while low-brightness regions may correspond to the darker substrate or areas where the scale is thinner;

[0137] 5) Extract the gray values of different regions on the oxide layer surface:

[0138] Identify the oxide layer region: By setting a gray threshold or using image segmentation techniques, the oxide layer region in the image can be identified. These regions usually have different brightnesses, may be brighter than the substrate part, or may have different brightness manifestations due to different thicknesses of the scale;

[0139] Statistical gray values of the oxide layer region: By extracting the gray values of the oxide layer region, the average brightness of this region can be further calculated and analyzed. In this way, the thickness, reflection characteristics, and uniformity of the oxide layer can be quantified;

[0140] S212. Extract the texture features of the surface image using the local binary pattern method, and identify the boundary between the oxide layer and the substrate through the edge detection method; based on the pre-constructed oxide layer thickness identification model, output the oxide layer thickness corresponding to the texture features of the surface image, and obtain the first distribution result of the oxide layer thickness at different positions on the surface of the wire rod; specifically including:

[0141] 1) Extract texture features using the local binary pattern method: The local binary pattern (LBP) is a commonly used texture analysis method that can extract local texture features by comparing the gray values of each pixel in the image with those of its neighboring pixels. It has strong robustness to textures and good adaptability to illumination changes. The specific steps are as follows:

[0142] 1.1) Gray image conversion: Convert the surface image into a gray image (if the image is a color image) for texture analysis;

[0143] 1.2) Calculate LBP features: For each pixel in the image, calculate the local binary pattern by comparing the pixel value with the pixel values in its surrounding neighborhood. Generally, the neighborhood around each pixel is a 3x3 area, that is, 8 neighboring pixels;

[0144] 1.3) LBP texture mapping: Calculate the LBP texture features of the entire image. Usually, a histogram is used to represent these features, that is, count the frequency of each LBP value. This histogram can reflect the texture pattern of the image;

[0145] 1.4) LBP feature matrix: An LBP feature matrix can be obtained to describe the texture features of the surface image. The texture features can help analyze the oxide layer thickness in different regions later;

[0146] For the neighboring pixels of each pixel, compare them from left to right and from top to bottom. If the value of the neighboring pixel is greater than the central pixel, assign 1, otherwise 0;

[0147] Convert the obtained binary number to decimal to get the LBP value of this pixel;

[0148] 2) Identify the boundary between the oxide layer and the substrate through the edge detection method: To accurately extract the thickness of the oxide layer, it is necessary to first identify the boundary between the oxide layer and the substrate. Edge detection is a common image processing method. Common edge detection algorithms include Canny edge detection and Sobel operator;

[0149] 2.1) Gray image edge detection: In the grayscale image of the image, apply the edge detection method to identify the boundary between the oxide layer and the substrate. For example, Canny edge detection can be used, which detects edges by calculating the gradient in the image;

[0150] Steps of Canny edge detection algorithm:

[0151] Gaussian filtering: First, use a Gaussian filter to remove noise;

[0152] Calculate the gradient: Calculate the gradient of the image through the Sobel operator, etc., to obtain the intensity and direction of the edges;

[0153] Non-maximum suppression: Refine the edges and remove spurious edges;

[0154] Double-threshold detection: Determine the real edges by setting high and low thresholds;

[0155] 2.2) Edge extraction: After edge detection, the boundary between the oxide layer and the substrate will be clearly identified. In the edge detection image, the edge of the oxide layer usually appears as a high-contrast region;

[0156] 2.3) Extract boundary information: Combine the edge detection results with the original image to accurately determine the boundary position of the oxide layer;

[0157] 3) Output the oxide layer thickness based on the oxide layer thickness recognition model:

[0158] 3.1) Build an oxide layer thickness recognition model: Based on the texture features in the image (such as LBP features) and the edge detection results, an oxide layer thickness recognition model can be built; this model can be a machine learning-based model (such as support vector machine, decision tree, random forest, etc.), or an estimation based on a physical model; when training the model, a set of sample data with calibrated oxide layer thicknesses is required to learn the relationship between the texture features and the oxide layer thickness;

[0159] 3.2) Model input: The features input into the model are the texture features (such as LBP histogram) and edge information extracted from the image; the model predicts the oxide layer thickness at each position based on these inputs;

[0160] 3.3) Oxide layer thickness output: Use the trained oxide layer thickness model, combined with the texture features extracted from the image, to output the oxide layer thickness value at each position, and the oxide layer thickness at each position will correspond to the texture features;

[0161] 4) Obtain the first distribution result of the oxide layer thickness at different positions on the surface of the wire rod:

[0162] 4.1) Thickness distribution map generation: According to the oxide layer thickness data output by the model, a thickness distribution map can be generated to show the oxide layer thickness at different positions on the surface of the wire rod. This distribution map can represent the oxide layer thickness through color coding, so as to visually display the uniformity and thickness variation of the oxide layer;

[0163] 4.2) Analysis and visualization: Analyze the thickness distribution of the oxide layer and identify areas where the oxide layer is thinner or thicker. This can help optimize the subsequent laser cleaning process. Heat maps, contour maps, or 3D maps can be used to visualize the thickness distribution of the oxide layer;

[0164] 4.3) Statistics of the oxide layer thickness: Based on the thickness distribution map, statistical data of the oxide layer thickness can be calculated, such as average thickness, maximum thickness, minimum thickness, etc.;

[0165] S22. Use terahertz technology to identify the thickness distribution of the oxide layer on the surface of the bar to be processed, and obtain the second distribution result of the oxide layer thickness;

[0166] Specifically, the use of terahertz technology to identify the thickness distribution of the oxide layer on the surface of the bar to be processed and obtain the second distribution result of the oxide layer thickness includes the following steps:

[0167] S221. Irradiate the surface of the bar to be processed with ultrashort pulse terahertz waves, measure the reflection, transmission, or absorption of the terahertz waves by the oxide layer, and record the amplitude, phase, and time delay information of the terahertz waves;

[0168] Irradiating the surface of the bar to be processed with ultrashort pulse terahertz waves and measuring the reflection, transmission, or absorption of the terahertz waves by the oxide layer is an efficient surface detection and surface property analysis technique. By recording the amplitude, phase, and time delay information of the terahertz waves, the physical properties, thickness, and uniformity of the oxide layer can be obtained. The following are the detailed steps:

[0169] 1) Principle of ultrashort pulse terahertz wave irradiation

[0170] Generation of terahertz waves: Ultrashort pulse terahertz waves are usually generated by laser-excited materials or using a terahertz pulse generator (such as a photoconductive antenna). This terahertz wave usually has a very short time scale (picosecond or femtosecond level), so it has very high time resolution and is suitable for rapid dynamic analysis of surfaces and materials;

[0171] Properties of terahertz waves: The frequency range of terahertz waves is usually between 0.1 THz and 10 THz. Electromagnetic waves in this frequency range have strong penetration characteristics and can penetrate many materials, such as non-metals, composite materials, insulating materials, etc. Therefore, it is very suitable to use terahertz waves for surface and oxide layer analysis, especially when detecting on the surface of metals. The reflection of terahertz waves on metals and the absorption and transmission of terahertz waves by the oxide layer provide important information;

[0172] Ultrashort pulse: Ultrashort pulse terahertz waves usually have a very high peak power and a short pulse width, which can provide sufficient time resolution to detect the detailed changes of the oxide layer;

[0173] 2) Interaction between the irradiated terahertz wave and the surface of the wire rod

[0174] Irradiation method: The terahertz wave is irradiated onto the surface of the wire rod to be processed. The interaction between the terahertz wave and the oxide layer mainly occurs through reflection, transmission, and absorption:

[0175] Reflection: After the terahertz wave irradiates the surface, part of the wave will be reflected back to the sensor, which depends on the electromagnetic properties (such as conductivity, thickness, etc.) of the surface and the oxide layer;

[0176] Transmission: If the oxide layer is thin, part of the terahertz wave may pass through the oxide layer and continue to propagate into the substrate. At this time, the properties of the oxide layer can be analyzed by measuring the intensity and time delay of the transmitted wave;

[0177] Absorption: The oxide layer will absorb the terahertz wave, especially in a specific frequency range. The absorption peak of the terahertz wave by the oxide layer may be related to the thickness and composition of the oxide layer;

[0178] Influence of the oxide layer: The influence of the oxide layer on the terahertz wave is mainly reflected in its physical properties such as thickness, dielectric constant, and conductivity. Changes in the oxide layer will cause changes in the reflectivity, transmittance, and absorption characteristics of the terahertz wave;

[0179] 3) Recording the amplitude, phase, and time delay of the terahertz wave

[0180] Amplitude measurement: Use a terahertz wave detector (such as a photoconductive antenna or a terahertz time-domain spectrometer) to measure the amplitude of the terahertz wave. This can reveal the reflection, transmission, and absorption characteristics of the oxide layer on the wave;

[0181] Reflection amplitude: Record the amplitude of the reflected terahertz wave, which reflects the properties of the interface between the oxide layer and the substrate;

[0182] Transmission amplitude: Record the amplitude of the terahertz wave passing through the oxide layer, which can reflect the physical properties such as the thickness, density, and dielectric constant of the oxide layer;

[0183] Phase information: By analyzing the phases of the irradiated, reflected, and transmitted terahertz waves, the thickness, surface uniformity, and dielectric properties of the oxide layer can be obtained. The greater the thickness of the oxide layer, the greater the phase delay of the terahertz wave. Using the interference method or the Fourier transform method, the phase change of the terahertz wave after passing through the oxide layer can be accurately calculated;

[0184] Time delay information: Record the time delay from the irradiation of terahertz waves to reflection / transmission. The time delay is related to characteristics such as the thickness and dielectric constant of the oxide layer. A thicker oxide layer will cause a longer delay for terahertz waves when passing through the oxide layer; Use time-domain spectroscopy (TDS) to measure this time delay. By extracting the propagation time and time delay of the oxide layer from the reflected signal, the thickness of the oxide layer can be inferred;

[0185] 4) Analyze the reflection, transmission, and absorption characteristics of the oxide layer

[0186] Analysis of reflection characteristics: Through the amplitude of the reflected signal, the thickness and interface quality of the oxide layer can be estimated. A thinner oxide layer usually results in a lower reflected signal, while a thicker or uneven oxide layer will enhance the reflected signal. The conductivity and other electromagnetic properties of the oxide layer can also be deduced based on the changes in the amplitude and phase of the reflection;

[0187] Analysis of transmission characteristics: The intensity and phase changes of the transmitted signal can provide information about the thickness of the oxide layer. A thinner oxide layer transmits more terahertz waves, while a thicker oxide layer may absorb or reflect more waves. Based on the time delay and phase of the transmitted signal, the actual thickness of the oxide layer can be calculated;

[0188] Analysis of absorption characteristics: The absorption characteristics can be analyzed through the attenuation of terahertz waves. A thicker oxide layer usually causes more energy absorption of terahertz waves during penetration. The absorption characteristics are closely related to the composition of the oxide layer. By studying the changes in the absorption rate, the chemical composition and thickness information of the oxide layer can be revealed;

[0189] S222. Calculate the reflection and transmission time differences of terahertz waves at different oxide layer thicknesses through time-domain analysis, use Fourier transform to convert the time-domain signal into a frequency-domain signal, extract the spectral characteristics of different frequencies, and calculate the absorption and reflectance of the oxide layer for terahertz waves of specific frequencies to determine its thickness information;

[0190] To calculate the reflection and transmission time differences of terahertz waves at different oxide layer thicknesses through time-domain analysis, use Fourier transform to convert the time-domain signal into a frequency-domain signal, extract the spectral characteristics of different frequencies, then calculate the absorption and reflectance of the oxide layer for terahertz waves of specific frequencies, and finally determine the thickness information of the oxide layer, the following steps can be taken:

[0191] 1) Analyze the reflection and transmission time differences of terahertz waves in the time domain

[0192] 1.1) Irradiation, reflection, and transmission of terahertz waves: When terahertz waves irradiate the surface of the wire rod to be processed, part of the waves will be reflected back to the sensor, and the other part may penetrate the oxide layer and enter the substrate. Characteristics such as the thickness, dielectric constant, and conductivity of the oxide layer will affect the time-domain performance of the reflected and transmitted waves;

[0193] 1.2) Measure the reflected and transmitted signals:

[0194] Reflected signal: The arrival time of the reflected wave can be directly measured, and the amplitude and phase of the reflected signal change with the thickness of the oxide layer;

[0195] Transmitted signal: Similarly, the time delay and intensity of the transmitted signal are also affected by the thickness of the oxide layer. A thicker oxide layer usually results in an increased time delay of the transmitted signal and may cause attenuation of the transmitted signal;

[0196] 1.3) Calculate the time difference: By measuring the arrival times of the reflected and transmitted waves, the time difference between them can be calculated. This time difference is closely related to the thickness of the oxide layer. The thicker the oxide layer, the longer the propagation time of the wave (especially the signal passing through the oxide layer), resulting in a time delay; The calculation formula can be derived from the speed of light and the dielectric constant of the oxide layer;

[0197] 2) Fourier transform converts the time-domain signal into a frequency-domain signal

[0198] 2.1) Fourier transform principle: The time-domain signal can be transformed into a frequency-domain signal through Fourier transform. Fourier transform can reveal the response of terahertz waves at different frequencies and help us analyze the influence of the oxide layer on terahertz waves of different frequencies;

[0199] 2.2) Conversion of time-domain signal to frequency-domain signal: The frequency-domain signal obtained through Fourier transform contains information on each frequency component. For terahertz waves, their frequency range is generally from 0.1 THz to 10 THz. By analyzing the responses at different frequencies, the absorption and reflection characteristics of the oxide layer for different frequency bands can be understood;

[0200] 2.3) Spectrum analysis: The frequency-domain signal contains the amplitude and phase information of terahertz waves at each frequency, which in turn helps us identify the influence of the oxide layer on specific frequencies. Each material (including the oxide layer) has a different response to different frequencies. Through spectrum feature analysis, the properties of the oxide layer can be deeply understood;

[0201] 3) Extract spectral features at different frequencies

[0202] 3.1) Spectrum feature extraction: For the frequency-domain signal obtained through Fourier transform, extract the magnitude spectrum and phase spectrum at different frequencies. The magnitude spectrum represents the intensity of the signal at each frequency, while the phase spectrum provides the relative phase information of the frequency components; These two features can reflect the absorption, reflection, or transmission characteristics of the oxide layer at different frequencies:

[0203] Amplitude spectrum: The absorption and reflection of terahertz waves with different frequencies by the oxide layer usually show significant differences within a specific frequency range. Therefore, the amplitude spectrum can reveal characteristics such as the thickness and uniformity of the oxide layer;

[0204] Phase spectrum: The change in phase can further reveal the change in the dielectric constant of the material. An oxide layer with a larger thickness usually causes a larger phase change;

[0205] 3.2) Frequency characteristic analysis:

[0206] High-frequency part: The thin layer or surface non-uniformity of the oxide layer usually has a greater impact on the reflection of high-frequency terahertz waves. By analyzing the response in the high-frequency part, the microstructure of the oxide layer can be understood;

[0207] Low-frequency part: When the oxide layer is thicker, its absorption of low-frequency terahertz waves is stronger. Therefore, the low-frequency response is also of great significance for estimating the thickness of the oxide layer;

[0208] 4) Calculate the absorption and reflectivity of the oxide layer to terahertz waves at specific frequencies

[0209] Reflectivity: Reflectivity represents the proportion of terahertz waves incident on the surface that are reflected back. The reflectivity R(f) can usually be obtained by measuring the ratio of the amplitude of the reflected signal to the amplitude of the incident signal, that is , where Sref(f) represents the frequency-domain representation of the reflected signal, and Sinc(f) represents the frequency-domain representation of the incident signal;

[0210] Absorptivity: Absorptivity represents the absorbed part of the terahertz waves by the oxide layer. The absorptivity A(f) can usually be calculated by the following formula: A(f) = 1 - R(f) - T(f), where T(f) represents the transmittance, which represents the proportion of waves that pass through the oxide layer. The absorptivity reflects the absorption characteristics of the oxide layer, and the position and size of the absorption peak are related to the thickness and dielectric constant of the oxide layer;

[0211] Frequency dependence of absorption and reflection: By analyzing the absorption and reflection characteristics at different frequencies, the thickness of the oxide layer can be identified. Usually, the absorption and reflection characteristics of the oxide layer at specific frequencies change with the thickness, so these changes can be used to estimate the thickness of the oxide layer;

[0212] 5) Determine the thickness information of the oxide layer

[0213] 5.1) Using model fitting: Compare the spectral characteristics obtained from the Fourier transform with the spectral data of oxide layer samples with known thicknesses. An inversion algorithm or a machine learning model can be used to extract the thickness information of the oxide layer from the frequency-domain data;

[0214] 5.2) Thickness calculation formula: By analyzing the changing trends of reflectivity, transmittance, and absorptance, and combining the response of the oxide layer to specific frequencies, the following model can be used to estimate the thickness of the oxide layer:

[0215] d = f(Δt, A(f), R(f), T(f))

[0216] where d is the thickness of the oxide layer, Δt is the time delay, and A(f), R(f), and T(f) are the functions of absorptance, reflectivity, and transmittance at different frequencies, respectively;

[0217] 5.3) Thickness distribution: Through the above process, the distribution map of the oxide layer thickness at different positions on the surface can be finally obtained, which helps to further analyze the uniformity and quality of the oxide layer;

[0218] S223. Calculate the thickness of the oxide layer based on the time difference between the transmission and reflection of terahertz waves, and combine the refractive index of the oxide layer to calculate the thickness of the oxide layer, and use a mathematical model to invert the thickness distribution of the oxide layer to obtain the second distribution result of the oxide layer thickness at different positions on the surface of the wire rod;

[0219] In order to calculate the thickness of the oxide layer based on the time difference between the transmission and reflection of terahertz waves, and combine the refractive index of the oxide layer to calculate the thickness of the oxide layer, and use a mathematical model to invert the thickness distribution of the oxide layer, we can proceed as follows:

[0220] 1) Calculate the thickness of the oxide layer based on the time difference

[0221] 1.1) Time difference between reflection and transmission:

[0222] When terahertz waves irradiate the surface of the wire rod, part of the waves will be reflected by the oxide layer, while the other part will pass through the oxide layer. If the thickness of the oxide layer is d, the refractive index is n, and considering the time delay of the wave propagation in the oxide layer, the thickness of the oxide layer can be deduced from the time difference between reflection and transmission;

[0223] The time difference between the arrival times of the reflected wave and the transmitted wave is closely related to the thickness of the oxide layer. When terahertz waves pass through the oxide layer, the wave velocity will be affected due to the change in refractive index;

[0224] Time delay formula: When terahertz waves pass through the oxide layer, the time delay Δt can be calculated by the following formula: , where d is the thickness of the oxide layer, cg is the speed of light in vacuum, and n is the refractive index of the oxide layer;

[0225] By measuring the time difference Δt between the reflected wave or the transmitted wave and the refractive index n of the oxide layer, the thickness of the oxide layer can be calculated;

[0226] 2) Influence of refractive index on wave propagation

[0227] Influence of refractive index: The refractive index of the oxide layer directly affects the propagation speed of terahertz waves in the oxide layer. The refractive index is closely related to the electromagnetic properties of the medium and can usually be obtained through experimental measurement or by using known material property data;

[0228] Wave speed calculation: In the oxide layer, the propagation speed of terahertz waves is the ratio of the speed of light to the refractive index of the oxide layer;

[0229] Refractive index inversion: If the refractive index of the oxide layer is unknown, the value of the refractive index can be inversely calculated by comparing the wave speeds at different frequencies or by analyzing the influence of the oxide layer on terahertz waves of different frequencies;

[0230] 3) Inverting the thickness distribution of the oxide layer using a mathematical model

[0231] 3.1) Establishing a mathematical model: Based on the time difference between reflection and transmission, an inversion model can be established to calculate the thickness distribution of the oxide layer according to the experimentally measured time delay. This model can use the method of solving inverse problems to deduce the thickness of the oxide layer at each position on the surface of the wire rod through the known time difference data and refractive index data;

[0232] Inversion model: By establishing an inversion algorithm, such as based on the least squares method or optimization algorithms (such as gradient descent, Newton's method, etc.), the optimal solution of the oxide layer thickness can be obtained;

[0233] 3.2) Distribution inversion: Using this inversion model, based on the time difference data obtained at multiple positions (such as through a sensor array or scanning), the thickness of the oxide layer at each position can be estimated, and finally the thickness distribution of the oxide layer can be obtained;

[0234] 3.3) Selection of optimization algorithm:

[0235] Least squares method: Based on the calculation of the time difference between reflection and transmission, the least squares method can be used to obtain the best estimate of the thickness distribution. The least squares method inversely calculates the thickness of the oxide layer by minimizing the error between the predicted value and the actual measured value;

[0236] Finite element method (FEM): For more complex geometries or non-uniform oxide layer distributions, the finite element method can be used to simulate the interaction process between the oxide layer and terahertz waves and perform inversion calculations;

[0237] 4) Results of the oxide layer thickness distribution

[0238] 4.1) Generating a thickness distribution map: Based on the inversely obtained oxide layer thickness (thickness estimation at different positions), a distribution map of the surface oxide layer thickness can be generated. These data can be visualized through graphical tools (such as heat maps, contour maps, etc.) to show the thickness differences of the oxide layer at each position on the surface of the wire rod;

[0239] 4.2) Further analysis: Analyze the thickness uniformity of the oxide layer: By observing the distribution map, check whether the oxide layer is uniform and whether there are areas with locally thicker or thinner parts. Optimize subsequent processes: Based on the results of the thickness distribution, subsequent laser cleaning processes, surface treatment processes, etc. can be optimized;

[0240] 4.3) Statistical analysis:

[0241] Average thickness: Calculate the average thickness of the oxide layer;

[0242] Maximum / minimum thickness: Find the maximum and minimum thicknesses of the oxide layer;

[0243] Standard deviation: Evaluate the thickness uniformity of the oxide layer;

[0244] 5) Verification and application

[0245] 5.1) Experimental verification: The thickness distribution of the oxide layer obtained by inversion can be verified by other methods, such as mechanical measurement, scanning electron microscopy (SEM), or other surface detection methods to verify the accuracy of the inversion results;

[0246] 5.2) Practical application: The results of this thickness distribution can provide a basis for subsequent laser cleaning, coating treatment, or other surface treatment processes, and optimize the process parameters to achieve a uniform surface quality;

[0247] S23. Determine the thickness distribution of the oxide layer on the surface of the wire rod to be processed based on the first distribution result and the second distribution result of the oxide layer thickness;

[0248] Specifically, determining the thickness distribution of the oxide layer on the surface of the wire rod to be processed based on the first distribution result and the second distribution result of the oxide layer thickness includes the following steps:

[0249] S231. Obtain the first distribution result and the second distribution result of the oxide layer thickness at different positions on the surface of the wire rod, and perform data alignment and preprocessing; According to the grid division method on the surface of the wire rod, establish the spatial correspondence relationship between the first distribution result data and the second distribution result data, and use the feature matching method to spatially align the thickness distribution data of the first distribution result and the second distribution result;

[0250] S232. Use the weighted average method to combine the first distribution result data and the second distribution result data with a preset weight coefficient to obtain the final thickness value, and generate a thickness distribution map of the oxide layer at different positions on the surface of the wire rod according to the final thickness value;

[0251] Specifically, in this embodiment, the weight of the oxide layer thickness recognition model: This model can adapt to relatively complex surface structures and can provide real-time feedback on the effect of oxide layer removal. Therefore, in the case of complex surfaces and irregular oxide layers, it has greater advantages. In this case, a higher weight is assigned to it, that is, the weight coefficient of the first distribution result data is 0.8;

[0252] The weight of terahertz technology: Terahertz wave technology has high accuracy in the case of relatively uniform thickness and thin oxide layers, but for complex surfaces or thicker oxide layers, errors may occur. Therefore, the weight of terahertz waves is slightly lower, that is, the weight coefficient of the second distribution result data is 0.2;

[0253] In addition, after data fusion, the errors between the fusion result and the original first and second distribution results can be calculated to analyze the deviation between the two; set a threshold range. If the error exceeds the allowable range, adjust the fusion weight or recalibrate the measurement data;

[0254] S24. Dynamically adjust the initial laser process parameters at different positions on the surface of the to-be-processed wire rod according to the thickness distribution of the oxide layer on the surface of the to-be-processed wire rod to obtain the optimal laser process parameters based on the oxide layer thickness;

[0255] In order to dynamically adjust the initial laser process parameters at different positions on the surface of the to-be-processed wire rod according to the thickness distribution of the oxide layer on the surface of the to-be-processed wire rod and finally obtain the optimal laser process parameters based on the oxide layer thickness, it can be achieved according to the following steps:

[0256] 1) Analyze the characteristics of the oxide layer thickness distribution

[0257] Uniformity of the oxide layer: Check the uniformity of the oxide layer on the surface. Thinner oxide layers may require stronger laser power or longer laser action time, while thicker oxide layers may require lower power or shorter laser action time; The change of the oxide layer not only affects the cleaning effect but also affects the absorption, reflection, and scattering characteristics of the laser;

[0258] Identification of high-thickness and thin-thickness regions: Identify the regions where the oxide layer is thinner and thicker. These regions may have different effects on the absorption characteristics of the laser, and it is necessary to adjust process parameters such as laser power, pulse duration, and scanning speed according to the thickness of the oxide layer;

[0259] 2) Adjust the laser process parameters according to the oxide layer thickness distribution

[0260] 2.1) Establish a relationship model between laser process parameters and oxide layer thickness: There is a specific relationship between the process parameters (such as laser power, pulse width, scanning speed, pulse frequency, etc.) during the laser cleaning process and the thickness of the oxide layer. A mathematical model can be established or machine learning algorithms can be used to learn the relationship between the oxide layer thickness and laser parameters;

[0261] 2.2) Adjust process parameters: For areas with a thinner oxide layer, it is necessary to increase the laser power, extend the pulse width, or slow down the scanning speed to ensure effective removal of the oxide layer. For areas with a thicker oxide layer, reduce the laser power, shorten the pulse width, or increase the scanning speed to prevent over-etching or over-cleaning;

[0262] 2.3) Spatial adjustment of process parameters: According to the distribution of the oxide layer thickness at different positions, dynamically adjust the laser process parameters at different positions. This process can be achieved through a control system, where the laser equipment can adjust parameters such as the laser power, pulse width, and scanning speed according to real-time oxide layer thickness data;

[0263] 3) Optimize laser process parameters

[0264] Optimization of laser power: For a thin oxide layer, using a lower laser power can avoid over-cleaning and reduce damage to the substrate. For a thick oxide layer, using a higher laser power can ensure sufficient removal of the oxide layer, but damage to the substrate should be avoided;

[0265] Optimization of pulse width: In areas with a thinner oxide layer, a shorter pulse width can avoid excessive heat accumulation and reduce the impact of thermal effects on the substrate. In areas with a thicker oxide layer, a longer pulse width may help to more effectively remove the oxide layer because the thicker oxide layer absorbs more energy;

[0266] Optimization of scanning speed: In areas with a thinner oxide layer, a higher scanning speed may be appropriate because a thin oxide layer requires a shorter laser irradiation time. In areas with a thicker oxide layer, a lower scanning speed may be appropriate because a thicker oxide layer requires a longer laser irradiation time to be effectively removed;

[0267] Adjustment of other parameters: The pulse frequency can be adjusted according to the thickness of the oxide layer. In areas with a thinner oxide layer, a lower pulse frequency can be used; while in areas with a thicker oxide layer, a higher pulse frequency can be used to enhance the removal effect;

[0268] S3. Based on the optimal laser process parameters, use a rotating laser head to perform laser treatment on the scale on the surface of the wire rod, and collect metal powder through negative pressure technology; use image recognition technology to detect the removal effect of the scale on the surface of the wire rod, and dynamically optimize the model parameters according to the detection results;

[0269] Specifically, in this embodiment, a plurality of rotating laser heads (rotating around the center line of the wire rod) are arranged in three rows on the circumference of the wire rod. Each rotating laser head is connected to a laser generator. Each rotating laser head contains a plurality of laser heads arranged in the circumferential direction. The plurality of rotating laser heads are respectively fixed on a bracket, and the rotating laser head is provided with a protective cover. Moreover, the angle between the laser beam generated by the rotating laser head on the circumference of the wire rod and the wire rod is 60-80 degrees. The width of the laser beam of the wire rod is matched with the circumferential width of the surface of the wire rod, so that the laser beam can achieve full coverage in the circumferential direction of the wire rod. The rotating laser head contains a plurality of laser beams, and through its own rotation, the scale on the surface of the wire rod can be removed and cleaned from multiple angles and multiple times, improving the efficiency and quality of scale cleaning.

[0270] Specifically, when cleaning the scale of the wire rod, in combination with equipment such as a suction fan and an electrode dust collector, a negative pressure condition in the area of the laser cleaning machine is realized to prevent the spilled metal powder from polluting the environment, and the metal powder is recycled and reused so that the particles do not return to the clean surface again.

[0271] Specifically, image recognition technology is used to detect the removal effect of the scale on the surface of the wire rod, and the model parameters are dynamically optimized according to the detection results, including:

[0272] Principle of image recognition technology:

[0273] Image recognition technology (such as computer vision-based technology) is used to detect the removal effect of the scale on the surface of the wire rod in real time. A high-resolution camera, a laser scanner or an infrared imaging device is installed to comprehensively detect the surface of the wire rod. Image recognition technology can analyze the removal degree, uniformity, residual situation, etc. of the surface scale. Common image processing technologies include edge detection, texture analysis and color contrast analysis, etc.;

[0274] Detection of removal effect:

[0275] Removal rate: Image recognition technology can calculate the removal rate of the scale by comparing with the image before processing. If the removal rate is low, it may be necessary to increase the laser power or reduce the scanning speed;

[0276] Uniformity detection: Image recognition can also detect the uniformity of the removal to ensure that there are no untreated areas left on the surface and guarantee the consistency of the processing effect;

[0277] Surface damage: Image recognition can also detect whether the surface of the substrate is damaged during the laser cleaning process, such as excessive ablation or other physical damage. This can avoid unnecessary damage to the wire rod during the processing;

[0278] Result feedback of image processing: The image recognition system generates feedback data by detecting the removal effect in real time, indicating which areas need further adjustment of laser parameters. The processed image results will be transmitted to the central control system;

[0279] Data feedback and optimization model: The removal effect data obtained through image recognition (such as removal rate, uniformity, surface damage condition) will be fed back to the central control system. Based on these real-time detection results, the parameters of the oxide layer thickness recognition model will be dynamically adjusted.

[0280] According to another embodiment of the present invention, a system for online removal of scale on wire rods based on laser technology is provided, including an initial laser process parameter determination module, a laser process parameter optimization module, and an online scale removal module;

[0281] Among them, the initial laser process parameter determination module is used to establish the relationship between the detachment stress and the adhesion force according to the temperature distribution and displacement distribution caused by laser heating, determine the relationship between different process parameters and the scale removal effect, and combine the current wire rod material properties and scale performance to determine the initial laser process parameters;

[0282] The laser process parameter optimization module is used to utilize the oxide layer thickness recognition model combined with terahertz technology to identify the thickness distribution of the oxide layer on the wire rod surface, and dynamically adjust the initial laser process parameters according to the thickness distribution of the oxide layer on the wire rod surface to obtain the optimal laser process parameters based on the thickness distribution;

[0283] The online scale removal module is used to perform laser treatment on the scale on the wire rod surface by using a rotating laser head based on the optimal laser process parameters, and collect metal powder through negative pressure technology; use image recognition technology to detect the removal effect of the scale on the wire rod surface, and dynamically optimize the model parameters according to the detection results.

[0284] In summary, with the above technical solutions of the present invention, through the high energy density and precise control of the laser, the present invention can quickly heat the scale to the falling-off temperature, realize the rapid removal of the scale, and can remove the scale and stains on the wire rod surface during the continuous production process at one time, solving the pain points in the prior art that the scale on the wire rod surface cannot be completely removed and the scale on the workpiece surface cannot be efficiently processed online. Compared with the traditional method, the laser removal process does not require chemical substances or a large amount of mechanical force, has a high removal efficiency, and will not cause damage to the base material.

[0285] In addition, in combination with the precise adjustment of laser power and scanning speed, the present invention can precisely process mill scale with different thicknesses and properties. By dynamically adjusting process parameters, it can ensure that the removal process of mill scale always remains in the best state, avoiding overheating or damaging the surface of the wire rod.

[0286] In addition, the present invention integrates an advanced oxide layer thickness identification model and terahertz technology to real-time monitor the thickness distribution of the oxide layer on the surface of the wire rod, and dynamically adjusts the laser process parameters according to the actual situation. This real-time feedback mechanism ensures that the removal effect of mill scale during the laser treatment process always remains at the best level.

[0287] In addition, by combining with the material properties of the wire rod, the properties of the mill scale, and laser technology, the present invention can adapt to wire rod materials of different types and sizes. Whether it is a thin oxide layer or a thick oxide layer, ideal removal effects can be obtained by flexibly adjusting process parameters.

[0288] In addition, by combining image recognition technology and negative pressure collection technology, the present invention can real-time detect the removal effect and perform dynamic optimization to ensure the thoroughness and uniformity of the removal of mill scale, further improving the automation and intelligence level of the process.

[0289] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for removing wire rod iron oxide scale online based on laser technology, characterized in that: The following steps are involved: S1. According to the temperature distribution and displacement distribution caused by laser heating, the relationship between the detachment stress and the adhesion force is established, the relationship between different process parameters and the oxide scale removal effect is determined, and the initial laser process parameters are determined in combination with the current wire rod material properties and oxide scale performance; S2. Using the oxide layer thickness recognition model combined with terahertz technology, the thickness distribution of the oxide layer on the wire rod surface is identified, and the initial laser process parameters are dynamically adjusted according to the thickness distribution of the oxide layer on the wire rod surface to obtain the optimal laser process parameters based on the thickness distribution; S3. Based on the optimal laser process parameters, the iron oxide scale on the surface of the wire rod is laser treated by a rotating laser head, and the metal powder is collected by negative pressure technology. The image recognition technology is used to detect the removal effect of the iron oxide scale on the surface of the wire rod, and the model parameters are dynamically optimized according to the detection results.

2. The method for online removal of wire rod iron scale based on laser technology according to claim 1 is characterized in that: The method of establishing the relationship between the detachment stress and the adhesion force according to the temperature distribution and displacement distribution caused by laser heating, determining the relationship between different process parameters and the oxide scale removal effect, and determining the initial laser process parameters in combination with the current wire rod material properties and oxide scale performance includes the following steps: S11. According to the laser power, material properties and heat conduction characteristics, a temperature distribution model caused by laser heating is established; based on the temperature gradient caused by laser heating and the thermal expansion characteristics of the material, a displacement and stress distribution model caused by laser heating is established; S12, determining the detachment stress between the iron oxide scale and the substrate according to the thermal stress, and calculating the adhesion between the iron oxide scale and the substrate, and determining the critical value between the detachment stress and the adhesion according to the calculation results of the adhesion and the detachment stress; S13. Use finite element analysis to determine the relationship between laser parameters, material parameters and the cleaning and removal effect of iron oxide scale, and determine the initial laser process parameters based on the current wire rod material properties and oxide scale performance.

3. The method for online removal of wire rod iron scale based on laser technology according to claim 2 is characterized in that: The expression of the temperature distribution model caused by laser heating is: ； The expression of the displacement distribution model caused by the laser is: ； ； In the formula, T represents temperature, t represents time, represents the Laplace operator of temperature, represents the thermal diffusion coefficient of the material, Q(r) represents the distribution of laser power at position r, ρ is the density of the material, c represents the specific heat capacity of the material, represents thermal stress, represents the temperature change, E represents the elastic modulus, represents the thermal expansion coefficient of the material, u(x,y,z) represents the distribution of displacement in space, and L represents the length of the path.

4. The method for online removal of wire rod iron scale based on laser technology according to claim 2, characterized in that: The formula for calculating the adhesion between the iron oxide scale and the substrate is: ； In the formula, F represents the adhesion between the iron oxide scale and the substrate. It represents the adhesion strength between the iron oxide scale and the substrate, and A represents the contact area between the iron oxide scale and the substrate.

5. The method for online removal of wire rod iron oxide scale based on laser technology according to claim 2, characterized in that: The method of using the finite element analysis method to determine the relationship between laser parameters, material parameters and the cleaning and removal effect of iron oxide scale, and combining the current wire rod material properties and oxide scale performance, determining the initial laser process parameters includes the following steps: S131. Based on the temperature distribution model, displacement and stress distribution model, combined with the critical value between the detachment stress and the adhesion force, simulate the influence of different laser process parameters on the removal effect of the oxide scale, and determine the basic laser process parameters according to the simulation results; S132. Obtain the property data of the wire rod material to be processed, the property data of the iron oxide scale and the laser parameter data, optimize the process conditions of the laser power and scanning speed in the basic laser process parameters through finite element simulation, and determine the initial laser process parameters according to the optimization results.

6. The method for online removal of wire rod iron scale based on laser technology according to claim 1, characterized in that: The method of using the oxide layer thickness recognition model in combination with terahertz technology to recognize the thickness distribution of the oxide layer on the surface of the wire rod, and dynamically adjusting the initial laser process parameters according to the thickness distribution of the oxide layer on the surface of the wire rod to obtain the optimal laser process parameters based on the thickness distribution includes the following steps: S21, based on a pre-constructed oxide layer thickness identification model, identifying the thickness distribution of the oxide layer on the surface of the wire rod to be processed, and obtaining a first distribution result of the oxide layer thickness; S22, using terahertz technology to identify the thickness distribution of the oxide layer on the surface of the wire rod to be processed, and obtain a second distribution result of the oxide layer thickness; S23, determining the thickness distribution of the oxide layer on the surface of the wire rod to be processed based on the first distribution result and the second distribution result of the oxide layer thickness; S24. According to the thickness distribution of the oxide layer on the surface of the wire rod to be processed, the initial laser process parameters at different positions on the wire rod surface are dynamically adjusted to obtain the optimal laser process parameters based on the thickness of the oxide layer.

7. The method for online removal of wire rod iron scale based on laser technology according to claim 6, characterized in that: The method of identifying the thickness distribution of the oxide layer on the surface of the wire rod to be processed based on the pre-constructed oxide layer thickness identification model and obtaining a first distribution result of the oxide layer thickness comprises the following steps: S211, obtaining a surface image of the wire rod to be processed and performing preprocessing, analyzing the brightness difference in the surface image, extracting the grayscale values ​​of different areas on the surface of the oxide layer, calculating the local brightness distribution in the surface image, obtaining the average brightness of the local area by a sliding window method, and calculating the global brightness distribution to generate a grayscale histogram; S212, extracting the texture features of the surface image using the local binary pattern method, and identifying the boundary between the oxide layer and the substrate using the edge detection method; based on a pre-constructed oxide layer thickness recognition model, outputting the oxide layer thickness corresponding to the texture features of the surface image, and obtaining a first distribution result of the oxide layer thickness at different positions on the wire rod surface.

8. The method for online removal of wire rod iron scale based on laser technology according to claim 6, characterized in that: The method of using terahertz technology to identify the thickness distribution of the oxide layer on the surface of the wire rod to be processed and obtaining a second distribution result of the oxide layer thickness comprises the following steps: S221, using ultrashort pulse terahertz waves to irradiate the surface of the wire rod to be processed, measuring the reflection, transmission or absorption of the oxide layer to the terahertz waves, and recording the amplitude, phase and time delay information of the terahertz waves; S222. Calculate the reflection and transmission time difference of the terahertz wave at different oxide layer thicknesses by time domain analysis, convert the time domain signal into a frequency domain signal by Fourier transform, extract the spectral characteristics of different frequencies, and calculate the absorption and reflectivity of the oxide layer to the terahertz wave of a specific frequency to determine its thickness information; S223. Based on the transmission or reflection time difference of the terahertz wave and in combination with the refractive index of the oxide layer, the thickness of the oxide layer is calculated, and a mathematical model is used to invert the thickness distribution of the oxide layer to obtain a second distribution result of the oxide layer thickness at different positions on the wire rod surface.

9. The method for online removal of wire rod iron scale based on laser technology according to claim 6, characterized in that: Determining the thickness distribution of the oxide layer on the surface of the wire rod to be processed based on the first distribution result and the second distribution result of the oxide layer thickness comprises the following steps: S231, obtaining a first distribution result and a second distribution result of the oxide layer thickness at different positions on the wire rod surface, and performing data alignment and preprocessing; establishing a spatial correspondence between the first distribution result data and the second distribution result data according to a meshing method of the wire rod surface, and spatially aligning the thickness distribution data of the first distribution result with the second distribution result using a feature matching method; S232. The first distribution result data and the second distribution result data are merged by using a weighted average method in combination with a preset weight coefficient to obtain a final thickness value, and a distribution diagram of the oxide layer thickness at different positions on the wire rod surface is generated according to the final thickness value.

10. A system for online removal of wire rod iron scale based on laser technology, used to implement the steps of the method for online removal of wire rod iron scale based on laser technology according to any one of claims 1 to 9, characterized in that: It includes an initial laser process parameter determination module, a laser process parameter optimization module and an oxide scale online removal module; The initial laser process parameter determination module is used to establish the relationship between the detachment stress and the adhesion force according to the temperature distribution and displacement distribution caused by laser heating, determine the relationship between different process parameters and the iron oxide scale removal effect, and determine the initial laser process parameters in combination with the current wire rod material properties and oxide scale performance; The laser process parameter optimization module is used to identify the thickness distribution of the oxide layer on the surface of the wire rod by using the oxide layer thickness recognition model combined with terahertz technology, and dynamically adjust the initial laser process parameters according to the thickness distribution of the oxide layer on the surface of the wire rod to obtain the optimal laser process parameters based on the thickness distribution; The online iron oxide scale removal module is used to laser process the iron oxide scale on the surface of the wire rod using a rotating laser head based on the optimal laser process parameters, and collect metal powder through negative pressure technology; use image recognition technology to detect the removal effect of the iron oxide scale on the surface of the wire rod, and dynamically optimize the model parameters according to the detection results.

Images