Laser-assisted rolling forming fillet radius control method based on prediction model

Abstract

The invention provides a laser-assisted rolling forming fillet radius control method based on a prediction model, and the method comprises the steps: carrying out an ultrahigh-strength steel laser-assisted rolling fillet experiment through designing a plurality of groups of laser power, scanning rate and light spot offset parameter combinations, measuring an actual fillet radius, and analyzing a quantitative relation between the actual fillet radius and process parameters. Based on this, a process parameter prediction model is established by using machine learning. When a specific target fillet radius is needed, reverse calculation can be carried out through the model, and a process parameter combination is directly obtained. The parameter combination is used for machining, and then a target part can be obtained. The quantitative prediction model is applied to the laser-assisted rolling forming process for the first time, accurate control over the small fillet radius of the ultra-high-strength steel is achieved, the fillet radius can be set as needed within the range of 1.0-2.5 times of the plate thickness, and the control precision of + / -5% is guaranteed. According to the method, the limitation of a traditional laser-assisted rolling forming process is broken through, the production efficiency and the part quality are greatly improved, and the method has wide industrial application prospects.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of roll forming, and in particular to a laser-assisted roll forming fillet radius control method based on a prediction model. BACKGROUND

[0002] Roll forming process is a new type of sheet bending forming process developed on the basis of traditional roll forming process. It can finally obtain the target part through programmed control of robot forming trajectory for multi-pass progressive processing of sheet. Compared with traditional roll forming process, it has higher flexibility and can form parts with different cross-sectional shapes through standard tool heads. At the same time, it has high flexibility and can meet the needs of small batch and customization. Based on the idea of progressive forming, laser-assisted robot precision roll forming process realizes the forming of parts with specific cross-sectional profiles through multi-pass bending. The sheet accumulates plastic deformation during bending to complete part forming. Due to multi-pass forming, the forming load is relatively smaller than that of one-time forming process. For metal materials with poor forming performance such as ultra-high strength steel, titanium alloy and aluminum alloy, the laser head installed on the side arm of the robot preheats the bending area of the sheet during roll forming, softens the material in the deformation area, and improves the forming performance of the material.

[0003] However, in the existing laser-assisted roll forming technology, although small fillet forming can be achieved, it is difficult to accurately, quantitatively and flexibly control the fillet radius, thereby meeting the design requirements of different parts of the same part or different batches of parts for the diversification of the fillet radius. Moreover, for some thick sheets, the pursuit of small fillet forming may cause excessive stress and strain concentration at the fillet, leading to cracking and other problems. Therefore, it is necessary to find a method to control the fillet radius.

[0004] Existing documents such as Liu et al. "Research on Laser-Assisted Roll Forming of Martensitic Steel" study the effects of laser power, scanning speed and other parameters on the forming results, but only stay at the stage of exploring the influence law and fail to propose specific prediction control methods. Min et al. in "Laser-Assisted Roll Forming of Ultra-High Strength Steel QP1180" (published in the journal Materials in 2023) also studied the effects of process parameters, but lacked quantitative description of these effects and prediction models. These studies have failed to achieve accurate and controllable fillet radius. SUMMARY

[0005] In view of the above problems, a laser-assisted roll forming fillet radius control method based on a prediction model is proposed to overcome the above problems or at least partially solve the above problems, comprising:

[0006] establishing a process parameter prediction model, the process parameter prediction model being a mathematical function R = f(P, V, L) representing a relationship between a process parameter and a formed fillet radius in the laser-assisted roll forming process, where R is the fillet radius, P is the laser power, V is the laser scanning speed, L is the spot offset distance, f is a function relationship obtained by fitting experimental data, and the process parameter prediction model is established by a process including:

[0007] setting a plurality of groups of process parameter combinations including different laser power, laser scanning speed, spot offset distance, and roll forming force;

[0008] performing laser-assisted roll forming experiments of different process parameter combinations on a plurality of groups of metal plates;

[0009] measuring actual fillet radii of formed parts obtained in each group of experiments;

[0010] establishing a process parameter prediction model based on a relationship between the laser power, the laser scanning speed, the spot offset distance, and the actual fillet radius;

[0011] solving a target process parameter combination using the process parameter prediction model according to a target fillet radius;

[0012] performing laser-assisted roll forming operations of the target process parameter combination on a target metal plate to obtain a formed part;

[0013] wherein the target fillet radius ranges from 1T to 2.5T, T is the thickness of the target metal plate, the metal plate is ultra-high strength steel, the tensile strength of the ultra-high strength steel is not less than 1500MPa, the elongation is not less than 5%, and the thickness ranges from 1mm to 2mm, and the process parameter prediction model is a quadratic polynomial model.

[0014] Optionally, before performing laser-assisted roll forming experiments of different process parameter combinations on a plurality of groups of metal plates and performing laser-assisted roll forming operations of the target process parameter combination on a target metal plate, the method further includes:

[0015] polishing the metal plate in a rolling direction using sandpaper and cleaning the metal plate twice with deionized water and drying the metal plate.

[0016] Optionally, the process parameter prediction model is:

[0017]

[0018] wherein R is the target fillet radius, P is the laser power, V is the laser scanning speed, L is the spot offset distance, and b0-b9 are coefficients of the process parameter prediction model.

[0019] Optionally, the fitting degree of the process parameter prediction model is greater than 0.95.

[0020] Optionally, the method further comprises:

[0021] measuring the formed fillet radius using a laser scanner and calculating an error between the formed fillet radius and the target fillet radius;

[0022] in the case that the error is greater than or equal to 2%, adjusting the process parameter prediction model coefficients based on the error.

[0023] Optionally, performing a laser-assisted roll forming operation on the target metal sheet using the target process parameter combination, comprising:

[0024] fixing the metal sheet to the fixture and inputting the target process parameter combination to the control system;

[0025] controlling the laser emitter to irradiate and heat the pre-formed bending area of the metal sheet, while using the roll forming tool head to roll form the metal sheet according to the target part shape.

[0026] Optionally, the spot irradiated on the surface of the metal sheet by the laser emitter is rectangular, with a size of 2mm x 4mm.

[0027] Optionally, the laser power range of the laser emitter is 1kw to 2kw, and the scanning rate range of the laser is 10mm / s to 30mm / s.

[0028] Optionally, during the forming operation, the forming force applied by the roll forming tool head ranges from 1000N to 4000N.

[0029] The present application also provides a laser-assisted roll forming device based on a prediction model, the device comprising:

[0030] a fixture, the fixture being used to fix the metal sheet;

[0031] a kuka six-degree-of-freedom robot, the kuka six-degree-of-freedom robot comprising a laser emitter and a roll forming tool head, the laser emitter being used to irradiate and heat the pre-formed bending area of the metal sheet, and the roll forming tool head being used to roll form the metal sheet according to the target part shape.

[0032] The embodiment of the present application sets a plurality of groups of process parameter combinations containing different laser power, laser scanning rate and spot offset; performs laser-assisted roll forming experiments of different process parameter combinations on a plurality of groups of metal plates; measures the actual fillet radius of the formed parts obtained in each group of experiments; establishes a process parameter prediction model based on the relationship between laser power, laser scanning rate, spot offset and actual fillet radius; solves the target process parameter combination by using the process parameter prediction model according to the target fillet radius; performs laser-assisted roll forming operation of the target process parameter combination on the target metal plate to obtain the formed part, deeply analyzes the quantitative relationship between laser power, laser scanning rate, spot offset distance and target fillet radius, and constructs a model by means of machine learning, so that the operator does not need to experience trial and error debugging in the traditional process, but directly obtains the optimal parameter combination by model reverse calculation, realizing flexible and accurate control of the fillet radius. BRIEF DESCRIPTION OF DRAWINGS

[0033] In order to more clearly illustrate the technical solutions of the present application, the following will briefly introduce the drawings needed to be used in the description of the present application. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0034] Figure 1 is a schematic diagram of a laser-assisted roll forming device for processing steel plates adopted by the embodiment of the present application;

[0035] Figure 2 is a schematic diagram of the appearance of a laser-assisted roll forming part provided by the embodiment of the present application;

[0036] Figure 3 is a schematic diagram of the appearance of a fillet part of a laser-assisted roll forming part provided by the embodiment of the present application;

[0037] Figure 4 is a schematic diagram of the scanning appearance of a laser scanner when measuring a fillet provided by the embodiment of the present application;

[0038] Figure 5 is a schematic diagram of the appearance of a fillet part of a laser-assisted roll forming part provided by the present application. DETAILED DESCRIPTION

[0039] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the following will further describe the present application in detail with reference to the drawings and specific embodiments. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0040] The embodiment of the present application provides a laser-assisted roll forming round radius control method based on a prediction model, which comprises the following steps:

[0041] A process parameter prediction model is established, which is a mathematical function R = f(P, V, L) representing the relationship between the process parameters and the forming round radius in the laser-assisted roll forming process, and is used to predict the quantitative relationship between the laser process parameters, the roll forming process parameters and the forming round radius in the laser-assisted roll forming process. Wherein R is the round radius, P is the laser power, V is the laser scanning speed, L is the spot offset distance, f is the function relationship obtained by fitting the experimental data, and the establishment process of the process parameter prediction model comprises the following steps: setting a plurality of process parameter combinations containing different laser power, laser scanning rate, spot offset and roll forming force; performing laser-assisted roll forming round forming experiments of different process parameter combinations on a plurality of metal plates; measuring the actual round radius of the formed parts obtained in each group of experiments; and establishing the process parameter prediction model based on the relationship between the laser power, the laser scanning rate, the spot offset and the actual round radius.

[0042] According to the target round radius, the target process parameter combination is solved by using the process parameter prediction model; and the laser-assisted roll forming round forming operation is performed on the target metal plate by using the target process parameter combination to obtain a formed part; wherein the target round radius ranges from 1T to 2.5T, T is the thickness of the target metal plate, the metal plate is ultra-high strength steel, the tensile strength of the ultra-high strength steel is not less than 1500MPa, the elongation is not less than 5%, the thickness ranges from 1mm to 2mm, and the process parameter prediction model is a quadratic polynomial model.

[0043] In order to systematically explore the influence law of process parameters on the forming round radius and realize accurate control, the embodiment of the present application firstly configures a plurality of different process parameter combinations, each of which contains specific values of the three variables of laser power, laser scanning rate and spot offset. Subsequently, laser-assisted roll forming round forming experiments corresponding to these preset process parameter combinations are respectively implemented on a plurality of metal plate samples of the same specification. After the experiment is completed, the actual round radius of the formed parts obtained in each group of experiments is measured. Based on this, the internal correlation and influence law between the input parameters such as laser power, laser scanning rate and spot offset and the round radius are analyzed and quantified in depth, and then a process parameter prediction model capable of predicting the forming round radius is constructed.

[0044] Specifically, a set of experimental schemes containing different parameter combinations can be designed through orthogonal or response surface experimental design methods; each group of experiments in the experimental scheme is executed, and the corresponding round radius value is measured; and a mathematical prediction model is established based on the experimental data by using multivariate regression analysis or machine learning algorithm.

[0045] Further, when a part with a specific target corner radius is needed, the target value is input into the established prediction model, and the process parameter combination that can achieve the target radius, i.e., the target process parameter combination, is inversely solved. Finally, the laser-assisted roll forming operation is performed on the final target metal sheet using the determined target process parameter combination, so that the formed part meeting the preset corner size requirement is successfully manufactured.

[0046] wherein the target corner radius ranges from 1T to 2.5T, T is the thickness of the target metal sheet, the metal sheet is an ultra-high strength steel, the tensile strength of the ultra-high strength steel is not less than 1500MPa, the elongation is not less than 5%, the thickness ranges from 1mm to 2mm, and the process parameter prediction model is a quadratic polynomial model.

[0047] It can be understood that the ultra-high strength steel with a tensile strength of ≥1500MPa has limited plasticity reserve, and too much pursuit of small corner forming may cause excessive stress and strain concentration at the corner, thereby causing cracking and a series of problems, i.e., a corner smaller than 1T may cause the fiber strain on the bending outside to exceed the material ductility limit, thereby causing micro-cracks or even breakage; and a corner larger than 2.5T, although it can avoid cracking, may lose the advantage of lightweight structure.

[0048] In the embodiments of the present application, in order to reduce the error of the corner radius after forming, laser-assisted roll forming experiments of different process parameter combinations can be performed on multiple groups of metal sheets, and before performing the laser-assisted roll forming operation on the target metal sheet using the target process parameter combination, the metal sheet is unidirectionally polished along the rolling direction using sandpaper and then cleaned twice with deionized water and dried.

[0049] Specifically, a sandpaper of a specific mesh number (which can be selected according to the material of the plate and the initial surface state) can be used to perform a one-way and uniform polishing operation along the inherent rolling direction of the metal plate, which can effectively remove the oxidation layer, small scratches, oil stains and other pollutants that can affect the subsequent process on the surface of the plate, and at the same time make the surface obtain a consistent and controllable roughness, providing an ideal basic condition for the subsequent laser absorption and rolling contact. It should be noted to maintain the singleness of the polishing direction to avoid complex surface texture. After polishing, the plate surface is immediately cleaned thoroughly with high-purity deionized water. The cleaning process can be performed twice to completely remove the metal dust, sandpaper particle residues and any possible residual oil or impurities generated during the polishing process. Each cleaning needs to ensure that the water flow fully covers the entire treatment surface. After the cleaning step is completed, the metal plate is placed in a clean environment for complete drying, which can be naturally dried in a clean bench or low-dust environment or dried with clean and oil-free compressed air / nitrogen. It can ensure that the plate surface is free of any water stains or liquid residues, preventing water from interfering with the subsequent laser processing or rolling process.

[0050] In the embodiments of the present application, the process parameter prediction model can be:

[0051]

[0052] wherein R is the target fillet radius, P is the laser power, V is the laser scanning rate, L is the spot offset, and b0-b9 are the process parameter prediction model coefficients.

[0053] It can be understood that, in order to ensure the reliability of the model, the fitting degree of the model needs to be calculated, and when the fitting degree of the process parameter prediction model is greater than 0.95, it is proved that the model can be used to inversely calculate the process parameters.

[0054] In the embodiments of the present application, the laser scanner can also be used to measure the formed fillet radius and calculate the error between the formed fillet radius and the target fillet radius; in the case that the error is greater than or equal to 2%, the process parameter prediction model coefficients are adjusted based on the error.

[0055] To ensure the accuracy of the formed fillet size and achieve closed-loop optimization control of the process parameters, after the laser-assisted roll forming operation is completed, a high-precision laser scanner can also be used to measure the three-dimensional topography of the formed fillet area of the metal plate to obtain the actual fillet radius. Subsequently, the actual measured value is compared with the target fillet radius, and the relative error between the two is calculated. When it is determined that the calculated error value is greater than or equal to the preset 2% threshold, it indicates that the prediction result of the current process parameter prediction model is not accurate enough. At this time, the system will automatically or according to the error information, intelligently adjust or optimize update the related coefficients b0~b9 of the process parameter prediction model. This model coefficient dynamic correction mechanism based on the measured error can improve the prediction accuracy of the model for future process parameters, and achieve higher precision and more stable and reliable process parameter output results.

[0056] In the embodiments of the present application, the laser-assisted roll forming operation on the target metal plate using the target process parameter combination can be performed in the following way:

[0057] The metal plate is fixed to the clamp, and the target process parameter combination is input to the control system; the laser emitter irradiates and heats the pre-formed bending area of the metal plate, and the roll forming tool head is used to roll form the metal plate according to the target part shape.

[0058] Specifically, the metal plate to be processed can be clamped on the clamp first to ensure that the plate does not shift or vibrate during subsequent processing. Then, through the man-machine interface or the upper computer software, the selected target process parameter combination is input to the integrated control system. After receiving the instruction, the control system synchronously coordinates the laser emitter and the roll forming tool head to work: the laser emitter moves above the predetermined bending area (i.e. the part where the fillet is planned to be formed) of the metal plate according to the preset parameters, and emits a focused laser beam to scan and irradiate, the purpose of which is to form a controllable high-temperature softening zone on the surface layer of the plate, significantly reducing the deformation resistance of the area. At the same time, at a set distance behind the laser beam, the roll forming tool head is driven by a servo motor to apply a controllable radial pressure to the metal area in the laser-induced softening state according to the target part shape and perform roll forming.

[0059] In actual application, the spot irradiated by the laser emitter on the surface of the metal plate is rectangular, and the size can be 2mm x 4mm. The laser power range of the laser emitter can be 1kw to 2kw, and the scanning rate of the laser can be 10mm / s to 30mm / s. During the forming operation, the forming force applied by the roll forming tool head can be in the range of 1000N to 4000N.

[0060] To optimize the energy input distribution and adapt to different plate thicknesses and forming requirements, the laser spot shape used in the embodiments of the present application can be rectangular. The rectangular spot uniformly irradiates the surface of the pre-formed target bending area of the metal plate, and its size can be configured as 2mmx4mm. It can be understood that such a rectangular spot helps to form a more uniform and wider heat-affected zone on the rolling path, promoting material plastic flow. The laser power can be flexibly adjusted between 1kW and 2kW according to the material properties and plate thickness to provide sufficient heat to make the surface layer material reach the appropriate softening state, while avoiding excessive melting or thermal damage; the scanning rate of the laser beam can be set in the range of 10mm / s to 30mm / s, and this rate range needs to be accurately matched with the laser power and rolling speed to ensure that the heat input rate is coordinated with the rolling deformation rhythm, both ensuring sufficient softening effect and maintaining high production efficiency. In the synchronous rolling forming operation in the laser heat softening area, the forming force applied by the rolling tool head can be set in the range of 1000N to 4000N. The selection of this force value range can be determined according to factors such as the target fillet radius, plate thickness and yield strength, aiming to provide sufficient plastic deformation driving force to make the actual profile of the softened material close to the target profile, while avoiding problems such as plate crushing or residual stress concentration caused by excessive pressure, ensuring the forming accuracy and dimensional accuracy of the part.

[0061] The present application also provides a laser-assisted rolling forming device based on a prediction model, the device comprising:

[0062] a clamp for fixing the metal plate;

[0063] a Kuka six-degree-of-freedom robot, which includes a laser emitter for irradiating and heating the pre-formed bending area of the metal plate and a rolling tool head for rolling forming the metal plate according to the shape of the target part.

[0064] It should be noted that the laser emitter and the rolling tool head need to move simultaneously during processing without relative displacement.

[0065] The embodiments of the present application construct a systematic scientific method for converting abstract fillet radius design requirements into concrete and executable process parameters, changing the production mode of relying on experience and trial-and-error in traditional laser-assisted rolling forming technology. This method deeply analyzes the quantitative relationship between laser power, laser scanning rate, spot offset distance and target fillet radius through experimental design methods such as response surface method, and constructs a model by means of machine learning, so that the operator does not need to go through the trial-and-error debugging in the traditional process, but directly obtains the optimal parameter combination through the model inverse calculation, realizing the flexible and accurate control of the fillet radius.

[0066] In terms of production efficiency and quality control, the open-loop control mechanism based on the prediction model shortens the new product process debugging period by more than 60%. Taking a certain automobile beam as an example, the traditional debugging needs to consume 3-5 days, and the present application can complete parameter determination in only 4 hours through model pre-computation. The matching closed-loop feedback system uses a line laser scanner to collect profile data of the formed part in real time, which further eliminates the influence of factors such as material batch difference and environmental temperature fluctuation, and ensures the stability and consistency of the production process, with high yield.

[0067] In terms of technology inheritance and intelligent manufacturing construction, the present application changes the traditional manufacturing experience inheritance mode by converting implicit process knowledge into explicit mathematical model. The machine learning iteration mechanism built in the model can automatically absorb each batch of production data to continuously optimize the coefficient matrix, so that the prediction accuracy is continuously improved as the production scale expands. This data-driven technology sedimentation mode not only builds a core technology barrier that is difficult to replicate for the enterprise, but also becomes a key engine to promote the construction of intelligent factory.

[0068] For the convenience of understanding, the present application also provides the following examples and comparative examples:

[0069] Example 1

[0070] A laser-assisted roll forming round corner control method based on a prediction model is provided. In the experiment, a part is made of martensitic steel, different process parameters are set through multiple experiments, and the round corner radius of the formed part under each process parameter condition is measured by a laser scanner. A prediction model is established, and each process parameter under the condition of processing a specific required round corner is inversely solved through the prediction model, so as to finally realize accurate control of the round corner radius of the processed part, with an error controlled within 2%.

[0071] In this embodiment, the tensile strength of the ultra-high strength steel used is about 1500 MPa, the room temperature elongation is not less than 5%, and the thickness is about 1 mm.

[0072] Taking an L-shaped steel part as an example, the laser-assisted roll forming round corner radius control method based on a prediction model specifically includes the following steps:

[0073] S1, determine the process parameters related to the prediction model;

[0074] S2, design an experimental scheme;

[0075] S3, fix the sheet on the tooling and set the relevant process parameters;

[0076] S4, perform multiple laser-assisted roll forming experiments and measure the round corners;

[0077] S5, construct a mathematical model according to the experimental data;

[0078] S6, according to the model inverse solution under the specific fillet requirements of process parameters;

[0079] S7, the realization of specific fillet requirements of parts roll forming.

[0080] In this embodiment, the laser assisted roll forming process can be used as Figure 1 : the steel plate 4 is clamped and fixed by the clamp 1, and then the roll forming tool head 2 is used to move according to the target part shape, and the steel plate is roll formed. In the process of roll forming, the laser head 3 emitting laser is controlled to scan synchronously to assist heating in the bending forming area. Figure 1 The arrow in the figure indicates the direction of movement.

[0081] In step S1, the selected independent variable parameters are laser power (P), scanning speed (V), and spot offset per pass (L). Each factor is set to three levels.

[0082] In step S2, a total of 10 experimental schemes are designed by using the Box-Behnken experimental design method.

[0083] In step S4, the appearance of the part after laser-assisted roll forming is shown in Figure 2 ;

[0084] In step S4, the appearance of the part after laser-assisted roll forming is shown in Figure 3 ;

[0085] In step S4, the laser scanner measures the fillet as shown in Figure 4 ;

[0086] In step S5, the selected mathematical model is a quadratic polynomial model:

[0087]

[0088] Where R is the target fillet radius, P is the laser power, V is the laser scanning speed, L is the spot offset, and b0-b9 are the process parameter prediction model coefficients.

[0089] In step S7, the target fillet is set to 1.10T, 1.80T, and 2.30T.

[0090] From the experimental data obtained by the above method, when the remaining conditions do not change, the higher the laser power, the slower the scanning speed, the smaller the fillet radius of the formed part; the larger the spot offset, the larger the fillet radius of the formed part.

[0091] According to the analysis, it can be obtained that the three main parameters (laser power, scanning speed, and spot offset) in the prediction model are highly significant, indicating that they have a great influence on the corner radius; two cross terms are significant, indicating that there is an interaction effect between the process parameters, which should be considered comprehensively in the experiment; two quadratic terms are significant, indicating that the change of the corner radius presents a nonlinear trend. The overall model is significant, and the fitting degree R2=0.965, proving that the model is reliable.

[0092] This embodiment establishes a prediction model between the corner radius and the laser power, scanning speed, and spot offset, and verifies the accuracy of the model in realizing specific corner roll forming. The results show that the error of the corner radius obtained by using the process parameters obtained by the inverse solution of the model for roll forming is within ±1.8% of the target value, meeting the process requirements.

[0093] The process parameters of each group of experiments are shown in Table 1:

[0094] Experiment No. P(W) V(mm / s) L(mm) Measured R 1 1500 20 0.5 1.47T 2 1200 20 0.5 1.78T 3 1800 20 0.5 1.26T 4 1500 20 0.0 1.31T 5 1500 20 1.0 1.71T 6 1200 10 0.5 1.63T 7 1800 10 0.5 1.09T 8 1200 30 0.5 1.94T 9 1800 30 0.5 1.41T 10 1500 30 0.0 1.53T

[0095] Table 1

[0096] The process parameters for processing the target corner part are shown in Table 2:

[0097] Targeted Fillet R(T) Laser Power P(W) Scan Rate V(mm / s) Offset L(mm) 1.10 1720 11 0.1 1.80 1420 22 0.6 2.30 1280 28 0.9

[0098] Table 2

[0099] Embodiment 2

[0100] A laser-assisted roll forming corner control method based on a prediction model is used in the experiment. Martensitic steel is used to make parts, and different process parameters are set through multiple experiments. The corner radius of the formed parts under each process parameter condition is measured by a laser scanner, a prediction model is established, and each process parameter under the condition of processing a specific corner is inversely solved through the prediction model. Finally, the accuracy of the corner radius of the processed parts is controlled, and the error is controlled within 2%.

[0101] In this embodiment, the tensile strength of the ultra-high strength steel used is about 1500 MPa, the room temperature elongation is not less than 5%, and the thickness is about 1.5 mm.

[0102] Taking an L-shaped steel part as an example, the laser-assisted roll forming corner radius control method based on a prediction model specifically includes the following steps:

[0103] S1, determine the process parameters related to the prediction model;

[0104] S2, design an experimental scheme;

[0105] S3, fix the sheet on the tooling and set the relevant process parameters;

[0106] S4, performing a plurality of sets of laser-assisted roll forming experiments and measuring the fillets;

[0107] S5, constructing a mathematical model according to experimental data;

[0108] S6, inversely solving process parameters under specific fillet requirements according to the model;

[0109] S7, realizing roll forming of a part with specific fillet requirements.

[0110] In this embodiment, a laser-assisted roll forming process as shown in Figure 1 may be used: a steel plate 4 is clamped and fixed by a clamp 1, and then a roll forming tool head 2 is used to move according to the shape of a target part to perform roll forming on the steel plate. During the roll forming process, a laser head 3 emitting laser is controlled to scan synchronously to assist in heating the bending forming area. Figure 1 The arrow indicates the direction of movement.

[0111] In step S1, the selected independent variable parameters are laser power (P), scanning speed (V), and spot offset per pass (L), and each factor is set to three levels.

[0112] In step S2, a Box-Behnken experimental design method is used, and a total of 10 experimental schemes are designed.

[0113] In step S4, the appearance of the part after laser-assisted roll forming is as shown in Figure 2 .

[0114] In step S4, the appearance of the part at the fillet after laser-assisted roll forming is as shown in Figure 3 .

[0115] In step S4, the measurement of the fillet by a laser scanner is as shown in Figure 4 .

[0116] In step S5, the selected mathematical model is a quadratic polynomial model:

[0117]

[0118] where R is the target fillet radius, P is the laser power, V is the laser scanning speed, L is the spot offset, and b0-b9 are process parameter prediction model coefficients.

[0119] In step S7, the target fillet is set to 1.20T, 1.90T, and 2.40T, respectively.

[0120] From the analysis of the experimental data obtained from the above method, when the remaining conditions do not change, the higher the laser power, the slower the scanning rate, the smaller the corner radius of the formed part; the larger the spot offset, the larger the corner radius of the formed part.

[0121] According to the analysis, it can be obtained that the three main parameters (laser power, scanning rate, and spot offset) in the prediction model are highly significant, which indicates that they have a great influence on the corner radius; two cross terms are significant, which indicates that there is an interaction effect between the process parameters, which should be considered comprehensively in the experiment; two quadratic terms are significant, which indicates that the change of the corner radius presents a nonlinear trend. The model is significant in general, and the fitting degree R2=0.978, which proves that the model is reliable.

[0122] This embodiment establishes a prediction model between the corner radius and the laser power, the scanning rate, and the spot offset, and verifies the accuracy of the model in realizing specific corner roll forming. The results show that the error of the corner radius obtained by using the process parameters obtained by the inverse solution of the model for roll forming is within ±1.3% of the target value, which meets the process requirements.

[0123] The process parameters of each group of experiments are shown in Table 3:

[0124] Experiment No. P(W) V(mm / s) L(mm) Measured R 1 1500 20 0.5 1.55T 2 1200 20 0.5 1.86T 3 1800 20 0.5 1.32T 4 1500 20 0.0 1.36T 5 1500 20 1.0 1.78T 6 1200 10 0.5 1.69T 7 1800 10 0.5 1.13T 8 1200 30 0.5 2.01T 9 1800 30 0.5 1.47T 10 1500 30 0.0 1.57T

[0125] Table 3

[0126] The process parameters for processing the target corner part are shown in Table 4:

[0127] Targeted Fillet R(T) Laser Power P(W) Scan Rate V(mm / s) Offset L(mm) 1.20 1780 10 0.1 1.90 1480 20 0.4 2.40 1340 26 0.7

[0128] Table 4

[0129] Embodiment 3

[0130] A laser-assisted roll forming corner control method based on a prediction model is used in the experiment. Martensitic steel is used to make parts. Different process parameters are set through multiple experiments, and the corner radius of the formed part under each process parameter condition is measured by a laser scanner. A prediction model is established, and each process parameter under the condition of processing a specific corner requirement is inversely solved through the prediction model. Finally, the corner radius of the processed part is accurately controlled, and the error is controlled within 2%.

[0131] In this embodiment, the tensile strength of the ultra-high strength steel used is about 1500 MPa, the room temperature elongation is not less than 5%, and the thickness is about 2 mm.

[0132] Taking an L-shaped steel part as an example, the laser-assisted roll forming corner radius control method based on the prediction model specifically includes the following steps:

[0133] S1, determine the process parameters related to the prediction model;

[0134] S2, design an experimental scheme;

[0135] S3, fix the sheet metal on the tooling and set the relevant process parameters;

[0136] S4, perform multiple sets of laser-assisted roll forming experiments and measure the fillets;

[0137] S5, construct a mathematical model according to the experimental data;

[0138] S6, according to the model, inversely solve the process parameters under specific fillet requirements;

[0139] S7, realize roll forming of parts with specific fillet requirements.

[0140] In this embodiment, the laser-assisted roll forming process as shown in Figure 1 is adopted: the steel plate 4 is clamped and fixed by the clamp 1, and then the roll forming tool head 2 is used to move according to the target part shape to perform roll forming on the steel plate. During the roll forming process, the laser head 3 that emits laser is controlled to scan synchronously to assist in heating the bending forming area. Figure 1 The arrows in the figure indicate the direction of movement.

[0141] In step S1, the selected independent variable parameters are laser power (P), scanning speed (V), and spot offset per pass (L), and each factor is set to 3 levels.

[0142] In step S2, the Box-Behnken experimental design method is used, and a total of 10 experimental schemes are designed.

[0143] In step S4, the appearance of the part after laser-assisted roll forming is shown in Figure 2 .

[0144] In step S4, the appearance of the part at the fillet after laser-assisted roll forming is shown in Figure 3 .

[0145] In step S4, the laser scanner measures the fillet as shown in Figure 4 .

[0146] In step S5, the selected mathematical model is a quadratic polynomial model:

[0147]

[0148] where R is the target fillet radius, P is the laser power, V is the laser scanning speed, L is the spot offset, and b0-b9 are the process parameter prediction model coefficients.

[0149] In step S7, the target fillets are set to 1.30T, 2.00T and 2.50T respectively.

[0150] From the analysis of the experimental data obtained from the above method, when the remaining conditions do not change, the higher the laser power, the slower the scanning speed, the smaller the fillet radius of the formed part; the larger the spot offset, the larger the fillet radius of the formed part.

[0151] According to the analysis, it can be obtained that the three main parameters (laser power, scanning speed and spot offset) in the prediction model are highly significant, which indicates that they have a great influence on the fillet radius; two cross terms are significant, which indicates that there is an interaction effect between the process parameters, which should be considered comprehensively in the experiment; two quadratic terms are significant, which indicates that the change of the fillet radius presents a nonlinear trend. The model is significant in general, and the fitting degree R2=0.969, which proves that the model is reliable.

[0152] This embodiment establishes a prediction model between the fillet radius and the laser power, scanning speed and spot offset, and verifies the accuracy of the model in realizing specific fillet roll forming. The results show that the error of the fillet radius obtained by using the process parameters obtained by the model inverse solution for roll forming is within ±1.5% of the target value, which meets the process requirements.

[0153] The process parameters of each group of experiments are shown in Table 5:

[0154] Experiment No. P(W) V(mm / s) L(mm) Measured R 1 1500 20 0.5 1.62T 2 1200 20 0.5 1.94T 3 1800 20 0.5 1.38T 4 1500 20 0.0 1.42T 5 1500 20 1.0 1.84T 6 1200 10 0.5 1.74T 7 1800 10 0.5 1.18T 8 1200 30 0.5 2.07T 9 1800 30 0.5 1.52T 10 1500 30 0.0 1.62T

[0155] Table 5

[0156] The process parameters when processing the target fillet part are shown in Table 6:

[0157] Targeted Fillet R(T) Laser Power P(W) Scan Rate V(mm / s) Offset L(mm) 1.30 1980 8 0.08 2.00 1680 15 0.3 2.50 1540 21 0.5

[0158] Table 6

[0159] Comparative example

[0160] The difference between this comparative example and the embodiment is that this comparative example uses the traditional empirical method to set the process parameters for processing to achieve target fillet control.

[0161] The experimental target, experimental equipment and measuring tools are the same as in the embodiment.

[0162] This comparative example does not use the prediction model to calculate the process parameters, but the operator confirms the process parameter combination based on previous processing experience after repeated trial adjustment within the process parameter requirement range:

[0163] For R=1.10T, power 1720W, scanning speed 12.5mm / s, offset 0.2mm;

[0164] For R = 1.80T, power 1480W, scan rate 20.5mm / s, offset 0.6mm;

[0165] For R = 2.30T, power 1300W, scan rate 27.5mm / s, offset 1.0mm;

[0166] The schematic diagram of the appearance of the round corner part of the laser-assisted roll forming part is shown in Figure 5 ;

[0167] The experimental comparison results are shown in Table 7:

[0168] Targeted Fillet R(T) Example 1 Measured R(T) Example 1 Error (%) Comparative Measured R(T) Comparative Error (%) 1.10 1.12 +1.8 1.07 -2.7 1.80 1.77 -1.7 1.72 -4.4 2.30 2.28 -0.9 2.21 -3.9

[0169] Table 7

[0170] The round corner radius of the laser roll forming part prepared by the examples and the comparative examples is analyzed to further illustrate the present application.

[0171] 1. The forming precision is significantly improved: in the comparative examples, the traditional empirical method is used to adjust the process parameters through multiple tests within the process parameter range, and the results show that when the target round corner is 1.10T, 1.80T and 2.30T, the actual round corner is 1.07T, 1.72T and 2.21T respectively, and the corresponding errors are -2.7%, -4.4% and -3.9% respectively; while the process parameters calculated by the prediction model of the present application are directly processed, the actual round corner error is controlled within ±1.7%, and the maximum error is not more than 1.7%, and the forming precision is greatly improved.

[0172] 2. The debugging efficiency is significantly improved: the empirical method used in the comparative examples needs to rely on the experience of the operator to adjust repeatedly, and it takes an average of 3-4 tests for each target round corner to achieve a relatively close value, and the whole debugging process takes a long time; while the present application method based on the prediction model only needs to input the target round corner, and through calculation, the process parameter combination meeting the requirements can be directly obtained, realizing one-time forming, greatly shortening the debugging time and significantly improving the production efficiency.

[0173] 3. The material utilization rate is improved, and the waste is reduced: the repeated test adjustment in the comparative examples leads to a high scrap rate of test pieces, and the material waste is obvious, while the present application method can form once and meet the target round corner requirement, the test piece qualification rate is high, effectively reducing the raw material loss, and the material utilization rate is significantly improved.

[0174] 4. Better process stability and consistency: In the comparative example, the forming fillet fluctuates greatly due to instability in parameter adjustment during the processing process, and the consistency between different batches is poor; while the process parameters calculated by the prediction model in the method are stable and reliable, the forming consistency is good, the fluctuation between batches is small, and the method can better meet the requirements of modern manufacturing industry for high consistency and high stability.

[0175] The above describes in detail a laser-assisted roll forming fillet radius control method based on a prediction model. This paper uses specific examples to explain the principles and implementation methods of the present application. The above examples are only used to help understand the method and its core idea; at the same time, for those skilled in the art, according to the idea of the present application, the specific implementation and application range will be changed; in summary, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A method for controlling fillet radius of laser-assisted rolling forming based on a prediction model, characterized in that: The method comprises: A process parameter prediction model is established. The process parameter prediction model is a mathematical function R = f(P, V, L) that represents the relationship between process parameters and the forming fillet radius in the laser-assisted roll forming process, where R is the fillet radius, P is the laser power, V is the laser scanning speed, L is the spot offset distance, and f is a functional relationship obtained by fitting experimental data. The process of establishing the process parameter prediction model includes: Setting several sets of process parameter combinations including different laser powers, laser scanning rates, spot offsets, and rolling forces; Laser-assisted rolling fillet forming experiments with different process parameter combinations were performed on multiple groups of metal sheets. Measure the actual fillet radius of the formed parts obtained from each group of experiments; Establishing a process parameter prediction model based on the relationship between the laser power, laser scanning rate, spot offset and actual fillet radius; According to the target fillet radius, the target process parameter combination is solved using the process parameter prediction model; Performing a laser-assisted rolling fillet forming operation on a target metal sheet using the target process parameter combination to obtain a formed part; Among them, the target fillet radius range is 1T to 2.5T, T is the target metal plate thickness, the metal plate is ultra-high strength steel, the tensile strength of the ultra-high strength steel is not less than 1500MPa, the elongation is not less than 5%, the thickness range is 1mm to 2mm, and the process parameter prediction model is a quadratic polynomial model.

2. The method according to claim 1, characterized in that Before performing laser-assisted rolling fillet forming experiments with different process parameter combinations on multiple groups of metal sheets, and performing laser-assisted rolling fillet forming operations on target metal sheets using the target process parameter combination, the method includes: The metal sheet was unidirectionally polished along the rolling direction using sandpaper, and then washed twice with deionized water and dried.

3. The method according to claim 2, characterized in that The process parameter prediction model is: Among them, R is the target fillet radius, P is the laser power, V is the laser scanning rate, L is the spot offset, and b0~b9 are the process parameter prediction model coefficients.

4. The method according to claim 3, characterized in that The fitting degree of the process parameter prediction model is greater than 0.

95.

5. The method according to claim 4, characterized in that The method further comprises: Using a laser scanner to measure the fillet radius after forming and calculate the error between the fillet radius after forming and the target fillet radius; When the error is greater than or equal to 2%, the process parameter prediction model coefficients are adjusted based on the error.

6. The method according to claim 1 or 5, characterized in that The laser-assisted rolling fillet forming operation is performed on the target metal sheet using the target process parameter combination, comprising: Fixing the metal sheet on a fixture and inputting the target process parameter combination into a control system; The laser emitter is controlled to irradiate and heat the preformed bending area of ​​the metal sheet, and at the same time, a rolling tool head is used to roll-form the metal sheet according to the shape of the target part.

7. The method according to claim 6, characterized in that The light spot irradiated by the laser emitter on the surface of the metal plate is rectangular and has a size of 2 mm×4 mm.

8. The method according to claim 7, characterized in that The laser power of the laser transmitter ranges from 1 kW to 2 kW, and the laser scanning rate ranges from 10 mm / s to 30 mm / s.

9. The method according to claim 8, characterized in that During the forming operation, the rolling tool head applies a forming force ranging from 1000N to 4000N.

10. A laser-assisted roll forming device based on a prediction model, characterized in that: The device comprises: A clamp, wherein the clamp is used to fix the metal sheet; The kuka six-degree-of-freedom robot includes a laser emitter and a rolling tool head. The laser emitter is used to irradiate and heat the preformed bending area of ​​the metal sheet, and the rolling tool head is used to roll the metal sheet according to the shape of the target part.

Images