A method for optimizing process parameters of laser cladding on rail surface

Abstract

The invention provides a method for optimizing process parameters of laser cladding on a rail surface, which relates to the field of laser cladding process. Single factor analysis and response surface method are used to design an experimental scheme and conduct experiments. Firstly, the dilution rate, shape coefficient and microhardness are used as evaluation indicators of the forming quality of a single-channel cladding layer to determine the initial range of process parameters (laser power, scanning speed and powder feeding speed). Single factor experiments of a single-channel cladding layer are carried out to analyze the forming law of the single-channel cladding layer, and a better combination of process parameters is selected to narrow the selection range of process parameters for a response surface experiment, and the cladding experiment is completed according to a central composite experimental scheme. Finally, a least squares algorithm is used to fit data, a quadratic regression model and a response surface model are established, and an optimal combination of process parameters is obtained. By comparing theoretical predicted values ​​with actual values, it is verified that the model established by the method of the invention has high accuracy, reliability and good practicality.

Description

Technical Field

[0001] The invention belongs to the technical field of laser cladding technology, and in particular relates to a method for optimizing process parameters of laser cladding on a rail surface. Background Art

[0002] Railway transportation has become one of the main modes of transportation in my country with its advantages of high efficiency and convenience. However, with the increase in train speed and axle weight, the track load environment has become complex and harsh, which may cause more serious damage to the wheel-rail system, such as rolling contact fatigue and wear. The existence of these damages shortens the life and maintenance cycle of the rails, restricting the transportation efficiency of the railway, so it is urgent to seek rail strengthening and repair technologies. Although most fatigue damage to rails is currently eliminated through grinding operations, wear is usually irreparable. Especially at sharp turns on curved tracks, due to the direct contact between the wheel flange and the side of the rail, the side wear is more significant. The grinding amount of severely worn rails is large, which will lead to serious material loss and further shorten the service life of the rails.

[0003] Laser cladding is a surface modification technology that has emerged in recent years. It uses a high-energy laser beam to melt the cladding material and the thin layer on the surface of the substrate together, and finally forms a metallurgical coating. It can significantly improve the hardness, wear resistance and high temperature resistance of the base material, and has therefore been widely used, mainly in automobile manufacturing, aerospace, impeller blade manufacturing and repair, and other industrial fields. The applicability and feasibility of laser cladding technology in strengthening the surface of rails has attracted widespread attention from relevant researchers and manufacturers at home and abroad.

[0004] At present, the selection of appropriate combination of laser cladding process parameters is crucial to the quality and performance of the cladding layer. Since laser cladding is a complex nonlinear process involving the coupling of multiple process parameters, it is very difficult to qualitatively describe the mapping relationship between each process parameter and the quality of the cladding layer. Therefore, in order to obtain high-quality coatings, it is necessary to optimize the laser cladding process parameters. Summary of the invention

[0005] The technical problem solved by the present invention is to provide a method for optimizing the process parameters of laser cladding on the surface of a rail, and to establish a high-precision regression prediction model between process parameters and response values ​​by combining the single factor method and the response surface method, so as to obtain the optimal combination of laser cladding process parameters, and to verify the accuracy and reliability of the regression model through experiments.

[0006] The technical solution adopted by the present invention is: a method for optimizing the process parameters of laser cladding on the surface of a rail, comprising the following steps:

[0007] Step 1: Determine the initial range of each process parameter, select Ni45 powder as the cladding material, and use the dilution rate η, shape coefficient ξ and microhardness HV 0.3 As an evaluation index of the forming quality of a single-pass cladding layer;

[0008] Step 2: Perform a single-channel cladding layer single factor experiment on the rail surface according to the initial range of each process parameter in step 1, and then analyze the influence of each process parameter on the forming quality of the cladding layer based on the experimental results and preliminarily select a process parameter combination N1;

[0009] Step 3, by using the response surface methodology-central composite experimental design method, with the process parameter combination N1 as the central composite experimental design factor, and the dilution rate, shape coefficient and microhardness as the response value, a central composite experiment is performed, wherein the process parameter levels in the central composite experiment are 0, ±1 and ±1.68, the number of factors n is 3, each factor has 5 levels, a total of 20 experiments are performed, relevant data are measured and recorded, variance analysis is performed, and a second-order polynomial regression model between each process parameter and the response value is established;

[0010] Step 4: Determine the process parameter combination N2 based on the second-order polynomial regression model, and predict the response value. Verify the accuracy and reliability of the response surface model through experiments, so as to select the process parameters under actual working conditions.

[0011] Preferably, the dilution rate in step 1 is calculated using the area method, that is:

[0012]

[0013] Wherein A1 is the cross-sectional area of ​​the cladding layer, A2 is the cross-sectional area of ​​the molten substrate, h is the depth of the molten pool, and H is the height of the cladding layer; the shape factor is defined as ξ=w / H, i.e., the ratio of the width to the height of the cladding layer. Usually, when ξ≤2, the height is relatively high and the molten pool is prone to collapse; when ξ≥2, a cladding layer with a good shape can be obtained; and when ξ≥2, the cladding layer is too wide, the cladding efficiency is reduced, and the cladding quality is affected.

[0014] Preferably, the step 3 uses a least squares algorithm to fit the data and establish a second-order response surface model of each process parameter and response value.

[0015] Preferably, the variance analysis in step 3 is to perform a significance analysis on the response index of the cladding layer of each process parameter, and the significance and degree of influence of each factor are further characterized by using P value and F value, and the influencing factors with P value less than 0.05 are regarded as significant influencing factors.

[0016] Preferably, the process parameter combination N2 in step 4 is based on multi-objective optimization theory, seeking the maximum aspect ratio, the maximum microhardness and the minimum dilution rate, and setting the process parameter level range as a constraint condition to select the optimal process parameter combination.

[0017] Preferably, the process parameters selected by the process parameter combination N1 and the process parameter combination N2 include laser power A, scanning speed B, and powder feeding speed C.

[0018] Preferably, in the single-channel cladding layer single-factor experiment described in step 2, after each experiment is completed, a wire cutting is performed in a direction perpendicular to the scanning direction, and the cross section of the sample is preliminarily polished with 240 to 2000 mesh sandpaper in sequence, and the two polishing directions are perpendicular to each other to ensure that the scratches of the previous time are worn away, and then polished on a metallographic polishing machine. After polishing, the surface is smooth and scratch-free, and presents a mirror surface; after corrosion, the cladding layer morphology is observed under a microscope and the corresponding dimensional parameters are measured, the molten pool morphology photos are taken, and the average microhardness of each cladding layer is measured and calculated using a microhardness tester, and the data is recorded.

[0019] Preferably, the second-order polynomial regression model established in step 3 is used to fit the experimental data, and then evaluate the influence of the process parameters and the response index, and its functional relationship is as follows:

[0020] y u =β0+β1x 1u +β2x 2u +β 11 x 2 1u +β 22 x 2 2u +β 12 x 1u x 2u +ε u ;

[0021] Where y is the predicted response value, β0 is the intercept, and β i , β ii , β ij are the estimated values ​​of the regression coefficients, and ε is the experimental error.

[0022] Preferably, the second-order polynomial regression model for dilution rate, shape coefficient and microhardness established in step 3 is as follows:

[0023] η=+0.34+0.023A+0.020B-0.012C+9.505×10 -3 ×B 2 ;

[0024] ξ=+2.98+0.11A+0.14B-0.28A2 -0.093B 2 -0.25C 2 ;

[0025] HV 0.3 =+457.13-7.18A+12.40B+3.17C-11.84A 2 -2.76B 2 -4.13C 2 ;

[0026] A represents laser power, B represents scanning speed, and C represents powder feeding speed.

[0027] The beneficial effects of the present invention are:

[0028] First, through single-factor experiments, the influence of relevant process parameters on the dilution rate, shape parameters and microhardness of a single-pass cladding layer was explored, and the better parameter combination was selected to narrow the range of parameter selection; further, the response surface method was used, and the central composite experimental design was used to perform variance analysis on the experimental data obtained, and the second-order polynomial regression model of process parameters and response indicators was established by the least squares method. The process parameters were optimized by software and experimentally verified, and a set of process parameter combinations that can better meet the morphology and hardness requirements of the cladding layer were obtained.

[0029] Second, by analyzing the residuals and comparing the actual values ​​with the predicted values, the model fit analysis was performed. The results showed that the model established by the method of the present invention has high accuracy and credibility, can realize the prediction of relevant response indicators, and has good practicality.

[0030] 3. Explore the significance of the impact of each process parameter on the response index through variance analysis; obtain the significance arrangement results of the factors affecting the response index, and better guide the experimental operation based on the significance arrangement results. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 The figure is a flow chart of a method for optimizing process parameters of rail surface laser cladding in an embodiment of the present invention.

[0032] Figure 2 It is a schematic diagram of the cross section of a single-pass cladding in an embodiment of the present invention.

[0033] Figure 3 The cross-sectional morphology of the cladding layer as the process parameters change in the embodiment of the present invention.

[0034] Figure 4 3. The influence of various process parameters on the morphology parameters and microhardness of the cladding layer in the embodiment of the present invention.

[0035] Figure 5This is the cross-sectional morphology of the central composite experiment in an embodiment of the present invention.

[0036] Figure 6 This is a graph analyzing the accuracy of the shape coefficient, dilution rate and microhardness model in the embodiment of the present invention.

[0037] Figure 7 1 is a response surface diagram of laser power and scanning speed to shape coefficient in an embodiment of the present invention.

[0038] Figure 8 4 is a response surface diagram of laser power and scanning speed to dilution rate in an embodiment of the present invention.

[0039] Fig. 9 Surface diagram showing the response of laser power and powder feeding speed to dilution rate in an embodiment of the present invention.

[0040] Fig.10 It is a response surface diagram of powder feeding speed and scanning speed to microhardness in an embodiment of the present invention.

[0041] Fig.11 Flow chart of the model optimization steps in an embodiment of the present invention.

[0042] Fig.12 It is the cross section of the cladding layer under the optimal process parameter combination in the actual experiment in the embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0044] Example 1: Figure 1 As shown, the present invention provides a method for optimizing the process parameters of laser cladding on the surface of a rail; a relevant laser cladding experiment is carried out on a substrate made of a rail material using a KUKA robot, Ni45 powder is selected, and the powder feeding gas flow rate is set to 8.0 L / h and the shielding gas flow rate is set to 12 L / h.

[0045] Single factor experiments were carried out with laser power A, scanning speed B and powder feeding speed C as variables, and the values ​​of each process parameter are shown in Table 1. After completion, wire cutting was performed perpendicular to the scanning direction, and the cross section of the sample was preliminarily polished with sandpaper from 240 mesh to 2000 mesh in sequence. The two polishing directions were perpendicular to each other to ensure that the scratches of the previous polishing were removed. Then polish it on a metallographic polishing machine. After polishing, the surface is smooth and scratch-free, and it is mirror-like. After corrosion, the morphology of the cladding layer is observed under a microscope and the melting depth, melting height and melting width are measured. Photos of the molten pool morphology are taken as shown in the figure. Figure 3As shown, a microhardness tester is used to measure and calculate the average microhardness of each cladding layer, and the data are recorded.

[0046] Table 1 Single factor experiment value range

[0047]

[0048] like Figure 4 As shown in the figure, a curve graph was drawn according to the results of a single experiment, and the influence of each process parameter on the dilution rate, morphology coefficient and microhardness was analyzed. The optimal process parameter combination was initially selected as laser power 435W, scanning speed 5mm / s, and powder feeding rate 1r / min. Based on the above parameter combination, a central composite experiment was designed, and the corresponding relationship between the response surface experimental parameters and the coding values ​​is shown in Table 2.

[0049] Table 2 Process parameter levels

[0050]

[0051] The central composite experimental scheme and experimental results are shown in Table 3. The cross section of the cladding layer is shown in Figure 5 shown.

[0052] The second-order response surface model between each process parameter and the response value is proposed to be in the form of:

[0053] y u =β0+β1x 1u +β2x 2u +β 11 x 2 1u +β 22 x 2 2u +β 12 x 1u x 2u +ε u ;

[0054] Where y is the predicted response value, β0 is the intercept, and β i , β ii , β ij are the estimated values ​​of the regression coefficients, and ε is the experimental error.

[0055] Table 3 Central composite experimental design and results

[0056]

[0057] The least squares algorithm was used to fit the data and obtain the coefficients of the second-order response surface model. The following second-order response surface regression model was obtained, where A represents laser power, B represents scanning speed, and C represents powder feeding speed.

[0058] η=+0.34+0.023A+0.020B-0.012C+9.505×10 -3 ×B 2 ;

[0059] ξ=+2.98+0.11A+0.14B-0.28A 2 -0.093B 2 -0.25C 2 ;

[0060] HV 0.3 =+457.13-7.18A+12.40B+3.17C-11.84A 2 -2.76B 2 -4.13C 2 ;

[0061] Figure 6 The left side shows the residual diagrams of the aspect ratio, dilution rate and microhardness model. There are no abnormal values ​​in the diagram, indicating that the model has good credibility. Figure 6 On the right are comparison charts of measured and predicted values ​​of aspect ratio, dilution rate and microhardness. It can be found that the measured experimental values ​​are concentrated on both sides of the fitting straight line, which also shows the accuracy of the established model.

[0062] The variance analysis method was used to analyze the influence of each process parameter on the aspect ratio. The results are shown in Table 4:

[0063] Table 4 Aspect ratio analysis of variance

[0064]

[0065] From the variance analysis, we can see that the P values ​​of laser power and scanning speed are both less than 0.05, so these two factors will have a significant impact on the aspect ratio, while the effect of powder feeding speed on the aspect ratio is not significant. Figure 7 The contour diagram and 3D response surface diagram of the scanning speed and laser power on the aspect ratio show that a decrease in laser power and an increase in scanning speed will produce a larger aspect ratio. This is because as the scanning speed increases, the metal powder does not have enough time to accumulate in height.

[0066] The variance analysis method was used to analyze the influence of each process parameter on microhardness. The results are shown in Table 5:

[0067] Table 5 Analysis of variance of microhardness

[0068]

[0069] From the variance analysis, it can be seen that the P values ​​of laser power, scanning speed and powder feeding speed are all less than 0.05, indicating that the influence of these three parameters on microhardness is significant, and FC>FA>FB, among which the F value of scanning speed is the largest, indicating that it is the most significant factor; the contour map and 3D response surface map below are as follows Fig.10 As shown in the figure, it shows that with the increase of laser power, the microhardness of the cladding layer tends to decrease, while with the increase of scanning speed, the microhardness tends to increase. Therefore, a higher microhardness can be obtained by appropriately reducing the laser power and increasing the scanning speed.

[0070] The influence of each process parameter on the dilution rate was analyzed by variance analysis. The results are shown in Table 6:

[0071] Table 6 Analysis of variance of dilution rate

[0072]

[0073]

[0074] From the variance analysis, we can see that the P values ​​of laser power, scanning speed and powder feeding speed are all less than 0.05, and the P values ​​of the first two parameters are <0.0001, indicating that these two parameters have the most significant influence. Figure 8 As shown in the figure, when both the laser power and the scanning speed are reduced, the dilution rate decreases accordingly. In the actual cladding process, in order to make the dilution rate of the cladding layer low, it is necessary to set a low level of laser power and scanning speed. When the laser power is increased, the dilution rate increases. This is because as the power increases, the energy absorbed by the powder and the substrate increases; Fig. 9 It can be seen that increasing the powder feeding speed and reducing the laser power can also obtain a smaller dilution rate. This is because with the increase in the powder feeding speed, most of the energy is absorbed by more powder, and the substrate absorbs less energy.

[0075] Finally, based on the multi-objective optimization theory, the maximum aspect ratio, the maximum microhardness and the minimum dilution rate are sought, and the process parameter level range is set as the restriction condition as shown in Table 7. The multi-objective optimization study of laser cladding process parameters is realized with the help of Design Expert software. The model optimization process is shown in Fig.11 shown.

[0076] Table 7 Optimization conditions and objectives

[0077]

[0078] Finally, the optimal process parameter combination obtained by response surface methodology is: laser power 431W, scanning speed 5.34mm / s, powder disk speed 1.03r / min. Based on the above process parameter combination, a single-pass cladding experiment was carried out, and each index was measured three times to obtain the average value. The specific results are shown in Table 8:

[0079] Table 8 Comparison of prediction optimization results and experimental verification results

[0080]

[0081] The measured errors between the theoretical prediction value and the actual value are no more than 10%. The experimental results meet the expected requirements, indicating that the response surface model has a high prediction accuracy for the experimental results, which further illustrates the reliability of the model.

[0082] The cross-sectional shape of the cladding layer under the optimal process parameter combination in the actual experiment in the embodiment of the present invention is as follows: Fig.12 shown.

[0083] Although the present invention is disclosed as above, the protection scope of the present invention is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention, and these changes and modifications will fall within the protection scope of the present invention.

Claims

1. A method for optimizing process parameters of rail surface laser cladding, characterized in that: The following steps are involved: Step 1: Determine the initial range of each process parameter, select Ni45 powder as the cladding material, and use the dilution rate η, shape coefficient ξ and microhardness HV 0.3 As an evaluation index of the forming quality of a single-pass cladding layer; Step 2: Perform a single-pass cladding layer single factor experiment on the rail surface according to the initial range of each process parameter in step 1, and then analyze the influence of each process parameter on the forming quality of the cladding layer based on the experimental results and preliminarily select a process parameter combination N1; Step 3, by using the response surface methodology-central composite experimental design method, with the process parameter combination N1 as the central composite experimental design factor, and the dilution rate, shape coefficient and microhardness as the response value, a central composite experiment is performed, wherein the process parameter levels in the central composite experiment are 0, ±1 and ±1.68, the number of factors n is 3, each factor has 5 levels, a total of 20 experiments are performed, relevant data are measured and recorded, variance analysis is performed, and a second-order polynomial regression model between each process parameter and the response value is established; Step 4: Determine the process parameter combination N2 based on the second-order polynomial regression model, and predict the response value. Verify the accuracy and reliability of the response surface model through experiments, so as to select the process parameters under actual working conditions.

2. The method for optimizing the process parameters of rail surface laser cladding according to claim 1, characterized in that: The dilution rate described in step 1 is calculated using the area method, namely: Where A1 is the cross-sectional area of ​​the cladding layer, A2 is the cross-sectional area of ​​the molten substrate, h is the depth of the molten pool, and H is the height of the cladding layer; The shape factor is defined as ξ=w / H, ie, the ratio of the width to the height of the cladding layer.

3. The method for optimizing the process parameters of rail surface laser cladding according to claim 1, characterized in that: The step 3 adopts the least squares algorithm to fit the data and establish a second-order response surface model of each process parameter and response value.

4. The method for optimizing the process parameters of rail surface laser cladding according to claim 1, characterized in that: The variance analysis in step 3 is to perform a significance analysis on the response index of the cladding layer of each process parameter, and further characterize the significance and degree of influence of each factor by using P value and F value, and the influencing factors with P value less than 0.05 are regarded as significant influencing factors.

5. The method for optimizing the process parameters of rail surface laser cladding according to claim 1, characterized in that: The process parameter combination N2 described in step 4 is based on multi-objective optimization theory, seeking the maximum aspect ratio, maximum microhardness and minimum dilution rate, and setting the process parameter level range as a constraint condition to select the optimal process parameter combination.

6. The method for optimizing the process parameters of rail surface laser cladding according to claim 1, characterized in that: The process parameters selected by the process parameter combination N1 and the process parameter combination N2 include laser power A, scanning speed B, and powder feeding speed C.

7. The method for optimizing the process parameters of rail surface laser cladding according to claim 1, characterized in that: In the single-channel cladding layer single-factor experiment described in step 2, after each experiment is completed, a wire cutting is performed in a direction perpendicular to the scanning direction, and the cross section of the sample is preliminarily polished with 240 to 2000 mesh sandpaper in sequence. The two polishing directions are perpendicular to each other to ensure that the scratches of the previous time are worn away, and then polished on a metallographic polishing machine. After polishing, the surface is smooth and scratch-free, and presents a mirror surface; after corrosion, the cladding layer morphology is observed under a microscope and the corresponding dimensional parameters are measured, the molten pool morphology photos are taken, and the average microhardness of each cladding layer is measured and calculated using a microhardness tester, and the data is recorded.

8. The method for optimizing the process parameters of rail surface laser cladding according to claim 1, characterized in that: The second-order polynomial regression model established in step 3 is used to fit the experimental data and then evaluate the influence of process parameters and response indicators. The functional relationship is as follows: y u =β0+β1x 1u +β2x 2u +b 11 x 2 1u +b 22 x 2 2u +b 12 x 1u x 2u +e u ; Where y is the predicted response value, β0 is the intercept, and β i , β ii , β ij are the estimated values ​​of the regression coefficients, and ε is the experimental error.

9. The method for optimizing the process parameters of rail surface laser cladding according to claim 1, characterized in that: The second-order polynomial regression model for dilution rate, shape coefficient and microhardness established in step 3 is as follows: η=+0.34+0.023A+0.020B-0.012C+9.505×10 -3 ×B 2 ; ξ=+2.98+0.11A+0.14B-0.28A 2 -0.093B 2 -0.25C 2 ; HV 0.3 =+457.13-7.18A+12.40B+3.17C-11.84A 2 -2.76B 2 -4.13C 2 ; A represents laser power, B represents scanning speed, and C represents powder feeding speed.

Images