A method of arc additive manufacturing of a multi-structured large aluminum alloy vehicle body

Abstract

The application discloses an arc additive manufacturing method for a large aluminum alloy vehicle body with multiple structures and belongs to the field of arc additive manufacturing. The application adopts a pre-printing experiment to print only a special structure part of a workpiece or to print after reducing the whole workpiece, can test additive process parameters and additive software parameters under the condition of reducing cost and test time, reduces problems in the subsequent additive process of the aluminum alloy vehicle body workpiece, is favorable for improving the yield of the workpiece, and solves the problems of a too long delivery period and a too low single-piece yield of single pieces and small-batch pieces compared with a traditional casting mode, so that the rapid and customized delivery of the product is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of electric arc additive manufacturing, in particular to a method for electric arc additive manufacturing of a large aluminum alloy car body with multiple structures. BACKGROUND

[0002] In the production of aluminum alloy car bodies, the traditional casting method has the problems of long cycle, high cost and low yield. The additive manufacturing rapid forming technology does not need mold processing, can greatly shorten the production cycle, improve the production efficiency, and can save materials and production cost. Since the process parameters of electric arc additive manufacturing are not unique, the welding process parameters need to be debugged before electric arc additive manufacturing to reduce or avoid problems in formal welding, thereby improving the yield. Therefore, the present application provides a method for electric arc additive manufacturing of a large aluminum alloy car body with multiple structures. SUMMARY

[0003] The present application provides a method for electric arc additive manufacturing of a large aluminum alloy car body with multiple structures, which can solve the problems pointed out in the background art.

[0004] Technical scheme, according to one aspect of the present application, a method for electric arc additive manufacturing of a large aluminum alloy car body with multiple structures, comprising the following steps:

[0005] Step S1, constructing a three-dimensional model of the workpiece, simulating the additive path, collecting and analyzing the influencing factors of the process parameters, and establishing the feasible solution space of the process parameters;

[0006] Step S2, designing a pre-printed workpiece and performing a printing operation, detecting the composition and mechanical properties of the pre-printed workpiece, forming a detection result, selecting a feasible solution based on the detection result, optimizing the feasible solution space, and obtaining a workpiece printing optimal path solution set;

[0007] Step S3, constructing a car body workpiece and performing a printing operation, and performing detection, and post-processing the car body workpiece that passes the detection.

[0008] According to one aspect of the present application, the step 1 is further:

[0009] Step S11, constructing a three-dimensional model of the workpiece, simulating the additive path;

[0010] Step S12, determining the influencing factors of the process parameters according to the physical model and empirical data of electric arc additive manufacturing, quantifying and standardizing these factors, and the influencing factors include material performance, workpiece shape, weld appearance, thermal stress and deformation parameters;

[0011] Step S13, determine the range and resolution of the process parameters, and determine the impact index of the process parameters on the additive result according to the material properties and quality requirements of the workpiece, the process parameters at least including electrode parameter, welding current, arc voltage, focal length, preheating temperature, weld layer number and wire feeding speed; the impact index at least including mechanical properties, chemical composition and surface quality;

[0012] Step S14, establish the mutual influence and coupling relationship between the process parameters, analyze the influence mechanism and law of each process parameter on the printing result according to the above-mentioned physical model and empirical data, establish a multi-objective optimization model according to the influence degree and direction of each process parameter on the printing result, take the key indicators in the printing result as the optimization objective function; take the process parameters as the optimization variables, and take the mutual influence and coupling relationship between the process parameters as the constraint condition; the influence mechanism includes: the influence of current and voltage on the molten pool morphology and temperature field, the influence of wire feeding speed and layer height on the filling density and defect formation, the influence of speed and path mode on the weld overlap and surface roughness;

[0013] Step S15, according to the physical meaning and actual adjustable range of the process parameters, set the value interval for each process parameter, and discretize the interval into a predetermined number of discrete values;

[0014] Combine the discrete values of all process parameters to obtain a feasible solution space containing all possible process parameter combinations; or, use the Latin hypercube sampling method to generate a set of uniformly distributed process parameter combinations as initial samples of the feasible solution space according to the range and resolution of the process parameters;

[0015] Step S16, for the multi-objective optimization model, use a multi-objective optimization method to evaluate and sort the initial samples according to the impact index of the process parameters on the additive result, to obtain a set of Pareto optimal solutions as the preferred samples of the feasible solution space.

[0016] According to one aspect of the present application, the step 2 is further:

[0017] Step S21, design a pre-printed workpiece and perform a printing operation, detect the composition and mechanical properties of the pre-printed workpiece to form a detection result; according to the preferred samples of the feasible solution space, sequentially perform pre-printing experiments and obtain pre-printed parts; perform non-destructive testing, chemical composition testing and mechanical property testing on the pre-printed parts, and obtain the detection results;

[0018] Step S22, compare the printing result with the optimization objective function, and determine whether there is adverse effect, if the printing result meets the optimization objective function and there is no adverse effect, the set of process parameter combinations is retained; the adverse effect includes defects, deformation and cracks;

[0019] Step S23, if the printing result does not satisfy the optimization objective function or there is adverse effect, the set of process parameter combinations is deleted, and another set of process parameter combinations is selected from the feasible solution space for pre-printing experiment.

[0020] According to an aspect of the present application, each printing and detection process in step 21 is further:

[0021] Step S21a: pre-cleaning and substrate installation are performed, and welding process parameters are debugged;

[0022] Step S21b: pre-printing experiment is performed to obtain a pre-printed piece;

[0023] Step S21c: the pre-printed piece is subjected to heat treatment, and after the heat treatment, the pre-printed piece is subjected to non-destructive testing, chemical composition testing and mechanical property testing to obtain detection results;

[0024] Step S21d: if all the detection results are qualified, formal printing preparation is made; if one of the detection results is unqualified, the pre-printing scheme and process parameters are optimized, and the pre-printing experiment is performed again.

[0025] According to an aspect of the present application, it further includes step S24, optimizing the feasible solution space to obtain a workpiece printing optimal path solution set;

[0026] Step S241: multi-objective optimization is performed on the remaining feasible solution set to find process parameter combinations satisfying the Pareto optimal condition, i.e. process parameter combinations in which improvement in one objective function does not lead to deterioration in another objective function;

[0027] Step S242: genetic algorithm is used to further search and optimize the Pareto optimal solution set, new process parameter combinations are generated through simulation of natural selection, crossover and mutation operations, and their fitness, i.e. the value of the optimization objective function, is evaluated, and the process parameter combination with the highest fitness is selected as the final optimal solution;

[0028] Step S243: PCA algorithm is used to perform sensitivity analysis on the optimal solution, i.e. to analyze the contribution rate and influence degree of each process parameter on the optimization objective function, to determine the most critical process parameter, and to set a reasonable fluctuation range for it, so as to adjust and adapt in the actual printing process;

[0029] Step S244: the workpiece printing optimal path solution set is constructed according to the above calculation results.

[0030] According to an aspect of the present application, it further includes step S24, optimizing the feasible solution space to obtain a workpiece printing optimal path solution set;

[0031] Step S24a, constructing a deep reinforcement learning module, establishing an intelligent agent for learning and optimizing the selection strategy of process parameters according to the feasible solution space of process parameters and the detection results of pre-printing experiments;

[0032] Step S24b, using the multi-armed bandit technique, dynamically adjusting the exploration and exploitation ratio of process parameters according to the selection strategy of the intelligent agent and the detection results of pre-printing experiments, to balance the trade-off between exploring new feasible solutions and exploiting known excellent solutions;

[0033] Step S24c, until a predetermined number of iterations or a convergence condition is reached, obtaining the final preferred feasible solution, and constructing the printing preferred path solution set of the workpiece.

[0034] According to one aspect of the present application, the step S3 is further:

[0035] Step S31: start the formal printing of the aluminum alloy car body according to the pre-printing piece setting process parameters and scheme, and obtain the aluminum alloy car body;

[0036] Step S32: heat treatment is performed on the aluminum alloy car body workpiece, after heat treatment, size detection and non-destructive detection are performed on the aluminum alloy car body workpiece, and mechanical property detection and chemical composition detection are performed on the furnace sample, to obtain the detection results;

[0037] Step S33: if all the detection results are qualified, then machining is performed according to the size of the three-dimensional model;

[0038] If the size detection is unqualified, further processing is performed by means of repair welding or machining, if the processing is qualified, the next step is performed, if the processing is unqualified, the aluminum alloy car body workpiece is scrapped, and the aluminum alloy car body piece is re-printed;

[0039] If the non-destructive detection is unqualified, the aluminum alloy car body workpiece is scrapped, and the aluminum alloy car body workpiece is re-printed;

[0040] If the chemical composition detection of the furnace sample is unqualified, the aluminum alloy car body workpiece is scrapped, and the aluminum alloy car body workpiece is re-printed after replacing the welding wire;

[0041] If the mechanical property detection of the furnace sample is unqualified, the aluminum alloy car body workpiece is scrapped, and the aluminum alloy car body workpiece is re-printed;

[0042] Step S34: machining is performed on the aluminum alloy car body;

[0043] Step S35: size detection is performed on the machined aluminum alloy car body workpiece, if the size detection is qualified, the next step is performed, if the size detection is unqualified, steps two to ten are re-performed;

[0044] Step S36: treating the surface of the qualified aluminum alloy vehicle body workpiece to obtain a finished workpiece.

[0045] According to an aspect of the present application, in step S11, after the three-dimensional model of the workpiece is constructed, a lightweight processing procedure of the model is further included, specifically:

[0046] Step S11a: calling the three-dimensional model of the workpiece, and dividing the model into many small units for finite element analysis;

[0047] Step S11b: defining the target and constraints of the topology optimization, the target is usually to minimize the mass or flexibility of the structure, and the constraints include the stress, displacement, frequency and heat conduction of the structure, and the manufacturing constraints of the structure;

[0048] Step S11c: constructing and starting the topology optimization program module, automatically finding the best material distribution scheme, the topology optimization module iteratively updates the density distribution of the structure using the SIMP variable density method or the level set method, and evaluates the performance indicators of the structure until the preset target and constraints are met or the maximum number of iterations is reached.

[0049] According to an aspect of the present application, the non-destructive testing process of the aluminum alloy vehicle body workpiece in step S32 is further:

[0050] Step S32a: constructing an ultrasonic array and a phased array control system, an array is formed by arranging ultrasonic transmitters and receivers in a predetermined arrangement to cover the surface or internal area of the pre-printed workpiece;

[0051] Step S32b: connecting the ultrasonic transmitters and receivers to the phased array control system to control and adjust the phase and amplitude of each element;

[0052] Step S32c: setting the phased array detection parameters and corresponding threshold values on the phased array control system, the phased array detection parameters include ultrasonic frequency, focusing depth, scanning angle, scanning speed and scanning range;

[0053] Constructing and selecting appropriate algorithms and models to analyze and judge the ultrasonic signals to give the detection results.

[0054] According to an aspect of the present application, step S21b further includes:

[0055] At least two welding heads are installed on the electric arc additive manufacturing machine, and appropriate positions, angles and spacings are selected according to the shape and size of the workpiece;

[0056] A synchronous control device is installed on the electric arc additive manufacturing machine, and the moving speed and direction of the welding head are adjusted according to the control signal;

[0057] According to the structural characteristics of the workpiece, two working modes of simultaneous cooperation or alternating cooperation are selected; the process parameters of each welding head are received and welding is performed according to the working mode.

[0058] Compared with the prior art, the beneficial effects of the present application are: the present application adopts pre-printing experiments to print only the special structure part of the workpiece or to print after reducing the whole workpiece, which can test the additive process parameters and additive software parameters while reducing the cost and test time, reduce the problems in the subsequent aluminum alloy car body workpiece additive process, and is beneficial to improve the yield of the workpiece, the additive manufacturing method solves the problems of too long delivery cycle and too low single product yield compared with the traditional casting method, and realizes the rapid and customized delivery of such products. BRIEF DESCRIPTION OF DRAWINGS

[0059] Figure 1 It is a structural schematic diagram of the aluminum alloy car body of the present application.

[0060] Figure 2 It is an electric arc additive manufacturing flowchart of the first embodiment of the present application.

[0061] Figure 3 It is a whole simulation path diagram of the workpiece of the present application.

[0062] Figure 4 It is a first layer path diagram of the present application.

[0063] Figure 5 It is a square structure thick wall of the present application.

[0064] Figure 6 It is a simulation path diagram of four overhanging corners of the present application.

[0065] Figure 7 It is a printing overlap part of the present application.

[0066] Figure 8 It is a printing lap joint part of the present application.

[0067] Figure 9 It is a flowchart of the second embodiment of the present application.

[0068] Figure 10 It is a flowchart of step S1 of the second embodiment of the present application.

[0069] Figure 11 It is a flowchart of step S2 of the second embodiment of the present application. DETAILED DESCRIPTION

[0070] One specific embodiment of the present application will be described in detail below with reference to the accompanying drawings, but it should be understood that the protection scope of the present application is not limited by the specific embodiment.

[0071] Since the whole process is relatively very complex, a brief workflow is first given to understand the general processing flow. As shown in Figures 1 to 2 The arc additive manufacturing method for the large aluminum alloy car body with multiple structures provided by the embodiment of the present application comprises the following steps:

[0072] Step one: build a three-dimensional model of the workpiece, import the three-dimensional model into IungoPNT software to simulate the additive path, select ZigZag path for additive according to the shape structure characteristics of the workpiece, and select 45° for the first layer path direction;

[0073] Step two: perform pre-welding cleaning and substrate installation, and debug the welding process parameters;

[0074] Step three: perform a pre-printing experiment to obtain a pre-printing piece;

[0075] Step four: heat treat the pre-printing piece, and after heat treatment, perform non-destructive testing, chemical composition testing, and mechanical property testing on the pre-printing piece to obtain the testing results;

[0076] Step five: if all testing results are qualified, make preparations for formal printing;

[0077] If one of the testing results is unqualified, optimize the pre-printing scheme and process parameters, and perform a pre-printing experiment again;

[0078] Step six: start the formal printing of the aluminum alloy car body according to the pre-printing piece setting process parameters and scheme to obtain the aluminum alloy car body;

[0079] Step seven: heat treat the aluminum alloy car body workpiece, and after heat treatment, perform size detection and non-destructive testing on the aluminum alloy car body workpiece, and perform mechanical property testing and chemical composition testing on the furnace sample to obtain the testing results;

[0080] Step eight: if all testing results are qualified, machine according to the size of the three-dimensional model;

[0081] If the size detection is unqualified, further process through repair welding or machining, if the processing is qualified, proceed to the next step, if the processing is unqualified, the aluminum alloy car body workpiece is scrapped, and the aluminum alloy car body workpiece is printed again;

[0082] If the non-destructive testing is unqualified, the aluminum alloy car body workpiece is scrapped, and the aluminum alloy car body workpiece is printed again;

[0083] If the chemical composition testing of the furnace sample is unqualified, the aluminum alloy car body workpiece is scrapped, and the aluminum alloy car body workpiece is printed again after replacing the welding wire;

[0084] If the mechanical property detection of the furnace sample is unqualified, the aluminum alloy car body workpiece is scrapped, and the aluminum alloy car body workpiece is printed again;

[0085] Step nine: machining the aluminum alloy car body;

[0086] Step ten: size detection is performed on the machined aluminum alloy car body workpiece, if the size detection is qualified, the next step is performed, if the size detection is unqualified, steps two to ten are performed again;

[0087] Step eleven: surface treatment is performed on the qualified aluminum alloy car body workpiece to obtain a finished workpiece.

[0088] After X-ray irradiation, if the workpiece has a strip-shaped defect, a penetrating defect or a circular defect with a diameter greater than 2mm, the nondestructive testing is unqualified, the aluminum alloy car body workpiece is scrapped, and the aluminum alloy car body workpiece is printed again;

[0089] After detection, if the mass percentage of the multiple chemical components of the furnace sample reaches: Si≤0.400%; Cu≤0.100%; Zn≤0.250%; S≤0.300%; Ti≤0.150%; 0.430%≤Mg≤0.520%; 0.0500%≤Mn≤0.10%; 0.050%≤Cr≤0.250%; Fe≤0.400%; other single≤0.005%; other total≤0.015%; Al balance, the furnace sample chemical composition detection is qualified, if not, the furnace sample chemical composition detection is unqualified, the aluminum alloy car body workpiece is scrapped, and the aluminum alloy car body workpiece is printed again after replacing the welding wire.

[0090] Next, the detailed implementation process of the present application is given, that is, on the basis of the above embodiment, the related steps are described in detail, as follows: a multi-structure large aluminum alloy car body arc additive manufacturing method, comprising the following steps:

[0091] Step S1, constructing a three-dimensional model of the workpiece, simulating the additive path, collecting and analyzing the influencing factors of the process parameters, and establishing a feasible solution space of the process parameters;

[0092] Step S2, designing a pre-printed workpiece and performing printing operation, detecting the composition and mechanical property of the pre-printed workpiece, forming a detection result, screening feasible solutions based on the detection result, optimizing the feasible solution space, and obtaining a workpiece printing optimal path solution set;

[0093] Step S3, constructing a car body workpiece and performing printing operation, and performing detection, and performing post-treatment on the car body workpiece that passes the detection.

[0094] The embodiment can realize arc additive manufacturing of large-scale aluminum alloy car bodies with multiple structures. Compared with the traditional casting method, the embodiment has the advantages of short production cycle, low cost, high yield, and strong customization. By using the pre-printing experiment method, the special structure part of the workpiece or the workpiece is printed after being reduced, the additive process parameters and additive software parameters are tested and optimized, the problems and risks in the formal printing process are reduced, and the printing quality and efficiency are improved. The IungoPNT software is used to simulate the additive path, and according to the shape structure characteristics of the workpiece, the ZigZag path is selected for additive manufacturing to ensure the uniformity and continuity of the weld, and to avoid the generation of incomplete fusion defects. The aluminum alloy car body workpiece after printing is subjected to heat treatment, nondestructive testing, chemical composition testing and mechanical property testing to ensure that the performance and quality of the workpiece meet the requirements, and the unqualified workpiece is re-printed or machined to improve the qualified rate of the workpiece. The machined aluminum alloy car body workpiece is subjected to surface treatment to obtain a finished workpiece, realizing arc additive manufacturing of large-scale aluminum alloy car bodies with multiple structures.

[0095] According to one aspect of the present application, step 1 is further:

[0096] Step S11, constructing a three-dimensional model of the workpiece, simulating the additive path;

[0097] Step S12, according to the physical model and empirical data of arc additive manufacturing, determining the influencing factors of process parameters, quantifying and standardizing these factors, the influencing factors including material properties, workpiece shape, weld appearance, thermal stress and deformation parameters;

[0098] Among them, the material properties: such as melting point, thermal conductivity, thermal expansion coefficient, tensile strength, elongation rate, etc. These properties determine the melting, flowing, solidification and deformation behavior of the material under arc heating.

[0099] Workpiece shape: such as thickness, curvature, angle, etc. These shapes determine the stress state and thermal stress distribution of the workpiece during additive manufacturing.

[0100] Weld appearance: such as width, depth, height, angle, etc. These appearances determine the filling effect and surface quality of the weld during additive manufacturing.

[0101] Thermal stress and deformation parameters: such as temperature field, stress field, strain field, etc. These parameters reflect the thermodynamic behavior and deformation law of the workpiece during additive manufacturing.

[0102] Step S13, determine the range and resolution of the process parameters, and determine the influence index of the process parameters on the additive result according to the material performance and quality requirements of the workpiece, the process parameters at least including electrode parameter, welding current, arc voltage, focal length, preheating temperature, weld layer number and wire feeding speed; the influence index at least including mechanical property, chemical composition and surface quality;

[0103] Electrode parameter: such as diameter, type, coating, etc., these parameters determine the melting rate and composition of the electrode.

[0104] Welding current: is the main process parameter of electrode arc welding, directly affects the welding quality and production efficiency. The selection of welding current should be considered comprehensively according to the electrode diameter, electrode type, welding thickness, joint form, welding position and welding layer. Generally, the larger the electrode diameter, the more heat is needed to melt the electrode, and the welding current must be increased. When welding in the flat position, a larger welding current can be selected, and when welding in the non-flat position, in order to facilitate the control of the weld shape, the welding current is 10%~20% smaller than that in the flat position.

[0105] Arc voltage: refers to the voltage difference between the two ends of the arc, mainly determined by the arc length. The longer the arc length, the higher the arc voltage, and vice versa. The arc length should not be too long or too short, otherwise it will affect the stability of the arc and the penetration. Generally, the arc length is equal to or slightly larger than the electrode diameter, and the corresponding arc voltage is 16~25V.

[0106] Focal length: refers to the distance from the electrode tip to the workpiece surface. If the focal length is too long, the arc will be unstable, the spatter will increase, and the heat input will decrease; if the focal length is too short, the electrode tip will be easy to stick to the workpiece, causing short circuit or open circuit. Generally, the focal length should be equal to or slightly smaller than the arc length.

[0107] Preheating temperature: refers to the temperature of the base material before formal welding. The role of preheating temperature is to reduce the temperature difference between the base material and the molten pool, reduce the cooling speed of the base material to the molten pool, reduce the residual stress and deformation, and facilitate the escape of hydrogen from the molten pool, prevent the generation of hydrogen-induced cracks. The preheating temperature is considered comprehensively according to the chemical composition, performance, thickness of the base material, the restraint degree of the welding joint, the welding environment temperature and the technical standards of the relevant products, etc. Generally, for low carbon steel and low alloy steel, the preheating temperature is 100~200℃; for high carbon steel and high alloy steel, the preheating temperature is 200~400℃.

[0108] Number of weld layers: This refers to the number of layers required to fill the weld when welding thick plates, in order to ensure the quality and performance of the weld. The more weld layers, the thinner each layer, the finer the microstructure of the weld joint, and the narrower the heat-affected zone. Each weld pass preheats the subsequent weld pass, while the subsequent weld pass heats the previous one; therefore, the joint has better ductility and toughness. The selection of the number of weld layers should be based on a comprehensive consideration of factors such as the thickness of the base metal, the joint type, the groove shape, the electrode diameter, and the welding current.

[0109] Wire feed speed: refers to the speed at which the welding wire enters the arc region, that is, the length of welding wire consumed per unit time.

[0110] The wire feed speed should be carefully selected based on factors such as welding current, arc length, and electrode diameter. Too fast a feed speed will result in an unstable arc, increased spatter, and incomplete fill; too slow a feed speed will result in an excessively long arc, shallow penetration, and defects such as undercut and lack of fusion.

[0111] Mechanical properties, such as tensile strength, yield strength, elongation, and impact toughness, reflect the workpiece's ability to resist deformation and fracture under external forces.

[0112] Chemical composition: such as element content, phase composition, etc., these components determine the physical properties and microstructure of the workpiece.

[0113] Surface quality: such as surface roughness, surface defects, etc., these qualities affect the appearance and corrosion resistance of the workpiece.

[0114] Step S14: Establish the mutual influence and coupling relationships among process parameters. Based on the above physical model and empirical data, analyze the influence mechanism and laws of each process parameter on the printing results.

[0115] Based on the degree and direction of the influence of each process parameter on the printing results, a multi-objective optimization model is established, and the key indicators in the printing results are used as the optimization objective function.

[0116] Process parameters are used as optimization variables, and the mutual influence and coupling relationships between process parameters are used as constraints.

[0117] The influencing mechanisms include: the effects of current and voltage on the morphology and temperature field of the molten pool, the effects of wire feed speed and layer height on the filling density and defect formation, and the effects of speed and path mode on weld overlap and surface roughness.

[0118] The purpose of this step is to find the optimal combination of process parameters through mathematical modeling and optimization algorithms, so that the printing results reach or are close to optimal. This step can resolve multiple conflicting objectives and constraints in the arc additive manufacturing process, improving printing quality and efficiency.

[0119] Step S15, according to the physical meaning and actual adjustable range of the process parameters, set the value interval for each process parameter, and discretize the interval into a predetermined number of discrete values; combine the discrete values of all process parameters to obtain a feasible solution space containing all possible combinations of process parameters;

[0120] For example, according to the melting point of aluminum alloy and the type of welding rod, the value interval of welding current can be set to 100-300A, and it can be divided into 10 discrete values, i.e. 100A, 120A, 140A, …, 280A, 300A. Similarly, the corresponding value interval and discrete value can also be set for other process parameters. If there are 7 process parameters, each with 10 discrete values, the size of the feasible solution space is 10^7, i.e. one hundred million possible combinations. Such a feasible solution space is too large and is not conducive to optimization and solution.

[0121] Alternatively, using the Latin hypercube sampling method, a set of uniformly distributed process parameter combinations is generated according to the range and resolution of the process parameters as initial samples of the feasible solution space; if 1000 samples are to be generated, each process parameter is divided into 1000 parts on average within its value interval, and a value is randomly selected from each part as the parameter value of the sample. The samples generated in this way can cover the main characteristics of the feasible solution space while reducing the amount of calculation.

[0122] Step S16, for the multi-objective optimization model, using a multi-objective optimization method, evaluating and ranking the initial samples according to the influence indicators of the process parameters on the additive results, to obtain a set of Pareto optimal solutions as the preferred samples of the feasible solution space.

[0123] The multi-objective optimization method is a method for solving optimization problems with multiple conflicting objective functions, which can find a set of non-inferior solutions, i.e. solutions that do not deteriorate other objective functions when improving in a certain objective function. Pareto optimal solution is the best part of non-inferior solution, i.e. solution that deteriorates other objective functions when improving in a certain objective function. For example, genetic algorithm can be used as a multi-objective optimization method, the initial samples are used as the initial population, and the operations such as selection, crossover and mutation are used for iterative evolution until a set of Pareto optimal solutions are obtained. These optimal solutions can reflect the best process parameter combination, which can achieve the optimization of printing efficiency and cost under the premise of meeting the quality requirements.

[0124] In further embodiments, a simple case is used to illustrate the data processing flow,

[0125] Multi-objective optimization model: min f1(x) = ∑ n i=1 (x i ) 2minf2(x) = ∑ n i=1 (x) i -2) 2 ;

[0126] stx i ∈[-10,10],i=1,…,n); x is an n-dimensional vector of process parameters, and f1 and f2 are two influencing parameters, representing thermal stress and deformation parameters, respectively. The goal is to find a set of process parameters that minimizes both indices simultaneously.

[0127] If NSGA-II is used to solve the problem, the process is as follows: A set of initial solutions is randomly generated, each consisting of n random numbers uniformly distributed within the interval [-10, 10]. The two objective function values ​​f1 and f2 corresponding to each solution are calculated. According to the non-dominated sorting method, the solutions are divided into different levels: solutions in the first level are not dominated by other solutions, solutions in the second level are dominated only by solutions in the first level, and so on. According to the crowding distance method, the distance between each solution and its neighboring solutions in the objective space is calculated; a larger distance indicates greater dispersion. According to the roulette wheel method, based on the solution level and crowding distance, a subset of solutions are selected for crossover and mutation operations to generate new solutions. The newly generated solutions are merged with the original solutions, and steps two through five are repeated until the preset number of iterations or convergence conditions are reached. The first-level solutions in the last generation are output as the Pareto optimal solution set, along with their corresponding objective function values.

[0128] For example, to optimize the process parameters of an electric arc additive manufacturing system, which has three process parameters: current, voltage, and wire feed speed, and two impact indicators: mechanical properties and surface roughness. The goal is to find a set of process parameters that make the mechanical properties and surface roughness simultaneously optimal. The data flow is as follows: First, collect or generate some sample data, including different combinations of process parameters and corresponding impact indicator values. Experimental design methods such as Latin hypercube sampling can be used to evenly cover the value range of process parameters, and impact indicator values can be obtained through simulation or experiment. Second, non-dominated sorting is performed on the sample data, i.e. the sample is divided into different levels, the first level of sample is not dominated by other samples, i.e. there is no other sample that is better than it in both impact indicators; the second level of sample is only dominated by the first level of sample, and so on. In this way, a set of non-dominated solutions, i.e. the Pareto optimal solution set, can be obtained. Then, the non-dominated solutions are calculated by the crowding distance, i.e. the sum of the distance between each solution and the adjacent solution in the target space, the greater the distance, the more dispersed the solution, the more likely to be retained. In this way, a set of crowding distance values can be obtained. Finally, the non-dominated solutions are selected and displayed, i.e. according to the crowding distance value or other standards, a part of representative solutions is selected from the non-dominated solutions, and displayed in the form of charts or tables. In this way, the expert can choose the most suitable process parameter combination.

[0129] In further embodiments, parameter optimization can also be performed by machine learning, deep learning, etc. methods, such as using neural networks, support vector machines, decision trees, random forests, XGBoost, Gaussian processes, etc.

[0130] In some embodiments, the specific process is as follows: In the previous step, different combinations of process parameters and corresponding part processing quality data are collected, including mechanical properties, surface roughness, and geometric accuracy indicators. Using the XGBoost machine learning algorithm, a prediction model is established based on process parameters and processing quality data, which can predict the processing quality indicators of the part according to the given process parameters. Then use the genetic algorithm to find the optimal process parameter combination according to the prediction model and the user's quality target, so that the predicted processing quality indicators reach the optimal or meet the requirements. Finally, the effectiveness and reliability of the prediction model and optimization algorithm are verified through experiments.

[0131] In some embodiments, different combinations of process parameters and corresponding part processing quality data, including mechanical properties, surface roughness, and geometric accuracy, are collected. Using a Gaussian process machine learning method, a proxy model is established based on the additive process parameters and the forming quality data, which can predict the additive forming quality indicators according to the given additive process parameters. Using a genetic algorithm, according to the proxy model and the quality target set by the user, the optimal additive process parameter combination is found, so that the predicted additive forming quality indicators reach the optimal or meet the requirements.

[0132] According to one aspect of the present application, step 2 is further:

[0133] Step S21, design a pre-printed workpiece and perform a printing operation, detect the composition and mechanical properties of the pre-printed workpiece, and form a detection result; according to the preferred samples of the feasible solution space, sequentially perform pre-printing experiments, and obtain pre-printed parts; perform non-destructive testing, chemical composition testing, and mechanical property testing on the pre-printed parts, and obtain the detection results;

[0134] For example, according to the preferred samples of the feasible solution space, a set of process parameter combinations is selected, such as a welding rod diameter of 3.2 mm, a welding current of 200 A, an arc voltage of 20 V, a focal length of 3.5 mm, a preheating temperature of 150 °C, a weld layer number of 2 layers, and a wire feeding speed of 10 m / min. Then, using an arc additive manufacturing device, the pre-printed workpiece is printed according to the set of process parameter combinations. The pre-printed workpiece can be a special structural part of the workpiece, such as four overhanging corners, or a scaled-down version of the entire workpiece. After printing, the pre-printed workpiece is heat treated and detected. The purpose of heat treatment is to improve the mechanical properties and corrosion resistance of the workpiece. The purpose of detection is to evaluate the impact of process parameters on additive results, such as tensile strength, chemical composition, and surface quality.

[0135] Step S22, compare the printing result with the optimization objective function, and determine whether there is an adverse effect, if the printing result meets the optimization objective function and there is no adverse effect, then keep the set of process parameter combinations; the adverse effects include defects, deformation and cracks; if the tensile strength of the pre-printed workpiece reaches 300 MPa or more, the chemical composition meets the requirements, and there are no obvious defects on the surface, it is considered that the set of process parameter combinations is effective, and it is recorded.

[0136] Step S23, if the printing result does not satisfy the optimization objective function or there is a bad effect, delete the set of process parameter combinations, and select another set of process parameter combinations from the feasible solution space for pre-printing experiment. For example, if the tensile strength of the pre-printed workpiece is lower than 294 MPa, or the chemical composition does not meet the requirements, or there are defects such as incomplete fusion, porosity, cracks and the like on the surface, it is considered that the set of process parameter combinations is invalid, and it is removed from the feasible solution space. Then, another set of process parameter combinations is randomly or orderly selected in the remaining feasible solution space, and steps S21 and S22 are repeated.

[0137] According to one aspect of the present application, each printing and detection process in step 21 is further:

[0138] Step S21a: pre-weld cleaning and substrate installation are performed, and the welding process parameters are debugged;

[0139] According to the melting point of the aluminum alloy and the type of the electrode, the value interval of the welding current can be set to 100-300 A, and it can be divided into 10 discrete values, i.e. 100 A, 120 A, 140 A, …, 280 A, 300 A. Similarly, corresponding value intervals and discrete values can also be set for other process parameters. Then, using the arc additive manufacturing equipment, printing experiments are performed on the substrate according to different process parameter combinations, and the printing effect and quality are observed.

[0140] Step S21b: pre-printing experiment is performed to obtain a pre-printed workpiece; according to the preferred samples of the feasible solution space, a set of process parameter combinations is selected, such as an electrode diameter of 3.2 mm, a welding current of 200 A, an arc voltage of 20 V, a focal length of 3.5 mm, a preheating temperature of 150 ℃, a number of weld layers of 2 layers, and a wire feeding speed of 10 m / min. Then, using the arc additive manufacturing equipment, the pre-printed workpiece is printed according to the set of process parameter combinations. The pre-printed workpiece can be a special structural part of the workpiece, such as four overhanging corners, or a scaled-down version of the entire workpiece. After printing, the pre-printed workpiece is heat treated and detected.

[0141] Step S21c: the pre-printed workpiece is heat treated, and after heat treatment, the pre-printed workpiece is subjected to non-destructive testing, chemical composition testing and mechanical property testing to obtain the detection results; the pre-printed workpiece is subjected to solution and aging treatment to improve its mechanical properties and corrosion resistance. Then, using X-ray, spectrometer, tensile testing machine and other instruments and equipment, the pre-printed workpiece is subjected to non-destructive testing, chemical composition testing and mechanical property testing, and the detection results are recorded.

[0142] Step S21d: if all the test results are qualified, then make formal printing preparation; if one of the test results is unqualified, then optimize the pre-printing scheme and process parameters, and re-execute the pre-printing experiment. If the tensile strength of the pre-printed piece reaches 300 MPa or more, the chemical composition meets the requirements, and there is no obvious defect on the surface, it is considered that the process parameter combination is effective, and it is recorded. If the tensile strength of the pre-printed piece is less than 294 MPa, or the chemical composition does not meet the requirements, or there are defects such as incomplete fusion, pores, cracks and the like on the surface, it is considered that the process parameter combination is invalid, and it is eliminated from the feasible solution space. Then, another set of process parameter combination is randomly or orderly selected in the remaining feasible solution space, and steps S21b and S21c are repeated.

[0143] According to one aspect of the present application, it further comprises step S24, optimizing the feasible solution space to obtain a workpiece printing preferred path solution set;

[0144] Step S241, multi-objective optimization is performed on the remaining feasible solution set to find out process parameter combinations that meet the Pareto optimal condition, i.e. process parameter combinations that improve one objective function without deteriorating another objective function;

[0145] The NSGA-II algorithm can be used as a multi-objective optimization method, the feasible solution set is used as an initial population, and iterative evolution is performed through selection, crossover, mutation and the like until a set of Pareto optimal solutions are converged, which are used as the preferred solution set. These optimal solutions can reflect the best process parameter combination, which realizes the optimization of printing efficiency and cost under the premise of meeting the quality requirements.

[0146] Step S242, genetic algorithm is used to further search and optimize the Pareto optimal solution set, new process parameter combinations are generated through simulation of natural selection, crossover and mutation and the like, and their fitness, i.e. the value of the optimization objective function, is evaluated, and the process parameter combination with the highest fitness is selected as the final optimal solution;

[0147] The GA algorithm can be used as a genetic algorithm method, the Pareto optimal solution set is used as an initial population, and iterative evolution is performed through selection, crossover, mutation and the like until an optimal solution is converged, which is used as the final solution. This optimal solution can reflect the best process parameter combination, which realizes the maximization of printing efficiency and cost under the premise of meeting the quality requirements.

[0148] Step S243, PCA algorithm is used to perform sensitivity analysis on the optimal solution, i.e. to analyze the contribution rate and influence degree of each process parameter on the optimization objective function, to determine the most critical process parameter, and to set a reasonable fluctuation range for it, so as to adjust and adapt in the actual printing process;

[0149] The PCA algorithm can be used as a sensitivity analysis method, the optimal solution is used as input data, and the contribution rate and influence degree of each process parameter on the principal component variance are obtained by principal component analysis, and the most critical process parameter is obtained according to the contribution rate size. Then, according to the influence degree, a fluctuation range is set, and fine tuning and adaptation are carried out in the range.

[0150] Step S244, constructing the workpiece printing preferred path solution set according to the calculation result. The printing path is generated according to the three-dimensional model and the additive path software, and fine tuning and adaptation are carried out according to the adjustment amplitude. Finally, the printing path is saved as the workpiece printing preferred path solution set.

[0151] According to one aspect of the present application, it further comprises the step S24, optimizing the feasible solution space to obtain the workpiece printing preferred path solution set;

[0152] Step S24a, constructing an intelligent agent for learning and optimizing the selection strategy of process parameters by using a deep reinforcement learning module according to the feasible solution space of process parameters and the detection results of pre-printing experiments; the DQN algorithm can be used as the deep reinforcement learning module, the feasible solution space is used as the state space, the process parameter combination is used as the action space, the detection results of pre-printing experiments are used as the reward function, and the Q value function is fitted by a neural network, so that the intelligent agent can select the optimal action according to the current state and the reward, that is, the optimal process parameter combination.

[0153] Step S24b, dynamically adjusting the exploration and utilization ratio of process parameters according to the selection strategy of the intelligent agent and the detection results of pre-printing experiments by using the multi-armed bandit technique, to balance the trade-off between exploring new feasible solutions and utilizing known excellent solutions; the UCB algorithm can be used as the multi-armed bandit technique, the average reward and the number of selections of each process parameter combination are calculated to calculate the confidence upper bound of each process parameter combination, and the process parameter combination with the maximum confidence upper bound is selected for pre-printing experiment. In this way, while utilizing known excellent solutions, new potential excellent solutions can also be explored.

[0154] Step S24c, until the predetermined number of iterations or convergence condition is reached, the final preferred feasible solution is obtained, and the workpiece printing preferred path solution set is constructed. If the selection strategy of the intelligent agent does not change after N (N=100) iterations, and the detection results of pre-printing experiments reach the optimal or near-optimal, it is considered that the final preferred feasible solution has been converged, and the process parameter combination is determined according to the solution, the printing path is generated according to the three-dimensional model and the additive path software, and saved as the workpiece printing preferred path solution set.

[0155] According to one aspect of the present application, the step S3 further comprises:

[0156] Step S31: Start the formal printing of the aluminum alloy vehicle body according to the process parameters and scheme set during the pre-printing, to obtain the aluminum alloy vehicle body;

[0157] Step S32: Perform heat treatment on the aluminum alloy vehicle body workpiece, after heat treatment, perform size detection and non-destructive detection on the aluminum alloy vehicle body workpiece, perform mechanical property detection and chemical composition detection on the furnace sample, to obtain the detection results;

[0158] Step S33: If all detection results are qualified, then machine according to the size of the three-dimensional model;

[0159] If the size detection is unqualified, further processing is performed by means of repair welding or machining, if the processing is qualified, then proceed to the next step, if the processing is unqualified, the aluminum alloy vehicle body workpiece is scrapped, and the aluminum alloy vehicle body piece is re-printed;

[0160] If the non-destructive detection is unqualified, the aluminum alloy vehicle body workpiece is scrapped, and the aluminum alloy vehicle body workpiece is re-printed;

[0161] If the chemical composition detection of the furnace sample is unqualified, the aluminum alloy vehicle body workpiece is scrapped, and the aluminum alloy vehicle body workpiece is re-printed after replacing the welding wire;

[0162] If the mechanical property detection of the furnace sample is unqualified, the aluminum alloy vehicle body workpiece is scrapped, and the aluminum alloy vehicle body workpiece is re-printed;

[0163] Step S34: Machine the aluminum alloy vehicle body;

[0164] Step S35: Perform size detection on the machined aluminum alloy vehicle body workpiece, if the size detection is qualified, proceed to the next step, if the size detection is unqualified, re-perform steps two to ten;

[0165] Step S36: Perform surface treatment on the qualified aluminum alloy vehicle body workpiece to obtain the finished workpiece.

[0166] This step is basically the same as the above embodiment, and the specific advantages will not be described here.

[0167] According to one aspect of the present application, in step S11, after constructing the three-dimensional model of the workpiece, a lightweight processing process of the model is further included, specifically:

[0168] Step S11a, retrieve the three-dimensional model of the workpiece, and divide the model into many small units for finite element analysis;

[0169] Step S11b, define the target and constraint of topology optimization, the target is usually to minimize the mass or flexibility of the structure, the constraint includes stress, displacement, frequency and heat conduction of the structure, and manufacturing constraint of the structure;

[0170] Step S11c, constructing and starting a topology optimization program module, automatically finding the best material distribution scheme, the topology optimization module iteratively updates the density distribution of the structure using the SIMP variable density method or the level set method, and evaluates the performance index of the structure until the preset target and constraint are met or the maximum number of iterations is reached.

[0171] According to an aspect of the present application, the step S32 of non-destructive testing of the aluminum alloy vehicle body workpiece further comprises:

[0172] Step S32a, constructing an ultrasonic array and phased array control system, arranging the ultrasonic transmitters and receivers in a predetermined manner to form an array to cover the surface or internal area of the pre-printed workpiece;

[0173] Step S32b, connecting the ultrasonic transmitters and receivers to the phased array control system to control and adjust the phase and amplitude of each element;

[0174] Step S32c, setting the phased array detection parameters and corresponding threshold values on the phased array control system, the phased array detection parameters including ultrasonic frequency, focusing depth, scanning angle, scanning speed and scanning range;

[0175] Constructing and selecting appropriate algorithms and models to analyze and judge the ultrasonic signals and give the detection results.

[0176] Using the ultrasonic array and phased array control system, the surface or internal area of the pre-printed workpiece is fully covered and high-resolution scanning is achieved, improving the efficiency and accuracy of non-destructive testing; according to the phased array detection parameters and corresponding threshold values, the ultrasonic signals are analyzed and judged in real time to quickly find the defects and unqualified conditions of the pre-printed workpiece, providing a basis for optimizing process parameters; constructing and selecting appropriate algorithms and models to intelligently process the ultrasonic signals and give the detection results and evaluation, providing a reference for improving the quality and performance of the pre-printed workpiece.

[0177] According to an aspect of the present application, the step S21b further comprises:

[0178] Installing at least two welding heads on the electric arc additive manufacturing machine, selecting appropriate positions, angles and spacings according to the shape and size of the workpiece;

[0179] Installing a synchronous control device on the electric arc additive manufacturing machine, adjusting the moving speed and direction of the welding heads according to the control signals;

[0180] According to the structural characteristics of the workpiece, selecting two working modes of simultaneous cooperation or alternating cooperation; receiving the input process parameters of each welding head and welding according to the working mode.

[0181] The pre-printing experiment is performed simultaneously or alternately by multiple welding heads to improve the printing speed and efficiency and shorten the printing time; the synchronous control device is used to realize accurate control and coordination of the multiple welding heads, ensure the printing quality and consistency, and avoid printing errors and conflicts; the positions, angles and spacings of the welding heads are flexibly selected according to the shape and size of the workpiece to adapt to different printing requirements and difficulties and improve the flexibility and adaptability of printing.

[0182] The pre-printing experiment in step 21b needs to modify the three-dimensional model according to the pre-printing requirements before printing, and simulate the additive path;

[0183] The pre-printing requirements are special structures of the printed workpiece or reduction of the entire workpiece;

[0184] The process of optimizing the pre-printing scheme according to the detection results in step 21c is as follows:

[0185] If there are internal defects such as incomplete fusion, pores and cracks, the temperature and humidity of the printing environment and the test welding process parameters: current, voltage and wire feeding speed need to be re-modified until a qualified pre-printed part is printed to obtain appropriate additive software parameters: layer height, speed and path mode.

[0186] The welding process parameters used in the formal printing in step 21e are the welding process parameters of the qualified pre-printed part;

[0187] In step ten, if the workpiece cannot meet the required size by means of repair welding, the three-dimensional model is modified to increase the allowance of the unqualified position, and the printing is re-performed;

[0188] The unqualified requirements of mechanical properties in steps 21c and 32 are that the tensile strength at room temperature is less than 294MPa.

[0189] In step 32, the unqualified requirements of non-destructive testing are that there are strip defects, penetrating defects or circular defects with a diameter greater than 2mm.

[0190] After detection, if the tensile strength of the workpiece at room temperature is less than 294MPa, the mechanical property detection of the furnace sample is unqualified, the aluminum alloy car body workpiece is scrapped, and the aluminum alloy car body workpiece is re-printed.

[0191] After the three-dimensional model is built, the aluminum alloy car body workpiece model is optimized according to the characteristics of the electric arc additive manufacturing process; the thickness of the square structure at the middle height is 10 mm, and only two paths of the inner wall and the outer wall can meet the thickness requirement; because the four corners are in a suspended and unsupported state, the substrate needs to be increased when the additive height reaches 130 mm; and the coincidence degree of the filling path and the outer wall path will affect the forming quality of each layer of printing, too low coincidence degree will cause the welding seam to not overlap, resulting in un-melted defects, and too high coincidence degree will cause excessive accumulation, affecting the next layer of additive, the preset coincidence degree is currently 15%, which is adjusted slightly during the workpiece additive process; in order to prevent excessive accumulation at the arc starting and ending overlap and the welding corner, the distance between the arc starting point and the arc ending point is set to control the starting and ending arc parameters, and the solution to the corner of the welding seam is to increase the speed of the transition path and increase the welding speed at the corner; finally, the surface flatness is realized by polishing, which can ensure the efficiency and quality of additive manufacturing, after the printing of the aluminum alloy car body workpiece model is optimized, according to the shape structure characteristics of the aluminum alloy car body and the corresponding path process, the model is imported into the software using IungoPNT software, and the ZigZag path is selected for additive manufacturing according to the shape structure characteristics of the workpiece.

[0192] Before the pre-printing experiment and the formal printing of the aluminum alloy car body workpiece, the surface of the substrate needs to be cleaned to ensure that the substrate surface is free of oil stains, water stains, oxidation film and other factors affecting the additive quality.

[0193] The pre-printing experiment in the above step 21a includes the following steps:

[0194] S1: modify the three-dimensional model before printing (i.e. reduce the entire workpiece, or only print the special structure of the workpiece);

[0195] S2: simulate the additive path according to the pre-printing model;

[0196] S3: test the printing process to test the appropriate additive process parameters (current, voltage, wire feeding speed) and obtain appropriate additive software parameters: layer height, speed, path mode;

[0197] S4: print a qualified pre-printing piece according to the corresponding process parameters tested:

[0198] S5: during the additive manufacturing process, the main protective gas is argon, and the gas flow is 20L / min-30L / min.

[0199] The printed aluminum alloy car body workpiece and the pre-printing piece generally need to be subjected to solid solution and aging treatment to enhance the performance of the workpiece; a special heat treatment furnace corresponding to the material should be selected for the heat treatment furnace; according to the need, a stabilization treatment is added in the middle of the related process to release the internal stress of the workpiece;

[0200] In the pre-printing experiment, only the special structure part of the workpiece is printed or the workpiece is printed after being reduced as a whole, so that the additive process parameters and additive software parameters can be tested to reduce the cost and test time, reduce the problems in the subsequent aluminum alloy car body workpiece additive process, and improve the yield of the workpiece.

[0201] As shown in Figure 3 , the present application adopts the method of electric arc additive manufacturing, which needs to use lungoPNT software to simulate additive path, and match appropriate additive path; according to the shape structure characteristics of the workpiece, zigzag path is selected for additive manufacturing.

[0202] As shown in Figure 4 , the present application adopts the method of electric arc additive manufacturing, which needs to use lungoPNT software to simulate additive path, and match appropriate additive path; according to the shape structure characteristics of the workpiece, zigzag path is selected for additive manufacturing.

[0203] As shown in Figure 5 , the present application adopts the method of electric arc additive manufacturing, which needs to use lungoPNT software to simulate additive path, and match appropriate additive path; according to the shape structure characteristics of the workpiece, zigzag path is selected for additive manufacturing.

[0204] As shown in Figure 6 , the present application adopts the method of electric arc additive manufacturing, which needs to use lungoPNT software to simulate additive path, and match appropriate additive path; according to the shape structure characteristics of the workpiece, zigzag path is selected for additive manufacturing. Figure 6

[0205] As shown in Figure 7 , the present application adopts the method of electric arc additive manufacturing, and the coincidence degree of the filling path and the outer wall path shown in the figure will affect the forming quality of each layer of printing, too low coincidence degree will cause the lap joint of the weld to be not up, and the unmelted defect appears, too high coincidence degree will appear high accumulation, which affects the next layer of additive, and the preset coincidence degree is 15%, which is adjusted according to the actual situation during the workpiece additive process.

[0206] As shown in Figure 8 ​As shown, the present application adopts the method of arc additive manufacturing. In order to prevent excessive accumulation at the start and end of the arc lap joint and the weld corner, the distance between the start and end of the arc is set to control the welding start and end parameters. The solution to the corner of the weld is to increase the speed of the transition path and increase the welding speed at the corner.

[0207] In further embodiments, step 21a further comprises adaptive control techniques to adjust the welding process parameters, as follows:

[0208] Step SP1: Install sensors and feedback devices;

[0209] Step SP11: Install various sensors on the arc additive manufacturing machine, such as current sensors, voltage sensors, temperature sensors, weld topography sensors, etc., for real-time measurement of various parameters during the additive process;

[0210] Step SP12: Install feedback devices on the arc additive manufacturing machine, such as servo motors, frequency converters, etc., for adjusting welding process parameters such as current, voltage, wire feed speed, etc. according to control signals;

[0211] Step SP2: Establish an adaptive control model;

[0212] Step SP21: According to the physical laws and empirical data of arc additive manufacturing, establish an adaptive control model such as a neural network model, a fuzzy logic model, etc. to describe the input-output relationship in the additive process;

[0213] Step SP22: According to the type and structure of the adaptive control model, select appropriate algorithms such as gradient descent algorithm, genetic algorithm, etc. to train and optimize the parameters of the adaptive control model;

[0214] Step SP3: Perform adaptive control;

[0215] Step SP31: Before starting the additive, input the initial value of the welding process parameters, and start the adaptive control model and algorithm;

[0216] Step SP32: During the additive process, real-time collection of sensor data is performed, and the data is used as input to the adaptive control model;

[0217] Step SP33: According to the output of the adaptive control model, generate a control signal and send it to the feedback device;

[0218] Step SP34: Adjust the welding process parameters according to the control signal and continue the additive;

[0219] Step SP35: Update and optimize the parameters of the adaptive control model according to the algorithm, and repeat sub-step SP32 to sub-step SP35 until the additive is completed.

[0220] In another embodiment of the present application, in step S21c, laser post-processing technology can also be used to quickly heat and cool the surface of the pre-printed part. The surface quality and microstructure are improved, and the mechanical properties and corrosion resistance are improved.

[0221] The specific steps are as follows:

[0222] Step STP1: Select a suitable laser and optical system;

[0223] Sub-step STP11: According to the material type, shape and size of the pre-printed part, select a suitable laser, such as a nanosecond pulse laser, a femtosecond laser, etc., for quickly heating and cooling the surface of the pre-printed part;

[0224] Sub-step STP12: According to the surface characteristics and quality requirements of the pre-printed part, select a suitable optical system, such as a lens, a mirror, a scanning mirror, etc., for focusing, deflecting and scanning the laser beam;

[0225] Step STP2: Set the laser post-processing parameters;

[0226] Sub-step STP21: According to the material properties and microstructure of the pre-printed part, set the laser post-processing parameters, such as laser power, pulse width, pulse frequency, scanning speed, scanning interval, scanning times, etc., and click the "Start Post-processing" button;

[0227] Step STP3: Perform laser post-processing;

[0228] Sub-step STP31: Place the pre-printed part into the laser post-processing machine and connect it with the laser and optical system;

[0229] Sub-step STP32: According to the set laser post-processing parameters, control the laser to emit a laser beam, and through the optical system, quickly heat and cool the surface of the pre-printed part;

[0230] Sub-step STP33: Monitor the laser post-processing process; if abnormal conditions occur, such as burning, cracking, deformation, etc., stop the post-processing in time and check the cause;

[0231] Sub-step STP34: After the laser post-processing is completed, take out the pre-printed part and mark and record it;

[0232] In another embodiment of the present application, in step S36, when the surface of the qualified aluminum alloy vehicle body workpiece is treated, electrolytic polishing technology is used, using electrolyte as the medium, under the action of a direct current electric field, the workpiece surface is subjected to anodic dissolution, thereby removing surface defects such as burrs, oxide layers, stress concentration areas, etc., and forming a smooth, bright and corrosion-resistant surface.

[0233] The specific steps are as follows: Step STEP1: Select appropriate electrolyte and electrode;

[0234] Sub-step STEP11: According to the material type, shape and size of the pre-printed piece, select appropriate electrolyte such as sulfuric acid, phosphoric acid, nitric acid, etc. for conducting current and dissolving metal on the surface of the workpiece;

[0235] Sub-step STEP12: According to the surface characteristics and quality requirements of the pre-printed piece, select appropriate electrode such as copper, aluminum, stainless steel, etc. for use as anode or cathode corresponding to the workpiece;

[0236] Step STEP2: Connect the DC power supply and control system;

[0237] Sub-step STEP21: Connect the electrode and workpiece to the DC power supply such as transformer, rectifier, etc. for providing stable DC electric field;

[0238] Sub-step STEP22: Connect the DC power supply to the control system such as computer, controller, etc. for controlling and adjusting electrolytic parameters;

[0239] Step STEP3: Set electrolytic polishing parameters and thresholds;

[0240] Sub-step STEP31: On the control system, set electrolytic polishing parameters such as voltage, current, temperature, time, gap, etc. and set corresponding thresholds such as surface roughness, surface gloss, etc.;

[0241] Sub-step STEP32: On the control system, select appropriate algorithms and models such as feedback control algorithm, optimization model, etc. for adjusting electrolytic parameters according to real-time data and feedback information;

[0242] Step STEP4: Perform electrolytic polishing;

[0243] Sub-step STEP41: Before starting polishing, start the control system and DC power supply and click the "Start Polishing" button;

[0244] Sub-step STEP42: During polishing, according to the set electrolytic polishing parameters and thresholds, as well as the selected algorithms and models, control the DC power supply to output stable voltage and current, and adjust temperature, time, gap, etc. parameters through the control system, and measure and process the data and feedback information of the workpiece surface;

[0245] Sub-step STEP43: On the control system, according to the set thresholds, as well as the selected algorithms and models, analyze and judge the data and feedback information, generate adjustment signals and send them to the DC power supply or control system;

[0246] Sub-step STEP44: according to the adjustment signal, adjusting the voltage, current, temperature, time, gap and the like on the direct current power supply or control system, and continuing polishing;

[0247] In addition, it should be understood that although the present specification is described in terms of embodiments, not every embodiment contains only one independent technical solution, and the description of the specification is only for the sake of clarity, and those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can be combined appropriately to form other embodiments that those skilled in the art can understand. The method of the present application is also applicable to the printing of the body of other alloys.

Claims

1. A method of arc additive manufacturing of a multi-structured large aluminum alloy vehicle body, characterized by, Comprising the following steps: Step S1, constructing a three-dimensional model of the workpiece, simulating the additive path, collecting and analyzing the influencing factors of the process parameters, and establishing a feasible solution space of the process parameters; Step S2, designing a pre-printed workpiece and performing a printing operation, detecting the composition and mechanical properties of the pre-printed workpiece, forming a detection result, screening feasible solutions based on the detection result, optimizing the feasible solution space, and obtaining a workpiece printing optimal path solution set; Step S3, constructing a vehicle body workpiece and performing a printing operation, and performing detection, and post-processing the vehicle body workpiece that passes the detection; The step 1 is further: Step S11, constructing a three-dimensional model of the workpiece, simulating the additive path; Step S12, determining the influencing factors of the process parameters according to the physical model and empirical data of the electric arc additive manufacturing, and quantifying and standardizing them; Step S13, determining the range and resolution of the process parameters, and according to the material properties and quality requirements of the workpiece, determining the influence index of the process parameters on the additive result; Step S14, establishing the mutual influence and coupling relationship between the process parameters, analyzing the influence mechanism and law of each process parameter on the printing result according to the above physical model and empirical data, and establishing a multi-objective optimization model by taking the key indicators in the printing result as the optimization objective function; Taking the process parameters as optimization variables and the mutual influence and coupling relationship between the process parameters as constraint conditions; Step S15, setting the value interval for each process parameter according to the physical meaning and actual adjustable range of the process parameter, and discretizing the interval into a predetermined number of discrete values; Combining the discrete values of all process parameters to obtain a feasible solution space containing all possible process parameter combinations; Step S16, for the multi-objective optimization model, using a multi-objective optimization method, evaluating and sorting the initial samples according to the influence index of the process parameters on the additive result, to obtain a group of Pareto optimal solutions as the preferred samples of the feasible solution space; The step 2 is further: Step S21, designing a pre-printed workpiece and performing a printing operation, detecting the composition and mechanical properties of the pre-printed workpiece, and forming a detection result; according to the preferred samples of the feasible solution space, sequentially performing pre-printing experiments, and obtaining a pre-printed workpiece; Performing non-destructive testing, chemical composition testing, and mechanical property testing on the pre-printed workpiece, and obtaining the detection result; Step S22, comparing the printing result with the optimization objective function, and determining whether there is an adverse effect, if the printing result meets the optimization objective function and there is no adverse effect, the process parameter combination is retained; the adverse effects include defects, deformation and cracks; Step S23, if the printing result does not meet the optimization objective function or there is an adverse effect, delete the process parameter combination, and select another process parameter combination from the feasible solution space for pre-printing experiment; Step S24, optimizing the feasible solution space to obtain a workpiece printing optimal path solution set; Step S241, multi-objective optimization is performed on the remaining feasible solution set to find a process parameter combination satisfying the Pareto optimal condition, that is, a process parameter combination in which improvement in one objective function does not lead to deterioration in another objective function; Step S242, a genetic algorithm is used to further search and optimize the Pareto optimal solution set, new process parameter combinations are generated through simulation of natural selection, crossover and mutation and the like, and the fitness, that is, the value of the optimization objective function, of the new process parameter combinations is evaluated, and the process parameter combination with the highest fitness is selected as the final optimal solution; Step S243, a PCA algorithm is used to perform sensitivity analysis on the optimal solution, that is, to analyze the contribution rate and influence degree of each process parameter on the optimization objective function, to determine the most critical process parameter, and to set a reasonable fluctuation range for the process parameter, so as to adjust and adapt in the actual printing process; Step S244, a workpiece printing optimal path solution set is constructed according to the calculation results.

2. A method of arc additive manufacturing of a multi-structured large aluminum alloy vehicle body as defined in claim 1, wherein, Each printing and detection process in the step 21 further comprises: Step S21a: pre-weld cleaning and substrate installation are performed, and welding process parameters are debugged; Step S21b: a pre-printing experiment is performed to obtain a pre-printed workpiece; Step S21c: the pre-printed workpiece is subjected to heat treatment, and after the heat treatment, the pre-printed workpiece is subjected to non-destructive detection, chemical composition detection and mechanical property detection to obtain detection results; Step S21d: if all the detection results are qualified, formal printing preparation is performed; If one of the detection results is unqualified, the pre-printing scheme and process parameters are optimized, and the pre-printing experiment is performed again.

3. A method of arc additive manufacturing of a multi-structured large aluminum alloy vehicle body as defined in claim 1, wherein, The step S3 further comprises: Step S31: formal printing of the aluminum alloy vehicle body is started according to the process parameters and scheme set for the pre-printed workpiece to obtain the aluminum alloy vehicle body; Step S32: the aluminum alloy vehicle body workpiece is subjected to heat treatment, and after the heat treatment, the aluminum alloy vehicle body workpiece is subjected to size detection and non-destructive detection, and the furnace sample is subjected to mechanical property detection and chemical composition detection to obtain detection results; Step S33: if all the detection results are qualified, machining is performed according to the size of the three-dimensional model; If the size detection is unqualified, further processing is performed through repair welding or machining, if the processing is qualified, the next step is performed, if the processing is unqualified, the aluminum alloy vehicle body workpiece is scrapped, and the aluminum alloy vehicle body workpiece is printed again; If the non-destructive detection is unqualified, the aluminum alloy vehicle body workpiece is scrapped, and the aluminum alloy vehicle body workpiece is printed again; If the chemical composition detection of the furnace sample is unqualified, the aluminum alloy vehicle body workpiece is scrapped, and the aluminum alloy vehicle body workpiece is printed again after the welding wire is replaced; If the mechanical property detection of the furnace sample is unqualified, the aluminum alloy vehicle body workpiece is scrapped, and the aluminum alloy vehicle body workpiece is printed again; Step S34: the aluminum alloy vehicle body is machined; Step S35: the machined aluminum alloy vehicle body workpiece is subjected to size detection, if the size detection is qualified, the next step is performed, if the size detection is unqualified, steps two to ten are performed again; Step S36: the surface of the qualified aluminum alloy vehicle body workpiece is treated to obtain a finished workpiece.

4. The method of arc additive manufacturing of a multi-structured large aluminum alloy vehicle body of claim 1, wherein, In the step S11, after the three-dimensional model of the workpiece is constructed, a lightweight processing process of the model is further included, specifically: Step S11a, call the three-dimensional model of the workpiece, divide the model into many small units for finite element analysis; Step S11b, define the target and constraints of topology optimization, the target is usually to minimize the mass or flexibility of the structure, the constraints include stress, displacement, frequency and heat conduction of the structure, and manufacturing constraints of the structure; Step S11c, build and start the topology optimization program module, automatically find the best material distribution scheme, the topology optimization module uses SIMP variable density method or level set method to iteratively update the density distribution of the structure, and evaluates the performance index of the structure until the preset target and constraints are met or the maximum iteration number is reached.

5. A method of arc additive manufacturing of a multi-structured large aluminum alloy vehicle body as defined in claim 3 wherein, The step S32 of non-destructive testing of the aluminum alloy vehicle body workpiece further comprises: Step S32a, build an ultrasonic array and phased array control system, arrange the ultrasonic transmitters and receivers in a predetermined manner to form an array to cover the surface or internal area of the pre-printed workpiece; Step S32b, connect the ultrasonic transmitters and receivers to the phased array control system to control and adjust the phase and amplitude of each element; Step S32c, set the phased array detection parameters and corresponding threshold values on the phased array control system, the phased array detection parameters include ultrasonic frequency, focusing depth, scanning angle, scanning speed and scanning range; Build and select appropriate algorithms and models to analyze and judge the ultrasonic signal and give the detection result.

6. A method of arc additive manufacturing of a multi-structured large aluminum alloy vehicle body as defined in claim 2 wherein, The step S21b further comprises: Install at least two welding heads on the electric arc additive manufacturing machine, select appropriate positions, angles and distances according to the shape and size of the workpiece; Install a synchronous control device on the electric arc additive manufacturing machine, adjust the moving speed and direction of the welding heads according to the control signal; according to the structural characteristics of the workpiece, select two working modes of simultaneous cooperation and alternating cooperation; receive the input process parameters of each welding head and weld according to the working mode.

Images