Welding control method for manufacturing electric tricycle

Abstract

The invention provides a welding control method for manufacturing an electric tricycle, and relates to the technical field of intelligent welding, and the welding control method comprises the following steps: determining a plurality of adaptive welding parameters according to matching of a plurality of part features; optimizing the plurality of initial welding sequences according to the thermal evaluation function and the plurality of adaptive welding parameters by taking minimization of thermal influence as an expected condition, and outputting an optimal welding sequence; and according to the optimal welding sequence, the multiple adaptive welding parameters are sequenced to generate an adaptive welding parameter sequence, and welding control is conducted according to the optimal welding sequence and the adaptive welding parameter sequence. By means of the method, the technical problems that in an existing method, due to the fact that the matched part welding sequence cannot be set according to the structural characteristics of the electro-tricycle, the phenomenon of local thermal stress concentration is likely to occur, and the welding quality of the tricycle body frame is poor can be solved, thermal stress concentration can be effectively reduced, and the welding quality is improved. And deformation and cracks in the welding process are reduced, and the welding quality and the structural stability of the vehicle body frame are remarkably improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent welding technology, and in particular to a welding control method for manufacturing electric tricycles. Background Art

[0002] As an important means of transportation, electric tricycles are widely used in urban travel and cargo transportation. Their body frames are usually welded from multiple metal parts. These welded parts not only require high welding quality, but also must have good structural strength and stability to ensure that they can withstand different loads and stresses during long-term use.

[0003] However, as the manufacturing process of electric tricycles becomes increasingly complex, traditional welding methods have gradually exposed many problems. Existing welding technologies mostly rely on operating experience to determine the welding sequence, and do not fully consider the complex structural design of electric tricycles and the thermal influence characteristics of each component. Local thermal stress concentration is prone to occur, resulting in the welding quality and structural stability of the body frame failing to meet expected requirements. Summary of the invention

[0004] The purpose of this application is to provide a welding control method for the manufacture of electric tricycles, so as to solve the technical problem that the existing welding method cannot set a matching welding sequence according to the structural characteristics of the electric tricycle, resulting in the phenomenon of local thermal stress concentration and poor welding quality of the body frame.

[0005] In view of the above problems, the present application provides a welding control method for manufacturing electric tricycles, including: obtaining multiple welding parts and multiple part features of the electric tricycle, and determining multiple adaptive welding parameters according to the matching of the multiple part features; enumerating welding sequences according to the multiple welding parts to generate multiple initial welding sequences; based on the welding twin space, with minimizing thermal influence as the expected condition, optimizing the multiple initial welding sequences according to the thermal evaluation function and the multiple adaptive welding parameters, and outputting the optimal welding sequence; mapping and sorting the multiple adaptive welding parameters according to the optimal welding sequence to generate an adaptive welding parameter sequence, and performing welding control on the electric tricycle according to the optimal welding sequence and the adaptive welding parameter sequence.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: By acquiring multiple welding parts and multiple part features of an electric tricycle, multiple adaptive welding parameters are determined according to the matching of the multiple part features; then, the welding sequence is enumerated according to the multiple welding parts to generate multiple initial welding sequences; further, based on the welding twin space, with minimizing thermal influence as the expected condition, the multiple initial welding sequences are optimized according to the thermal evaluation function and the multiple adaptive welding parameters, and the optimal welding sequence is output; then, according to the optimal welding sequence, the multiple adaptive welding parameters are mapped and sorted to generate an adaptive welding parameter sequence; finally, the electric tricycle is welded and controlled according to the optimal welding sequence and the adaptive welding parameter sequence; the above method can improve the matching degree between the part welding sequence and the structure of the electric tricycle, effectively reduce the concentration of thermal stress, and reduce the occurrence of deformation and cracks during welding, thereby significantly improving the welding quality and structural stability of the body frame.

[0007] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented according to the contents of the specification, and in order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0009] Figure 1 This is a flow chart of a welding control method for manufacturing an electric tricycle according to the present application.

[0010] Figure 2 This is a schematic flow chart of determining multiple adaptive welding parameters in a welding control method for manufacturing an electric tricycle according to the present application. DETAILED DESCRIPTION

[0011] The present application provides a welding control method for manufacturing electric tricycles, which solves the technical problem that the existing welding method cannot set a matching welding sequence according to the structural characteristics of the electric tricycle, resulting in the phenomenon of local thermal stress concentration and poor welding quality of the body frame. The matching degree between the welding sequence and the structure of the electric tricycle can be improved, the concentration of thermal stress can be effectively reduced, and the occurrence of deformation and cracks during welding can be reduced, thereby significantly improving the welding quality and structural stability of the body frame.

[0012] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0013] For examples, please see the attached Figure 1 The present application provides a welding control method for manufacturing an electric tricycle, which specifically includes the following steps: Step 1: Acquire multiple welding parts and multiple part features of the electric tricycle, and determine multiple adaptive welding parameters according to the matching of the multiple part features.

[0014] Further, if Figure 2 As shown, step one of this application also includes: Acquire multiple welding parts and multiple part features of the electric tricycle, wherein the part features correspond to the welding parts one by one, and the part features at least include material type, joint type and component force; randomly select a first part feature, wherein the first part feature includes a first material type, a first joint type and a first component force; input the first material type, the first joint type and the first component force into a welding parameter analyzer for analysis, output a first adaptive welding parameter, and add it to the multiple adaptive welding parameters.

[0015] Specifically, according to the design drawings and structural requirements of the electric tricycle body frame, all parts that need to be welded are identified, which usually include frame connection points, supporting structures, battery trays, battery racks, and wheel hub and frame connection points, etc., to obtain multiple welding parts; further obtain multiple part features of the multiple welding parts, wherein each welding part corresponds to a part feature, and the part features at least include material type, joint type and component stress. The material type means that different parts of the body frame may use different types of metal materials, such as carbon steel, aluminum alloy, stainless steel, etc., and the thermal conductivity, melting point and expansion coefficient of different materials vary greatly; the joint types include butt joints, corner joints, T-joints, etc.; the component stress means that the stress conditions of each part are different, and some parts need to bear greater loads, such as the supporting parts of the body frame or the connection parts of the suspension system. The parts with greater stress require higher welding quality and greater welding strength to ensure the stability and safety of the body.

[0016] Next, any one of the plurality of part features is randomly selected as a first part feature, wherein the first part feature includes a first material type, a first joint type, and a first component force. The first material type, the first joint type, and the first component force are then input into a welding parameter analyzer constructed based on a BP neural network for analysis, and a first adaptive welding parameter is output, which is added to the plurality of adaptive welding parameters.

[0017] Further, to construct a welding parameter analyzer, the present application also includes: Taking the attribute parameters of welding equipment as constraints and the expected welding quality as conditional constraints, the operation logs of similar equipment are retrieved to obtain sample material type sets, sample joint type sets, sample component force sets and sample welding parameter sets, wherein the welding parameters include welding current, welding voltage and welding speed; the BP neural network is supervisedly trained by using the sample material type sets, sample joint type sets, sample component force sets and sample welding parameter sets until convergence, so as to obtain the welding parameter analyzer.

[0018] Specifically, the welding equipment attribute parameters and expected welding quality are obtained. The expected welding quality can be set according to the welding requirements of the parts, including weld strength, smoothness of the weld joint (there are no obvious defects on the surface of the weld joint, such as pores, cracks, etc.); the welding equipment attribute parameters include equipment type, operation parameters and other characteristics. Then, the welding equipment attribute parameters are used as equipment constraints, and the expected welding quality is used as a conditional constraint (that is, the retrieved sample data must meet the expected welding quality). The operation logs of existing similar welding equipment are retrieved to obtain sample data in different welding scenarios, where the sample data includes a sample material type set, a sample joint type set, a sample component force set and a sample welding parameter set. The welding parameters include welding current, welding voltage and welding speed, where the welding current affects the welding heat input and determines the depth and width of the welding; the welding voltage affects the stability of the arc and the morphology of the molten pool, further affecting the appearance and strength of the weld; the welding speed determines the length of time for the welding heat input, thereby affecting the quality of the weld joint and the heat-affected zone.

[0019] Then, a welding parameter analyzer is constructed based on the BP neural network. The welding parameter analyzer is a BP neural network model that can be iteratively optimized in machine learning. It is obtained through supervised training with a sample data set. The welding parameter analyzer includes an input layer, a hidden layer, and an output layer. The input data of the input layer are material type, joint type, and component force, and the output data of the output layer are welding parameters (welding current, welding voltage, and welding speed). Then, the sample material type set, sample joint type set, sample component force set and sample welding parameter set are used as training data to supervise the training of the welding parameter analyzer. First, the data of the input layer (material type, joint type, component force) is forward propagated through the neural network, and after the nonlinear transformation of the hidden layer, the welding parameter prediction value of the output layer is finally obtained; then, according to the difference between the welding parameter prediction value of the output layer and the actual welding parameter, the error is calculated, and the commonly used error calculation method is the mean square error; then, the back propagation algorithm is used to adjust the weights and biases in the neural network according to the error, and the parameters of the neural network are updated by calculating the gradient, so that the output of the network is closer to the actual welding parameters; iterative training is repeated, and the parameters are optimized each time by back propagation. When the training error is less than the set threshold or after a set number of iterations, the training process is stopped. At this time, the prediction accuracy of the model has reached the expected level, and a trained welding parameter analyzer is obtained.

[0020] By building a welding parameter analyzer, the welding parameters can be automatically adjusted and optimized through a machine learning model to provide accurate welding parameters. This automated control can avoid the uncertainty caused by manual adjustment, and effectively avoid welding defects caused by improper selection of welding parameters, such as porosity, cracks, uneven welds, etc., thereby improving the strength and stability of welded joints.

[0021] Step 2: Enumerate the welding sequences according to the multiple welding positions to generate multiple initial welding sequences.

[0022] Specifically, welding sequences are enumerated according to the multiple welding positions. The core purpose of welding sequence enumeration is to explore sequential combinations of welding of different positions. For example, assuming that there are N welding positions, all possible arrangements can be generated by enumeration. For N positions, the number of enumerations of welding sequences is N! (factorial of N), that is, all possible arrangements and combinations, and multiple initial welding sequences are obtained, wherein each welding sequence represents a different position arrangement scheme. These initial welding sequences will become the basis for subsequent welding sequence optimization.

[0023] Step 3: Based on the welding twin space, with minimizing the thermal influence as the expected condition, the multiple initial welding sequences are optimized according to the thermal evaluation function and the multiple adaptive welding parameters, and the optimal welding sequence is output.

[0024] Furthermore, step three of this application also includes: Based on the attribute parameters of welding equipment and the attribute characteristics of electric tricycle, a welding twin space is constructed by simulation; a first welding sequence is randomly selected from the multiple initial welding sequences, and the multiple adaptive welding parameters are mapped and arranged according to the first welding sequence to generate a first parameter sequence.

[0025] Specifically, the goal of constructing a welding twin space is to reflect the actual welding process through a virtual model and simulate various physical phenomena in the welding process to predict factors such as welding quality and thermal influence; then, based on the welding equipment attribute parameters (equipment type, operating parameters, etc.) and the electric tricycle attribute characteristics (structural characteristics, geometric shape, material type, etc.), based on digital twin technology, through welding simulation software (such as ANSYS, Abaqus, COMSOL, etc.) combined with heat conduction, deformation and stress analysis models, a complete welding process simulation is carried out to construct a welding twin space. Among them, the welding twin space is a virtual digital environment that can reflect the status of the actual welding process in real time, including material status, temperature distribution, deformation, etc.

[0026] Next, any one welding sequence is randomly selected from the multiple initial welding sequences as a first welding sequence, and the multiple adaptive welding parameters are mapped and arranged according to the first welding sequence, that is, according to the first welding sequence, the multiple adaptive welding parameters corresponding to the multiple welding parts are mapped and arranged to obtain a first parameter sequence.

[0027] In the welding twin space, welding simulation is performed according to the first welding sequence and the first parameter sequence, regional temperature distribution diagrams under a plurality of simulated welding nodes are obtained, and a first temperature distribution diagram set is constructed.

[0028] Furthermore, the present application also includes: Acquire several welding positions except the first welding position in the first welding sequence, and acquire several thermal influence coverage areas of the several welding positions, wherein the first welding position is the initial welding position in the first welding sequence; perform welding simulation according to the first welding sequence and the first parameter sequence, and based on the several thermal influence coverage areas, record regional temperature distribution diagrams of several welding starting nodes of the several welding positions, obtain several regional temperature distribution diagrams, and construct the first temperature distribution diagram set, wherein the regional temperature distribution diagram is the temperature distribution characteristics of the thermal influence coverage area of ​​the welding position.

[0029] Specifically, first, the first welding part is selected in the first welding sequence, wherein the first welding part is the initial welding part in the first welding sequence, that is, the part to be welded first. Then, several welding parts other than the first welding part in the first welding sequence are obtained, that is, after the first welding sequence is determined, the remaining welding parts need to be determined next, and these parts are usually affected by the thermal force of the first welding part; further, several thermal force influence coverage areas of the several welding parts are obtained, wherein the thermal force influence coverage areas correspond to the welding parts one by one, and each welding part will be affected by the transfer and diffusion of heat during the welding process, wherein the temperature within this range will affect the welding quality of the welding part, and this area is the thermal force influence coverage area of ​​the welding part.

[0030] Then, in the welding twin space, welding simulation is performed according to the first welding sequence and the first parameter sequence, and according to the several thermal influence coverage areas, the regional temperature distribution diagrams of several welding start time nodes of the several welding parts are recorded to obtain several regional temperature distribution diagrams, wherein the regional temperature distribution diagram corresponds to the welding part one by one, and the regional temperature distribution diagram is the temperature distribution characteristic of the thermal influence coverage area of ​​the welding part, that is, the temperature distribution diagram shows the temperature conditions of the area where the welding part is located and the surrounding area at the time node, and can reflect the distribution of heat sources, the diffusion of heat and the generation of thermal stress; the first temperature distribution atlas set is constructed according to the several regional temperature distribution diagrams. Through the analysis of the temperature distribution diagram, it is possible to identify which welding parts may be subject to greater thermal influence and which parts are prone to thermal stress concentration or deformation problems during the welding process. Based on these analysis results, the welding sequence can be optimized to avoid excessive concentration of thermal stress and reduce deformation and cracks.

[0031] A thermal influence calculation is performed on the first temperature distribution atlas according to the thermal evaluation function to obtain a first thermal influence coefficient.

[0032] Furthermore, the present application also includes: According to a predetermined temperature threshold value, regional screening is performed on several regional temperature distribution maps in the first temperature distribution map set, and areas with temperatures greater than the predetermined temperature threshold value are set as thermally affected areas to obtain several thermally affected area sets; regional area ratio calculation is performed on the several thermally affected area sets to obtain several affected area ratios, wherein the affected area ratio is the ratio of the sum of the areas of multiple thermally affected areas in the thermally affected area set to the area of ​​the corresponding thermally affected coverage area; regional distribution discreteness analysis is performed on the several thermally affected area sets to obtain several regional distribution discreteness.

[0033] Specifically, a predetermined temperature threshold value is obtained, and the predetermined temperature threshold value can be set according to the thermal properties of the material, the requirements of the welding process, and the melting point, yield point and other characteristics of the material, wherein different welding parts correspond to different predetermined temperature threshold values; then, according to the predetermined temperature threshold value, several regional temperature distribution maps in the first temperature distribution map set are respectively screened, and the areas with temperatures greater than the predetermined temperature threshold value are set as thermally affected areas, that is, high temperature areas that have an adverse effect on the welding quality, and several thermally affected area sets are obtained.

[0034] Next, for the several sets of thermally affected regions, computer-aided design (CAD) software or finite element simulation platform can be used to extract the actual area of ​​the thermally affected region. For example, in the welding simulation process, the area with a temperature greater than the set threshold value will be identified and marked as the thermally affected region. The area of ​​each region can be obtained by grid calculation or by integral calculation of two-dimensional / three-dimensional geometric shapes, and the areas of multiple thermally affected regions are summed to obtain several total thermally affected region areas; further, the area ratio is calculated based on several total thermally affected region areas and several thermally affected coverage areas to obtain several affected region ratios, wherein the affected region ratio is the ratio of the total thermally affected region area to the area of ​​the corresponding thermally affected coverage area. The affected region ratio reflects the degree of influence of heat transfer and heat diffusion of a certain welding part during the welding process. The larger the affected region ratio, the greater the degree of thermal influence of the welding part during the welding process, which may lead to problems such as thermal stress concentration, deformation or cracks in the part.

[0035] Then, the regional distribution discreteness analysis is performed on the several sets of thermally affected zones. For example, the spatial positions of the thermally affected zones can be used as data points, and the discreteness is measured by calculating the spatial distribution variance of these zones. The regional distribution discreteness is an important indicator for measuring the uniformity of the spatial distribution of the thermally affected zones during welding. Low discreteness means that the distribution of the thermally affected zones is relatively uneven, which may lead to increased risks of local thermal stress concentration, deformation and cracks. High discreteness indicates that the heat is more evenly distributed, which is conducive to reducing thermal stress concentration, thereby improving welding quality and stability, and obtaining several regional distribution discreteness.

[0036] Finally, based on the thermal evaluation function, the thermal influence calculation is performed according to the several affected area ratios and several regional distribution discretenesses to obtain the first thermal influence coefficient, wherein the thermal influence coefficient reflects the overall thermal influence degree of the welding sequence, is positively correlated with the affected area ratio, and is negatively correlated with the regional distribution discreteness, that is, the larger the affected area ratio, the greater the degree of thermal influence of the area during welding, and the higher the risk of thermal stress concentration; the lower the regional distribution discreteness, which may lead to local thermal stress concentration, increased risks of deformation and cracks, and the higher the risk of thermal stress concentration.

[0037] Based on the thermal evaluation function, the first thermal influence coefficient is calculated according to the several influence area proportions and the several area distribution discretenesses.

[0038] Furthermore, the present application also includes: The expression of the thermal evaluation function is: ;in, is the thermal influence coefficient, is the regional proportion weight, is the distribution discrete weight, N is the number of welding locations, is the importance of the nth welding part, is the ratio of the affected area of ​​the nth welding part, is the regional distribution discreteness of the nth welding position.

[0039] Specifically, in the thermal evaluation function, is the thermal influence coefficient, is the regional proportion weight, is the distribution discrete weight, where the sum of the regional proportion weight and the distribution discrete weight is 1. The weight can be configured according to the index influence. The greater the index influence, the greater the corresponding weight. N is the number of welding parts. is the importance of the nth welding part, where the greater the importance of the part, the greater the impact; is the ratio of the affected area of ​​the nth welding part, is the regional distribution discreteness of the nth welding position.

[0040] Continue to optimize other welding sequences in the multiple initial welding sequences until a predetermined convergence condition is reached, obtain multiple thermal influence coefficients, and output the initial welding sequence corresponding to the minimum thermal influence coefficient as the optimal welding sequence.

[0041] Specifically, the same method is used to continue to optimize other welding sequences in the multiple initial welding sequences until the multiple initial welding sequences are analyzed and multiple thermal influence coefficients of the multiple initial welding sequences are obtained; finally, the initial welding sequence corresponding to the minimum thermal influence coefficient is selected as the optimal welding sequence, wherein the optimal welding sequence can effectively reduce local thermal stress concentration and reduce the risk of deformation and cracks.

[0042] Step 4: mapping and sorting the multiple adaptive welding parameters according to the optimal welding sequence to generate an adaptive welding parameter sequence, and performing welding control on the electric tricycle according to the optimal welding sequence and the adaptive welding parameter sequence.

[0043] Specifically, the multiple adaptive welding parameters are mapped and sorted according to the optimal welding sequence, that is, the corresponding adaptive welding parameters are mapped and sorted according to the order of multiple welding parts in the optimal welding sequence to obtain an adaptive welding parameter sequence; finally, during the welding process, the electric tricycle is welded and controlled according to the optimal welding sequence and the adaptive welding parameter sequence.

[0044] Further, after the electric tricycle is welded and controlled according to the optimal welding sequence and the adapted welding parameter sequence, the present application further includes: During the welding control process, the gas cylinder pressure and gas temperature of the shielding gas are monitored at fixed points, and the pressure deviation and gas temperature deviation are calculated; the adaptive welding parameters under the current monitoring node are obtained, and the adaptive welding parameters are compensated and analyzed according to the pressure deviation and gas temperature deviation, and the compensated welding parameters are generated to update the adaptive welding parameters, wherein a parameter compensation model is constructed through machine learning, and welding parameter compensation analysis is performed.

[0045] Specifically, during the welding control process, at the predetermined monitoring time node (such as monitoring every 3 minutes), the gas cylinder pressure and gas temperature of the shielding gas are monitored at a fixed point. The gas cylinder pressure directly affects the welding atmosphere and the protection effect. If the gas pressure is too low, it may cause unstable airflow and affect the welding quality. The shielding gas is heated during the flow process, and its temperature increase may reduce the protection effect and affect the welding quality. Then, the monitored gas cylinder pressure and gas temperature are compared with the preset standard pressure and gas temperature to calculate the pressure deviation and gas temperature deviation.

[0046] A parameter compensation model is constructed based on machine learning, such as using existing historical data or experimental data to construct a welding parameter compensation model using a BP neural network. The input of the model is the gas pressure deviation and the gas temperature deviation, and the output is the adjustment amount of the welding parameter. The parameter compensation model is trained to convergence through sample data, wherein the construction principle and training process of the parameter compensation model are the same as those of the above-mentioned welding parameter analyzer. Those skilled in the art can refer to the construction and training steps of the above-mentioned welding parameter analyzer. Finally, the gas pressure deviation and the gas temperature deviation are input into the parameter compensation model, and the parameter compensation amount is output; further, the adaptive welding parameters under the current monitoring node are obtained, and the adaptive welding parameters are optimized and compensated according to the parameter compensation amount, that is, the two are added and summed to obtain the compensated welding parameters, and the adaptive welding parameters are updated by the compensated welding parameters. By real-time monitoring of the gas cylinder pressure and gas temperature, and using machine learning methods to compensate and optimize the welding parameters, a more accurate and stable welding process can be achieved. This intelligent compensation method can significantly improve the welding quality, reduce welding defects caused by changes in shielding gas, and improve production efficiency.

[0047] In summary, the welding control method for manufacturing an electric tricycle provided in this application has the following technical effects: By acquiring multiple welding parts and multiple part features of an electric tricycle, multiple adaptive welding parameters are determined according to the matching of the multiple part features; then, the welding sequence is enumerated according to the multiple welding parts to generate multiple initial welding sequences; further, based on the welding twin space, with minimizing thermal influence as the expected condition, the multiple initial welding sequences are optimized according to the thermal evaluation function and the multiple adaptive welding parameters, and the optimal welding sequence is output; then, according to the optimal welding sequence, the multiple adaptive welding parameters are mapped and sorted to generate an adaptive welding parameter sequence; finally, the electric tricycle is welded and controlled according to the optimal welding sequence and the adaptive welding parameter sequence; the above method can improve the matching degree between the part welding sequence and the structure of the electric tricycle, effectively reduce the concentration of thermal stress, and reduce the occurrence of deformation and cracks during welding, thereby significantly improving the welding quality and structural stability of the body frame.

[0048] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

[0049] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is also intended to include these modifications and variations.

Claims

1. A welding control method for manufacturing an electric tricycle, characterized in that: include: Acquire multiple welding parts and multiple part features of the electric tricycle, and determine multiple adaptive welding parameters according to the matching of the multiple part features; Enumerate the welding sequences according to the plurality of welding positions to generate a plurality of initial welding sequences; Based on the welding twin space, with minimizing thermal influence as the expected condition, optimizing the multiple initial welding sequences according to the thermal evaluation function and the multiple adaptive welding parameters, and outputting the optimal welding sequence; The multiple adaptive welding parameters are mapped and sorted according to the optimal welding sequence to generate an adaptive welding parameter sequence, and the welding of the electric tricycle is controlled according to the optimal welding sequence and the adaptive welding parameter sequence.

2. The welding control method for manufacturing an electric tricycle according to claim 1, characterized in that: According to the matching of the multiple part features, a plurality of adaptive welding parameters are determined, including: Acquire multiple welding locations and multiple location features of the electric tricycle, wherein the location features correspond to the welding locations one by one, and the location features at least include material type, joint type, and component stress; Randomly selecting a first part feature, wherein the first part feature includes a first material type, a first joint type, and a first component force; The first material type, the first joint type and the first component force are input into a welding parameter analyzer for analysis, and a first adaptive welding parameter is output and added to the plurality of adaptive welding parameters.

3. The welding control method for manufacturing an electric tricycle according to claim 2, characterized in that: Build a welding parameter analyzer, including: Taking the attribute parameters of welding equipment as constraints and the expected welding quality as conditional constraints, the operation logs of similar equipment are retrieved to obtain a sample material type set, a sample joint type set, a sample component force set and a sample welding parameter set, wherein the welding parameters include welding current, welding voltage and welding speed; The BP neural network is supervised and trained by using the sample material type set, the sample joint type set, the sample component force set and the sample welding parameter set until convergence, thereby obtaining the welding parameter analyzer.

4. The welding control method for manufacturing an electric tricycle according to claim 3, characterized in that: Based on the welding twin space, with minimizing the thermal influence as the expected condition, the multiple initial welding sequences are optimized according to the thermal evaluation function and the multiple adaptive welding parameters, and the optimal welding sequence is output, including: Based on the attribute parameters of welding equipment and the attribute characteristics of electric tricycle, the welding twin space is constructed by simulation; Randomly selecting a first welding sequence from the multiple initial welding sequences, mapping and arranging the multiple adaptive welding parameters according to the first welding sequence, and generating a first parameter sequence; In the welding twin space, welding simulation is performed according to the first welding sequence and the first parameter sequence, regional temperature distribution diagrams under a plurality of simulated welding nodes are obtained, and a first temperature distribution diagram set is constructed; Performing thermal influence calculation on the first temperature distribution atlas according to the thermal evaluation function to obtain a first thermal influence coefficient; Continue to optimize other welding sequences in the multiple initial welding sequences until a predetermined convergence condition is reached, obtain multiple thermal influence coefficients, and output the initial welding sequence corresponding to the minimum thermal influence coefficient as the optimal welding sequence.

5. The welding control method for manufacturing an electric tricycle according to claim 4, characterized in that: Performing welding simulation according to the first welding sequence and the first parameter sequence, obtaining regional temperature distribution diagrams under a plurality of simulated welding nodes, and constructing a first temperature distribution diagram set, including: Acquire a plurality of welding positions other than the first welding position in the first welding sequence, and acquire a plurality of thermal influence coverage areas of the plurality of welding positions, wherein the first welding position is an initial welding position in the first welding sequence; Welding simulation is performed according to the first welding sequence and the first parameter sequence, and based on the several thermal influence coverage areas, regional temperature distribution diagrams of several welding start nodes of the several welding parts are recorded to obtain several regional temperature distribution diagrams, and the first temperature distribution diagram set is constructed, wherein the regional temperature distribution diagram is the temperature distribution characteristic of the thermal influence coverage area of ​​the welding part.

6. The welding control method for manufacturing an electric tricycle according to claim 5, characterized in that: Performing thermal influence calculation on the first temperature distribution atlas according to the thermal evaluation function to obtain a first thermal influence coefficient includes: performing regional screening on a plurality of regional temperature distribution maps in the first temperature distribution map set according to a predetermined temperature threshold value, setting a region with a temperature greater than the predetermined temperature threshold value as a thermally affected region, and obtaining a plurality of thermally affected region sets; Calculating the area ratio of the plurality of thermally affected area sets to obtain a plurality of affected area ratios, wherein the affected area ratio is the ratio of the sum of the areas of the plurality of thermally affected areas in the thermally affected area set to the area of ​​the corresponding thermally affected coverage area; Performing regional distribution discreteness analysis on the plurality of thermal influence region sets respectively to obtain a plurality of regional distribution discretenesses; Based on the thermal evaluation function, the first thermal influence coefficient is calculated according to the several influence area proportions and the several area distribution discretenesses.

7. The welding control method for manufacturing an electric tricycle according to claim 6, characterized in that: The expression of the thermal evaluation function is: ； in, is the thermal influence coefficient, is the regional proportion weight, is the distribution discrete weight, N is the number of welding locations, is the importance of the nth welding part, is the ratio of the affected area of ​​the nth welding part, is the regional distribution discreteness of the nth welding position.

8. The welding control method for manufacturing an electric tricycle according to claim 1, characterized in that: The electric tricycle is welded and controlled according to the optimal welding sequence and the adapted welding parameter sequence, and then further includes: During the welding control process, the gas cylinder pressure and gas temperature of the shielding gas are monitored at fixed points, and the pressure deviation and gas temperature deviation are calculated; The adaptive welding parameters under the current monitoring node are obtained, compensation analysis is performed on the adaptive welding parameters according to the gas pressure deviation and the gas temperature deviation, and compensation welding parameters are generated to update the adaptive welding parameters, wherein a parameter compensation model is constructed through machine learning to perform welding parameter compensation analysis.

Images