Robot welding path optimization method for electric tricycle manufacturing

Abstract

The invention discloses a robot welding path optimization method for electric tricycle manufacturing, and relates to the technical field related to data processing, a welding path parameter space of a battery compartment structure in electric tricycle manufacturing is obtained, and a plurality of first welding path parameters are randomly generated in the welding path parameter space; and according to the multiple first welding path parameters, welding simulation prediction of the battery compartment structure is conducted, and multiple first deformation parameters and multiple first welding seam protrusion parameters are obtained. And battery looseness analysis and battery damage analysis are carried out after the welding operation of the battery compartment. And on the basis of the analysis result, the multiple first welding path parameters are subjected to bidirectional adjustment and screening to obtain multiple second welding path parameters, parameter optimization continues to be conducted, and the optimal welding path parameters are obtained to serve as the robot welding path optimization result. The technical problem that in the prior art, a systematic and accurate optimization means is lacked in the welding process of the battery compartment of the electric tricycle, and consequently the fault of the battery compartment is difficult to reduce is solved.

Description

Technical Field

[0001] The present application relates to the technical field related to data processing, and in particular to a robot welding path optimization method for electric tricycle manufacturing. Background Art

[0002] With the widespread use of electric tricycles in urban logistics and short-distance travel, the battery compartment, as one of its core components, has an important impact on the performance and safety of the entire vehicle. The battery compartment structure of electric tricycles is usually manufactured by robot welding to improve production efficiency and welding quality. However, in the actual welding process, due to the unreasonable selection of welding path parameters, the battery compartment structure may be deformed and the weld bulge may occur, causing the battery to loosen or be damaged, affecting the performance of the entire vehicle and the service life of the battery. Existing welding path optimization methods mostly rely on empirical adjustments, lack of systematic and precise optimization methods, and it is difficult to effectively reduce the probability of battery loosening and damage.

[0003] Therefore, in the prior art, the welding process of the battery compartment of the electric tricycle lacks systematic and precise optimization means, resulting in the technical problem that it is difficult to reduce battery compartment failures. Summary of the invention

[0004] This application provides a robot welding path optimization method for electric tricycle manufacturing, which solves the technical problem that the welding process of the battery compartment of the electric tricycle in the prior art lacks systematic and accurate optimization means, resulting in the difficulty in reducing battery compartment failures. By optimizing the robot welding path parameters, the deformation and weld defects of the battery compartment structure of the electric tricycle are significantly reduced, the risk of battery loosening and damage is reduced, and the technical effect of improving the performance of the whole vehicle and the safety of the battery is improved.

[0005] The present application provides a robot welding path optimization method for electric tricycle manufacturing, the method comprising: obtaining a welding path parameter space of a battery compartment structure in electric tricycle manufacturing, and randomly generating a plurality of first welding path parameters in the welding path parameter space; performing welding simulation prediction of the battery compartment structure according to the plurality of first welding path parameters, and obtaining a plurality of first deformation parameters and a plurality of first weld protrusion parameters; performing battery loosening analysis and battery damage analysis after battery compartment welding operation according to the plurality of first deformation parameters and the plurality of first weld protrusion parameters, and obtaining a plurality of first battery loosening probabilities and a plurality of first battery damage probabilities; bidirectionally adjusting the plurality of first welding path parameters according to the plurality of first battery loosening probabilities and the plurality of first battery damage probabilities, and screening to obtain a plurality of second welding path parameters, and continuing to optimize the welding path parameters to obtain the optimal welding path parameters as the robot welding path optimization result.

[0006] In the implementation method, the welding path parameter space of the battery compartment structure in the manufacture of electric tricycles is obtained, and multiple first welding path parameters are randomly generated in the welding path parameter space, including: obtaining the welding path parameter space of the battery compartment structure welded by a welding robot in the manufacture of electric tricycles, wherein the welding path parameter space consists of a welding path space and a welding parameter space; and randomly generating multiple first welding path parameters in the welding path parameter space.

[0007] In the implementation method, according to the multiple first welding path parameters, the welding simulation prediction of the battery compartment structure is performed to obtain multiple first deformation parameters and multiple first weld protrusion parameters, including: pre-training a welding simulation channel for the welding simulation prediction of the battery compartment structure, wherein the welding simulation channel includes a deformation simulation path and a weld protrusion simulation path; respectively inputting the multiple first welding path parameters into the welding simulation channel, performing the welding simulation prediction of the battery compartment structure, and obtaining multiple first deformation parameters and multiple first weld protrusion parameters.

[0008] In the implementation method, a welding simulation channel for welding simulation prediction of the battery compartment structure is pre-trained, including: collecting a sample welding path parameter set according to the welding data of the battery compartment structure in historical time, and collecting deformation position parameters and deformation size parameters of the battery compartment after welding according to different sample welding path parameters, as well as bulge position parameters and bulge size parameters of the weld bulge, which are marked as a sample deformation parameter set and a sample weld bulge parameter set; using the sample welding path parameter set as input training data, and using the sample deformation parameter set and the sample weld bulge parameter set as supervision data, respectively, to train the deformation simulation path and the weld bulge simulation path; after the deformation simulation path and the weld bulge simulation path are trained and converged, they are combined to obtain a welding simulation channel.

[0009] In the implementation method, battery looseness analysis and battery damage analysis are performed after the battery compartment welding operation according to the multiple first deformation parameters and the multiple first weld protrusion parameters, respectively, to obtain multiple first battery loosening probabilities and multiple first battery damage probabilities, including: training an operation abnormality prediction channel according to the operation data of the battery compartment structure after welding, wherein the operation abnormality prediction channel includes a battery loosening analysis path and a battery damage analysis path; inputting the multiple first deformation parameters into the battery loosening analysis path, analyzing and outputting to obtain multiple first battery loosening probabilities; inputting the multiple first weld protrusion parameters into the battery damage analysis path, analyzing and outputting to obtain multiple first battery damage probabilities.

[0010] In the implementation method, according to the operation data of the battery compartment structure after welding, an operation abnormality prediction channel is trained, including: according to the operation data of the battery compartment structure during and after welding, a sample deformation parameter set and a sample weld protrusion parameter set are collected, and the loosening ratio of the battery under different sample deformation parameters and the damage ratio of the battery under different sample weld protrusion parameters are collected, and marked as a sample battery loosening probability set and a sample battery damage probability set; the sample deformation parameter set and the sample battery loosening probability set are used as input training data and supervision data to train a battery loosening analysis path until the training converges; the sample weld protrusion parameter set and the sample battery damage probability set are used as input training data and supervision data to train a battery damage analysis path until the training converges; and the converged battery loosening analysis path and battery damage analysis path are combined to obtain an operation abnormality prediction channel.

[0011] In the implementation method, the multiple first welding path parameters are bidirectionally adjusted according to the multiple first battery loosening probabilities and the multiple first battery damage probabilities, multiple second welding path parameters are screened and obtained, and the welding path parameters are continuously optimized to obtain the optimal welding path parameters, including: bidirectionally adjusting the multiple first welding path parameters according to the multiple first battery loosening probabilities and the multiple first battery damage probabilities to obtain multiple adjustment welding path parameters; predicting and analyzing to obtain multiple adjustment battery loosening probabilities and multiple adjustment battery damage probabilities of the multiple adjustment welding path parameters; weighted calculation of the multiple adjustment battery loosening probabilities and the multiple adjustment battery damage probabilities to obtain multiple adjustment battery abnormality probabilities, screening and obtaining the adjustment welding path parameters corresponding to the smallest several adjustment battery abnormality probabilities as multiple second welding path parameters; continuing to optimize the multiple second welding path parameters until the optimization converges, and obtaining the welding path parameters with the smallest battery abnormality probability as the optimal welding path parameters.

[0012] In the implementation method, the multiple first welding path parameters are bidirectionally adjusted according to the multiple first battery loosening probabilities and the multiple first battery damage probabilities to obtain multiple adjusted welding path parameters, including: within the multiple first battery loosening probabilities and the multiple first battery damage probabilities, screening to obtain the minimum first battery loosening probability and the minimum first battery damage probability, and using the first welding path parameters corresponding to the minimum first battery loosening probability and the minimum first battery damage probability as the loosening adjustment direction and the damage adjustment direction; using the path parameter adjustment step size to adjust the other multiple first welding path parameters respectively towards the loosening adjustment direction and the damage adjustment direction to obtain multiple adjusted welding path parameters.

[0013] The robot welding path optimization method for electric tricycle manufacturing proposed in this application is proposed. By obtaining the welding path parameter space of the battery compartment structure in the manufacturing of electric tricycles, a plurality of first welding path parameters are randomly generated in the welding path parameter space; according to the plurality of first welding path parameters, the welding simulation prediction of the battery compartment structure is performed to obtain a plurality of first deformation parameters and a plurality of first weld bulge parameters; according to the plurality of first deformation parameters and the plurality of first weld bulge parameters, the battery loosening analysis and battery damage analysis after the battery compartment welding operation are performed to obtain a plurality of first battery loosening probabilities and a plurality of first battery damage probabilities; according to the plurality of first battery loosening probabilities and the plurality of first battery damage probabilities, the plurality of first welding path parameters are adjusted bidirectionally, a plurality of second welding path parameters are screened, and the welding path parameters are continuously optimized to obtain the optimal welding path parameters as the robot welding path optimization result. The technical problem that the welding process of the battery compartment of the electric tricycle lacks systematic and accurate optimization means in the prior art, which makes it difficult to reduce the failure of the battery compartment, is solved. By optimizing the robot welding path parameters, the deformation and weld defects of the battery compartment structure of the electric tricycle are significantly reduced, the risk of battery loosening and damage is reduced, and the technical effect of improving the performance of the entire vehicle and battery safety is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0015] Figure 1 A schematic flow chart of a robot welding path optimization method for electric tricycle manufacturing provided in an embodiment of the present application.

[0016] Figure 2 A schematic flow chart of a method for optimizing a robot welding path for manufacturing an electric tricycle provided in an embodiment of the present application for generating a plurality of first welding path parameters. DETAILED DESCRIPTION

[0017] 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 in accordance with the contents of the specification. 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 listed below.

[0018] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0019] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0020] The present application embodiment provides a robot welding path optimization method for electric tricycle manufacturing, such as Figure 1 As shown, the method includes: Acquire a welding path parameter space of a battery compartment structure in the manufacture of an electric tricycle, and randomly generate a plurality of first welding path parameters in the welding path parameter space; perform welding simulation prediction of the battery compartment structure according to the plurality of first welding path parameters, and obtain a plurality of first deformation parameters and a plurality of first weld protrusion parameters; The welding path parameter space of the battery compartment structure in the manufacture of electric tricycles is obtained, and the welding path parameter space contains welding path data in the manufacture of electric tricycles and specific welding execution parameters, such as welding speed, welding current, voltage and other parameters. Subsequently, a plurality of first welding path parameters are randomly generated in the welding path parameter space. Based on the plurality of first welding path parameters, the plurality of first welding path parameters are input into the welding simulation channel to perform welding simulation prediction of the battery compartment structure, predict the deformation and weld bulge that may occur in the battery compartment structure after welding, and obtain a plurality of first deformation parameters and a plurality of first weld bulge parameters, wherein the first deformation parameters include specific position parameters of deformation and deformation size parameters, and the first weld bulge parameters include bulge position and bulge size parameters.

[0021] like Figure 2As shown, the method provided in the embodiment of the present application further includes: obtaining a welding path parameter space for welding a battery compartment structure by a welding robot in the manufacture of an electric tricycle, wherein the welding path parameter space is composed of a welding path space and a welding parameter space; A plurality of first welding path parameters are randomly generated in the welding path parameter space.

[0022] Obtaining a welding path parameter space of a battery compartment structure in the manufacture of an electric tricycle, and randomly generating a plurality of first welding path parameters in the welding path parameter space, including: obtaining a welding path parameter space for a welding robot to weld a battery compartment structure in the manufacture of an electric tricycle, wherein the welding path parameter space is composed of a welding path space and a welding parameter space. The welding path parameter space is all the paths that the welding robot can walk on the battery compartment structure, including the location of the welding point, the length and direction of the path, etc. The welding parameter space is the parameters that need to be controlled during the welding process, such as welding speed, welding current, voltage, etc. Further, randomly generating a plurality of first welding path parameters in the welding path parameter space, that is, randomly selecting a set of welding paths and welding parameters in the welding path parameter space for matching, and generating a plurality of first welding path parameters.

[0023] The method provided in an embodiment of the present application also includes: pre-training a welding simulation channel for welding simulation prediction of the battery compartment structure, wherein the welding simulation channel includes a deformation simulation path and a weld protrusion simulation path; respectively inputting the multiple first welding path parameters into the welding simulation channel, performing welding simulation prediction of the battery compartment structure, and obtaining multiple first deformation parameters and multiple first weld protrusion parameters.

[0024] According to the multiple first welding path parameters, a welding simulation prediction of the battery compartment structure is performed to obtain multiple first deformation parameters and multiple first weld protrusion parameters, including: pre-training a welding simulation channel for welding simulation prediction of the battery compartment structure, wherein the welding simulation channel includes a deformation simulation path and a weld protrusion simulation path. The deformation simulation path is used to predict the possible deformation of the battery compartment structure after welding according to the welding path parameters, and the weld protrusion simulation path is used to predict the possible weld protrusion of the battery compartment structure after welding according to the welding path parameters. The multiple first welding path parameters are respectively input into the welding simulation channel, and a welding simulation prediction of the battery compartment structure is performed to obtain multiple first deformation parameters and multiple first weld protrusion parameters.

[0025] The method provided in the embodiment of the present application also includes: collecting a sample welding path parameter set based on the welding data of the battery compartment structure in historical time, and collecting deformation position parameters and deformation size parameters of the battery compartment after welding according to different sample welding path parameters, as well as bulge position parameters and bulge size parameters of the weld bulge, and marking them as a sample deformation parameter set and a sample weld bulge parameter set; using the sample welding path parameter set as input training data, and using the sample deformation parameter set and the sample weld bulge parameter set as supervision data, respectively, to train a deformation simulation path and a weld bulge simulation path; after the deformation simulation path and the weld bulge simulation path are trained and converged, combining them to obtain a welding simulation channel.

[0026] Pre-training a welding simulation channel for welding simulation prediction of the battery compartment structure includes: obtaining a sample welding path parameter set in the historical welding data record according to the welding data recorded by the battery compartment structure in the historical time, and obtaining the deformation position parameters and deformation size parameters of the battery compartment after welding according to different sample welding path parameters, as well as the bulge position parameters and bulge size parameters of the weld bulge, which are marked as a sample deformation parameter set and a sample weld bulge parameter set. The sample deformation parameter set includes the deformation position parameters and deformation size parameters corresponding to the marked sample welding path parameters, and the sample weld bulge parameter set includes the bulge position parameters and bulge size parameters of the weld bulge corresponding to the marked sample welding path parameters. When training the deformation simulation path and the weld bulge simulation path, the sample welding path parameter set is used as the input training data, and the sample deformation parameter set and the sample weld bulge parameter set are used as the supervision data respectively, and the neural network model is supervised and trained respectively until the model converges. After the deformation simulation path and the weld bulge simulation path are trained and converged, the deformation simulation path and the weld bulge simulation path are combined to obtain the welding simulation channel.

[0027] According to the multiple first deformation parameters and the multiple first weld protrusion parameters, the battery looseness analysis and the battery damage analysis are performed after the battery compartment welding operation to obtain multiple first battery loosening probabilities and multiple first battery damage probabilities; according to the multiple first battery loosening probabilities and the multiple first battery damage probabilities, the multiple first welding path parameters are bidirectionally adjusted to obtain multiple second welding path parameters through screening, and the welding path parameters are continuously optimized to obtain the optimal welding path parameters as the robot welding path optimization result.

[0028] When the electric tricycle is running, if there is welding deformation, the battery will become loose, thereby affecting the battery performance. If there is a weld bulge, the weld bulge may squeeze the battery, thereby affecting the battery safety. According to the predicted multiple first deformation parameters and multiple first weld bulge parameters, the battery looseness analysis and battery damage analysis after the battery compartment welding operation are performed to obtain multiple first battery loosening probabilities and multiple first battery damage probabilities. Finally, according to the multiple first battery loosening probabilities and multiple first battery damage probabilities, the multiple first welding path parameters are adjusted bidirectionally, multiple second welding path parameters are screened, and the welding path parameters are optimized to obtain the optimal welding path parameters as the robot welding path optimization result. The technical problem that the welding process of the battery compartment of the electric tricycle lacks systematic and accurate optimization means in the prior art, resulting in the difficulty in reducing the failure of the battery compartment, is solved. By optimizing the robot welding path parameters, the deformation and weld defects of the battery compartment structure of the electric tricycle are significantly reduced, the risk of battery loosening and damage is reduced, and the technical effect of improving the performance of the whole vehicle and the safety of the battery is improved.

[0029] The method provided in the embodiment of the present application also includes: training an operation abnormality prediction channel according to the operation data of the battery compartment structure after welding, wherein the operation abnormality prediction channel includes a battery loosening analysis path and a battery damage analysis path; inputting the multiple first deformation parameters into the battery loosening analysis path, and analyzing and outputting to obtain multiple first battery loosening probabilities; inputting the multiple first weld protrusion parameters into the battery damage analysis path, and analyzing and outputting to obtain multiple first battery damage probabilities.

[0030] According to the multiple first deformation parameters and the multiple first weld protrusion parameters, the battery looseness analysis and the battery damage analysis after the battery compartment welding operation are performed to obtain multiple first battery loosening probabilities and multiple first battery damage probabilities, including: according to the operation data recorded after the battery compartment structure is welded, the operation abnormality prediction channel is trained, wherein the operation abnormality prediction channel includes a battery loosening analysis path and a battery damage analysis path. Further, the multiple first deformation parameters are input into the battery loosening analysis path, and the multiple first battery loosening probabilities are obtained through the battery loosening analysis path output. The multiple first weld protrusion parameters are input into the battery damage analysis path, and the multiple first battery damage probabilities are obtained through the battery damage analysis path analysis output.

[0031] The method provided in the embodiment of the present application also includes: collecting a sample deformation parameter set and a sample weld protrusion parameter set according to the operation data of the welding and post-welding of the battery compartment structure, and collecting the loosening ratio of the battery under different sample deformation parameters, as well as the damage ratio of the battery under different sample weld protrusion parameters, and marking them as a sample battery loosening probability set and a sample battery damage probability set; using the sample deformation parameter set and the sample battery loosening probability set as input training data and supervision data to train the battery loosening analysis path until the training converges; using the sample weld protrusion parameter set and the sample battery damage probability set as input training data and supervision data to train the battery damage analysis path until the training converges; combining the converged battery loosening analysis path and battery damage analysis path to obtain an operation abnormality prediction channel.

[0032] According to the operation data after the battery compartment structure is welded, the operation abnormality prediction channel is trained, including: according to the operation data of the battery compartment structure welding and welding after the historical records, different deformation parameter sets and weld bulge parameter sets in the data are collected to obtain sample deformation parameter sets and sample weld bulge parameter sets, and the looseness ratio of the battery under different sample deformation parameters and the damage ratio of the battery under different sample weld bulge parameters are obtained, which are marked as sample battery loosening probability sets and sample battery damage probability sets. The sample deformation parameter set and the corresponding sample battery loosening probability set are used as input training data and supervision data to train the battery loosening analysis path until the training converges, and the battery loosening analysis path is constructed based on the neural network model. The sample weld bulge parameter set and the sample battery damage probability set are used as input training data and supervision data to train the battery damage analysis path until the training converges, and the battery damage analysis path is constructed based on the neural network model. Finally, the trained battery loosening analysis path and battery damage analysis path are combined to obtain the operation abnormality prediction channel. The deformation parameters and sample weld bulge parameters are analyzed through the abnormal prediction channel to obtain the corresponding battery looseness ratio and battery damage ratio, thereby providing data support for subsequent abnormal optimization.

[0033] The method provided in the embodiment of the present application also includes: bidirectionally adjusting the multiple first welding path parameters according to the multiple first battery loosening probabilities and the multiple first battery damage probabilities to obtain multiple adjustment welding path parameters; predicting and analyzing to obtain multiple adjustment battery loosening probabilities and multiple adjustment battery damage probabilities of the multiple adjustment welding path parameters; weighted calculation of the multiple adjustment battery loosening probabilities and the multiple adjustment battery damage probabilities to obtain multiple adjustment battery abnormality probabilities, screening and obtaining the adjustment welding path parameters corresponding to the smallest several adjustment battery abnormality probabilities as multiple second welding path parameters; continuing to optimize the multiple second welding path parameters until the optimization converges, and obtaining the welding path parameters with the smallest battery abnormality probability as the optimal welding path parameters.

[0034] According to the multiple first battery loosening probabilities and the multiple first battery damage probabilities, the multiple first welding path parameters are adjusted bidirectionally, multiple second welding path parameters are screened, and the welding path parameters are continuously optimized to obtain the optimal welding path parameters, including: according to the multiple first battery loosening probabilities and the multiple first battery damage probabilities, the multiple first welding path parameters are adjusted bidirectionally, and the welding path parameters corresponding to the minimum first battery loosening probability and the minimum first battery damage probability are used as the optimization direction, and the welding path and welding parameters in the first welding path parameters are adjusted with a fixed optimization parameter step length as the adjustment range to obtain multiple adjustment welding path parameters. Predictive analysis is performed on the multiple adjustment welding path parameters to obtain multiple adjustment battery loosening probabilities and multiple adjustment battery damage probabilities of the multiple adjustment welding path parameters. Weighted sum calculation is performed on the multiple adjustment battery loosening probabilities and multiple adjustment battery damage probabilities to obtain multiple adjustment battery abnormality probabilities, wherein the weights are set based on actual preference requirements. The adjustment battery abnormality probabilities are sorted from large to small, and the adjustment welding path parameters corresponding to several adjustment battery damage probabilities that are ranked later in the sorting from large to small are screened and obtained as multiple second welding path parameters. Continue to optimize the multiple second welding path parameters in the same steps until the optimization converges, and obtain the welding path parameters with the minimum battery abnormality probability as the optimal welding path parameters. The optimization convergence condition is to reach a preset number of iterations or adjust the battery damage probability to meet the preset value requirements.

[0035] The method provided in an embodiment of the present application also includes: screening and obtaining the minimum first battery loosening probability and the minimum first battery damage probability among the multiple first battery loosening probabilities and the multiple first battery damage probabilities, and using the first welding path parameters corresponding to the minimum first battery loosening probability and the minimum first battery damage probability as the loosening adjustment direction and the damage adjustment direction; using the path parameter adjustment step to adjust the other multiple first welding path parameters respectively towards the loosening adjustment direction and the damage adjustment direction to obtain multiple adjusted welding path parameters.

[0036] According to the multiple first battery loosening probabilities and the multiple first battery damage probabilities, the multiple first welding path parameters are bidirectionally adjusted to obtain multiple adjusted welding path parameters, including: obtaining multiple first battery loosening probabilities and multiple first battery damage probabilities from the historical records of the battery compartment structure welding and the operation data after welding, and among the multiple first battery loosening probabilities and the multiple first battery damage probabilities, extracting the battery loosening probability and the first battery damage probability whose sum of the first battery loosening probability and the first battery damage probability is the smallest and whose ratio is close to 1, to obtain the minimum first battery loosening probability and the minimum first battery damage probability, and using the first welding path parameters corresponding to the minimum first battery loosening probability and the minimum first battery damage probability as the loosening adjustment direction and the damage adjustment direction.

[0037] The technical solution provided by the embodiment of the present invention obtains the welding path parameter space of the battery compartment structure in the manufacturing of the electric tricycle, and randomly generates multiple first welding path parameters in the welding path parameter space; according to the multiple first welding path parameters, the welding simulation prediction of the battery compartment structure is performed to obtain multiple first deformation parameters and multiple first weld bulge parameters; according to the multiple first deformation parameters and multiple first weld bulge parameters, the battery loosening analysis and battery damage analysis after the battery compartment welding operation are performed to obtain multiple first battery loosening probabilities and multiple first battery damage probabilities; according to the multiple first battery loosening probabilities and multiple first battery damage probabilities, the multiple first welding path parameters are adjusted bidirectionally, multiple second welding path parameters are screened and obtained, and the welding path parameters are continuously optimized to obtain the optimal welding path parameters as the robot welding path optimization result. The technical problem that the welding process of the battery compartment of the electric tricycle lacks systematic and accurate optimization means in the prior art, which makes it difficult to reduce the battery compartment failure, is solved. By optimizing the robot welding path parameters, the deformation and weld defects of the battery compartment structure of the electric tricycle are significantly reduced, the risk of battery loosening and damage is reduced, and the technical effect of improving the performance of the whole vehicle and the safety of the battery is improved.

[0038] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A robot welding path optimization method for electric tricycle manufacturing, characterized in that: The method comprises: Acquire a welding path parameter space of a battery compartment structure in the manufacture of an electric tricycle, and randomly generate a plurality of first welding path parameters in the welding path parameter space; According to the plurality of first welding path parameters, a welding simulation prediction of the battery compartment structure is performed to obtain a plurality of first deformation parameters and a plurality of first weld protrusion parameters; According to the plurality of first deformation parameters and the plurality of first weld protrusion parameters, respectively, a battery loosening analysis and a battery damage analysis after the battery compartment welding operation are performed to obtain a plurality of first battery loosening probabilities and a plurality of first battery damage probabilities; According to the multiple first battery loosening probabilities and the multiple first battery damage probabilities, the multiple first welding path parameters are bidirectionally adjusted, multiple second welding path parameters are screened and obtained, and the welding path parameters are further optimized to obtain the optimal welding path parameters as the robot welding path optimization result.

2. The robot welding path optimization method for electric tricycle manufacturing according to claim 1 is characterized in that: Obtaining a welding path parameter space of a battery compartment structure in the manufacture of an electric tricycle, and randomly generating a plurality of first welding path parameters in the welding path parameter space, including: Obtaining a welding path parameter space for welding a battery compartment structure by a welding robot in the manufacture of an electric tricycle, wherein the welding path parameter space consists of a welding path space and a welding parameter space; A plurality of first welding path parameters are randomly generated in the welding path parameter space.

3. The robot welding path optimization method for electric tricycle manufacturing according to claim 1 is characterized in that: According to the plurality of first welding path parameters, a welding simulation prediction of the battery compartment structure is performed to obtain a plurality of first deformation parameters and a plurality of first weld protrusion parameters, including: Pre-training a welding simulation channel for welding simulation prediction of the battery compartment structure, wherein the welding simulation channel includes a deformation simulation path and a weld bulge simulation path; The multiple first welding path parameters are respectively input into the welding simulation channel to perform welding simulation prediction on the battery compartment structure, and obtain multiple first deformation parameters and multiple first weld protrusion parameters.

4. The robot welding path optimization method for electric tricycle manufacturing according to claim 3 is characterized in that: The welding simulation channel for pre-training the welding simulation prediction of the battery compartment structure includes: According to the welding data of the battery compartment structure in the historical time, a sample welding path parameter set is collected, and deformation position parameters and deformation size parameters of the battery compartment after welding according to different sample welding path parameters, as well as bulge position parameters and bulge size parameters of the weld bulge, are collected, and marked as a sample deformation parameter set and a sample weld bulge parameter set; The sample welding path parameter set is used as input training data, and the sample deformation parameter set and the sample weld bulge parameter set are used as supervision data to train the deformation simulation path and the weld bulge simulation path respectively; After the deformation simulation path and the weld bulge simulation path are trained and converged, a welding simulation channel is obtained by combining them.

5. The robot welding path optimization method for electric tricycle manufacturing according to claim 1 is characterized in that: According to the plurality of first deformation parameters and the plurality of first weld protrusion parameters, respectively, a battery loosening analysis and a battery damage analysis after the battery compartment welding operation are performed to obtain a plurality of first battery loosening probabilities and a plurality of first battery damage probabilities, including: According to the operation data of the battery compartment structure after welding, the operation abnormality prediction channel is trained, wherein the operation abnormality prediction channel includes a battery looseness analysis path and a battery damage analysis path; Inputting the plurality of first deformation parameters into the battery loosening analysis path, and analyzing and outputting to obtain a plurality of first battery loosening probabilities; The multiple first weld protrusion parameters are input into the battery damage analysis path, and the analysis output obtains multiple first battery damage probabilities.

6. The robot welding path optimization method for electric tricycle manufacturing according to claim 5 is characterized in that: According to the operation data of the battery compartment structure after welding, the abnormal operation prediction channel is trained, including: According to the operation data of the battery compartment structure welding and after welding, a sample deformation parameter set and a sample weld bulge parameter set are collected, and the looseness ratio of the battery under different sample deformation parameters and the damage ratio of the battery under different sample weld bulge parameters are collected, and marked as a sample battery looseness probability set and a sample battery damage probability set; Using the sample deformation parameter set and the sample battery loosening probability set as input training data and supervision data, training the battery loosening analysis path until the training converges; Using the sample weld bulge parameter set and the sample battery damage probability set as input training data and supervision data, training the battery damage analysis path until the training converges; The converged battery looseness analysis path and battery damage analysis path are combined to obtain an abnormal operation prediction channel.

7. The robot welding path optimization method for electric tricycle manufacturing according to claim 1 is characterized in that: According to the plurality of first battery loosening probabilities and the plurality of first battery damage probabilities, the plurality of first welding path parameters are adjusted bidirectionally, a plurality of second welding path parameters are screened and obtained, and the welding path parameters are continuously optimized to obtain the optimal welding path parameters, including: According to the plurality of first battery loosening probabilities and the plurality of first battery damage probabilities, bidirectionally adjusting the plurality of first welding path parameters to obtain a plurality of adjusted welding path parameters; Predictively analyzing and obtaining a plurality of adjusted battery loosening probabilities and a plurality of adjusted battery damage probabilities of the plurality of adjusted welding path parameters; Performing weighted calculation on the multiple adjustment battery loosening probabilities and the multiple adjustment battery damage probabilities to obtain multiple adjustment battery abnormality probabilities, and screening and obtaining adjustment welding path parameters corresponding to the smallest multiple adjustment battery abnormality probabilities as multiple second welding path parameters; Continue to optimize the multiple second welding path parameters until the optimization converges, and obtain the welding path parameters with the minimum probability of battery abnormality as the optimal welding path parameters.

8. The robot welding path optimization method for electric tricycle manufacturing according to claim 7 is characterized in that: According to the plurality of first battery loosening probabilities and the plurality of first battery damage probabilities, the plurality of first welding path parameters are bidirectionally adjusted to obtain a plurality of adjusted welding path parameters, including: Among the multiple first battery loosening probabilities and the multiple first battery damage probabilities, screen and obtain the minimum first battery loosening probability and the minimum first battery damage probability, and use the first welding path parameters corresponding to the minimum first battery loosening probability and the minimum first battery damage probability as the loosening adjustment direction and the damage adjustment direction; The path parameter adjustment step is adopted to adjust the other multiple first welding path parameters respectively toward the loose adjustment direction and the damage adjustment direction to obtain multiple adjusted welding path parameters.

Images