Method for constructing laser build-up welding joint interface fusion width prediction model, laser build-up welding joint interface fusion width prediction method, system and electronic equipment

Abstract

This invention proposes a method for constructing a laser lap welding interface weld width prediction model. The method includes: collecting surface weld width data, weld depth data, and interface weld width data at the interface between a first workpiece and a second workpiece that are laser-lapped together; constructing a dataset using the collected surface weld width data and weld depth data as input data and the interface weld width data as output data; and training the interface weld width prediction model using the dataset. According to some embodiments of this invention, a laser lap welding interface weld width prediction model can be constructed, enabling convenient and non-destructive prediction of interface weld width in production, thereby aiding in the evaluation of laser lap welding quality. This invention also proposes corresponding laser lap welding interface weld width prediction methods, systems, electronic devices, and computer program products.

Description

Technical Field

[0001] This invention relates to the field of laser processing and inspection technology, and in particular to a method for constructing a laser lap welding interface weld width prediction model, a laser lap welding interface weld width prediction method, a laser lap welding interface weld width prediction system, and electronic equipment and computer program products for laser lap welding interface weld width prediction. Background Technology

[0002] In laser welding, the characteristics of the weld, especially the dimensional features of certain parts, often reflect the quality of the weld. Particularly in laser lap welding for batteries, the weld must meet conductivity requirements. Accordingly, the weld width at the interface between the two welded workpieces is a crucial indicator for evaluating the weld's conductivity, and thus, the quality of this type of weld. However, current technology cannot obtain data on the weld width without destructive methods. Therefore, a non-destructive technique for obtaining weld width data from laser lap welding is needed. Summary of the Invention

[0003] The purpose of this invention is to provide a method for constructing a laser lap weld joint width prediction model, a laser lap weld joint width prediction method, a laser lap weld joint width prediction system, an electronic device and a computer program product for laser lap weld joint width prediction, so as to at least partially solve the problems existing in the prior art.

[0004] According to a first aspect of the present invention, a method for constructing a weld width prediction model for laser cladding joints is provided, the method comprising:

[0005] Data on surface weld width and weld depth of multiple welds from laser-welded overlapping first and second workpieces under predetermined beam characteristic parameters, as well as weld width data at the interface between the first and second workpieces, are collected; and

[0006] A dataset is constructed using the collected surface melt width and melt depth data as input data and the interface melt width data as output data. The interface melt width prediction model is then trained using the dataset.

[0007] According to the first aspect of the present invention, the method for constructing a laser lap weld joint width prediction model can construct a laser lap weld joint width prediction model, so that during or after welding, only two measurable data, the surface weld width and weld depth, can be simply obtained, and the prediction model can be used to achieve non-destructive joint width prediction, especially real-time joint width prediction, thereby helping to evaluate the quality of laser lap welds, especially in real time, without considering various related operating parameters of the laser lap weld operation or the characteristics or parameters of the workpiece.

[0008] According to an optional embodiment of the present invention, the dataset is divided into a training dataset and a validation dataset. The training dataset is used to train the interface weld width prediction model and adjust its parameters. The validation dataset is then used to validate the trained interface weld width prediction model until the deviation between the interface weld width data output by the trained model and the corresponding interface weld width data in the validation dataset is less than a preset threshold. This results in a successfully trained interface weld width prediction model. This embodiment enables the interface weld width prediction model to converge effectively, thus allowing it to effectively predict interface weld width.

[0009] According to an optional embodiment of the present invention, surface weld width data, weld depth data, and interface weld width data of multiple welds in the laser lap welding under different process control parameters are collected. This embodiment enables the acquisition of weld data of different sizes by changing laser power, welding speed, etc., under the same laser beam characteristic parameters, enriching the diversity of the collected data. This allows for better construction of a dataset to train the interface weld width prediction model and avoids model overfitting.

[0010] According to an optional embodiment of the present invention, surface weld width data is obtained by photographing and measuring the weld using a camera. This embodiment allows for convenient, non-destructive acquisition of weld surface weld width data, particularly through real-time acquisition using a coaxial camera with a machined lens.

[0011] According to an optional embodiment of the present invention, weld penetration data is acquired using an OCT (Optical Coherence Tomography) sensor. This embodiment allows for the convenient acquisition of weld penetration data in a non-destructive manner.

[0012] According to an optional embodiment of the present invention, the weld width data is obtained by destructively treating the weld to expose the junction between the first workpiece and the second workpiece and measuring the weld width at the junction.

[0013] According to an optional embodiment of the present invention, the interface width prediction model is a linear regression model or a neural network model.

[0014] According to a second aspect of the present invention, a method for predicting the weld width at the interface of a laser lap weld is provided. The method is performed using a weld width prediction model constructed by the method for constructing a weld width prediction model at the interface of a laser lap weld according to a first aspect of the present invention. The method includes:

[0015] Obtain surface weld width and weld depth data of the laser-welded seam of the first and second workpieces stacked on top of each other under the predetermined beam characteristic parameters; and

[0016] The interface weld width prediction model is used to predict the interface weld width data at the interface location between the first workpiece and the second workpiece based on the surface weld width data and weld depth data.

[0017] According to the laser lap welding interface width prediction method of the second aspect of the present invention, when it is desired to obtain the interface width of the weld, the surface width and depth data of the weld can be obtained only by relevant sensor equipment, and the interface width prediction model is used to predict the interface width data based on these two measurable data, without having to use destructive means to damage the material, and without having to consider various relevant operating parameters of the laser lap welding operation or the characteristics or parameters of the workpiece.

[0018] According to a third aspect of the present invention, a laser cladding joint width prediction system is provided, comprising:

[0019] A data acquisition module is configured to acquire surface weld width and weld depth data of a laser-welded weld between two stacked workpieces; and

[0020] The prediction module is configured to invoke a weld width prediction model to predict the weld width data at the weld width position between the first workpiece and the second workpiece based on the surface weld width data and weld depth data. The weld width prediction model is constructed by the method for constructing a laser lap weld weld width prediction model according to the first aspect of the present invention and for the laser lap weld of the first workpiece and the second workpiece.

[0021] By means of the laser lap welding interface weld width prediction system according to the third aspect of the present invention, it is possible to predict the interface weld width data of the weld in a non-destructive manner.

[0022] According to a fourth aspect of the present invention, an electronic device for predicting the weld width at the interface of laser cladding is provided, comprising:

[0023] At least one processor; and

[0024] A memory storing program instructions executable by the at least one processor, which, when executed by the at least one processor, implement the method according to the first or second aspect of the present invention.

[0025] According to a fifth aspect of the present invention, a computer program product, such as a computer-readable program carrier, is provided, comprising or storing computer program instructions that, when executed by a processor, at least assist in implementing the steps of the method described according to the first or second aspect of the present invention. Attached Figure Description

[0026] The invention will now be described in more detail with reference to the accompanying drawings, which will provide a better understanding of its principles, features, and advantages. The drawings include:

[0027] Figure 1 A flowchart illustrating the main steps of a method for constructing a weld width prediction model for laser cladding interface according to an exemplary embodiment of the present invention is shown.

[0028] Figure 2 The diagram schematically illustrates the weld formed after laser welding of a first and a second workpiece stacked on top of each other.

[0029] Figure 3 A flowchart illustrating the main steps of a laser cladding interface weld width prediction method according to an exemplary embodiment of the present invention is shown.

[0030] Figure 4 A schematic block diagram of a laser cladding joint width prediction system according to an exemplary embodiment of the present invention is shown.

[0031] Figure 5 A schematic block diagram of an electronic device for predicting the weld width of a laser lap bonding interface is shown according to an exemplary embodiment of the present invention. Detailed Implementation

[0032] To make the technical problems to be solved, the technical solutions, and the beneficial technical effects of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and several exemplary embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of protection of this invention.

[0033] Figure 1 A flowchart illustrating the main steps of a method for constructing a weld width prediction model for laser cladding interface according to an exemplary embodiment of the present invention is shown schematically for ease of explanation. Figure 2 The diagram schematically illustrates the weld formed after laser welding of a first workpiece W1 and a second workpiece W2 that are stacked on top of each other.

[0034] like Figure 1 As shown, the method for constructing a laser cladding joint width prediction model according to an exemplary embodiment of the present invention mainly includes the following steps:

[0035] S101: Collect surface weld width data Ws, weld depth data D, and intersection weld width data Wi at position P between the first workpiece W1 and the second workpiece W2 under predetermined beam characteristic parameters for laser lap welding of multiple weld seams; and

[0036] S102: Construct a dataset using the collected surface weld width data Ws and weld depth data D as input data and the interface weld width data Wi as output data, and use the dataset to train the interface weld width prediction model.

[0037] In step S101, the beam characteristic parameters involve parameters related to the spatial characteristics and energy distribution of the laser beam, such as beam quality, spot characteristics, beam mode, focusing parameters, and polarization state. These parameters determine the interaction between the laser beam and the material, thus determining the weld formation mechanism. Using predetermined beam characteristic parameters for laser lap welding can obtain welds of the same type, thereby ensuring that the mathematical relationships between the surface weld width data Ws, weld depth data D, and interface weld width data Wi are the same, regardless of the specific size of the weld, which is determined by process control parameters (such as laser power and dwell time / welding speed).

[0038] On the other hand, preferably, multiple laser lap welding operations are performed under different process control parameters to obtain surface weld width data, weld depth data, and interface weld width data for multiple welds in each laser lap welding operation. This allows for the acquisition of weld data of different sizes but with the same mathematical relationship between related data by changing laser power, welding speed, etc., under the same beam characteristic parameters. This enriches the diversity of the collected data, enabling better construction of datasets in subsequent steps to train the interface weld width prediction model and avoids model overfitting. For example, for a first workpiece W1 and a second workpiece W2 made of aluminum alloy with a thickness of 3mm, laser lap welding is performed on the two workpieces using laser powers of 2000W, 2500W, 3000W, 3500W, and 4000W respectively under predetermined beam characteristic parameters. This can obtain multiple welds of different sizes, and the weld width data Ws, weld depth data D, and interface weld width data Wi of these welds can be collected to construct a dataset for model training. Alternatively or additionally, multiple welds of different sizes can also be obtained and data collected by changing the welding speed. Furthermore, those skilled in the art will understand that in these laser cladding processes, auxiliary parameters or variables such as shielding gas type / flow rate, defocusing amount, and workpiece clamping method are also the same.

[0039] The surface weld width data Ws can be directly obtained by photographing the weld seam using a camera. Specifically, the surface weld width data Ws can be measured in real time during laser lap welding by taking real-time photographs of the weld seam using a coaxial camera with a processing lens. Furthermore, the weld penetration data D of each weld seam can be directly measured using an OCT sensor. Since the interface P between the first workpiece W1 and the second workpiece W2 is located inside the welded workpiece, the interface P can be exposed by destructive treatment of the weld seam, allowing for the measurement of the interface weld width data Wi at interface P. Obviously, the surface weld width data Ws and weld penetration data D can also be measured after destructive treatment of the weld seam to expose the interface P; various well-known and appropriate methods can be used for this measurement.

[0040] After data collection, a dataset can be constructed in step S102 to train the interface weld width prediction model. The collected surface weld width and weld depth data are used as input data, and the corresponding interface weld width data is used as output data. The input and output data are grouped together, and multiple datasets are divided into training and validation datasets. For example, 80% of the data can be randomly selected from multiple datasets as the training dataset, and the remaining 20% ​​as the validation dataset. The training dataset is then input into the interface weld width prediction model to train it and adjust its parameters. After training on the training dataset, the trained interface weld width prediction model is validated using the validation dataset. Specifically, the input data from the validation dataset is input into the interface weld width prediction model to obtain the model's output data. The deviation between the model's output data and the corresponding output data in the validation dataset is then calculated. If the deviation is less than or equal to a preset threshold, the trained interface weld width prediction model is considered to have converged, and the training of the interface weld width prediction model is complete. If there is a deviation greater than the preset threshold, the training rounds can be increased. That is, the training dataset is input into the interface weld width prediction model again to train the model, and then the training model is validated again using the validation dataset until the deviation between the interface weld width data output by the trained interface weld width prediction model and the corresponding interface weld width data in the validation dataset is less than the preset threshold. In this way, the trained interface weld width prediction model is obtained.

[0041] The data partitioning and training methods described above are merely illustrative. For example, if the model fails to converge after one round of training, the validation dataset can be partially merged back into the training dataset, a new validation dataset can be partitioned, and the model can be trained and validated again until the trained model converges. Furthermore, other well-known methods are also feasible.

[0042] Note that a data pre-cleaning step may also be included after step S101, which involves identifying obvious deviation data or erroneous data by checking the data based on experience or calculation, deleting them, or re-measuring and collecting data on the corresponding welds to ensure the reliability of the dataset constructed in the subsequent step S102, thereby improving training efficiency and ensuring the reliability of the trained interface weld width prediction model.

[0043] In a preferred embodiment, the weld width prediction model for the interface is a linear regression model. Linear regression models have low algorithmic complexity, fast training speed, and are suitable for large-scale data. Furthermore, the strong constraint of the linear assumption reduces the risk of overfitting, especially when the feature dimension is much smaller than the sample size. For the predetermined first and second workpieces, the weld width at the interface of their lap welds has an approximately linear relationship with the surface weld width and weld depth. Therefore, selecting a suitable linear regression model can meet the requirements for weld width prediction at the interface. Alternatively, a neural network model, such as a shallow neural network model, can also be used as the weld width prediction model for the interface.

[0044] The above describes the process of constructing a laser lap welding interface weld width prediction model. The following describes the method of using this model to predict the weld width of the laser lap welding interface.

[0045] Figure 3 A flowchart illustrating the main steps of a laser cladding joint width prediction method according to an exemplary embodiment of the present invention is shown. Figure 3 As shown, the laser cladding joint width prediction method according to an exemplary embodiment of the present invention mainly includes the following steps:

[0046] S301: Obtain the surface weld width data Ws and weld depth data D of the laser-welded seam of the first workpiece W1 and the second workpiece W2 that are stacked on top of each other; and

[0047] S302: The interface weld width prediction model constructed according to the model construction method described above predicts the interface weld width data Wi at the interface position P between the first workpiece W1 and the second workpiece W2 based on the surface weld width data Ws and weld depth data D.

[0048] Similarly, in step S301, the surface weld width data Ws of the corresponding weld can be directly obtained by taking pictures and measuring the weld seam using a camera. In particular, the surface weld width data Ws of the weld seam can be obtained in real time by taking pictures and measuring the weld seam in real time using a coaxial camera of the processing lens during laser lap welding. And the penetration depth data D of each weld seam can be directly measured using an OCT sensor. In this way, two input data can be obtained in a non-destructive manner, and in particular, two input data can be obtained in real time.

[0049] In step S302, since the weld width prediction model has been trained, after inputting the surface weld width data Ws and weld depth data D, the weld width prediction model can predict and output the corresponding weld width data Wi. Therefore, the weld width data Wi can be obtained without damaging the welded object, and in particular, the weld width data Wi can be acquired in real time, thus providing better assistance to the production process.

[0050] Figure 4 A schematic block diagram of a laser cladding joint width prediction system 1 according to an exemplary embodiment of the present invention is shown. Figure 4 As shown, a laser cladding joint width prediction system 1 according to an exemplary embodiment of the present invention includes:

[0051] Data acquisition module 11, configured to acquire surface weld width and weld depth data of laser-welded seams of a first and second workpiece stacked on top of each other; and

[0052] The prediction module 12 is configured to: invoke the method described above for constructing a laser lap welding interface width prediction model for the laser lap welding of the first workpiece and the second workpiece to construct an interface width prediction model for the laser lap welding, and predict the interface width data at the interface position between the first workpiece and the second workpiece based on the surface width data and depth data of the weld.

[0053] Figure 5 A schematic block diagram of an electronic device 2 for predicting the weld width at the interface of laser cladding is shown, according to an exemplary embodiment of the present invention. Figure 5 As shown, an electronic device 2 for predicting the weld width at the laser cladding interface according to an exemplary embodiment of the present invention includes:

[0054] At least one processor 21; and

[0055] The memory 22 stores program instructions that can be executed by the at least one processor 21. When the program instructions are executed by the at least one processor 21, they implement a method for constructing a laser lap weld joint width prediction model according to an exemplary embodiment of the present invention, or a laser lap weld joint width prediction method according to an exemplary embodiment of the present invention.

[0056] Although specific embodiments of the invention have been described in detail herein, they are given for illustrative purposes only and should not be construed as limiting the scope of the invention. Various substitutions, alterations, and modifications can be conceived without departing from the spirit and scope of the invention.

Claims

1. A method for constructing a prediction model for the weld width of a laser-coated weld joint, comprising: Data on surface weld width and depth, as well as weld width at the interface between the first and second workpieces, are collected from multiple welds obtained by laser lap welding under predetermined beam characteristic parameters. The weld width at the interface of the lap weld between the first and second workpieces exhibits an approximately linear relationship with the surface weld width and depth. A dataset is constructed using only the collected surface weld width and weld depth data as input data, and the weld width data at the interface as output data. This dataset is then used to train the interface weld width prediction model. Surface weld width data is obtained by taking photographs of the weld seam using a camera. The melting depth data is obtained with the help of an OCT sensor.

2. The method for constructing a laser cladding joint width prediction model according to claim 1, wherein, The dataset is divided into a training dataset and a validation dataset. The interface weld width prediction model is trained using the training dataset and its parameters are adjusted. The trained interface weld width prediction model is then validated using the validation dataset until the deviation between the interface weld width data output by the trained interface weld width prediction model and the corresponding interface weld width data in the validation dataset is less than a preset threshold. This process yields a successfully trained interface weld width prediction model.

3. The method for constructing a laser cladding joint width prediction model according to claim 1 or 2, wherein, Data on surface weld width, weld depth, and weld width at the interface of multiple welds in the laser lap welding process were collected under different process control parameters.

4. The method for constructing a weld width prediction model for laser cladding joints according to claim 1 or 2, wherein, The weld width data is obtained by destructively treating the weld to expose the interface between the first and second workpieces and measuring the weld width at the interface.

5. The method for constructing a laser cladding joint width prediction model according to claim 1 or 2, wherein, The interface width prediction model is a linear regression model or a neural network model.

6. A method for predicting the weld width at the interface of a laser-coated cladding joint, wherein the method is performed using a weld width prediction model constructed by the method for constructing a weld width prediction model at the interface according to any one of claims 1 to 5, the method comprising: Obtain surface weld width and weld depth data of the laser-welded seam of the first and second workpieces stacked on top of each other under the predetermined beam characteristic parameters; and The interface weld width prediction model is used to predict the interface weld width data at the interface location between the first workpiece and the second workpiece based on the surface weld width data and weld depth data.

7. A laser cladding joint width prediction system (1), comprising: The data acquisition module (11) is configured to acquire surface weld width and weld depth data of the laser-welded weld of the first and second workpieces stacked on top of each other. and The prediction module (12) is configured to call the interface weld width prediction model to predict the interface weld width data at the interface position between the first workpiece and the second workpiece based on the surface weld width data and weld depth data of the weld. The interface weld width prediction model is constructed by the method of constructing a laser lap weld interface weld width prediction model according to any one of claims 1 to 5 and for the laser lap weld of the first workpiece and the second workpiece.

8. An electronic device (2) for predicting the weld width at the interface of laser cladding, comprising: At least one processor (21); and A memory (22) storing program instructions executable by the at least one processor (21), which, when executed by the at least one processor (21), implement the method according to any one of claims 1 to 6.

9. A computer program product comprising or storing computer program instructions, which, when executed by a processor, at least assist in implementing the steps of the method according to any one of claims 1 to 6.

10. The computer program product of claim 9, wherein, The computer program product is a computer-readable program carrier.

Images