An aluminum alloy laser cutting quality prediction method, device, equipment and medium

Abstract

The application discloses an aluminum alloy laser cutting quality prediction method, device, equipment and medium, comprising: obtaining sample data obtained after cutting of multiple different aluminum alloy materials by multiple different laser processing parameters; identifying the cutting quality parameters corresponding to the sample data, generating a laser cutting data set according to the multiple laser processing parameters, aluminum alloy material parameters and cutting quality parameters; using the laser cutting data set to iteratively train the boosting tree model, and taking the trained boosting tree model as an aluminum alloy laser cutting quality prediction model. The technical scheme of the embodiment of the application can reduce the complexity of the aluminum alloy laser cutting quality prediction process, improve the prediction efficiency and the accuracy of the prediction result.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a method, apparatus, equipment and medium for predicting the quality of laser cutting of aluminum alloys. Background Technology

[0002] Aluminum alloys are important lightweight alloy materials in aerospace applications, and related laser cutting technology is widely used in the machining of precision parts. If the cutting quality of aluminum alloys does not meet standards after laser cutting, repeated process tests are necessary. Therefore, predicting the quality of laser-cut aluminum alloys is particularly important.

[0003] In existing technologies, the morphological changes of laser cutting are mainly simulated by coupling material properties, thermodynamics and optical parameters. However, this method is based on idealized boundary conditions for modeling, which results in data errors. Furthermore, since the accuracy of the simulation model depends on the setting of physical parameters, a large number of process experiments are required to verify the reliability of the model. This leads to high complexity and low prediction efficiency in the process of predicting the quality of aluminum alloy laser cutting. Summary of the Invention

[0004] This invention provides a method, apparatus, equipment, and medium for predicting the quality of aluminum alloy laser cutting, which can reduce the complexity of the aluminum alloy laser cutting quality prediction process and improve prediction efficiency and accuracy of prediction results.

[0005] According to one aspect of the present invention, a method for predicting the quality of laser cutting of aluminum alloys is provided, the method comprising: To obtain sample data after cutting multiple different aluminum alloy materials using multiple different laser processing parameters; Identify the cutting quality parameters corresponding to the sample data, and generate a laser cutting dataset based on multiple laser processing parameters, aluminum alloy material parameters, and cutting quality parameters; The lift-up tree model was iteratively trained using the laser cutting dataset, and the trained lift-up tree model was used as a prediction model for the quality of aluminum alloy laser cutting.

[0006] Optionally, the lift-up tree model is iteratively trained using the laser cutting dataset, and the trained lift-up tree model is used as a prediction model for the quality of aluminum alloy laser cutting, including: The laser cutting dataset is divided into a training set and a test set using the K-fold cross-validation algorithm. The boosting tree model is iteratively trained using the training set, and its performance is tested using the test set after the training is completed. If the performance of the trained boosting tree model meets the preset standard, then the trained boosting tree model will be used as the quality prediction model for aluminum alloy laser cutting.

[0007] Optionally, iteratively training the boosting tree model using the training set includes: The training set is input into the initial boosting tree model, and the negative residual corresponding to each training sample in the training set is determined according to the loss function corresponding to the initial boosting tree model. The new regression tree model is obtained by fitting the negative residuals, the optimal weights corresponding to the new regression tree model are determined, and the new regression tree model is added to the ensemble model according to the optimal weights. The model parameters of the ensemble model are iteratively optimized until the ensemble model meets the training termination condition.

[0008] Optionally, after inputting the training set into the initial boosting tree model, the method further includes: Non-numerical feature target input data is selected from the training set, and the target input data is encoded into binary vectors so that the initial boosting tree model can process the encoded target input data.

[0009] Optionally, sample data can be obtained after cutting multiple different aluminum alloy materials using multiple different laser processing parameters, including: Obtain aluminum alloy materials with multiple different grades and sheet thicknesses; Multiple aluminum alloy materials were cut using multiple different laser processing parameters to obtain multiple sample data; the laser processing parameters included at least laser power, cutting speed, defocusing amount and gas pressure.

[0010] Optionally, before iteratively training the boosting tree model using the laser cutting dataset, the method further includes: The laser cutting dataset is cleaned; the cleaning process includes at least one of the following: Missing value handling, outlier value handling, duplicate value handling, and error value correction.

[0011] Optionally, after using the trained boosting tree model as a prediction model for the quality of aluminum alloy laser cutting, the following steps are also included: Obtain the target aluminum alloy material parameters and target laser processing parameters to be evaluated; Input the target aluminum alloy material parameters and target laser processing parameters into the aluminum alloy laser cutting quality prediction model; The target cutting quality prediction result is output through the aluminum alloy laser cutting quality prediction model.

[0012] According to another aspect of the present invention, an aluminum alloy laser cutting quality prediction device is provided, the device comprising: The sample data acquisition module is used to acquire sample data obtained after cutting multiple different aluminum alloy materials with multiple different laser processing parameters; The dataset generation module is used to identify the cutting quality parameters corresponding to the sample data, and generate a laser cutting dataset based on multiple laser processing parameters, aluminum alloy material parameters, and cutting quality parameters. The model training module is used to iteratively train the lift-up tree model using the laser cutting dataset, and to use the trained lift-up tree model as a prediction model for the quality of aluminum alloy laser cutting.

[0013] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and a memory communicatively connected to said at least one processor; The memory stores a computer program that can be executed by the at least one processor, which is then executed by the at least one processor to enable the at least one processor to perform the aluminum alloy laser cutting quality prediction method according to any embodiment of the present invention.

[0014] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the aluminum alloy laser cutting quality prediction method according to any embodiment of the present invention.

[0015] The technical solution provided by this invention involves acquiring sample data obtained after cutting multiple different aluminum alloy materials using multiple different laser processing parameters, identifying the cutting quality parameters corresponding to the sample data, generating a laser cutting dataset based on multiple laser processing parameters, aluminum alloy material parameters, and cutting quality parameters, using the laser cutting dataset to iteratively train a lifting tree model, and using the trained lifting tree model as a prediction model for aluminum alloy laser cutting quality. This provides a method for automatically predicting the cutting quality of aluminum alloy materials after laser cutting using machine learning algorithms. Compared to the existing technology that simulates the morphological changes of laser cutting by coupling material properties, thermodynamics, and optical parameters, this method can reduce the complexity of the aluminum alloy laser cutting quality prediction process and improve prediction efficiency and accuracy.

[0016] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a method for predicting the quality of laser cutting of aluminum alloys according to an embodiment of the present invention; Figure 2 This is a flowchart of another method for predicting the quality of laser cutting of aluminum alloys according to an embodiment of the present invention; Figure 3 This is a schematic diagram of a sample obtained by laser cutting an aluminum alloy material according to an embodiment of the present invention. Figure 4 This is a schematic diagram of the structure of an aluminum alloy laser cutting quality prediction device according to an embodiment of the present invention; Figure 5 This is a schematic diagram of the structure of an electronic device that implements the aluminum alloy laser cutting quality prediction method according to an embodiment of the present invention. Detailed Implementation

[0019] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0020] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0021] Figure 1This is a flowchart illustrating a method for predicting the quality of laser cutting of aluminum alloys according to an embodiment of the present invention. This embodiment is applicable to predicting the cutting quality of aluminum alloy materials after laser cutting. The method can be executed by an aluminum alloy laser cutting quality prediction device, which can be implemented in hardware and / or software and configured in an electronic device, such as... Figure 1 As shown, the method includes: Step 110: Obtain sample data after cutting multiple different aluminum alloy materials using multiple different laser processing parameters.

[0022] In this embodiment, the laser processing parameters refer to the processing attribute indicators that need to be set when using a laser to cut aluminum alloy materials. Specifically, the laser processing parameters may include laser-related parameters (such as laser power, wavelength, etc.) and processing control parameters (such as cutting speed, etc.), which are not limited in this embodiment.

[0023] Optionally, after obtaining the above-mentioned multiple different laser processing parameters, the different types of laser processing parameters can be arranged and combined to obtain multiple parameter groups. Then, each parameter group can be used to laser cut aluminum alloy materials with different properties to obtain samples. Finally, the structural data (including shape, size data, etc.) corresponding to the samples can be obtained.

[0024] Step 120: Identify the cutting quality parameters corresponding to the sample data, and generate a laser cutting dataset based on multiple laser processing parameters, aluminum alloy material parameters, and cutting quality parameters.

[0025] In this embodiment, after obtaining the laser-cut sample data through the above steps, the burr height, surface roughness, and heat-affected zone width of the preset area in each sample data can be measured sequentially, and the burr height, surface roughness, and heat-affected zone width can be used as cutting quality parameters.

[0026] The burr height refers to the vertical distance from the cut reference surface to the tip of the burr after the aluminum alloy material is laser-cut, which is the irregular protrusion and residual slag at the cut edge. The surface roughness refers to the quantitative index of the microscopic unevenness of the aluminum alloy material after cutting. The smaller the value, the smoother the surface; the larger the value, the more obvious the unevenness of the surface. The heat-affected zone width refers to the lateral dimension of the area of ​​the aluminum alloy material whose properties change due to heat, that is, the distance from the cutting edge to the area where the material properties are not affected.

[0027] In this step, multiple cutting data samples can be established based on the mapping relationship between different laser processing parameters, aluminum alloy material parameters and cutting quality parameters. Then, the multiple cutting data samples are combined to obtain a laser cutting dataset.

[0028] Step 130: Use the laser cutting dataset to iteratively train the boosting tree model, and use the trained boosting tree model as a prediction model for the quality of aluminum alloy laser cutting.

[0029] In this embodiment, the boosting tree model is a machine learning model based on the idea of ​​ensemble learning. It trains multiple weak decision trees step by step and combines the prediction results of these weak trees in a weighted manner to form a powerful prediction model.

[0030] In one alternative implementation, when iteratively training the boosting tree model using a laser cutting dataset, a simple decision tree can be trained first as an initial model using the dataset. Then, the prediction error of the current model on the dataset is calculated, and training samples with inaccurate predictions are obtained. A new decision tree model is retrained for these inaccurate training samples, and the newly trained decision tree model is added to the model set. The optimal weights of the decision tree model are then determined. Finally, after multiple iterations, all the trained decision tree models are weighted and summed according to their respective weights to obtain the final aluminum alloy laser cutting quality prediction model.

[0031] The advantage of this setup is that, during the training process of the boost tree model, by adopting an iterative error correction mechanism, it is possible to gradually capture complex nonlinear relationships and subtle features in the data, thereby ensuring the training performance of the model and improving the accuracy of the model's prediction results.

[0032] The technical solution provided by this invention involves acquiring sample data obtained after cutting multiple different aluminum alloy materials using multiple different laser processing parameters, identifying the cutting quality parameters corresponding to the sample data, generating a laser cutting dataset based on multiple laser processing parameters, aluminum alloy material parameters, and cutting quality parameters, using the laser cutting dataset to iteratively train a lifting tree model, and using the trained lifting tree model as a prediction model for aluminum alloy laser cutting quality. This provides a method for automatically predicting the cutting quality of aluminum alloy materials after laser cutting using machine learning algorithms. Compared to the existing technology that simulates the morphological changes of laser cutting by coupling material properties, thermodynamics, and optical parameters, this method can reduce the complexity of the aluminum alloy laser cutting quality prediction process and improve prediction efficiency and accuracy.

[0033] Figure 2 A flowchart of another method for predicting the quality of laser cutting of aluminum alloys provided in an embodiment of the present invention is shown below. Figure 2 As shown, the method includes: Step 210: Obtain multiple aluminum alloy materials of different grades and thicknesses, and cut each aluminum alloy material using multiple different laser processing parameters to obtain multiple sample data.

[0034] In this embodiment, the first digit in the grade of the aluminum alloy material represents the alloy series. Different series correspond to different main alloying elements, which determine the core characteristics of the material. The following numbers are the serial numbers within the same series, used to distinguish subtle differences in composition and performance. Optionally, in this step, aluminum alloy materials with different sheet thicknesses and grades 2024 and 7075 can be obtained, and then the aluminum alloy materials can be cut using multiple different laser processing parameters to obtain multiple sample data.

[0035] Optionally, the laser processing parameters may include at least laser power, cutting speed, defocusing amount, and gas pressure.

[0036] Step 220: Identify the cutting quality parameters corresponding to the sample data, and generate a laser cutting dataset based on multiple laser processing parameters, aluminum alloy material parameters, and cutting quality parameters.

[0037] In this step, multiple laser processing parameters and aluminum alloy material parameters can be used as input variables, and the cutting quality parameters corresponding to the sample data can be used as output variables. Then, the input variables and output variables are integrated to obtain the laser cutting dataset.

[0038] Step 230: Clean the laser cutting dataset and use the K-fold cross-validation algorithm to divide the cleaned laser cutting dataset into a training set and a test set.

[0039] In this embodiment, the cleaning process includes at least one of the following: missing value processing, outlier processing, duplicate value processing, and error value correction. After cleaning the laser cutting dataset, the dataset can be divided into K training sets and K test sets using a K-fold cross-validation algorithm. Optionally, the value of K can be set to 3 or 5, and the specific value can be adjusted according to the actual situation; this embodiment does not impose any restrictions on this.

[0040] Step 240: Iteratively train the boosting tree model using the training set. After the boosting tree model is trained, use the test set to test the performance of the trained boosting tree model.

[0041] In one embodiment of this example, iterative training of the boosting tree model using the training set includes: Step 241: Input the training set into the initial boosting tree model, and determine the negative residual corresponding to each training sample in the training set according to the loss function corresponding to the initial boosting tree model. Step 242: Obtain the new regression tree model based on the negative residual fitting, determine the optimal weights corresponding to the new regression tree model, and add the new regression tree model to the ensemble model according to the optimal weights. Step 243: Iteratively optimize the model parameters of the ensemble model until the ensemble model meets the training termination condition.

[0042] In a specific embodiment, an initial boosting tree model for the loss function can be set first. and through Determine the parameters of the next decision tree, where, Indicates the first training samples, Represents the loss function. Indicates the predicted value. Represents the actual value; Then in the given function and its corresponding gradient Through iterative updates To minimize a given function for the current ensemble model The first one can be calculated using the following formula. A sample about loss function Negative residuals: , By employing the regression loss function in the root mean square error algorithm, the difference between the predicted and actual values ​​can be fitted, as shown in the following formula: , At the same time It can obtain the residual between the current predicted value and the true value, and iteratively update the approximate value of the gradient fitting in each round; Then, a new regression tree model is trained using the entire training set. , obtained the For each leaf node of the tree, obtain the corresponding value that better predicts the residuals (i.e., the residuals of all samples at that leaf); and use the following formula to find new regression tree models. Optimal learning rate : , Add a regression tree model Multiply by a smaller shrinkage rate (Values ​​within the interval [0.01, 0.3]) are added to the current ensemble model to prevent overfitting. The specific formula is as follows: , Among them, shrinkage rate This is used to control the contribution of each newly added regression tree model to the ensemble model, so that the ensemble model can achieve better generalization ability; After M iterations of the ensemble model, the model weight parameters calculated in each iteration can be updated into the ensemble model, ultimately generating a trained ensemble model. .

[0043] In this embodiment, optionally, after inputting the training set into the initial boosting tree model, a method for preprocessing non-numerical feature data is also provided. Specifically, non-numerical feature target input data can be selected from the training set, and the target input data can be encoded into binary vectors so that the initial boosting tree model can process the encoded target input data.

[0044] Specifically, for target input data in string format, one-hot encoding can be combined with feature frequency to calculate the corresponding binary vector. Simultaneously, a ranking boosting algorithm can be used to handle noisy points in the training set, avoiding bias in the gradient estimation step and thus addressing the prediction offset problem.

[0045] Step 250: If the performance of the trained boosting tree model meets the preset standard, then the trained boosting tree model will be used as the quality prediction model for aluminum alloy laser cutting.

[0046] In this step, specifically, when the performance of the trained boosting tree model can reach the desired metric R... 2 When the value is reached, the performance of the trained boosting tree model can be considered to meet the preset standard.

[0047] Step 260: Obtain the target aluminum alloy material parameters and target laser processing parameters to be evaluated, and input the target aluminum alloy material parameters and target laser processing parameters into the aluminum alloy laser cutting quality prediction model.

[0048] Step 270: Output the target cutting quality prediction result through the aluminum alloy laser cutting quality prediction model.

[0049] The technical solution provided in this invention involves obtaining multiple aluminum alloy materials of different grades and thicknesses, cutting each aluminum alloy material using multiple different laser processing parameters to obtain multiple sample data, identifying the cutting quality parameters corresponding to the sample data, generating a laser cutting dataset based on the multiple laser processing parameters, aluminum alloy material parameters, and cutting quality parameters, cleaning the laser cutting dataset, using a K-fold cross-validation algorithm to divide the cleaned laser cutting dataset into a training set and a test set, using the training set to iteratively train a boosting tree model, and using the test set to perform performance testing on the trained boosting tree model after training. If the model performance meets the preset standard, the model is used as the aluminum alloy laser cutting quality prediction model. This method, which involves obtaining the target aluminum alloy material parameters and target laser processing parameters to be evaluated, inputting these parameters into the aluminum alloy laser cutting quality prediction model, and outputting the target cutting quality prediction result through the aluminum alloy laser cutting quality prediction model, can reduce the complexity of the aluminum alloy laser cutting quality prediction process and improve prediction efficiency and accuracy.

[0050] Based on the above implementation methods, in order to better explain the technical solution of this embodiment, taking 2024-O, 2024-T3, 7075-O and 7075-T6 aluminum alloy materials with thicknesses of 1.6mm and 3.175mm as examples, the training process of the aluminum alloy laser cutting quality prediction model can refer to the following steps: Step 1: Use the laser processing parameters shown in Table 1 to cut aluminum alloy materials with a thickness of 1.6 mm and grade 2024 and 7075, and use the laser processing parameters shown in Table 2 to cut aluminum alloy materials with a thickness of 3.175 mm and grade 2024 and 7075, and obtain multiple sample data. Table 1 Table 2 Tables 1 and 2 only show some values ​​of the laser processing parameters. The specific values ​​can be expanded according to the actual situation, and this embodiment does not limit them.

[0051] Step 2: Identify the cutting quality parameters corresponding to the sample data, including the burr height, surface roughness, and heat-affected zone width in the preset area of ​​the sample data; In this step, Figure 3 This is a schematic diagram of a sample obtained after laser cutting aluminum alloy material. After obtaining the laser-cut sample, it can be based on... Figure 3The marked areas and marked positions (e.g., position "1", position "2", position "3") are used to measure the burr height, surface roughness, and heat-affected zone width, respectively. Based on these measurement data, laser processing parameters, and aluminum alloy material parameters, a laser cutting dataset is generated.

[0052] Step 3: Divide the laser cutting dataset into training and testing sets using either a 3-fold or 5-fold method.

[0053] Step 4: Import the training set into the initial boosting tree model. First, calculate the negative residuals of the model with respect to the loss function according to the formula: , Residual fitting is performed on all leaf nodes of the trained boosting tree model to obtain the corresponding values ​​of all better predictive residuals, and the optimal weights of the newly added regression tree model are calculated: , All newly added regression tree models After adding it to the ensemble model, update the ensemble model's shrinkage rate to prevent overfitting: , Finally, after M rounds of iteration, the weights of the tree model calculated in each round are updated into the ensemble model, and the trained ensemble model is saved.

[0054] Figure 4 This is a schematic diagram of the structure of an aluminum alloy laser cutting quality prediction device provided in an embodiment of the present invention, as shown below. Figure 4 As shown, the device includes: a sample data acquisition module 410, a dataset generation module 420, and a model training module 430.

[0055] The sample data acquisition module 410 is used to acquire sample data obtained after cutting multiple different aluminum alloy materials with multiple different laser processing parameters; The dataset generation module 420 is used to identify the cutting quality parameters corresponding to the sample data, and generate a laser cutting dataset based on multiple laser processing parameters, aluminum alloy material parameters and cutting quality parameters. The model training module 430 is used to iteratively train the lift-up tree model using the laser cutting dataset, and to use the trained lift-up tree model as a prediction model for the quality of aluminum alloy laser cutting.

[0056] The technical solution provided by this invention involves acquiring sample data obtained after cutting multiple different aluminum alloy materials using multiple different laser processing parameters, identifying the cutting quality parameters corresponding to the sample data, generating a laser cutting dataset based on multiple laser processing parameters, aluminum alloy material parameters, and cutting quality parameters, using the laser cutting dataset to iteratively train a lifting tree model, and using the trained lifting tree model as a prediction model for aluminum alloy laser cutting quality. This provides a method for automatically predicting the cutting quality of aluminum alloy materials after laser cutting using machine learning algorithms. Compared to the existing technology that simulates the morphological changes of laser cutting by coupling material properties, thermodynamics, and optical parameters, this method can reduce the complexity of the aluminum alloy laser cutting quality prediction process and improve prediction efficiency and accuracy.

[0057] Based on the above embodiments, the sample data acquisition module 410 includes: The material processing unit is used to obtain multiple aluminum alloy materials of different grades and sheet thicknesses; to cut each aluminum alloy material using multiple different laser processing parameters to obtain multiple sample data; the laser processing parameters include at least laser power, cutting speed, defocusing amount and gas pressure.

[0058] Model training module 430 includes: A dataset partitioning unit is used to divide the laser cutting dataset into a training set and a test set using a K-fold cross-validation algorithm. An iterative training unit is used to iteratively train the boosting tree model using the training set, and to perform performance testing on the trained boosting tree model using the test set after the training of the boosting tree model is completed. The model application unit is used to use the trained boosting tree model as a quality prediction model for aluminum alloy laser cutting if the performance of the trained boosting tree model meets the preset standard. The model input unit is used to input the training set into the initial boosting tree model and determine the negative residual corresponding to each training sample in the training set according to the loss function corresponding to the initial boosting tree model. The weight determination unit is used to obtain a new regression tree model based on the negative residual fitting, determine the optimal weights corresponding to the new regression tree model, and add the new regression tree model to the ensemble model according to the optimal weights. The parameter optimization unit is used to iteratively optimize the model parameters of the ensemble model until the ensemble model meets the training termination condition. A data encoding unit is used to filter target input data with non-numerical features in the training set and encode the target input data into binary vectors so that the initial boosting tree model can process the encoded target input data. A data cleaning unit is used to clean the laser cutting dataset; the cleaning process includes at least one of the following: missing value processing, outlier processing, duplicate value processing, and error value correction.

[0059] The device further includes: The cutting quality prediction module is used to acquire the target aluminum alloy material parameters and target laser processing parameters to be evaluated; input the target aluminum alloy material parameters and target laser processing parameters into the aluminum alloy laser cutting quality prediction model; and output the target cutting quality prediction result through the aluminum alloy laser cutting quality prediction model.

[0060] The above-described apparatus can execute the methods provided in all the foregoing embodiments of the present invention, and has the corresponding functional modules and beneficial effects for executing the above methods. Technical details not described in detail in the embodiments of the present invention can be found in the methods provided in all the foregoing embodiments of the present invention.

[0061] Figure 5 A schematic diagram of an electronic device 10, which can be used to implement embodiments of the present invention, is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0062] like Figure 5 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) or a random access memory (RAM), communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the read-only memory 12 or loaded from the storage unit 18 into the random access memory 13. The random access memory 13 can also store various programs and data required for the operation of the electronic device 10. The processor 11, the read-only memory 12, and the random access memory 13 are interconnected via a bus 14. Input / output (I / O) interfaces are also connected to the bus 14.

[0063] Multiple components in electronic device 10 are connected to input / output interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of monitors, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0064] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, digital signal processing (DSP) processors, and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as the aluminum alloy laser cutting quality prediction method.

[0065] In some embodiments, the aluminum alloy laser cutting quality prediction method can be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on electronic device 10 via read-only memory 12 and / or communication unit 19. When the computer program is loaded into random access memory 13 and executed by processor 11, one or more steps of the aluminum alloy laser cutting quality prediction method described above can be performed. Alternatively, in other embodiments, processor 11 can be configured to perform the aluminum alloy laser cutting quality prediction method by any other suitable means (e.g., by means of firmware).

[0066] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-chips (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0067] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0068] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, RAM, ROM, erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0069] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a cathode ray tube (CRT) or a liquid crystal display (LCD)) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0070] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or middleware components (e.g., application servers), or frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0071] A computing system can include clients and servers. Clients and servers are generally geographically separated and typically interact via communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system. It addresses the shortcomings of traditional physical hosts and Virtual Private Servers (VPS) in terms of management difficulty and weak business scalability.

[0072] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0073] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A method for predicting the quality of laser cutting of aluminum alloys, characterized in that, The method includes: To obtain sample data after cutting multiple different aluminum alloy materials using multiple different laser processing parameters; Identify the cutting quality parameters corresponding to the sample data, and generate a laser cutting dataset based on multiple laser processing parameters, aluminum alloy material parameters, and cutting quality parameters; The lift-up tree model was iteratively trained using the laser cutting dataset, and the trained lift-up tree model was used as a prediction model for the quality of aluminum alloy laser cutting.

2. The method according to claim 1, characterized in that, The lift-up tree model is iteratively trained using the laser cutting dataset, and the trained lift-up tree model is used as a prediction model for the quality of aluminum alloy laser cutting, including: The laser cutting dataset is divided into a training set and a test set using the K-fold cross-validation algorithm. The boosting tree model is iteratively trained using the training set, and its performance is tested using the test set after the training is completed. If the performance of the trained boosting tree model meets the preset standard, then the trained boosting tree model will be used as the quality prediction model for aluminum alloy laser cutting.

3. The method according to claim 2, characterized in that, Iterative training of the boosting tree model using the training set includes: The training set is input into the initial boosting tree model, and the negative residual corresponding to each training sample in the training set is determined according to the loss function corresponding to the initial boosting tree model. The new regression tree model is obtained by fitting the negative residuals, the optimal weights corresponding to the new regression tree model are determined, and the new regression tree model is added to the ensemble model according to the optimal weights. The model parameters of the ensemble model are iteratively optimized until the ensemble model meets the training termination condition.

4. The method according to claim 3, characterized in that, After inputting the training set into the initial boosting tree model, the process further includes: Non-numerical feature target input data is selected from the training set, and the target input data is encoded into binary vectors so that the initial boosting tree model can process the encoded target input data.

5. The method according to claim 1, characterized in that, Obtain sample data after cutting multiple different aluminum alloy materials using multiple different laser processing parameters, including: Obtain aluminum alloy materials with multiple different grades and sheet thicknesses; Multiple aluminum alloy materials were cut using multiple different laser processing parameters to obtain multiple sample data; the laser processing parameters included at least laser power, cutting speed, defocusing amount and gas pressure.

6. The method according to claim 1, characterized in that, Before iteratively training the boosting tree model using the laser cutting dataset, the following is also included: The laser cutting dataset is cleaned; the cleaning process includes at least one of the following: Missing value handling, outlier value handling, duplicate value handling, and error value correction.

7. The method according to claim 1, characterized in that, After using the trained boosting tree model as a prediction model for the quality of aluminum alloy laser cutting, the following is also included: Obtain the target aluminum alloy material parameters and target laser processing parameters to be evaluated; Input the target aluminum alloy material parameters and target laser processing parameters into the aluminum alloy laser cutting quality prediction model; The target cutting quality prediction result is output through the aluminum alloy laser cutting quality prediction model.

8. A device for predicting the quality of laser cutting of aluminum alloys, characterized in that, The device includes: The sample data acquisition module is used to acquire sample data obtained after cutting multiple different aluminum alloy materials with multiple different laser processing parameters; The dataset generation module is used to identify the cutting quality parameters corresponding to the sample data, and generate a laser cutting dataset based on multiple laser processing parameters, aluminum alloy material parameters, and cutting quality parameters. The model training module is used to iteratively train the lift-up tree model using the laser cutting dataset, and to use the trained lift-up tree model as a prediction model for the quality of aluminum alloy laser cutting.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and a memory communicatively connected to said at least one processor; The memory stores a computer program that can be executed by the at least one processor, which is then executed by the at least one processor to enable the at least one processor to perform the aluminum alloy laser cutting quality prediction method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the aluminum alloy laser cutting quality prediction method according to any one of claims 1-7.

Images