A method, device, equipment and storage medium for determining thermal spraying parameters
A pre-trained model for determining thermal spraying parameters automates the process, addressing inefficiencies in existing methods by directly deriving optimal parameters from metallographic images, thus reducing costs and time.
Patent Information
- Application Number
- CN202410640780.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-22
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-05-22
AI Technical Summary
Existing methods for determining thermal spray parameters require human experience and large-scale experiments, which are costly and inefficient.
A method using a pre-trained thermal spraying parameter determination model to automatically determine coating process parameters based on target metallographic images, reducing the need for manual experimentation and human intervention.
This approach significantly reduces development costs and time by directly determining optimal thermal spraying parameters, improving efficiency and reducing R&D costs.
Smart Images

Figure CN118429323B_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present invention relate to the technical field of surface coating processing, and in particular to a method, device, equipment and storage medium for determining thermal spraying parameters. Background Art
[0002] Thermal spraying is a processing method that heats the spray material to a molten or semi-molten state, and then sprays the molten or semi-molten spray material on the surface of the treated workpiece through a high-speed airflow, thereby depositing and forming a coating. The coating gives the workpiece various special functions such as corrosion resistance, wear resistance, insulation, and heat insulation. In the metallographic inspection used in the thermal spraying industry, the production samples are first prepared, polished, and placed under a microscope for observation and photography, and the different structures in the metallographic photos taken are qualitatively and quantitatively analyzed, evaluated and graded according to relevant national standards, and then the product is determined to be qualified or not. The thermal spraying parameters of the corresponding prepared coating will also be indirectly reflected according to the final metallographic results to see whether they meet the parameter curing requirements and authorization conditions for batch stable production of product quality.
[0003] For each specific thermal spray material, in the current process of developing and preparing coatings, there are more than ten parameter variables in the thermal spray process that need to find the best range for optimization. The metallographic results of the coating also need to be compared and judged one by one by multiple technical indicators to ensure that it meets the requirements of the standard metallographic diagram. However, this development process requires a considerable amount of time, manpower, and material costs. For example, it requires extremely high professional and experience requirements for development test designers, large-scale DOE tests, and large-scale tedious sample preparation, etc. High R&D costs. Summary of the invention
[0004] The embodiments of the present invention provide a thermal spraying parameter determination method, device, equipment and storage medium, which can determine the coating process parameters corresponding to the target metallographic image based on the target thermal spraying parameter determination model, improve the efficiency of determining the spraying process parameters, and reduce research and development costs.
[0005] In a first aspect, an embodiment of the present invention provides a method for determining thermal spraying parameters, the method comprising:
[0006] Acquire a target metallographic image of the coating to be sprayed;
[0007] The target metallographic image is input into a pre-trained target thermal spraying parameter determination model to obtain coating process parameters corresponding to the target metallographic image; wherein the coating process parameters are process parameters required for spraying the coating to be sprayed.
[0008] In a second aspect, an embodiment of the present invention provides a device for determining thermal spraying parameters, the device comprising:
[0009] A metallographic image acquisition module, used to acquire a target metallographic image of the coating to be sprayed;
[0010] The coating process parameter determination module is used to input the target metallographic image into a pre-trained target thermal spraying parameter determination model to obtain the coating process parameters corresponding to the target metallographic image; wherein the coating process parameters are the process parameters required for spraying the coating to be sprayed.
[0011] In a third aspect, an embodiment of the present invention provides a computer device, the computer device comprising:
[0012] one or more processors;
[0013] A memory for storing one or more programs;
[0014] When the one or more programs are executed by the one or more processors, the one or more processors implement the thermal spraying parameter determination method described in any embodiment.
[0015] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the thermal spraying parameter determination method described in any embodiment.
[0016] The technical solution provided by the embodiment of the present invention obtains a target metallographic image of the coating to be sprayed; inputs the target metallographic image into a pre-trained target thermal spray parameter determination model to obtain coating process parameters corresponding to the target metallographic image; wherein the coating process parameters are process parameters required for spraying the coating to be sprayed. The technical solution of the embodiment of the present invention solves the problem that the prior art requires human experience and a large number of experiments to determine the spray process parameters corresponding to the standard metallographic image, resulting in a long development cycle and high R&D costs for the coating. The coating process parameters corresponding to the target metallographic image can be determined based on the target thermal spray parameter determination model, thereby improving the efficiency of determining the spray process parameters and reducing R&D costs. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] Figure 1 is a flow chart of a method for determining thermal spraying parameters provided by an embodiment of the present invention;
[0018] Figure 2 is a flow chart of another method for determining thermal spraying parameters provided by an embodiment of the present invention;
[0019] Figure 3 It is a workflow diagram for determining thermal spraying parameters provided by an embodiment of the present invention;
[0020] Figure 4 is a structural schematic diagram of a thermal spraying parameter determination device provided by an embodiment of the present invention;
[0021] Figure 5 It is a structural schematic diagram of a computer device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0022] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0023] Figure 1 It is a flow chart of a thermal spraying parameter determination method provided by an embodiment of the present invention. The embodiment of the present invention can be applied to the scenario of determining the spraying process parameters corresponding to the target metallographic phase diagram. The method can be executed by a thermal spraying parameter determination device, which can be implemented by software and / or hardware.
[0024] like Figure 1 As shown, the thermal spraying parameter determination method includes the following steps:
[0025] S110, acquiring a target metallographic image of the coating to be sprayed.
[0026] The coating to be sprayed may be a coating to be sprayed by thermal spraying. The target metallographic image may be a metallographic image corresponding to the coating to be sprayed. The embodiment of the present invention needs to analyze the target metallographic image to determine the process parameters required for thermal spraying the coating to be sprayed.
[0027] S120, inputting the target metallographic image into a pre-trained target thermal spraying parameter determination model to obtain coating process parameters corresponding to the target metallographic image.
[0028] Among them, the target thermal spraying parameter determination model can be a model for determining the thermal spraying parameters corresponding to the coating to be sprayed. Specifically, the target thermal spraying parameter determination model can be obtained by training a selected neural network. Further, the coating process parameters can be process parameters required for thermally spraying the coating to be sprayed. Specifically, the coating quality characteristics can be input into the target thermal spraying parameter determination model so that the target thermal spraying parameter determination model performs image analysis on the target metallographic image, thereby obtaining the coating process parameters corresponding to the target metallographic image.
[0029] The technical solution provided by the embodiment of the present invention obtains a target metallographic image of the coating to be sprayed; inputs the target metallographic image into a pre-trained target thermal spray parameter determination model to obtain coating process parameters corresponding to the target metallographic image; wherein the coating process parameters are process parameters required for spraying the coating to be sprayed. The technical solution of the embodiment of the present invention solves the problem that the prior art requires human experience and a large number of experiments to determine the spray process parameters corresponding to the standard metallographic image, resulting in a long development cycle and high R&D costs for the coating. The coating process parameters corresponding to the target metallographic image can be determined based on the target thermal spray parameter determination model, thereby improving the efficiency of determining the spray process parameters and reducing R&D costs.
[0030] Figure 2 It is another flow chart of a thermal spraying parameter determination method provided by an embodiment of the present invention. The embodiment of the present invention can be applied to the scenario of determining the spraying process parameters corresponding to the target metallographic diagram. Based on the above embodiment, this embodiment further illustrates how to train to obtain the target thermal spraying parameter determination model; and how to determine the coating process parameters based on the target thermal spraying parameter determination model. The device can be implemented by software and / or hardware and integrated in a computer device with application development function.
[0031] like Figure 2 As shown, the thermal spraying parameter determination method includes the following steps:
[0032] S210, inputting a preset metallographic sample image into a preliminary thermal spraying parameter determination model to obtain sample process parameters.
[0033] The preset metallographic sample image may be a preset sample image of a metallographic image. The preliminary thermal spraying parameter determination model may be an untrained model for determining thermal spraying process parameters. The sample process parameters may be predicted process parameters required for thermally spraying a coating corresponding to the preset metallographic sample image. Specifically, the preset metallographic sample image may be input into the preliminary thermal spraying parameter determination model, so that the preliminary thermal spraying parameter determination model analyzes the features in the preset metallographic sample image, and then obtains the sample process parameters.
[0034] S220 , determining a model loss function according to the sample process parameters and a preset process parameter label of the preset metallographic sample image.
[0035] Among them, the preset process parameter label can be a label of the spraying process parameter corresponding to the preset metallographic sample image. Optionally, the preset process parameter label can be manually calibrated in advance. Furthermore, the model loss function can be an error loss function between the process parameters predicted by the model and the actual process parameters. Specifically, the sample process parameters and the parameters in the preset process parameter label can be substituted into the preset formula or algorithm for determining the loss function, thereby obtaining the model loss function. By determining the model loss function, the error between the process parameters predicted by the model and the actual process parameters can be known, which facilitates the subsequent adjustment of the preliminary thermal spraying parameter determination model based on the model loss function.
[0036] S230. Adjust the preliminary thermal spraying parameter determination model according to the model loss function to obtain the target thermal spraying parameter determination model.
[0037] The target thermal spraying parameter determination model may be a thermal spraying parameter determination model obtained after training. Specifically, the model parameters in the preliminary thermal spraying parameter determination model may be adjusted based on the model loss function to obtain the target thermal spraying parameter determination model.
[0038] S240, obtaining a target metallographic image of the coating to be sprayed, and inputting the target metallographic image into a feature extraction network in a target thermal spraying parameter determination model to obtain coating quality features.
[0039] The coating to be sprayed may be a coating to be sprayed by thermal spraying. The target metallographic image may be a metallographic image corresponding to the coating to be sprayed. The embodiment of the present invention needs to analyze the target metallographic image to determine the process parameters required for thermal spraying the coating to be sprayed.
[0040] Further, the target thermal spraying parameter determination model includes a feature extraction network and a process parameter determination network. Among them, the feature extraction network is used to extract the coating features corresponding to the target metallographic image. The process parameter determination network is used to analyze the coating features, and then determine the process parameters required for thermal spraying the coating to be sprayed. Further, the coating quality features can be features related to the coating quality of the coating to be sprayed. Among them, the coating quality features include but are not limited to the melting point, melting rate, thermal enthalpy, freezing point, solidification rate, condensation phase change characteristics or oxidation rate related to the material, the degree of flattening of powder particles impacting the workpiece surface, the degree of sputtering of powder particles impacting the workpiece surface, the degree of wetting and the paving state of a single layer of powder particles. Specifically, the target metallographic image can be input into the feature extraction network in the target thermal spraying parameter determination model so that the feature extraction network analyzes the target metallographic image to obtain the coating quality features.
[0041] Optionally, the feature extraction network includes: an image feature extraction subnetwork, inputting the target metallographic image into the feature extraction network to obtain coating quality features, including: inputting the target metallographic image into the image feature extraction subnetwork to obtain metallographic image quality features; determining the coating quality features based on the metallographic image quality features and a preset mapping relationship table; wherein the preset mapping relationship is used to represent the mapping relationship between the metallographic image quality features and the coating quality features.
[0042] Among them, the image feature extraction subnetwork is used to extract the quality features of the metallographic image. Further, the metallographic image quality features can be features related to the image quality of the target metallographic image. Specifically, the metallographic image quality features include but are not limited to gray-black pores, microcracks, irregular abnormal tissues, normal tissues, bright or dark tissues with different contrasts, mechanical mixed tissues in condensation, and other morphological features, texture features, and statistical features.
[0043] Furthermore, the preset mapping relationship may be a mapping relationship between the quality characteristics of the metallographic image and the quality characteristics of the coating. Specifically, the preset mapping relationship may be manually set in advance, or may be constructed by a mathematical relationship between the quality characteristics of the metallographic image and the quality characteristics of the coating. The specific method for obtaining the preset mapping relationship is not limited here. Furthermore, the metallographic image quality characteristics may be mapped in the direction of the coating quality characteristics based on the preset mapping relationship, thereby obtaining the coating quality characteristics of the coating to be sprayed. Optionally, in order to quantify the quality of the coating quality characteristics, after determining the coating quality characteristics, the coating quality characteristics may be converted into a corresponding vector, and the converted feature vector may be used as the coating quality characteristics.
[0044] Optionally, the target thermal spray parameter determination model also includes: a preparation method determination network, after obtaining the coating quality characteristics, it also includes: inputting the coating quality characteristics into the preparation method determination network to obtain the coating preparation method; wherein the coating preparation method is the preparation method required to spray the coating to be sprayed.
[0045] Among them, the preparation method determination network is used to determine the target preparation method required for thermal spraying the coating to be sprayed. Specifically, the preparation method includes but is not limited to flame spraying, supersonic spraying, plasma spraying, vacuum plasma, low-pressure plasma spraying and supersonic plasma spraying. Further, the coating preparation method can be the target preparation method required for thermal spraying the coating to be sprayed. Specifically, the coating quality characteristics can be input into the preparation method determination network so that the network analyzes the coating quality characteristics and then determines the coating preparation method required to obtain the coating quality characteristics.
[0046] Optionally, the preparation method determination network includes a flame flow characteristic determination subnetwork and a preparation method determination subnetwork, and the coating quality characteristics are input into the preparation method determination network to obtain the coating preparation method, including: inputting the coating quality characteristics into the flame flow characteristic determination subnetwork to obtain the flame flow characteristics; wherein the flame flow characteristics are state characteristics of the flame flow required for spraying the coating to be sprayed; inputting the flame flow characteristics into the preparation method determination subnetwork to obtain the coating preparation method.
[0047] Among them, the flame flow characteristic determination subnetwork is used to determine the flame flow characteristics required for the quality characteristics of the thermal spray coating. The preparation method determination subnetwork is used to determine the coating preparation method that is compatible with the flame flow characteristics. Specifically, the coating quality characteristics can be input into the flame flow characteristic determination subnetwork to obtain the flame flow characteristics. Among them, the flame flow characteristics include but are not limited to the flame flow enthalpy value, the flame flow velocity, the Reynolds coefficient of the flame flow beam, the relationship between the temperature distribution state of the flame flow cross section and the state of the powder conveying position in the flame flow. Further, the flame flow characteristics can be input into the preparation method determination subnetwork to obtain the coating preparation method.
[0048] S250, inputting the coating quality characteristics into a process parameter determination network in a target thermal spraying parameter determination model to obtain coating process parameters.
[0049] The coating process parameters may be process parameters required for thermally spraying the coating to be sprayed. Specifically, the coating quality characteristics may be input into a process parameter determination network to obtain the coating process parameters.
[0050] Optionally, after obtaining the coating process parameters and the coating preparation method, it also includes: determining spray control information according to the coating process parameters and the coating preparation method, and sending the spray control information to the target spray equipment.
[0051] Among them, the spray control information can be control parameter information for controlling the spray equipment. Specifically, the corresponding spray control information can be determined based on the coating process parameters and the coating preparation method. Furthermore, the target spray equipment can be a spray equipment for spraying the coating. By sending the spray control information to the target spray equipment, the target spray equipment can adjust its own spray control parameters according to the spray control information, and thermally spray the coating to be sprayed based on the adjusted spray control parameters.
[0052] For example, Figure 3 FIG. 1 is a flow chart for determining thermal spraying parameters provided by an embodiment of the present invention. Figure 3 As shown, the workflow for determining thermal spraying parameters includes the following steps:
[0053] 1. Input one or more standard metallographic diagrams of the specified thermal spray material into the thermal spray system module.
[0054] 2. Use convolutional neural network (CNN) to identify and preprocess the layers of standard metallographic images, including denoising, enhancement and standardization, to improve the accuracy of subsequent feature extraction. (For feature extraction and preliminary classification).
[0055] 3. Feature extraction module for standard metallographic images: Convolutional neural network (CNN) is used to extract features closely related to coating quality from preprocessed metallographic images. These features include but are not limited to melting point, melting rate, thermal enthalpy, solidification point, solidification rate, condensation phase change characteristics or oxidation rate related to materials, degree of flattening of powder particles, degree of sputtering of powder particles on the workpiece surface, degree of wetting, paving state of single-layer powder particles, etc., and gray-black pores, microcracks, irregular abnormal tissues, normal tissues, bright or dark tissues with different contrast, mechanical mixed tissues in condensation, and other morphological features, texture features, and statistical features associated with metallographic images, including but not limited to gray-black pores, microcracks, irregular abnormal tissues, normal tissues, bright or dark tissues with different contrasts, and mechanical mixed tissues in condensation. These features can be gradually abstracted and extracted at different levels of CNN to form a set of feature vectors that can characterize the intrinsic properties of the image.
[0056] 4. Process parameter mapping module: The extracted feature vector is input into the machine learning algorithm to map the feature to the coating process parameters. Using algorithms such as support vector machine, gradient boosting tree or supervised learning method for deep learning simulation training, this module can learn how to predict the best coating process parameters based on the input feature vector, such as current, voltage, gas flow and pressure, spray powder feed rate, spray distance, spray step, spray surface scanning speed, substrate temperature, etc. These parameters are crucial to controlling the microstructure and performance of the coating.
[0057] 5. Preparation method association module: In addition to process parameters, the preparation method of the coating will also affect its final performance. An additional classification task is introduced into the machine learning algorithm, and the extracted process feature vector is input into the machine learning algorithm of the gradient boosting tree, including but not limited to the flame enthalpy value, flame velocity, Reynolds coefficient of the flame beam, the relationship between the temperature distribution state of the flame flow section and the state of the powder transport position in the flame flow, etc. The most suitable coating preparation method is predicted based on the input feature vector, such as flame spraying, supersonic spraying, plasma spraying, vacuum plasma, low-pressure plasma spraying, supersonic plasma spraying, etc.
[0058] 6. Model optimization and interpretation module: Improve prediction accuracy and generalization ability by training and optimizing the hybrid model. At the same time, use the explanatory tools provided by the machine learning algorithm to interpret and analyze the prediction results of the model, as follows:
[0059] 6.1. Multi-modal parameter group fitting, obtain the output thermal spraying parameter group and key control information of coating process equipment.
[0060] 6.2 Simulate the spraying process and obtain the target preset metallographic effect diagram and automatically compare the results with the standard metallographic diagram.
[0061] 6.3. If the preset metallographic rendering reaches the standard metallographic rendering comparison result, the final simulation process equipment parameter group is output; if not, return to step 4 to continue the simulation training, and iterate until the preset metallographic rendering of step 6.3 meets the standard metallographic rendering comparison result, and then output the final process equipment parameter group result.
[0062] Finally, the prediction accuracy and generalization ability of the hybrid model can be further improved by training and optimizing it. At the same time, the prediction results of the model can be explained and analyzed by using the explanatory tools provided by the machine learning algorithm (such as feature importance scores, partial dependence plots, etc.), thereby helping to understand the influence mechanism of different features and process parameters on coating performance.
[0063] The invention has the advantage that it can automatically extract features closely related to coating quality from metallographic images and accurately map them to coating process parameters and preparation methods. By optimizing process parameters and selecting appropriate preparation methods, the quality and performance of the coating can be significantly improved. In addition, the invention also has good versatility and expansibility, and can adapt to different types of coating materials and preparation requirements.
[0064] Furthermore, in the present invention, the core task of the process parameter mapping module is to map the feature vector extracted from the input image data to the corresponding process parameters. To achieve this goal, we will adopt a supervised learning algorithm. Specifically, linear regression is used as an example to illustrate the algorithm details and the corresponding mathematical formulas.
[0065] Feature vectors and process parameters:
[0066] Assume that the feature vector extracted from the input data is represented as
[0067] (\mathbf{x}=(x_1,x_2,\ldots,x_n)), where (n) is the number of features,
[0068] The corresponding process parameter vector is expressed as
[0069] (\mathbf{y}=(y_1,y_2,\ldots,y_m)), where (m) is the number of process parameters.
[0070] It should be noted that the process parameters here can be multidimensional.
[0071] Linear Regression Model:
[0072] The linear regression model assumes that there is a linear relationship between the feature vector (\mathbf{x}) and the process parameter vector (\mathbf{y}), which can be expressed as:
[0073] [\hat{\mathbf{y}}=\mathbf{W}^T\mathbf{x}+\mathbf{b}]
[0074] Among them, (\hat{\mathbf{y}}) is the process parameter vector predicted by the model, (\mathbf{W}) is the weight matrix with dimension ((n,m)), and (\mathbf{b}) is the bias vector with dimension ((m,)).
[0075] Loss function:
[0076] In order to measure the accuracy of model prediction, we define a loss function, usually using Mean Squared Error (MSE):
[0077] [J(\mathbf{W},\mathbf{b})=\frac{1}{2N}\sum_{i=1}^{N}|\mathbf{y}_i-\hat{\mathbf{y}}_i|^2]
[0078] Where (N) is the number of training samples, (\mathbf{y}_i) is the true process parameter vector of the (i)th sample, and (\hat{\mathbf{y}}_i) is the model's predicted value for the (i)th sample.
[0079] Parameter optimization:
[0080] Our goal is to find the optimal weight matrix (\mathbf{W}) and bias vector (\mathbf{b}) so that the loss function (J(\mathbf{W},\mathbf{b})) is minimized. This is usually achieved through the gradient descent algorithm or its variants (such as stochastic gradient descent, batch gradient descent, etc.).
[0081] Gradient descent update rule:
[0082] The update rule for the weight matrix and bias vector can be expressed as:
[0083] [\mathbf{W}\leftarrow\mathbf{W}-\alpha\frac{\partial J}{\partial\mathbf{W}}]
[0084] [\mathbf{b}\leftarrow\mathbf{b}-\alpha\frac{\partial J}{\partial\mathbf{b}}]
[0085] Where (\alpha) is the learning rate, a positive decimal that controls the step size of parameter updates. The gradients (\frac{\partial J}{\partial\mathbf{W}}) and (\frac{\partial J}{\partial\mathbf{b}}) can be calculated using the chain rule and matrix calculus.
[0086] Notes:
[0087] The feature vector (\mathbf{x}) is a set of features related to process parameters extracted from the input data and can be multidimensional.
[0088] The process parameter vector (\mathbf{y}) is the target variable that we want the model to predict and can also be multidimensional.
[0089] The linear regression model assumes that there is a linear relationship between the feature vector and the process parameters, which is a simplifying assumption that is applicable to many practical problems.
[0090] The loss function is used to quantify the accuracy of model predictions. Mean squared error is a commonly used loss function suitable for regression problems.
[0091] Gradient descent is an optimization algorithm used to find parameter values that minimize the loss function.
[0092] The learning rate (\alpha) is a hyperparameter that needs to be adjusted based on the specific problem and data to ensure convergence and stability of the algorithm.
[0093] In the preparation method association module of the present invention, we will use a classification algorithm based on support vector machine (SVM) to realize the prediction of the coating preparation method. The following is the specific mathematical formula and expression of the algorithm:
[0094] Feature vector: First, we assume that the feature vector extracted from the metallographic image is expressed as
[0095] (\mathbf{x}=(x_1,x_2,\ldots,x_n)), where (n) is the number of features.
[0096] SVM classifier: The goal of the SVM classifier is to find a hyperplane that can maximize the separation of samples of different categories. For a binary classification problem, the hyperplane can be expressed as:
[0097] [\mathbf{w}\cdot\mathbf{x}+b=0]
[0098] Among them, (\mathbf{w}) is the weight vector and (b) is the bias term. For multi-classification problems, they can be solved by building multiple binary classifiers (such as one-to-one or one-to-many strategies).
[0099] Decision function: For a given feature vector (\mathbf{x}), the decision function of SVM can be expressed as:
[0100] [f(\mathbf{x})=\text{sign}(\mathbf{w}\cdot\mathbf{x}+b)]
[0101] Among them, (\text{sign}) is the sign function, and the return value is (+1) or (-1), representing two different categories.
[0102] Note: In practical applications, since we may be faced with a multi-classification problem (multiple preparation methods), we need to use a multi-classification SVM. The decision function at this time will be slightly different, and methods such as probability estimation may be used to determine the final category.
[0103] Training process: During the training phase, we need to solve the following optimization problem to find the optimal (\mathbf{w}) and (b):
[0104] [\min_{\mathbf{w},b}\frac{1}{2}|\mathbf{w}|^2+C\sum_{i=1}^{m}\xi_i]
[0105] [\text{st}\quad y_i(\mathbf{w}\cdot\mathbf{x}_i+b)\geq 1-\xi_i,\quad\xi_i\geq 0,\quad i=1,2,\ldots,m]
[0106] Where (m) is the number of training samples, (y_i) is the true class label of sample (i) ((\pm 1)), (\xi_i) is a slack variable that allows some samples to be misclassified, and (C) is a penalty parameter that controls the degree of penalty for misclassification.
[0107] Kernel function: For nonlinear separable problems, the kernel function can be introduced:
[0108] (K(\mathbf{x}_i,\mathbf{x}_j)) maps the original feature space to a higher-dimensional feature space, making the samples linearly separable in the new feature space. Commonly used kernel functions include linear kernels, polynomial kernels, radial basis function (RBF) kernels, etc.
[0109] Note: In the embodiments of the present invention, a suitable kernel function can be selected according to the complexity of the actual problem and the characteristics of the data.
[0110] Notes:
[0111] The feature vector (\mathbf{x}) is a set of features extracted from the metallographic image that is closely related to the coating quality.
[0112] The SVM classifier is a powerful supervised learning algorithm suitable for small sample, high dimensional and nonlinear classification problems.
[0113] The decision function is used to determine the most appropriate coating preparation method based on the extracted feature vectors.
[0114] The training process finds the optimal classification hyperplane by minimizing the structured risk while taking into account the classification margin and the penalty for misclassification.
[0115] The kernel function is used to deal with nonlinear problems and map the original feature space to a higher-dimensional feature space to achieve linear classification.
[0116] The technical solution provided by the embodiment of the present invention obtains a target metallographic image of the coating to be sprayed; inputs the target metallographic image into a feature extraction network in a target thermal spraying parameter determination model to obtain coating quality characteristics; inputs the coating quality characteristics into a process parameter determination network in a target thermal spraying parameter determination model to obtain coating process parameters. The technical solution of the embodiment of the present invention solves the problem that the prior art requires human experience and a large number of experiments to determine the spraying process parameters corresponding to the standard metallographic image, resulting in a long development cycle and high R&D costs for the coating. The coating process parameters corresponding to the target metallographic image can be determined based on the target thermal spraying parameter determination model, thereby improving the efficiency of determining the spraying process parameters and reducing R&D costs.
[0117] Figure 4 It is a structural schematic diagram of a thermal spraying parameter determination device provided in an embodiment of the present invention. The embodiment of the present invention can be applied to the scenario of determining the spraying process parameters corresponding to the target metallographic phase diagram. The device can be implemented by software and / or hardware and integrated into a computer device with application development function.
[0118] like Figure 4 As shown, the thermal spraying parameter determination device includes: a metallographic image acquisition module 310 and a coating process parameter determination module 320.
[0119] Among them, the metallographic image acquisition module 310 is used to obtain the target metallographic image of the coating to be sprayed; the coating process parameter determination module 320 is used to input the target metallographic image into a pre-trained target thermal spray parameter determination model to obtain the coating process parameters corresponding to the target metallographic image; wherein the coating process parameters are the process parameters required for spraying the coating to be sprayed.
[0120] The technical solution provided by the embodiment of the present invention obtains a target metallographic image of the coating to be sprayed; inputs the target metallographic image into a pre-trained target thermal spray parameter determination model to obtain coating process parameters corresponding to the target metallographic image; wherein the coating process parameters are process parameters required for spraying the coating to be sprayed. The technical solution of the embodiment of the present invention solves the problem that the prior art requires human experience and a large number of experiments to determine the spray process parameters corresponding to the standard metallographic image, resulting in a long development cycle and high R&D costs for the coating. The coating process parameters corresponding to the target metallographic image can be determined based on the target thermal spray parameter determination model, thereby improving the efficiency of determining the spray process parameters and reducing R&D costs.
[0121] In an optional embodiment, the target thermal spray parameter determination model includes: a feature extraction network and a process parameter determination network, and the coating process parameter determination module 320 is specifically used to: input the target metallographic image into the feature extraction network to obtain the coating quality characteristics; input the coating quality characteristics into the process parameter determination network to obtain the coating process parameters.
[0122] In an optional embodiment, the feature extraction network includes: an image feature extraction subnetwork, and the coating process parameter determination module 320 includes: a coating quality feature determination unit, used to: input the target metallographic image into the image feature extraction subnetwork to obtain metallographic image quality features; determine the coating quality features based on the metallographic image quality features and a preset mapping relationship table; wherein the preset mapping relationship is used to represent the mapping relationship between the metallographic image quality features and the coating quality features.
[0123] In an optional embodiment, the target thermal spraying parameter determination model also includes: a preparation method determination network, and the thermal spraying parameter determination device also includes: a preparation method determination module, which is used to: after obtaining the coating quality characteristics, input the coating quality characteristics into the preparation method determination network to obtain the coating preparation method; wherein the coating preparation method is the preparation method required to spray the coating to be sprayed.
[0124] In an optional embodiment, the preparation method determination network includes a flame flow characteristic determination subnetwork and a preparation method determination subnetwork, and the preparation method determination module is specifically used to: input the coating quality characteristics into the flame flow characteristic determination subnetwork to obtain the flame flow characteristics; wherein the flame flow characteristics are the state characteristics of the flame flow required for spraying the coating to be sprayed; input the flame flow characteristics into the preparation method determination subnetwork to obtain the coating preparation method.
[0125] In an optional embodiment, the thermal spraying parameter determination device also includes: a spraying control information determination module, which is used to: after obtaining the coating process parameters and the coating preparation method, determine the spraying control information according to the coating process parameters and the coating preparation method, and send the spraying control information to the target spraying equipment.
[0126] In an optional embodiment, the thermal spraying parameter determination device also includes: a thermal spraying parameter determination model training module, which is used to: input a preset metallographic sample image into a preliminary thermal spraying parameter determination model to obtain sample process parameters; determine a model loss function based on the sample process parameters and the preset process parameter labels of the preset metallographic sample image; adjust the preliminary thermal spraying parameter determination model based on the model loss function to obtain the target thermal spraying parameter determination model.
[0127] The thermal spraying parameter determination device provided in the embodiment of the present invention can execute the thermal spraying parameter determination method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.
[0128] Figure 5 A schematic diagram of the structure of a computer device provided in an embodiment of the present invention. Figure 5 A block diagram of an exemplary computer device 12 suitable for use in implementing embodiments of the present invention is shown. Figure 5 The computer device 12 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention. The computer device 12 can be any terminal device with computing capability and can be configured in the thermal spraying parameter determination device.
[0129] like Figure 5 As shown, the computer device 12 is in the form of a general-purpose computing device. The components of the computer device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that connects various system components (including the system memory 28 and the processing unit 16).
[0130] The bus 18 may be one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. For example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.
[0131] The computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 12, including volatile and non-volatile media, removable and non-removable media.
[0132] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache 32. The computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 5 not shown, usually called a "hard drive"). Although Figure 5 Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The system memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.
[0133] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. Program modules 42 generally perform the functions and / or methods of the embodiments described herein.
[0134] The computer device 12 may also communicate with one or more external devices 14 (e.g., keyboards, pointing devices, displays 24, etc.), one or more devices that enable a user to interact with the computer device 12, and / or any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed through an input / output (I / O) interface 22. Furthermore, the computer device 12 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) through a network adapter 20. Figure 5 As shown, the network adapter 20 communicates with other modules of the computer device 12 via the bus 18. It should be understood that although Figure 5 Not shown, other hardware and / or software modules may be used in conjunction with computer device 12, including but not limited to microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.
[0135] The processing unit 16 executes various functional applications and data processing by running the program stored in the system memory 28, for example, implementing the thermal spraying parameter determination method provided in the embodiment of the present invention, the method comprising:
[0136] Acquire a target metallographic image of the coating to be sprayed;
[0137] The target metallographic image is input into a pre-trained target thermal spraying parameter determination model to obtain coating process parameters corresponding to the target metallographic image; wherein the coating process parameters are process parameters required for spraying the coating to be sprayed.
[0138] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the method for determining thermal spraying parameters provided in any embodiment of the present invention is implemented, including:
[0139] Acquire a target metallographic image of the coating to be sprayed;
[0140] The target metallographic image is input into a pre-trained target thermal spraying parameter determination model to obtain coating process parameters corresponding to the target metallographic image; wherein the coating process parameters are process parameters required for spraying the coating to be sprayed.
[0141] The computer storage medium of the embodiment of the present invention can adopt any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium can be, for example, but not limited to: an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.
[0142] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.
[0143] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.
[0144] Computer program code for performing the operations of the present invention may be written in one or more programming languages or combinations thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).
[0145] It should be understood by those skilled in the art that the modules or steps of the present invention described above can be implemented by a general-purpose computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, optionally, they can be implemented by a program code executable by a computer device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.
[0146] 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 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 method for determining thermal spraying parameters, characterized in that: include: Acquire a target metallographic image of the coating to be sprayed; Inputting the target metallographic image into a feature extraction network in a target thermal spraying parameter determination model to obtain coating quality features; Inputting the coating quality characteristics into a process parameter determination network in a target thermal spraying parameter determination model to obtain coating process parameters; wherein the coating process parameters are process parameters required for spraying the coating to be sprayed; Wherein, the feature extraction network includes: an image feature extraction subnetwork, and the target metallographic image is input into the feature extraction network to obtain the coating quality feature, including: Inputting the target metallographic image into the image feature extraction subnetwork to obtain metallographic image quality features; Determining the coating quality feature based on the metallographic image quality feature and a preset mapping relationship table; wherein the preset mapping relationship is used to represent a mapping relationship between the metallographic image quality feature and the coating quality feature; Wherein, the target thermal spraying parameter determination model further includes: a preparation method determination network, after obtaining the coating quality characteristics, further includes: Inputting the coating quality characteristics into the flame flow characteristic determination subnetwork in the preparation method determination network to obtain the flame flow characteristics; wherein the flame flow characteristics are the state characteristics of the flame flow required for spraying the coating to be sprayed; The flame flow characteristics are input into the preparation method determination subnetwork in the preparation method determination network to obtain a coating preparation method; wherein the coating preparation method is a preparation method required for spraying the coating to be sprayed.
2. The method according to claim 1, characterized in that After obtaining the coating process parameters and the coating preparation method, the method further includes: Spraying control information is determined according to the coating process parameters and the coating preparation method, and the spraying control information is sent to a target spraying device.
3. The method according to claim 1, characterized in that The training process of the target thermal spraying parameter determination model includes: Inputting the preset metallographic sample image into the preliminary thermal spraying parameter determination model to obtain the sample process parameters; Determining a model loss function according to the sample process parameters and the preset process parameter labels of the preset metallographic sample images; The preliminary thermal spraying parameter determination model is adjusted according to the model loss function to obtain the target thermal spraying parameter determination model.
4. A thermal spraying parameter determination device, characterized in that: The device comprises: A metallographic image acquisition module, used to acquire a target metallographic image of the coating to be sprayed; A coating process parameter determination module is used to input the target metallographic image into a feature extraction network in a target thermal spraying parameter determination model to obtain coating quality features; input the coating quality features into a process parameter determination network in a target thermal spraying parameter determination model to obtain coating process parameters; wherein the coating process parameters are process parameters required for spraying the coating to be sprayed; The feature extraction network includes: an image feature extraction subnetwork, and the coating process parameter determination module includes: a coating quality feature determination unit, which is used to: input the target metallographic image into the image feature extraction subnetwork to obtain metallographic image quality features; determine the coating quality features based on the metallographic image quality features and a preset mapping relationship table; wherein the preset mapping relationship is used to represent the mapping relationship between the metallographic image quality features and the coating quality features; Wherein, the target thermal spraying parameter determination model includes: a preparation method determination network; the thermal spraying parameter determination device also includes: a preparation method determination module, which is used to input the coating quality characteristics into the flame flow characteristic determination subnetwork in the preparation method determination network to obtain the flame flow characteristics; wherein, the flame flow characteristics are the state characteristics of the flame flow required for spraying the coating to be sprayed; the flame flow characteristics are input into the preparation method determination subnetwork in the preparation method determination network to obtain the coating preparation method; wherein, the coating preparation method is the preparation method required for spraying the coating to be sprayed.
5. A computer device, characterized in that: The computer device comprises: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the thermal spraying parameter determination method as described in any one of claims 1 to 3.
6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method for determining thermal spraying parameters as described in any one of claims 1 to 3 is implemented.
