Automatic cutting machining system based on aluminum alloy production

Abstract

The invention relates to the technical field of automatic cutting of aluminum alloy through computer aided design, in particular to an automatic cutting processing system based on aluminum alloy production, which comprises the following steps: conveying aluminum alloy to a specified cutting area, and acquiring a multi-angle image of the surface of the aluminum alloy; performing feature extraction on the aluminum alloy surface multi-angle image, inputting the extracted features into a pre-trained aluminum alloy cutting parameter calculation model for solving, judging whether a specified cutting area needs to be cut or not according to a solving result, and if not, conveying the aluminum alloy to a next specified cutting area; if yes, the solving result is transmitted to a cutting device control system; and the cutting device control system adjusts the cutting device according to the cutting parameters in the solving result, and the aluminum alloy is cut until cutting is completed.

Description

Technical Field

[0001] The present invention relates to the technical field of automatic cutting of aluminum alloys through computer-aided design, and in particular to an automatic cutting processing system based on aluminum alloy production. Background Art

[0002] Aluminum alloys are widely used in aerospace, automobile manufacturing, and construction due to their light weight, high strength, and corrosion resistance. However, the cutting of aluminum alloys has always been a challenge for the manufacturing industry. With the development of industry and intelligent manufacturing, the requirements for precision, efficiency, and intelligence in aluminum alloy cutting are constantly increasing. Traditional aluminum alloy cutting technology mainly relies on manual experience and fixed parameter settings, which has many defects. On the one hand, manual judgment and parameter adjustment are easily affected by subjective factors, making it difficult to ensure the consistency of cutting quality. On the other hand, fixed parameters cannot adapt to the changes in the surface characteristics of aluminum alloys in different batches or regions, resulting in unstable cutting effects. Traditional aluminum alloy cutting technology is difficult to adapt to the complex and changeable surface characteristics of aluminum alloys, which can easily lead to unstable cutting quality, material waste, and low production efficiency of aluminum alloys. Therefore, the development of an intelligent system that can automatically adjust cutting parameters according to the surface characteristics of aluminum alloys has become a current research hotspot to meet the needs of modern manufacturing for high-quality and high-efficiency aluminum alloy processing. Summary of the invention

[0003] The purpose of the present invention is to provide an automatic cutting processing system based on aluminum alloy production, which can dynamically calculate cutting requirements and parameters according to the specific properties of aluminum alloy materials through image feature extraction and aluminum alloy cutting parameter calculation model, significantly improve cutting accuracy and efficiency, and real-time adjustment of cutting parameters under different conditions can reduce material loss, and can achieve lower operating costs and higher quality control. It is easy to integrate with existing production systems, which promotes the intelligence and automation of the overall manufacturing process.

[0004] The present invention is achieved through the following technical solutions:

[0005] An automatic cutting processing system based on aluminum alloy production, comprising:

[0006] The acquisition unit transports the aluminum alloy to the designated cutting area and obtains multi-angle images of the aluminum alloy surface;

[0007] The solving unit extracts features from the multi-angle images of the aluminum alloy surface, and inputs the extracted features into a pre-trained aluminum alloy cutting parameter calculation model for solving, and determines whether the designated cutting area needs to be cut according to the solving result. If not, the aluminum alloy is transported to the next designated cutting area; if yes, the solving result is transmitted to the cutting device control system;

[0008] The control unit and the cutting device control system adjust the cutting device according to the cutting parameters in the solution results, and execute cutting on the aluminum alloy until the cutting is completed.

[0009] Optionally, feature extraction is performed on the multi-angle images of the aluminum alloy surface to obtain surface texture features, surface roughness features, surface edge features, surface corner features and surface color features of the aluminum alloy.

[0010] Optionally, the surface texture characteristics of the aluminum alloy are calculated by the formula:

[0011]

[0012] Where T is the surface texture characteristic of aluminum alloy, count m (i, j) is the number of times the grayscale values ​​i and j appear together at a specific distance and direction in the aluminum alloy surface image at the mth angle, M is the number of multi-angle images of the aluminum alloy surface, m is the index of the multi-angle image of the aluminum alloy surface, and i and j are the grayscale values, respectively.

[0013] Optionally, the surface roughness characteristic of the aluminum alloy is calculated by the following formula:

[0014]

[0015] Among them, R is the surface roughness characteristic of aluminum alloy, M is the number of multi-angle images of aluminum alloy surface, P is the number of neighborhood pixels, p is the index of neighborhood pixels, g is the index of neighborhood pixels, p,m , g c,m are the neighborhood and center pixel grayscale values ​​of the aluminum alloy surface image at the mth angle, and s() is the threshold function.

[0016] Optionally, the surface edge features of the aluminum alloy include:

[0017] Solve the gradients of the aluminum alloy surface image m at each angle in the x direction and the y direction to obtain a first gradient and a second gradient;

[0018] The surface edge characteristics of the aluminum alloy are calculated by combining the first gradient and the second gradient;

[0019] The calculation formula of the first gradient is:

[0020]

[0021] The calculation formula of the second gradient is:

[0022]

[0023] The surface edge characteristics of the aluminum alloy are calculated by combining the first gradient and the second gradient, and the calculation formula is:

[0024]

[0025] Where E is the surface edge feature of the aluminum alloy, M is the number of multi-angle images of the aluminum alloy surface, and f m is the two-dimensional array of aluminum alloy surface images at the mth angle, is the first gradient, is the second gradient.

[0026] Optionally, the surface corner feature of the aluminum alloy is calculated by the following formula:

[0027]

[0028] Among them, C is the surface corner feature of aluminum alloy, M is the number of multi-angle images of the aluminum alloy surface, k is an empirical constant, det is the determinant, and tr is the trace, that is, the sum of the main diagonal elements.

[0029] Optionally, the surface color characteristics of the aluminum alloy are calculated by the following formula:

[0030]

[0031]

[0032]

[0033] Among them, H, S, and V are the comprehensive hue, saturation, and brightness characteristics of aluminum alloy, respectively; M is the number of multi-angle images of the aluminum alloy surface; N is the number of pixels of the aluminum alloy surface image at each angle; and H m,i , S m,i 、V m,i are the HSV values ​​of the i-th pixel in the aluminum alloy surface image at the m-th angle.

[0034] Optionally, the aluminum alloy cutting parameter calculation model specifically takes maximizing cutting quality as the optimization goal, and its calculation formula is:

[0035]

[0036] p i =W (out) H (L) +b (out)

[0037]

[0038] H (0) =X = [x1,x2,...,x N ] T

[0039] xi =[T i ,R i ,E i ,C i ,H i ,S i ,V i ]

[0040] Among them, max θ J(θ) is the objective function, and θ is the set of all training parameters in the aluminum alloy cutting parameter calculation model, including W (l) , W (out) 、b (out) and W k , N is the total number of samples, i is the sample index, K is the number of cutting quality indicators, k is the index of the cutting quality indicator, and w k is the weight of the kth cutting quality index, x i is the input feature vector of the i-th sample, W k is the parameter matrix of the kth cutting quality index, p i is the cutting parameter vector of the i-th sample, d is the dimension of the feature vector, λ is the L2 regularization coefficient, ‖p i ‖2 is the L2 norm of the cutting parameter vector, W (out) is the weight matrix of the output layer of the aluminum alloy cutting parameter calculation model, H (L) is the output of the last layer of the aluminum alloy cutting parameter calculation model, b (out) is the bias vector of the output layer of the aluminum alloy cutting parameter calculation model, l is the index of the aluminum alloy cutting parameter calculation model layer, H (l) is the characteristic matrix of the calculation model of the aluminum alloy cutting parameters at the first layer, W (l) is the weight matrix of the calculation model of the aluminum alloy cutting parameters at the lth layer, is the adjacency matrix after adding self-connection, yes degree matrix, X is the input feature matrix, including the feature vectors of all samples, T i , R i 、E i , C i , H i , S i 、V i are the surface texture feature, surface roughness feature, surface edge feature, surface corner feature, hue value in HSV color space, saturation value in HSV color space and brightness value in HSV color space of the i-th sample, ReLU is the rectified linear unit activation function, and tanh is the hyperbolic tangent function.

[0041] The technical solution of the present invention has at least the following advantages and beneficial effects:

[0042] The present invention can dynamically calculate cutting requirements and parameters according to the specific properties of the aluminum alloy material through image feature extraction and aluminum alloy cutting parameter calculation model, significantly improving cutting accuracy and efficiency. In addition, real-time adjustment of cutting parameters under different conditions can reduce material loss, achieve lower operating costs and higher quality control, and is easy to integrate with existing production systems, promoting the intelligence and automation of the overall manufacturing process. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A logical schematic diagram of an automatic cutting processing system based on aluminum alloy production provided by the present invention. DETAILED DESCRIPTION

[0044] In order to make the purpose, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be described clearly and completely below in conjunction with the drawings in the present invention. Obviously, what is described is only a part of the present invention, not all of it. Generally, the components of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0045] Reference Figure 1 As shown, Figure 1 A logical schematic diagram of an automatic cutting processing system based on aluminum alloy production provided by the present invention.

[0046] In one embodiment, an automatic cutting processing system based on aluminum alloy production includes:

[0047] The acquisition unit transports the aluminum alloy to the designated cutting area and obtains multi-angle images of the aluminum alloy surface;

[0048] The solving unit extracts features from the multi-angle images of the aluminum alloy surface, and inputs the extracted features into a pre-trained aluminum alloy cutting parameter calculation model for solving, and determines whether the designated cutting area needs to be cut according to the solving result. If not, the aluminum alloy is transported to the next designated cutting area; if yes, the solving result is transmitted to the cutting device control system;

[0049] The control unit and the cutting device control system adjust the cutting device according to the cutting parameters in the solution results, and execute cutting on the aluminum alloy until the cutting is completed.

[0050] This embodiment can be integrated into an intelligent system, which includes an aluminum alloy cutting parameter calculation model, an image acquisition system, a data processing unit, a cutting device control system, an aluminum alloy conveying system and a human-computer interaction interface. In the application of this embodiment, the aluminum alloy material is conveyed to the designated position by a conveyor belt system, and the aluminum alloy material is ensured to be positioned stably and accurately during the conveying process. After arriving at the designated area, the surface image of the aluminum alloy is captured from multiple angles by a high-resolution camera to ensure full coverage. Then, the collected aluminum alloy image is subjected to noise reduction, contrast enhancement and other processing, and key features are extracted at the same time, and the extracted key features are input into the trained aluminum alloy cutting parameter calculation model. Based on the output result of the aluminum alloy cutting parameter calculation model, it is determined whether the current position needs to be cut. If not, it returns to the conveying step and moves to the next position. If cutting is required, the accurate cutting parameters are calculated according to the output of the aluminum alloy cutting parameter calculation model, and the calculated parameters are transmitted to the cutting device control system. The control system adjusts the position and setting of the cutting device, starts the cutting device, performs the actual cutting operation, and monitors the cutting process in real time during the cutting process to ensure safety and accuracy.

[0051] During the implementation process, the multi-angle images of the aluminum alloy surface are subjected to feature extraction to obtain the surface texture features, surface roughness features, surface edge features, surface corner features and surface color features of the aluminum alloy.

[0052] The surface texture characteristics of the aluminum alloy are calculated by the following formula:

[0053]

[0054] Where T is the surface texture characteristic of aluminum alloy, count m (i, j) is the number of times the grayscale values ​​i and j appear together at a specific distance and direction in the aluminum alloy surface image at the mth angle, M is the number of multi-angle images of the aluminum alloy surface, m is the index of the multi-angle image of the aluminum alloy surface, and i and j are the grayscale values, respectively.

[0055] The surface roughness characteristics of the aluminum alloy are calculated as follows:

[0056]

[0057] Among them, R is the surface roughness characteristic of aluminum alloy, M is the number of multi-angle images of aluminum alloy surface, P is the number of neighborhood pixels, p is the index of neighborhood pixels, g is the index of neighborhood pixels, p,m , g c,m are the neighborhood and center pixel grayscale values ​​of the aluminum alloy surface image at the mth angle, and s() is the threshold function.

[0058] The surface edge features of the aluminum alloy include:

[0059] Solve the gradients of the aluminum alloy surface image m at each angle in the x direction and the y direction to obtain a first gradient and a second gradient;

[0060] The surface edge characteristics of the aluminum alloy are calculated by combining the first gradient and the second gradient;

[0061] The calculation formula of the first gradient is:

[0062]

[0063] The calculation formula of the second gradient is:

[0064]

[0065] The surface edge characteristics of the aluminum alloy are calculated by combining the first gradient and the second gradient, and the calculation formula is:

[0066]

[0067] Where E is the surface edge feature of the aluminum alloy, M is the number of multi-angle images of the aluminum alloy surface, and f m is the two-dimensional array of aluminum alloy surface images at the mth angle, is the first gradient, is the second gradient.

[0068] The surface corner feature of the aluminum alloy is calculated by the following formula:

[0069]

[0070] Among them, C is the surface corner feature of aluminum alloy, M is the number of multi-angle images of the aluminum alloy surface, k is an empirical constant, det is the determinant, and tr is the trace, that is, the sum of the main diagonal elements.

[0071] The surface color characteristics of the aluminum alloy are calculated by the following formula:

[0072]

[0073]

[0074]

[0075] Among them, H, S, and V are the comprehensive hue, saturation, and brightness characteristics of aluminum alloy, respectively; M is the number of multi-angle images of the aluminum alloy surface; N is the number of pixels of the aluminum alloy surface image at each angle; and H m,i , S m,i 、V m,i are the HSV values ​​of the i-th pixel in the aluminum alloy surface image at the m-th angle.

[0076] In a further implementation of this embodiment, the aluminum alloy cutting parameter calculation model constructed in this embodiment is specifically constructed based on the GCN model. The construction of the aluminum alloy cutting parameter calculation model is as follows: define input feature X = [x1, x2, ..., x N ] T ; Construct the adjacency matrix A to represent the relationship between features and calculate I is the identity matrix, and the degree matrix is ​​calculated Define L GCN layers and calculate each layer: Initialize the weight matrix W ( l) ; Define the output layer weight W (out) And bias b(out), calculate the cutting parameters: p i =W (out) H (L) +b (out) ; Define the objective function with maximizing cutting quality as the optimization goal.

[0077] For the training of the aluminum alloy cutting parameter calculation model, it mainly uses aluminum alloy historical data, which includes historical aluminum alloy surface feature data and the corresponding optimal cutting parameters, and is divided into a training set and a validation set. The specific training process is to evaluate the performance of the aluminum alloy cutting parameter calculation model after forward propagation and back propagation, and output the trained aluminum alloy cutting parameter calculation model. It can be understood that this process is a standard training step for a neural network model, which will not be elaborated in this embodiment.

[0078] The objective function of the aluminum alloy cutting parameter calculation model is calculated as follows:

[0079]

[0080] p i =W (out) H (L) +b (out)

[0081]

[0082] H (0) =X = [x1,x2,...,x N ] T

[0083] x i =[T i ,R i ,E i ,C i ,H i ,S i ,V i ]

[0084] Among them, max θ J(θ) is the objective function, and θ is the set of all training parameters in the aluminum alloy cutting parameter calculation model, including W (l) , W (out) 、b (out) and W k , N is the total number of samples, i is the sample index, K is the number of cutting quality indicators, k is the index of the cutting quality indicator, and w k is the weight of the kth cutting quality index, x i is the input feature vector of the i-th sample, W k is the parameter matrix of the kth cutting quality index, p i is the cutting parameter vector of the i-th sample, d is the dimension of the feature vector, λ is the L2 regularization coefficient, ‖p i ‖2 is the L2 norm of the cutting parameter vector, W (out) is the weight matrix of the output layer of the aluminum alloy cutting parameter calculation model, H (L) is the output of the last layer of the aluminum alloy cutting parameter calculation model, b (out) is the bias vector of the output layer of the aluminum alloy cutting parameter calculation model, l is the index of the aluminum alloy cutting parameter calculation model layer, H (l) is the characteristic matrix of the calculation model of the aluminum alloy cutting parameters at the first layer, W (l) is the weight matrix of the calculation model of the aluminum alloy cutting parameters at the lth layer, is the adjacency matrix after adding self-connection, yes degree matrix, X is the input feature matrix, including the feature vectors of all samples, T i , R i 、E i , C i , H i , S i 、V i are the surface texture feature, surface roughness feature, surface edge feature, surface corner feature, hue value in HSV color space, saturation value in HSV color space and brightness value in HSV color space of the i-th sample, ReLU is the rectified linear unit activation function, and tanh is the hyperbolic tangent function.

[0085] During implementation, the input features include surface texture, roughness, edges, corners, and H, S, and V values ​​of the HSV color space. These input features are processed by the aluminum alloy cutting parameter calculation model to generate cutting parameters. The cutting parameters are used together with the input features to calculate the cutting quality score. The objective function defined in this embodiment comprehensively considers the cutting quality score and parameter complexity, and optimizes the model parameters by maximizing this function. It fully considers the characteristics of the aluminum alloy cutting process and the optimization requirements of the machine learning model, and provides a theoretical basis for high-quality and efficient aluminum alloy cutting parameter calculation.

[0086] The above is only the preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An automatic cutting processing system based on aluminum alloy production, characterized in that: include: The acquisition unit transports the aluminum alloy to the designated cutting area and obtains multi-angle images of the aluminum alloy surface; The solving unit extracts features from the multi-angle images of the aluminum alloy surface, and inputs the extracted features into a pre-trained aluminum alloy cutting parameter calculation model for solving, and determines whether the designated cutting area needs to be cut according to the solution result. If not, the aluminum alloy is transported to the next designated cutting area; If yes, the solution result is transmitted to the cutting device control system; The control unit and the cutting device control system adjust the cutting device according to the cutting parameters in the solution results, and execute cutting on the aluminum alloy until the cutting is completed.

2. The automatic cutting processing system based on aluminum alloy production according to claim 1 is characterized in that: The feature extraction is performed on the multi-angle image of the aluminum alloy surface to obtain the surface texture feature, surface roughness feature, surface edge feature, surface corner feature and surface color feature of the aluminum alloy.

3. The automatic cutting processing system based on aluminum alloy production according to claim 2 is characterized in that: The surface texture characteristics of the aluminum alloy are calculated by the following formula: Where T is the surface texture characteristic of aluminum alloy, count m (i, j) is the number of times the grayscale values ​​i and j appear together at a specific distance and direction in the aluminum alloy surface image at the mth angle, M is the number of multi-angle images of the aluminum alloy surface, m is the index of the multi-angle image of the aluminum alloy surface, and i and j are the grayscale values, respectively.

4. The automatic cutting processing system based on aluminum alloy production according to claim 3 is characterized in that: The surface roughness characteristics of the aluminum alloy are calculated as follows: Among them, R is the surface roughness characteristic of aluminum alloy, M is the number of multi-angle images of aluminum alloy surface, P is the number of neighborhood pixels, p is the index of neighborhood pixels, g is the index of neighborhood pixels, p,m , g c,m are the neighborhood and center pixel grayscale values ​​of the aluminum alloy surface image at the mth angle, and s() is the threshold function.

5. The automatic cutting processing system based on aluminum alloy production according to claim 4 is characterized in that: The surface edge features of the aluminum alloy include: Solve the gradients of the aluminum alloy surface image m at each angle in the x direction and the y direction to obtain a first gradient and a second gradient; The surface edge characteristics of the aluminum alloy are calculated by combining the first gradient and the second gradient; The calculation formula of the first gradient is: The calculation formula of the second gradient is: The surface edge characteristics of the aluminum alloy are calculated by combining the first gradient and the second gradient, and the calculation formula is: Where E is the surface edge feature of the aluminum alloy, M is the number of multi-angle images of the aluminum alloy surface, and f m is the two-dimensional array of aluminum alloy surface images at the mth angle, is the first gradient, is the second gradient.

6. The automatic cutting processing system based on aluminum alloy production according to claim 5 is characterized in that: The surface corner feature of the aluminum alloy is calculated by the following formula: Among them, C is the surface corner feature of aluminum alloy, M is the number of multi-angle images of the aluminum alloy surface, k is an empirical constant, det is the determinant, and tr is the trace, that is, the sum of the main diagonal elements.

7. The automatic cutting processing system based on aluminum alloy production according to claim 6 is characterized in that: The surface color characteristics of the aluminum alloy are calculated by the following formula: Among them, H, S, and V are the comprehensive hue, saturation, and brightness characteristics of aluminum alloy, respectively; M is the number of multi-angle images of the aluminum alloy surface; N is the number of pixels of the aluminum alloy surface image at each angle; and H m,i , S m,i 、V m,i are the HSV values ​​of the i-th pixel in the aluminum alloy surface image at the m-th angle.

8. The automatic cutting processing system based on aluminum alloy production according to any one of claims 1 to 7, characterized in that: The aluminum alloy cutting parameter calculation model specifically takes maximizing cutting quality as the optimization goal, and its calculation formula is: p i =W (out) H (L) +b (out) H (0) =X=[x1,x2,…,x N ] T x i =[T i ,R i ,E i ,C i ,H i ,S i ,V i ] Among them, max θ J(θ) is the objective function, and θ is the set of all training parameters in the aluminum alloy cutting parameter calculation model, including W (l) , W (out) , b (out) and W k , N is the total number of samples, i is the sample index, K is the number of cutting quality indicators, k is the index of the cutting quality indicator, and w k is the weight of the kth cutting quality index, x i is the input feature vector of the i-th sample, W k is the parameter matrix of the kth cutting quality index, p i is the cutting parameter vector of the i-th sample, d is the dimension of the feature vector, λ is the L2 regularization coefficient, ‖p i ‖2 is the L2 norm of the cutting parameter vector, W (out) is the weight matrix of the output layer of the aluminum alloy cutting parameter calculation model, H (L) is the output of the last layer of the aluminum alloy cutting parameter calculation model, b (out) is the bias vector of the output layer of the aluminum alloy cutting parameter calculation model, l is the index of the aluminum alloy cutting parameter calculation model layer, H (l) is the characteristic matrix of the calculation model of the aluminum alloy cutting parameters at the first layer, W (l) is the weight matrix of the calculation model of the aluminum alloy cutting parameters at the lth layer, is the adjacency matrix after adding self-connection, yes degree matrix, X is the input feature matrix, including the feature vectors of all samples, T i , R i 、E i , C i , H i , S i 、V i are the surface texture feature, surface roughness feature, surface edge feature, surface corner feature, hue value in HSV color space, saturation value in HSV color space and brightness value in HSV color space of the i-th sample, ReLU is the rectified linear unit activation function, and tanh is the hyperbolic tangent function.

Images