A kind of anti-edge collapse numerical control panel saw and its cutting control method

Abstract

The application discloses a kind of anti-edge collapse numerical control panel saw and its cutting control method, it is related to cutting control technical field, the numerical control panel saw running data and board image data in the numerical control panel area are collected by the monitoring equipment installed in the numerical control panel area in the present application, after collection is completed, by data processing mode, processing is completed, by combination analysis mode, the numerical control panel saw running data and board image data after processing are combined and analyzed, according to the numerical control panel saw historical data after combination analysis, adaptive cutting control model is constructed.Finally, according to adaptive cutting control model, numerical control panel saw is adjusted, and real-time verification monitoring is carried out, improve the real-time of numerical control panel saw control.

Description

Technical Field

[0001] This invention relates to the field of cutting control technology, specifically to a chip-resistant CNC panel saw and its cutting control method. Background Technology

[0002] Traditional CNC panel saws are prone to chipping and burrs on the edges of solid wood or high-density fiberboard, which affects the appearance quality of furniture, and the cutting accuracy is greatly affected by fluctuations in the feed speed. Summary of the Invention

[0003] The purpose of this invention is to provide a chip-resistant CNC panel saw and its cutting control method, which solves the problems existing in the background art.

[0004] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: The present invention provides a CNC panel saw with anti-chipping feature and its cutting control method, specifically including the following steps: S1. Install monitoring equipment in the CNC panel cutting area to collect historical operating data of CNC panel saws and historical image data of panels in the CNC panel cutting area. S2. The collected historical operating data of CNC panel saw and historical image data of board are processed by data processing method to obtain the processed historical operating data of CNC panel saw and historical image data of board. S3. Combine and analyze the processed historical operating data of the CNC panel saw and the historical image data of the board material through data analysis to obtain the combined and analyzed historical data of the CNC panel saw. S4. Based on the historical data of the CNC panel saw after combined analysis, construct an adaptive cutting control model; S5. Adjust the CNC panel saw according to the adaptive cutting control model and perform real-time verification and monitoring.

[0005] Preferably, the step of installing monitoring equipment within the CNC cutting area and collecting historical operating data of the CNC panel saw and historical image data of the board material within the CNC cutting area through the installed monitoring equipment includes the following steps: Install monitoring equipment in the CNC cutting area and set the data collection cycle for the monitoring equipment; Set the data collection cycle set for the monitoring equipment , This indicates the first data acquisition cycle. Indicates the total number of data collection cycles; Based on the set acquisition cycle of the monitoring equipment, historical operating data of CNC panel saws and historical image data of boards are collected within the CNC panel cutting area.

[0006] Preferably, the step of processing the collected historical operating data of the CNC panel saw and the historical image data of the board material through data processing to obtain the processed historical operating data of the CNC panel saw and the historical image data of the board material includes the following steps: S21. The collected historical operating data of the CNC panel saw is processed by data processing method to obtain the processed historical operating data of the CNC panel saw. The historical operating data of the CNC panel saw is set to include feed pressure data, feed speed data, and saw blade rotation speed data; The historical operating data of CNC panel saws is standardized using data standardization methods; The standardized historical operation data of the CNC panel saw is set as the processed historical operation data of the CNC panel saw. S22. The collected historical image data of the board material is processed by image processing to obtain the processed historical image data of the board material.

[0007] Preferably, the step of processing the acquired historical image data of the board material using image processing methods to obtain processed historical image data of the board material includes the following steps: The acquired historical image data of the board material was enhanced by an adaptive local contrast enhancement method to obtain the enhanced historical image data of the board material. The adaptive local contrast enhancement formula is shown below: ; in, Represents the coordinates in the enhanced historical image data of the board material. pixel grayscale values, Represents the coordinates in the historical image data of the board material. pixel grayscale values, For The average gray level of pixels within a local area centered on the target. For The standard deviation of pixel grayscale within a local area centered on the target. This is a contrast limiting parameter. It is a constant; The enhanced historical image data of the board is set to the processed historical image data of the board.

[0008] Preferably, the step of combining and analyzing the processed historical operating data of the CNC panel saw and the historical image data of the board material through data analysis to obtain the combined and analyzed historical data of the CNC panel saw includes the following steps: S31. Arrange the processed CNC panel saw historical operation data and board historical image data in time series and combine them to obtain the combined CNC panel saw historical data. S32. Construct a decision tree based on the combined CNC panel saw historical data, and classify the combined CNC panel saw historical data by constructing the decision tree to obtain the classified CNC panel saw historical data. S33. Collect the scoring data of different experts on the historical data of CNC panel saw after classification, and use the weighted average method to aggregate the scoring data to obtain the scoring data of the historical data of CNC panel saw for each category. S34. Summarize the scoring data of historical data of CNC panel saws of each category and the historical data of CNC panel saws after classification to obtain the historical data of CNC panel saws after combined analysis.

[0009] Preferably, the step of constructing a decision tree based on the combined historical data of CNC panel saws, and classifying the combined historical data of CNC panel saws by constructing the decision tree to obtain the classified historical data of CNC panel saws includes the following steps: S321. Summarize the combined CNC panel saw historical data to obtain a sample set, and set each sample data in the sample set to represent a set of combined CNC panel saw historical data. For the sample data in the sample set, M sample data are randomly selected with replacement, and a decision tree is constructed based on the selected M sample data. The selected M sample data are set as the sample data at the root node of the decision tree. S322. Divide the root node of the decision tree according to the attributes of the sample data at the root node; When the sample data at the root node of the decision tree has Z attributes, one attribute is randomly selected from these Z attributes. Then, a range threshold is set for the selected attribute, and the set range threshold is used as the classification attribute of the node. Each category attribute can only be selected once, and each category will only generate two nodes; S323. Classify each node in the decision tree according to step S322 until each sample data in each node of the decision tree has only one attribute. Then, stop the classification, output the classified nodes, and build a decision tree based on the classified nodes. S324. Input the combined CNC panel saw historical data into the constructed decision tree for classification to obtain the classified CNC panel saw historical data.

[0010] Preferably, the process of constructing an adaptive cutting control model based on the historical data of the CNC panel saw after combined analysis includes the following steps: S41. Construct a cutting control parameter model based on the historical data of the CNC panel saw after combined analysis; S42. Construct a BP neural network and use the combined analysis of the historical data of the CNC panel saw as training samples to train the BP neural network, thus obtaining a trained BP neural network. S43. Based on the trained BP neural network, the adaptive cutting control model is determined by error comparison. The trained BP neural network is compared with the cutting control parameter model to analyze and determine whether the trained BP neural network and the cutting control parameter model meet the error requirements. By collecting historical operating data of CNC panel saws and historical image data of boards within the CNC panel cutting area, inputting them into the trained BP neural network and cutting control parameter model respectively, comparing the output results, and setting an error threshold; If the comparison shows that the trained BP neural network meets the error requirements of the cutting control parameter model, the trained BP neural network is set as the adaptive cutting control model; otherwise, the training parameters are increased and the BP neural network is trained again.

[0011] Preferably, the step of constructing a BP neural network and training the BP neural network using the combined analysis of historical CNC panel saw data as training samples to obtain a trained BP neural network includes the following steps: A BP neural network was constructed, and the historical data of the CNC panel saw after combinatorial analysis was used as training samples to train the BP neural network. The structure of a BP neural network includes: an input layer, hidden layers, and an output layer; The learning algorithms for BP neural networks include: Set the input of the input layer as ,in This represents the historical data of the CNC panel saw after the first input combination. This represents the historical data of the CNC panel saw after the k-th combination of inputs. The output of the output layer is... ,in This represents the historical data of the CNC panel saw after the first predictive analysis. This represents the historical data of the CNC panel saw after the j-th prediction analysis, where v is the weight from the input layer to the hidden layer and w is the weight from the hidden layer to the output layer. The process of a BP neural network includes two stages: forward propagation of the signal and backward propagation of the error. Forward propagation training involves data entering from the input layer, passing through the hidden layer, and finally reaching the output layer. Error backpropagation training involves sequentially adjusting the weights from the hidden layer to the output layer, and the weights from the input layer to the hidden layer. The data forward propagation training steps are as follows: Suppose that a BP neural network has One input node, There are one output node, and the output of the input layer node is equal to its input. Data is passed sequentially from input to output. After receiving the combined CNC panel saw historical data transmitted from the hidden layer, the output layer directly outputs the received combined CNC panel saw historical data to obtain the CNC panel saw historical data after predictive analysis. Set the historical data of the CNC panel saw after the combined analysis as the expected value; Calculate the error between the historical data and expected value of the CNC panel saw after predictive analysis, and set the error threshold between the historical data and expected value of the CNC panel saw after predictive analysis. When the error between the historical data and expected value of the CNC panel saw after predictive analysis is greater than or equal to the set error threshold, adjust the weights from the hidden layer to the output layer and the weights from the input layer to the hidden layer in sequence through the back propagation stage of the error. When the value of the weight adjustment during backpropagation training decreases to less than a set threshold, the BP neural network converges, and a well-trained BP neural network is obtained.

[0012] Preferably, the step of adjusting the CNC panel saw according to the adaptive cutting control model and performing real-time verification and monitoring includes the following steps: Real-time acquisition of CNC panel saw operation data and board image data within the CNC panel cutting area, and input of the real-time acquisition of CNC panel saw operation data and board image data within the CNC panel cutting area into the adaptive cutting control model; The accuracy of the current CNC panel saw cutting predictive control is calculated based on the adaptive cutting control model. Set the accuracy and recall thresholds for cut prediction control; if the calculated accuracy and recall are lower than the set accuracy and recall thresholds, it indicates that the current cut prediction control model is not accurate enough, and the sample set is increased to recalculate the cut prediction control model.

[0013] The beneficial effects of this invention are as follows: (1) This invention improves the real-time performance of CNC panel saw control by installing monitoring equipment within the CNC panel cutting area and collecting CNC panel saw operation data and board image data within the area. After data collection, the data is processed using a data processing method. After processing, the processed CNC panel saw operation data and board image data are combined and analyzed using a combined analysis method. At the same time, an adaptive cutting control model is constructed based on the historical data of the CNC panel saw after combined analysis. Finally, the CNC panel saw is adjusted according to the adaptive cutting control model, and real-time verification and monitoring are performed, which improves the real-time performance of CNC panel saw control.

[0014] (2) The present invention solves the edge chipping problem through an adaptive cutting control model, improves the cutting accuracy by ≥5%, adapts to various thicknesses and materials of plates, and reduces the workload of subsequent grinding processes. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the 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.

[0016] Figure 1 This is a schematic diagram of the anti-chipping CNC panel saw and its cutting control method of the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. 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 are within the scope of protection of the present invention.

[0018] In a specific embodiment of the present invention, Reference Figure 1 As shown, the present invention provides a chip-resistant CNC panel saw and its cutting control method, comprising: S1. Install monitoring equipment in the CNC panel cutting area to collect historical operating data of CNC panel saws and historical image data of panels in the CNC panel cutting area. S2. The collected historical operating data of CNC panel saw and historical image data of board are processed by data processing method to obtain the processed historical operating data of CNC panel saw and historical image data of board. S3. Combine and analyze the processed historical operating data of the CNC panel saw and the historical image data of the board material through data analysis to obtain the combined and analyzed historical data of the CNC panel saw. S4. Based on the historical data of the CNC panel saw after combined analysis, construct an adaptive cutting control model; S5. Adjust the CNC panel saw according to the adaptive cutting control model and perform real-time verification and monitoring.

[0019] Furthermore, referring to Figure 1 As shown, installing monitoring equipment within the CNC panel cutting area and collecting historical operating data of the CNC panel saw and historical image data of the board material within that area using the installed monitoring equipment includes the following steps: Install monitoring equipment in the CNC cutting area and set the data collection cycle for the monitoring equipment; Set the data collection cycle set for the monitoring equipment , This indicates the first data acquisition cycle. Indicates the total number of data collection cycles; Furthermore, based on the set acquisition cycle of the monitoring equipment, historical operating data of the CNC panel saw and historical image data of the board are collected within the CNC panel cutting area; Furthermore, referring to Figure 1 As shown, the collected historical operating data of the CNC panel saw and historical image data of the board are processed using data processing methods to obtain the processed historical operating data of the CNC panel saw and historical image data of the board, including the following steps: S21. The collected historical operating data of the CNC panel saw is processed by data processing method to obtain the processed historical operating data of the CNC panel saw. The historical operating data of the CNC panel saw is set to include feed pressure data, feed speed data, and saw blade rotation speed data; The historical operating data of CNC panel saws is standardized using data standardization methods; The data standardization formula is as follows: ; in, This represents the standardized historical operating data of the CNC panel saw. This represents the collected historical operating data of the CNC panel saw. This represents the minimum value of the collected historical operating data of the CNC panel saw. This represents the maximum value of the historical operating data of the CNC panel saw collected; The standardized historical operation data of the CNC panel saw is set as the processed historical operation data of the CNC panel saw. S22. The collected historical image data of the board material is processed by image processing to obtain the processed historical image data of the board material. The acquired historical image data of the board material was enhanced by an adaptive local contrast enhancement method to obtain the enhanced historical image data of the board material. The adaptive local contrast enhancement formula is shown below: ; in, Represents the coordinates in the enhanced historical image data of the board material. pixel grayscale values, Represents the coordinates in the historical image data of the board material. pixel grayscale values, For The average gray level of pixels within a local area centered on the target. For The standard deviation of pixel grayscale within a local area centered on the target. This is a contrast limiting parameter. It is a constant; Furthermore, the enhanced historical image data of the board material is set as the processed historical image data of the board material; Furthermore, referring to Figure 1 As shown, the historical operating data of the CNC panel saw and the historical image data of the board are combined and analyzed using data analysis methods to obtain the combined analysis of the historical data of the CNC panel saw, which includes the following steps: S31. Arrange the processed CNC panel saw historical operation data and board historical image data in time series and combine them to obtain the combined CNC panel saw historical data. S32. Construct a decision tree based on the combined CNC panel saw historical data, and classify the combined CNC panel saw historical data by constructing the decision tree to obtain the classified CNC panel saw historical data. S321. Summarize the combined CNC panel saw historical data to obtain a sample set, and set each sample data in the sample set to represent a set of combined CNC panel saw historical data. For the sample data in the sample set, M sample data are randomly selected with replacement, and a decision tree is constructed based on the selected M sample data. The selected M sample data are set as the sample data at the root node of the decision tree. S322. Divide the root node of the decision tree according to the attributes of the sample data at the root node; When the sample data at the root node of the decision tree has Z attributes, one attribute is randomly selected from these Z attributes. Then, a range threshold is set for the selected attribute, and the set range threshold is used as the classification attribute of the node. Each category attribute can only be selected once, and each category will only generate two nodes; S323. Classify each node in the decision tree according to step S322 until each sample data in each node of the decision tree has only one attribute. Then, stop the classification, output the classified nodes, and build a decision tree based on the classified nodes. S324. Input the combined CNC panel saw historical data into the constructed decision tree for classification to obtain the classified CNC panel saw historical data. S33. Collect the scoring data of different experts on the historical data of CNC panel saw after classification, and use the weighted average method to aggregate the scoring data to obtain the scoring data of the historical data of CNC panel saw for each category. The weighted average method is shown below: ; in, This represents the expert's score data for the historical data of CNC panel saws after the z-th category. This represents the score data of expert r on the historical data of CNC panel saws after the z-th classification. This represents the weight of expert r. Indicates the total number of experts; S34. Summarize the scoring data of historical data of CNC panel saws in each category and the historical data of CNC panel saws after classification to obtain the historical data of CNC panel saws after combined analysis. Furthermore, referring to Figure 1 As shown, based on the historical data of the CNC panel saw after combined analysis, the adaptive cutting control model is constructed through the following steps: S41. Construct a cutting control parameter model based on the historical data of the CNC panel saw after combined analysis;

[0020] in, This is the scoring data for historical data of CNC panel saws. This indicates the weights of parameters in the historical operating data of the CNC panel saw. Indicates the weights of parameters in the historical image data of the board material; S42. Construct a BP neural network and use the combined analysis of the historical data of the CNC panel saw as training samples to train the BP neural network, thus obtaining a trained BP neural network. A BP neural network was constructed, and the historical data of the CNC panel saw after combinatorial analysis was used as training samples to train the BP neural network. The structure of a BP neural network includes: an input layer, hidden layers, and an output layer; Furthermore, the learning algorithm for a BP neural network includes: Set the input of the input layer as ,in This represents the historical data of the CNC panel saw after the first input combination. This represents the historical data of the CNC panel saw after the k-th combination of inputs. The output of the output layer is... ,in This represents the historical data of the CNC panel saw after the first predictive analysis. This represents the historical data of the CNC panel saw after the j-th prediction analysis, where v is the weight from the input layer to the hidden layer and w is the weight from the hidden layer to the output layer. The process of a BP neural network includes two stages: forward propagation of the signal and backward propagation of the error. Forward propagation training involves data entering from the input layer, passing through the hidden layer, and finally reaching the output layer. Error backpropagation training involves sequentially adjusting the weights from the hidden layer to the output layer, and the weights from the input layer to the hidden layer. The data forward propagation training steps are as follows: Suppose that a BP neural network has One input node, There are one output node, and the output of the input layer node is equal to its input. Data is passed sequentially from input to output. After receiving the combined CNC panel saw historical data transmitted from the hidden layer, the output layer directly outputs the received combined CNC panel saw historical data to obtain the CNC panel saw historical data after predictive analysis. Set the historical data of the CNC panel saw after the combined analysis as the expected value; Furthermore, the error between the historical data and the expected value of the CNC panel saw after the predictive analysis is calculated, and an error threshold between the historical data and the expected value of the CNC panel saw after the predictive analysis is set. When the error between the historical data and the expected value of the CNC panel saw after the predictive analysis is greater than or equal to the set error threshold, the weights from the hidden layer to the output layer and the weights from the input layer to the hidden layer are adjusted sequentially through the back propagation stage of the error. Furthermore, the weights are adjusted through backpropagation training, and the function expression is as follows: ; Where B represents the backpropagation training adjustment weight of the combined CNC panel saw historical data. This represents the hidden layer in a BP neural network. This indicates the number of times the combined CNC panel saw's historical data was trained during the propagation process. Indicates the first This training session This indicates the historical data of the combined CNC panel saw at the node. Backpropagation; When the value of the weight adjustment during backpropagation training decreases to less than a set threshold, the BP neural network converges, and a well-trained BP neural network is obtained. S43. Based on the trained BP neural network, the adaptive cutting control model is determined by error comparison. The trained BP neural network is compared with the cutting control parameter model to analyze and determine whether the trained BP neural network and the cutting control parameter model meet the error requirements. By collecting historical operating data of CNC panel saws and historical image data of boards within the CNC panel cutting area, inputting them into the trained BP neural network and cutting control parameter model respectively, comparing the output results, and setting an error threshold; If the comparison shows that the trained BP neural network meets the error requirements of the cutting control parameter model, the trained BP neural network is set as the adaptive cutting control model; otherwise, the training parameters are increased and the BP neural network is trained again. Furthermore, referring to Figure 1 As shown, adjusting the CNC panel saw according to the adaptive cutting control model and performing real-time verification and monitoring includes the following steps: Real-time acquisition of CNC panel saw operation data and board image data within the CNC panel cutting area, and input of the real-time acquisition of CNC panel saw operation data and board image data within the CNC panel cutting area into the adaptive cutting control model; The accuracy of the current CNC panel saw cutting predictive control is calculated based on the adaptive cutting control model. ; ; Where TP represents the number of predictions that match the actual results; FP represents the number of predictions that do not match the actual results; FN represents the number of missed detections; Q is the precision rate; and R is the recall rate. Set the accuracy and recall thresholds for cut prediction control; if the calculated accuracy and recall are lower than the set accuracy and recall thresholds, it indicates that the current cut prediction control model is not accurate enough, and the sample set size is increased to recalculate the cut prediction control model. In one specific embodiment, the anti-chipping CNC panel saw and its cutting control method further include: a data acquisition module, a data processing module, a data analysis module, an adaptive cutting control module, and a vision inspection module; The data acquisition module is used to collect CNC panel saw operation data and panel image data within the CNC panel cutting area; The data processing module is used to process the collected CNC panel saw operation data and panel image data; The data analysis module is used to analyze the processed CNC panel saw operation data and panel image data; The adaptive cutting control module is used to construct an adaptive cutting control model based on the analyzed CNC panel saw operation data and board image data. The visual inspection module is used to detect the quality of the cut edges in real time.

[0021] It should be noted that the above content is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined by the present invention, they should all fall within the protection scope of the present invention.

Claims

1. A CNC panel saw with anti-chipping feature and its cutting control method, characterized in that, Includes the following steps: S1. Install monitoring equipment in the CNC panel cutting area to collect historical operating data of CNC panel saws and historical image data of panels in the CNC panel cutting area. S2. The collected historical operating data of CNC panel saw and historical image data of board are processed by data processing method to obtain the processed historical operating data of CNC panel saw and historical image data of board. S3. Combine and analyze the processed historical operating data of the CNC panel saw and the historical image data of the board material through data analysis to obtain the combined and analyzed historical data of the CNC panel saw. S4. Based on the historical data of the CNC panel saw after combined analysis, construct an adaptive cutting control model; S5. Adjust the CNC panel saw according to the adaptive cutting control model and perform real-time verification and monitoring.

2. The anti-chipping CNC panel saw and its cutting control method according to claim 1, characterized in that, The process of installing monitoring equipment within the CNC panel cutting area and collecting historical operating data of the CNC panel saw and historical image data of the board material within the CNC panel cutting area using the installed monitoring equipment includes the following steps: Install monitoring equipment in the CNC cutting area and set the data collection cycle for the monitoring equipment; Set the data collection cycle set for the monitoring equipment , This indicates the first data acquisition cycle. Indicates the total number of data collection cycles; Based on the set acquisition cycle of the monitoring equipment, historical operating data of CNC panel saws and historical image data of boards are collected within the CNC panel cutting area.

3. The anti-chipping CNC panel saw and its cutting control method according to claim 1, characterized in that, The process of processing the collected historical operating data of the CNC panel saw and the historical image data of the board material to obtain the processed historical operating data of the CNC panel saw and the historical image data of the board material includes the following steps: S21. The collected historical operating data of the CNC panel saw is processed by data processing method to obtain the processed historical operating data of the CNC panel saw. The historical operating data of the CNC panel saw is set to include feed pressure data, feed speed data, and saw blade rotation speed data; The historical operating data of CNC panel saws is standardized using data standardization methods; The standardized historical operation data of the CNC panel saw is set as the processed historical operation data of the CNC panel saw. S22. The collected historical image data of the board material is processed by image processing to obtain the processed historical image data of the board material.

4. The anti-chipping CNC panel saw and its cutting control method according to claim 3, characterized in that, The process of processing the acquired historical image data of the board material to obtain the processed historical image data of the board material includes the following steps: The acquired historical image data of the board material was enhanced by an adaptive local contrast enhancement method to obtain the enhanced historical image data of the board material. The adaptive local contrast enhancement formula is shown below: ； in, Represents the coordinates in the enhanced historical image data of the board material. pixel grayscale values, Represents the coordinates in the historical image data of the board material. pixel grayscale values, For The average gray level of pixels within a local area centered on the target. For The standard deviation of pixel grayscale within a local area centered on the target. This is a contrast limiting parameter. It is a constant; The enhanced historical image data of the board is set to the processed historical image data of the board.

5. The anti-chipping CNC panel saw and its cutting control method according to claim 1, characterized in that, The process of combining and analyzing the processed historical operating data of the CNC panel saw and the historical image data of the board material to obtain the combined historical data of the CNC panel saw includes the following steps: S31. Arrange the processed CNC panel saw historical operation data and board historical image data in time series and combine them to obtain the combined CNC panel saw historical data. S32. Construct a decision tree based on the combined CNC panel saw historical data, and classify the combined CNC panel saw historical data by constructing the decision tree to obtain the classified CNC panel saw historical data. S33. Collect the scoring data of different experts on the historical data of CNC panel saw after classification, and use the weighted average method to aggregate the scoring data to obtain the scoring data of the historical data of CNC panel saw for each category. S34. Summarize the scoring data of historical data of CNC panel saws of each category and the historical data of CNC panel saws after classification to obtain the historical data of CNC panel saws after combined analysis.

6. The anti-chipping CNC panel saw and its cutting control method according to claim 5, characterized in that, The process of constructing a decision tree based on the combined historical data of CNC panel saws, and classifying the combined historical data of CNC panel saws using the decision tree to obtain the classified historical data of CNC panel saws includes the following steps: S321. Summarize the combined CNC panel saw historical data to obtain a sample set, and set each sample data in the sample set to represent a set of combined CNC panel saw historical data. For the sample data in the sample set, M sample data are randomly selected with replacement, and a decision tree is constructed based on the selected M sample data. The selected M sample data are set as the sample data at the root node of the decision tree. S322. Divide the root node of the decision tree according to the attributes of the sample data at the root node; When the sample data at the root node of the decision tree has Z attributes, one attribute is randomly selected from these Z attributes. Then, a range threshold is set for the selected attribute, and the set range threshold is used as the classification attribute of the node. Each category attribute can only be selected once, and each category will only generate two nodes; S323. Classify each node in the decision tree according to step S322 until each sample data in each node of the decision tree has only one attribute. Then, stop the classification, output the classified nodes, and build a decision tree based on the classified nodes. S324. Input the combined CNC panel saw historical data into the constructed decision tree for classification to obtain the classified CNC panel saw historical data.

7. The anti-chipping CNC panel saw and its cutting control method according to claim 1, characterized in that, The process of constructing an adaptive cutting control model based on the historical data of the CNC panel saw after combined analysis includes the following steps: S41. Construct a cutting control parameter model based on the historical data of the CNC panel saw after combined analysis; S42. Construct a BP neural network and use the combined analysis of the historical data of the CNC panel saw as training samples to train the BP neural network, thus obtaining a trained BP neural network. S43. Based on the trained BP neural network, the adaptive cutting control model is determined by error comparison. The trained BP neural network is compared with the cutting control parameter model to analyze and determine whether the trained BP neural network and the cutting control parameter model meet the error requirements. By collecting historical operating data of CNC panel saws and historical image data of boards within the CNC panel cutting area, inputting them into the trained BP neural network and cutting control parameter model respectively, comparing the output results, and setting an error threshold; If the comparison shows that the trained BP neural network meets the error requirements of the cutting control parameter model, the trained BP neural network is set as the adaptive cutting control model; otherwise, the training parameters are increased and the BP neural network is trained again.

8. The anti-chipping CNC panel saw and its cutting control method according to claim 7, characterized in that, The process of constructing a BP neural network and training it using historical data from a CNC panel saw after combined analysis as training samples to obtain a trained BP neural network includes the following steps: A BP neural network was constructed, and the historical data of the CNC panel saw after combinatorial analysis was used as training samples to train the BP neural network. The structure of a BP neural network includes: an input layer, hidden layers, and an output layer; The learning algorithms for BP neural networks include: Set the input of the input layer as ,in This represents the historical data of the CNC panel saw after the first input combination. This represents the historical data of the CNC panel saw after the k-th combination of inputs. The output of the output layer is... ,in This represents the historical data of the CNC panel saw after the first predictive analysis. This represents the historical data of the CNC panel saw after the j-th prediction analysis, where v is the weight from the input layer to the hidden layer and w is the weight from the hidden layer to the output layer. The process of a BP neural network includes two stages: forward propagation of the signal and backward propagation of the error. Forward propagation training involves data entering from the input layer, passing through the hidden layer, and finally reaching the output layer. Error backpropagation training involves sequentially adjusting the weights from the hidden layer to the output layer, and the weights from the input layer to the hidden layer. The data forward propagation training steps are as follows: Suppose that a BP neural network has One input node, There are one output node, and the output of the input layer node is equal to its input. Data is passed sequentially from input to output. After receiving the combined CNC panel saw historical data transmitted from the hidden layer, the output layer directly outputs the received combined CNC panel saw historical data to obtain the CNC panel saw historical data after predictive analysis. Set the historical data of the CNC panel saw after the combined analysis as the expected value; Calculate the error between the historical data and expected value of the CNC panel saw after predictive analysis, and set the error threshold between the historical data and expected value of the CNC panel saw after predictive analysis. When the error between the historical data and expected value of the CNC panel saw after predictive analysis is greater than or equal to the set error threshold, adjust the weights from the hidden layer to the output layer and the weights from the input layer to the hidden layer in sequence through the back propagation stage of the error. When the value of the weight adjustment during backpropagation training decreases to less than a set threshold, the BP neural network converges, and a well-trained BP neural network is obtained.

9. The anti-chipping CNC panel saw and its cutting control method according to claim 1, characterized in that, The process of adjusting the CNC panel saw according to the adaptive cutting control model and performing real-time verification and monitoring includes the following steps: Real-time acquisition of CNC panel saw operation data and board image data within the CNC panel cutting area, and input of the real-time acquisition of CNC panel saw operation data and board image data within the CNC panel cutting area into the adaptive cutting control model; The accuracy of the current CNC panel saw cutting predictive control is calculated based on the adaptive cutting control model. Set the accuracy and recall thresholds for cut prediction control; if the calculated accuracy and recall are lower than the set accuracy and recall thresholds, it indicates that the current cut prediction control model is not accurate enough, and the sample set is increased to recalculate the cut prediction control model.

Images