A method and system for monitoring and controlling the molding dimensions of gypsum board

Abstract

This invention discloses a method and system for monitoring and controlling the forming dimensions of gypsum board, comprising the following steps: acquiring first defect information; acquiring second defect information; filtering out defect information representing the same size defect in the gypsum board from the first and second defect information as third defect information; acquiring first control information; training a control model based on the first size monitoring data corresponding to the third defect information and the first control information. The control model is used to obtain real-time control information for the gypsum board production line based on real-time size monitoring data during the gypsum board production process. This invention constructs a control model to obtain real-time control information for the gypsum board production line based on real-time size monitoring data during the gypsum board production process, realizing real-time monitoring of gypsum board dimensions during production, and based on the monitoring results, avoiding non-standard gypsum board dimensions in real time, achieving standardized gypsum board production, and reducing defect losses.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent production of gypsum board, and particularly relates to a gypsum board forming size monitoring and regulation method and system. BACKGROUND

[0002] Paper gypsum board may be damaged or have broken corners during production, cutting, and internal transportation in the drying machine, which may cause non-standard sizes in the gypsum board forming process. The gypsum board forming size is mainly determined by the instrument measurement of the quality inspection post personnel after the production of the gypsum board is completed. However, the gypsum board is in a formed product state at this time, so this size monitoring is lagging, and it is difficult to avoid the gypsum board with non-standard sizes. The non-standard size causes irreversible production loss. SUMMARY

[0003] The present application aims to provide a gypsum board forming size monitoring and regulation method and system to solve the technical problem of the lagging size monitoring in the prior art, which is difficult to avoid the gypsum board with non-standard sizes.

[0004] To solve the above technical problems, the present application specifically provides the following technical solutions:

[0005] A gypsum board forming size monitoring and regulation method, comprising the following steps:

[0006] Obtain first defect information, which is obtained by comparing first size monitoring data with standard size data, wherein the first size monitoring data is the size data of the gypsum board before forming obtained during the forming production process of the gypsum board, and the size data includes one or more of the following data: gypsum board appearance image data, gypsum board appearance text data, or gypsum board appearance numerical value data;

[0007] Obtain second defect information, which is obtained by comparing second size monitoring data with standard size data, wherein the second size monitoring data is the size data of the gypsum board after forming obtained after the completion of the forming production of the gypsum board;

[0008] Screen out the defect information representing the same size defect of the gypsum board in the first defect information and the second defect information as third defect information;

[0009] Obtain first regulation information, which is matched from the third defect information according to a defect regulation relationship;

[0010] Train a regulation model according to the first size monitoring data corresponding to the third defect information and the first regulation information, wherein the regulation model is used to obtain real-time regulation information of the gypsum board production line according to real-time size monitoring data during the production of the gypsum board.

[0011] As a preferred scheme of the present application, the first defect information is obtained, comprising:

[0012] The first size monitoring data is subjected to similarity comparison with the standard size data, wherein,

[0013] When the similarity of the first size monitoring data with the standard size data is lower than a preset threshold, the first size monitoring data is assigned a defect label;

[0014] When the similarity of the first size monitoring data with the standard size data is higher than or equal to the preset threshold, the first size monitoring data is assigned a non-defect label;

[0015] The first size monitoring data with the defect label is subjected to defect information recognition by using a pre-established defect information recognition model, to obtain first defect information representing defect information in the first size monitoring data with the defect label;

[0016] The similarity of the first size monitoring data with the standard size data is calculated by using one or more of Euclidean distance, cosine similarity, and correlation coefficient.

[0017] As a preferred scheme of the present application, the second defect information is obtained, comprising:

[0018] The second size monitoring data is subjected to similarity comparison with the standard size data, wherein,

[0019] When the similarity of the second size monitoring data with the standard size data is lower than a preset threshold, the second size monitoring data is assigned a defect label;

[0020] When the similarity of the second size monitoring data with the standard size data is higher than or equal to the preset threshold, the second size monitoring data is assigned a non-defect label;

[0021] The second size monitoring data with the defect label is subjected to defect information recognition by using a pre-established defect information recognition model, to obtain second defect information representing defect information in the second size monitoring data with the defect label;

[0022] The similarity of the second size monitoring data with the standard size data is calculated by using one or more of Euclidean distance, cosine similarity, and correlation coefficient.

[0023] As a preferred scheme of the present application, defect information representing the same size defect of the gypsum board in the first defect information and the second defect information is screened out as third defect information, comprising:

[0024] The first defect information and the second defect information are subjected to similarity comparison, wherein,

[0025] When the similarity of the first defect information and the second defect information is higher than or equal to a preset threshold, it indicates that the first defect information and the second defect information represent the same size defect of the gypsum board;

[0026] When the similarity of the first defect information and the second defect information is lower than the preset threshold, it indicates that the first defect information and the second defect information represent different size defects of the gypsum board;

[0027] The first defect information is extracted from the first defect information and the second defect information representing the same size defect of the gypsum board and marked as third defect information;

[0028] The similarity of the first defect information and the second defect information is calculated by using one or more of Euclidean distance, cosine similarity, and correlation coefficient.

[0029] As a preferred scheme of the present application, the preset threshold in the first defect information, the preset threshold in the second defect information, and the preset threshold in the third defect information are consistent.

[0030] As a preferred scheme of the present application, the pre-establishment of the defect information recognition model comprises:

[0031] A plurality of size monitoring data are obtained, and defect information is marked in each size monitoring data, wherein the size monitoring data comprises one or more of first size monitoring data and second size monitoring data;

[0032] The size monitoring data is used as an input item of a target detection network YOLO V3, the defect information is used as an output item of the target detection network YOLO V3, the input item of the target detection network YOLO V3 and the output item of the target detection network YOLO V3 are learned and trained by using the target detection network YOLO V3, and the defect information recognition model is obtained.

[0033] As a preferred scheme of the present application, the first defect information comprises one or more of the following data: gypsum board forming width defect, gypsum board forming thickness defect, and gypsum board forming corner defect;

[0034] The second defect information comprises one or more of the following data: gypsum board forming width defect, gypsum board forming thickness defect, and gypsum board forming corner defect;

[0035] The third defect information comprises one or more of the following data: gypsum board forming width defect, gypsum board forming thickness defect, and gypsum board forming corner defect;

[0036] The defect control relationship includes one or more of the following matching relationships: the gypsum board forming width defect corresponds to controlling the distance of the forming blade in the gypsum board production line; the gypsum board forming thickness defect corresponds to controlling the height of the forming board in the gypsum board production line; and the gypsum board forming corner defect corresponds to controlling the angle of the forming blade in the gypsum board production line.

[0037] The first control information includes one or more of the following data: controlling the distance of the forming blades in the gypsum board production line, controlling the height of the forming board in the gypsum board production line, and controlling the angle of the forming blades in the gypsum board production line.

[0038] As a preferred embodiment of the present invention, a control model is trained based on the first size monitoring data corresponding to the third defect information and the first control information, including:

[0039] The first size monitoring data corresponding to the third defect information is used as the input of the neural network, and the first regulation information is used as the output of the neural network. The regulation model is obtained by convolution training of the neural network input and output.

[0040] The model expression for the regulation model is:

[0041] H = network(S.data);

[0042] Where H represents control information, S.data represents size monitoring data, and network represents a neural network.

[0043] As a preferred embodiment of the present invention, it further includes: obtaining real-time control information for the gypsum board production line based on real-time size monitoring data during the gypsum board production process.

[0044] The real-time control information of the gypsum board production line obtained based on real-time dimensional monitoring data during the gypsum board production process includes:

[0045] Real-time dimensional monitoring data during the gypsum board production process is input into the control model, and the control model outputs real-time control data for the gypsum board production line.

[0046] The real-time size monitoring data refers to the size data of the gypsum board before molding, which is monitored during the gypsum board molding production process.

[0047] As a preferred embodiment of the present invention, the present invention provides a monitoring and control system applying the above-described gypsum board molding dimension monitoring and control method, comprising:

[0048] The size monitoring unit is installed on the gypsum board production line to monitor real-time size monitoring data during the gypsum board production process. The real-time size monitoring data is the size data of the gypsum board before molding, which is monitored during the gypsum board molding production process.

[0049] Information storage unit, storing control models;

[0050] The control and calculation unit uses the control model to determine the real-time control information of the gypsum board production line based on real-time size monitoring data;

[0051] The control and calculation unit is connected to the molding main control room of the gypsum board production line to feed back real-time control information to the molding main control room of the gypsum board production line to control the distance of the molding blades, the height of the molding board, or the angle of the molding blades in the gypsum board production line, so as to achieve standardized production of gypsum board molding dimensions.

[0052] The control and calculation unit is communicatively connected to the size monitoring unit and the information storage unit to realize data interaction with the size monitoring unit and the information storage unit.

[0053] Compared with the prior art, the present invention has the following advantages:

[0054] This invention constructs a control model to obtain real-time control information for the gypsum board production line based on real-time size monitoring data during the gypsum board production process. This enables real-time monitoring of gypsum board dimensions during production and allows for real-time avoidance of gypsum boards with non-standard dimensions based on the monitoring results, thereby achieving standardized production of gypsum boards and reducing defect losses. Attached Figure Description

[0055] To more clearly illustrate the embodiments of the present invention or the technical solutions in 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 merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0056] Figure 1 A flowchart of the gypsum board forming dimension monitoring and control method provided in the embodiments of the present invention;

[0057] Figure 2 A block diagram of a monitoring and control system provided in an embodiment of the present invention. Detailed Implementation

[0058] 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.

[0059] The dimensions of gypsum board are primarily determined by quality control personnel using instruments after production. However, at this point, the gypsum board is already a finished product, resulting in a lag in dimensional monitoring. This makes it difficult to avoid gypsum board with non-standard dimensions, and the irreversible production losses caused by these non-standard dimensions exacerbate production losses. Therefore, this invention provides a method for monitoring and controlling the dimensional dimensions of gypsum board. It constructs a control model that obtains real-time control information for the gypsum board production line based on real-time dimensional monitoring data during the production process. This model enables real-time control of dimensional non-standard dimensions during gypsum board production, reducing the rate of defective products due to dimensional discrepancies.

[0060] like Figure 1 As shown, the present invention provides a method for monitoring and controlling the molding dimensions of gypsum board, comprising the following steps:

[0061] The first defect information is obtained by comparing the first size monitoring data with the standard size data. The first size monitoring data is the size data of the gypsum board before molding, which is monitored during the gypsum board molding production process. The size data includes one or more of the following: gypsum board appearance image data, gypsum board appearance text data, or gypsum board appearance numerical data.

[0062] The second defect information is obtained by comparing the second dimension monitoring data with the standard dimension data. The second dimension monitoring data is the dimension data of the gypsum board after molding, which is monitored after the gypsum board molding production is completed.

[0063] The defect information that represents the same size defect in the gypsum board from the first and second defect information is selected as the third defect information.

[0064] First control information is obtained, which is matched with third defect information based on defect control relationship;

[0065] The control model is trained based on the first size monitoring data and the first control information corresponding to the third defect information. The control model is used to obtain real-time control information of the gypsum board production line based on real-time size monitoring data during the gypsum board production process.

[0066] To achieve early identification of dimensional defects in gypsum board molding, i.e., to identify dimensional defects in gypsum board molding after production based on dimensional monitoring data during the production process, this invention compares and analyzes the dimensional data of the gypsum board before molding (first dimensional monitoring data) monitored during the gypsum board molding production process and the dimensional data of the gypsum board after molding (second dimensional monitoring data) monitored after the gypsum board molding process. The dimensional defects corresponding to the dimensional defects in the gypsum board before molding are extracted, and the dimensional defect information after molding is compared with the dimensional data before molding. This allows for the early detection of dimensional defects in the gypsum board after molding based on the dimensional data before molding. The specific process of comparing the dimensional defect information after molding with the dimensional data before molding is as follows:

[0067] Obtain first defect information, including:

[0068] The first size monitoring data was compared with the standard size data for similarity.

[0069] If the similarity between the first size monitoring data and the standard size data is lower than a preset threshold, the first size monitoring data will be labeled with a defect.

[0070] If the similarity between the first dimension monitoring data and the standard dimension data is higher than or equal to a preset threshold, the first dimension monitoring data will be assigned a non-defect label.

[0071] The first dimension monitoring data with defect labels is used to identify defect information using a pre-established defect information identification model to obtain the first defect information that represents the defect information in the first dimension monitoring data with defect labels.

[0072] The similarity between the first-dimensional monitoring data and the standard-dimensional data is calculated using one or more of Euclidean distance, cosine similarity, and correlation coefficient.

[0073] Obtain second defect information, including:

[0074] The second-size monitoring data was compared with the standard-size data for similarity.

[0075] If the similarity between the second-dimensional monitoring data and the standard-dimensional data is lower than a preset threshold, the second-dimensional monitoring data will be labeled as defective.

[0076] If the similarity between the second dimension monitoring data and the standard dimension data is higher than or equal to a preset threshold, the second dimension monitoring data will be assigned a non-defect label.

[0077] The second-dimensional monitoring data with defect labels is used to identify defect information using a pre-established defect information identification model, thereby obtaining second defect information that characterizes the defect information in the second-dimensional monitoring data with defect labels.

[0078] The similarity between the second-dimensional monitoring data and the standard-dimensional data is calculated using one or more of Euclidean distance, cosine similarity, and correlation coefficient.

[0079] The defect information representing the same size defect in the gypsum board from the first and second defect information is selected as the third defect information, including:

[0080] The similarity between the first defect information and the second defect information is compared.

[0081] When the similarity between the first defect information and the second defect information is higher than or equal to a preset threshold, it indicates that the first defect information and the second defect information represent defects of the same size in the gypsum board.

[0082] When the similarity between the first defect information and the second defect information is lower than a preset threshold, it indicates that the first defect information and the second defect information represent defects of different sizes in the gypsum board.

[0083] The first defect information is extracted and marked as the third defect information from the first and second defect information that characterize the same size defect in gypsum board.

[0084] The similarity between the first and second defect information is calculated using one or more of Euclidean distance, cosine similarity, and correlation coefficient.

[0085] The preset thresholds obtained from the first defect information, the second defect information, and the third defect information are consistent.

[0086] This invention extracts pre-forming defect information (first defect information) from the dimensional data of gypsum board before molding (first dimensional monitoring data) obtained during the gypsum board molding production process, and extracts post-forming dimensional defect information (second defect information) from the dimensional data of gypsum board after molding (second dimensional monitoring data) obtained after the gypsum board molding production process. The invention also extracts common terms from the pre-forming and post-forming dimensional defect information, indicating that the post-forming dimensional defect information is already present in the pre-forming dimensional defect information, which is contained within the first dimensional monitoring data. Therefore, this invention enables the early detection of post-forming dimensional defect information through the first dimensional monitoring data, achieving predictive monitoring of dimensional defects and avoiding monitoring lag.

[0087] The pre-establishment of the defect information identification model includes:

[0088] Multiple dimensional monitoring data are acquired, and defect information is marked in each dimensional monitoring data. The dimensional monitoring data includes one or more of the first dimensional monitoring data and the second dimensional monitoring data.

[0089] Size monitoring data is used as the input to the YOLO V3 target detection network, and defect information is used as the output of the YOLO V3 target detection network. The YOLO V3 target detection network is used to learn and train the input and output of the YOLO V3 target detection network to obtain the defect information recognition model.

[0090] The first defect information includes one or more of the following data: gypsum board forming width defect, gypsum board forming thickness defect, and gypsum board forming corner defect.

[0091] The second defect information includes one or more of the following data: gypsum board forming width defect, gypsum board forming thickness defect, and gypsum board forming corner defect.

[0092] The third defect information includes one or more of the following data: gypsum board forming width defect, gypsum board forming thickness defect, and gypsum board forming edge and corner defect.

[0093] Defect control relationships include one or more of the following matching relationships: gypsum board forming width defects correspond to controlling the distance of forming blades in the gypsum board production line; gypsum board forming thickness defects correspond to controlling the height of the forming board in the gypsum board production line; and gypsum board forming corner defects correspond to controlling the angle of forming blades in the gypsum board production line.

[0094] The first control information includes one or more of the following data: controlling the distance of the forming blades in the gypsum board production line, controlling the height of the forming board in the gypsum board production line, and controlling the angle of the forming blades in the gypsum board production line.

[0095] To achieve real-time control of gypsum board production line equipment based on pre-forming dimensional data (first dimensional monitoring data), this invention aims to standardize dimensional production and prevent dimensional defects after gypsum board molding. Therefore, this invention converts the pre-forming dimensional data (first dimensional monitoring data) into a correlation relationship between first control information and third defect information. This correlation relationship is then used to train and construct a defect information recognition model. This allows for real-time control based on the post-forming defect information reflected in the pre-forming dimensional data, preventing post-forming defects and further enhancing the predictability of dimensional defect control, avoiding control lag. Specifically:

[0096] A control model is trained based on the first size monitoring data and first control information corresponding to the third defect information, including:

[0097] The first size monitoring data corresponding to the third defect information is used as the input of the neural network, and the first regulation information is used as the output of the neural network. The regulation model is obtained by convolution training of the neural network input and output.

[0098] The model expression for the regulation model is:

[0099] H = network(S.data);

[0100] Where H represents control information, S.data represents size monitoring data, and network represents a neural network.

[0101] This also includes: obtaining real-time control information for the gypsum board production line based on real-time dimensional monitoring data during the gypsum board production process.

[0102] Real-time control information for the gypsum board production line is obtained based on real-time dimensional monitoring data during the gypsum board production process, including:

[0103] Real-time dimensional monitoring data during the gypsum board production process is input into the control model, and the control model outputs real-time control data for the gypsum board production line.

[0104] Real-time dimensional monitoring data refers to the dimensional data of gypsum board before molding, obtained during the gypsum board molding production process.

[0105] like Figure 2 As shown, the present invention provides a monitoring and control system for an applied method of monitoring and controlling the molding dimensions of gypsum board, comprising:

[0106] The size monitoring unit is installed on the gypsum board production line to monitor real-time size monitoring data during the gypsum board production process. The real-time size monitoring data is the size data of the gypsum board before molding, which is monitored during the gypsum board molding production process.

[0107] Information storage unit, storing control models;

[0108] The control and calculation unit uses the control model to determine the real-time control information of the gypsum board production line based on real-time size monitoring data;

[0109] The control and calculation unit is connected to the molding main control room of the gypsum board production line to feed back real-time control information to the molding main control room of the gypsum board production line to control the distance of the molding blades, the height of the molding board, or the angle of the molding blades in the gypsum board production line, so as to achieve standardized production of gypsum board molding dimensions.

[0110] The control and calculation unit is communicatively connected to the size monitoring unit and the information storage unit to realize data interaction with the size monitoring unit and the information storage unit.

[0111] This invention constructs a control model to obtain real-time control information for the gypsum board production line based on real-time size monitoring data during the gypsum board production process. This enables real-time monitoring of gypsum board dimensions during production and allows for real-time avoidance of gypsum boards with non-standard dimensions based on the monitoring results, thereby achieving standardized production of gypsum boards and reducing defect losses.

[0112] The above embodiments are merely exemplary embodiments of this application and are not intended to limit this application. The scope of protection of this application is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to this application within its substance and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of this application.

Claims

1. A method of monitoring and regulating the sizing of a gypsum board, the method comprising: The method comprises the following steps: ​ obtaining first defect information, wherein the first defect information is obtained by comparing first size monitoring data with standard size data, the first size monitoring data is size data of a gypsum board before forming obtained by monitoring during a gypsum board forming production process, and the size data comprises one or more of the following data: gypsum board appearance image data, gypsum board appearance text data, or gypsum board appearance numerical value data; obtaining second defect information, wherein the second defect information is obtained by comparing second size monitoring data with standard size data, the second size monitoring data is size data of a gypsum board after forming obtained by monitoring after a gypsum board forming production process is completed; screening out defect information representing the same size defect of a gypsum board from the first defect information and the second defect information as third defect information; obtaining first regulation information, wherein the first regulation information is matched according to a defect regulation relationship based on the third defect information; training a regulation model according to the first size monitoring data corresponding to the third defect information and the first regulation information, wherein the regulation model is used to obtain real-time regulation information of a gypsum board production line according to real-time size monitoring data during a gypsum board production process.

2. A method of monitoring and regulating the sizing of a gypsum board according to claim 1, characterized in that: The method for obtaining the first defect information comprises: comparing the first size monitoring data with the standard size data in terms of similarity, wherein when the similarity of the first size monitoring data and the standard size data is lower than a preset threshold, the first size monitoring data is assigned a defect label; when the similarity of the first size monitoring data and the standard size data is higher than or equal to the preset threshold, the first size monitoring data is assigned a non-defect label; performing defect information recognition on the first size monitoring data with the defect label by using a pre-established defect information recognition model to obtain first defect information representing the defect information in the first size monitoring data with the defect label; the similarity of the first size monitoring data and the standard size data is calculated by using one or more of the following: Euclidean distance, cosine similarity, and correlation coefficient.

3. A method of monitoring and regulating the sizing of a gypsum board according to claim 2, characterized in that: The method for obtaining the second defect information comprises: comparing the second size monitoring data with the standard size data in terms of similarity, wherein when the similarity of the second size monitoring data and the standard size data is lower than a preset threshold, the second size monitoring data is assigned a defect label; when the similarity of the second size monitoring data and the standard size data is higher than or equal to the preset threshold, the second size monitoring data is assigned a non-defect label; performing defect information recognition on the second size monitoring data with the defect label by using a pre-established defect information recognition model to obtain second defect information representing the defect information in the second size monitoring data with the defect label; the similarity of the second size monitoring data and the standard size data is calculated by using one or more of the following: Euclidean distance, cosine similarity, and correlation coefficient.

4. A method of monitoring and regulating the sizing of a gypsum board according to claim 3, wherein: The method for screening out the third defect information comprises: comparing the first defect information and the second defect information in terms of similarity, wherein When the similarity of the first defect information and the second defect information is higher than or equal to a preset threshold, it indicates that the first defect information and the second defect information represent the same size defect of the gypsum board; When the similarity of the first defect information and the second defect information is lower than the preset threshold, it indicates that the first defect information and the second defect information represent different size defects of the gypsum board; The first defect information is extracted from the first defect information and the second defect information representing the same size defect of the gypsum board and is marked as third defect information; The similarity of the first defect information and the second defect information is calculated by using one or more of Euclidean distance, cosine similarity, and correlation coefficient.

5. A method of monitoring and regulating the sizing of a gypsum board according to claim 4, characterized in that: The preset threshold in the first defect information, the preset threshold in the second defect information, and the preset threshold in the third defect information are consistent.

6. A method of monitoring and regulating the sizing of a gypsum board according to claim 5, wherein: The pre-establishment of the defect information recognition model comprises: Obtaining a plurality of size monitoring data, and marking defect information in each size monitoring data, wherein the size monitoring data comprises one or more of first size monitoring data and second size monitoring data; Taking the size monitoring data as an input item of a target detection network YOLO V3, taking the defect information as an output item of the target detection network YOLO V3, and learning and training the input item of the target detection network YOLO V3 and the output item of the target detection network YOLO V3 by using the target detection network YOLO V3 to obtain the defect information recognition model.

7. A method of monitoring and regulating the sizing of a gypsum board according to claim 6, wherein: The first defect information comprises one or more of the following data: gypsum board forming width defect, gypsum board forming thickness defect, and gypsum board forming corner defect; The second defect information comprises one or more of the following data: gypsum board forming width defect, gypsum board forming thickness defect, and gypsum board forming corner defect; The third defect information comprises one or more of the following data: gypsum board forming width defect, gypsum board forming thickness defect, and gypsum board forming corner defect; The defect regulation relationship comprises one or more of the following matching relationships: the gypsum board forming width defect corresponds to the distance of the forming knife in the gypsum board production line, the gypsum board forming thickness defect corresponds to the height of the forming plate in the gypsum board production line, and the gypsum board forming corner defect corresponds to the angle of the forming knife in the gypsum board production line; The first regulation information comprises one or more of the following data: the distance of the forming knife in the gypsum board production line, the height of the forming plate in the gypsum board production line, and the angle of the forming knife in the gypsum board production line.

8. A method of monitoring and regulating the sizing of a gypsum board according to claim 7, characterized in that: Training the regulation model according to the first size monitoring data corresponding to the third defect information and the first regulation information comprises: Taking the first size monitoring data corresponding to the third defect information as an input item of a neural network, taking the first regulation information as an output item of the neural network, and performing convolution training on the input item of the neural network and the output item of the neural network by using the neural network to obtain the regulation model; The model expression of the regulation model is: H = network(S.data); Wherein, H is the regulation information, S.data is the size monitoring data, and network is the neural network.

9. A method of monitoring and regulating the sizing of a gypsum board according to claim 8, wherein: Further comprising: Real-time regulation information of the gypsum board production line is obtained according to real-time size monitoring data in the gypsum board production process, The real-time regulation information of the gypsum board production line obtained according to real-time size monitoring data in the gypsum board production process comprises: The real-time size monitoring data in the gypsum board production process is input into the regulation model, and real-time regulation data of the gypsum board production line is output by the regulation model; The real-time size monitoring data is size data of the gypsum board before forming obtained by monitoring in the gypsum board forming production process.

10. A monitoring and regulating system applying the monitoring and regulating method of the formed size of the gypsum board according to any one of claims 1 to 9, characterized in that, Comprise: A size monitoring unit is arranged on the gypsum board production line and is used for monitoring real-time size monitoring data in the gypsum board production process, wherein the real-time size monitoring data is size data of the gypsum board before forming obtained by monitoring in the gypsum board forming production process; An information storage unit stores a regulation model; A regulation calculation unit determines real-time regulation information of the gypsum board production line based on the real-time size monitoring data by using the regulation model; The regulation calculation unit is in communication connection with a forming main control room of the gypsum board production line, so as to feed back the real-time regulation information to the forming main control room of the gypsum board production line to regulate the distance of a forming knife in the gypsum board production line, regulate the height of a forming plate in the gypsum board production line or regulate the angle of the forming knife in the gypsum board production line, so as to achieve standardized production of the gypsum board forming size; The regulation calculation unit is in communication connection with the size monitoring unit and the information storage unit respectively, so as to realize data interaction with the size monitoring unit and the information storage unit.

Images