Abnormal monitoring method and system for corrugated cardboard production control system

Abstract

The present invention relates to the field of corrugated paper production control technology, specifically to an abnormal monitoring method and system for a corrugated paperboard production control system; identifying the abnormal monitoring position and abnormal monitoring cycle of a corrugated paperboard production system, determining a paperboard area to be monitored; monitoring the paperboard area to be monitored, obtaining a first data set and a second data set; constructing a first abnormal recognition model to identify the first data set, obtaining a first abnormal coefficient; constructing a second abnormal recognition model to identify the second data set, obtaining a second abnormal coefficient; calculating a third abnormal coefficient of the paperboard area to be monitored based on the first abnormal coefficient and the second abnormal coefficient; obtaining multiple third abnormal coefficients of the same abnormal monitoring position and the same production cycle, calculating a comprehensive abnormal coefficient, and giving an early warning based on a comprehensive abnormal threshold. The present invention realizes accurate identification and early warning of abnormal conditions in the paperboard area to be monitored through the comprehensive abnormal coefficient.

Description

Technical Field

[0001] The invention relates to the technical field of corrugated paper production control, and in particular to an abnormality monitoring method and system for a corrugated paperboard production control system. Background Art

[0002] Corrugated cardboard is a multi-layer adhesive body, usually composed of two layers of cardboard and one layer of corrugated paper; the unique structure of corrugated cardboard can effectively absorb impact and vibration, protecting the internal items from damage during transportation; at the same time, the density of corrugated cardboard is low, which can reduce the total weight during transportation and handling, thereby reducing transportation costs; therefore, it is widely used in food packaging, electronic product packaging, industrial packaging and other fields.

[0003] The production process of corrugated cardboard includes steps such as raw material preparation, corrugated paper production, cardboard gluing, drying, cutting and packaging; among them, corrugated paper production and cardboard gluing are important links in the production process of corrugated cardboard.

[0004] In the corrugated paper production process, the shape and structure of the corrugated paper are important factors affecting the performance of the corrugated paperboard; if the shape of the corrugated paper is uneven or irregular, it will affect the compressive strength of the paperboard, causing the paperboard to deform or break during use; in the paperboard bonding process, if the glue is unevenly applied or the amount is insufficient, the paperboard and the corrugated paper will not be firmly bonded, reducing the strength of the paperboard and affecting the durability and stability of the paperboard. Therefore, it is necessary to monitor and warn the corrugated paperboard in production.

[0005] The existing technical solutions mainly collect data from various processing equipment on the corrugated cardboard production line and monitor abnormalities based on the data; not much attention is paid to the bonding conditions and shape of corrugated paper and cardboard during the production process, resulting in inaccurate abnormal monitoring and early warning in the corrugated cardboard production process.

[0006] Therefore, an abnormality monitoring method and system for a corrugated board production control system are proposed. Summary of the invention

[0007] The object of the present invention is to provide an abnormality monitoring method and system for a corrugated cardboard production control system, which determines a cardboard area to be monitored through an abnormality monitoring position and an abnormality monitoring period of a corrugated cardboard production system; monitors the cardboard area to be monitored to obtain a first data set and a second data set; constructs a first abnormality recognition model to identify the first data set to obtain a first abnormality coefficient; constructs a second abnormality recognition model to identify the second data set to obtain a second abnormality coefficient; calculates a third abnormality coefficient of the cardboard area to be monitored based on the first abnormality coefficient and the second abnormality coefficient; obtains multiple third abnormality coefficients of the same abnormal monitoring position and the same production period, calculates a comprehensive abnormality coefficient, and issues an early warning based on the comprehensive abnormality threshold, thereby realizing accurate identification and early warning of abnormal conditions in the cardboard area to be monitored.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] A method for monitoring abnormality of a corrugated paperboard production control system, comprising:

[0010] S10. Obtain the production process and production cycle of the corrugated board production system, determine the abnormal monitoring position according to the production process, and determine the abnormal monitoring cycle according to the production cycle;

[0011] S20. Determine the cardboard area to be monitored according to the abnormal monitoring position and the abnormal monitoring period; acquire data of the cardboard area to be monitored to obtain a first data set and a second data set; the first data set includes a first panel image, a first panel infrared image, a second panel image and a second panel infrared image; the second data set includes a first interlayer image and a second interlayer image;

[0012] S30. Construct a first anomaly recognition model to recognize the first data set and obtain a first anomaly coefficient; the first anomaly recognition model includes an image anomaly recognition unit, a temperature anomaly recognition unit and a first anomaly fusion unit;

[0013] S40. Construct a second anomaly recognition model to recognize the second data set to obtain a second anomaly coefficient; the second anomaly recognition model includes a first interlayer recognition unit, a second interlayer recognition unit, a bonding point recognition unit, and a second anomaly fusion unit;

[0014] S50. Calculate the third abnormality coefficient of the cardboard area to be monitored based on the first abnormality coefficient and the second abnormality coefficient;

[0015] S60. Obtain multiple third abnormality coefficients of the same abnormal monitoring position and the same production cycle, calculate the comprehensive abnormality coefficient, and issue an early warning based on the comprehensive abnormality threshold.

[0016] The process of determining the cardboard area to be monitored based on the abnormal monitoring position and abnormal monitoring cycle is as follows:

[0017] The abnormal monitoring cycle includes an abnormal monitoring time and an abnormal monitoring duration; at the abnormal monitoring time, the corrugated cardboard passing through the abnormal monitoring position is marked to obtain a first cross section, and after the abnormal monitoring duration has passed, the corrugated cardboard passing through the abnormal monitoring position is marked to obtain a second cross section;

[0018] The corrugated cardboard area between the first cross section and the second cross section is used as the cardboard area to be monitored;

[0019] The paperboard area to be monitored includes a first panel, a second panel, a first interlayer area and a second interlayer area.

[0020] The first anomaly recognition model includes an image anomaly recognition unit, a temperature anomaly recognition unit and a first anomaly fusion unit;

[0021] The image abnormality recognition unit recognizes the first panel image and the second panel image by color features, texture features and shape features, and determines the corresponding first image abnormal area and second image abnormal area;

[0022] The temperature anomaly recognition unit recognizes the first panel infrared image and the second panel infrared image by temperature distribution to obtain corresponding first temperature anomaly area and second temperature anomaly area;

[0023] The first abnormal fusion unit determines a first abnormal coefficient according to the first image abnormal area, the second image abnormal area, the first temperature abnormal area, and the second temperature abnormal area.

[0024] The second anomaly recognition model includes a first interlayer recognition unit, a second interlayer recognition unit, an adhesion point recognition unit, and a second anomaly fusion unit;

[0025] The first interlayer recognition unit recognizes the corrugated paper shape in the first interlayer image and the second interlayer image to obtain a first corrugated paper abnormality coefficient and a second corrugated paper abnormality coefficient; the first interlayer abnormality coefficient is calculated according to the first corrugated paper abnormality coefficient and the second corrugated paper abnormality coefficient;

[0026] The second interlayer recognition unit recognizes the difference between the shapes of the corrugated paper in the first interlayer image and the second interlayer image to obtain a second interlayer abnormality coefficient;

[0027] The bonding point identification unit identifies the bonding point between the corrugated paper and the panel to obtain a bonding degree coefficient and a bonding position coefficient; the bonding abnormality coefficient is obtained by identifying the bonding degree coefficient and the bonding position coefficient;

[0028] The second abnormal fusion unit calculates the second abnormal coefficient according to the first interlayer abnormal coefficient, the second interlayer abnormal coefficient and the bonding abnormal coefficient.

[0029] The recognition process of the first interlayer recognition unit is:

[0030] Acquire a standard corrugated paper shape of corrugated paper; recognize the corrugated paper shape in the first sandwich image to acquire a first corrugated paper shape; recognize the corrugated paper shape in the second sandwich image to acquire a second corrugated paper shape;

[0031] According to the difference between the first corrugated paper shape and the standard corrugated paper shape, a first corrugated paper anomaly coefficient is obtained; according to the difference between the second corrugated paper shape and the standard corrugated paper shape, a second corrugated paper anomaly coefficient is obtained;

[0032] The first interlayer abnormal coefficient is calculated according to the first corrugated paper abnormal coefficient and the second corrugated paper abnormal coefficient.

[0033] The bonding point identification unit includes a bonding degree identification layer and a bonding position identification layer;

[0034] The bonding degree identification layer is used to identify the bonding length of the bonding point between the corrugated paper and the panel, and obtain the bonding degree coefficient through the bonding length and the standard bonding length; the bonding position identification layer is used to obtain the coordinates of the midpoint of the bonding length as the bonding point coordinates; the bonding position coefficient is obtained by the difference between the bonding point coordinates and the standard coordinates; the point coordinates are obtained by the first coordinate system and the second coordinate system constructed respectively based on the first interlayer image and the second interlayer image.

[0035] The process of calculating the comprehensive abnormality coefficient according to the first abnormality coefficient and the second abnormality coefficient is as follows:

[0036] Collect and obtain the first abnormal coefficient and the second abnormal coefficient of the same production cycle at the same abnormal monitoring position;

[0037] A third abnormal coefficient is calculated based on the corresponding first abnormal coefficient and second abnormal coefficient, and a third abnormal coefficient sequence is obtained by arranging them in chronological order; in the third abnormal coefficient sequence, the third abnormal coefficient is weighted by the change rate of the third abnormal coefficient, and a comprehensive abnormal coefficient is calculated; a comprehensive abnormal threshold is set to judge the comprehensive abnormal coefficient, and when the comprehensive abnormal threshold is exceeded, a warning message is issued.

[0038] An abnormality monitoring system for a corrugated board production control system, comprising:

[0039] A process determination module, which obtains the production process and production cycle of the corrugated board production system, determines the abnormality monitoring position according to the production process, and determines the abnormality monitoring cycle according to the production cycle;

[0040] A data acquisition module, which determines a cardboard area to be monitored according to the abnormal monitoring position and the abnormal monitoring period; acquires data on the cardboard area to be monitored to obtain a first data set and a second data set; the first data set includes a first panel image, a first panel infrared image, a second panel image, and a second panel infrared image; the second data set includes a first interlayer image and a second interlayer image;

[0041] A first recognition module constructs a first anomaly recognition model to recognize the first data set and obtain a first anomaly coefficient; the first anomaly recognition model includes an image anomaly recognition unit, a temperature anomaly recognition unit and a first anomaly fusion unit;

[0042] A second recognition module constructs a second anomaly recognition model to recognize the second data set to obtain a second anomaly coefficient; the second anomaly recognition model includes a first interlayer recognition unit, a second interlayer recognition unit, a bonding point recognition unit, and a second anomaly fusion unit;

[0043] An abnormality identification module, which calculates a third abnormality coefficient according to the first abnormality coefficient and the second abnormality coefficient;

[0044] The abnormal warning module obtains multiple third abnormal coefficients of the same abnormal monitoring position and the same production cycle, calculates the comprehensive abnormal coefficient, and issues an early warning based on the comprehensive abnormal threshold.

[0045] The process of determining the cardboard area to be monitored based on the abnormal monitoring position and abnormal monitoring cycle is as follows:

[0046] The abnormal monitoring cycle includes an abnormal monitoring time and an abnormal monitoring duration; at the abnormal monitoring time, the corrugated cardboard passing through the abnormal monitoring position is marked to obtain a first cross section, and after the abnormal monitoring duration has passed, the corrugated cardboard passing through the abnormal monitoring position is marked to obtain a second cross section;

[0047] The corrugated cardboard area between the first cross section and the second cross section is used as the cardboard area to be monitored;

[0048] The paperboard area to be monitored includes a first panel, a second panel, a first interlayer area and a second interlayer area.

[0049] The process of calculating the comprehensive abnormality coefficient according to the first abnormality coefficient and the second abnormality coefficient is as follows:

[0050] The first abnormality coefficient and the second abnormality coefficient of the same production cycle at the same abnormal monitoring position are collected; the third abnormality coefficient is calculated based on the corresponding first abnormality coefficient and the second abnormality coefficient, and the third abnormality coefficient sequence is obtained by arranging them in chronological order; in the third abnormality coefficient sequence, the third abnormality coefficient is weighted by the change rate of the third abnormality coefficient to calculate the comprehensive abnormality coefficient; a comprehensive abnormality threshold is set to judge the comprehensive abnormality coefficient, and when the comprehensive abnormality threshold is exceeded, an early warning message is issued.

[0051] Compared with the prior art, the present invention has the following beneficial effects:

[0052] 1. The present invention constructs a first abnormality recognition model to recognize a first data set; wherein, an image abnormality area of ​​a panel is recognized by an image abnormality recognition unit, and an abnormal temperature area of ​​the panel is recognized by a temperature abnormality recognition unit, and a first abnormality coefficient is obtained according to the recognition of the image abnormality area and the temperature abnormality area; the panel abnormality in the monitored area is accurately recognized by the first abnormality coefficient.

[0053] 2. The present invention constructs a second anomaly recognition model to recognize the second data set; wherein, the first interlayer anomaly recognition unit is used to recognize the shape anomaly of the corrugated paper to obtain the first interlayer anomaly coefficient; the second interlayer anomaly recognition unit is used to recognize the difference in the shape of the corrugated paper between the first interlayer image and the second interlayer image to obtain the second interlayer anomaly coefficient; the bonding point anomaly of the corrugated paper and the panel is recognized by the bonding point recognition unit to obtain the bonding anomaly coefficient; the second anomaly coefficient is obtained by comprehensively measuring the anomaly of the corrugated paper; the abnormality of the corrugated paper is accurately recognized by the second anomaly coefficient.

[0054] 3. The present invention obtains a third abnormal coefficient by measuring the first abnormal coefficient and the second abnormal coefficient, and obtains a comprehensive abnormal coefficient by weighting the third abnormal coefficient according to the rate of change of the third abnormal coefficient; a threshold value is judged according to the comprehensive abnormal coefficient, and an abnormal warning is issued according to the judgment result, so as to accurately reflect the abnormal situation of corrugated paper production. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is a flow chart of an abnormality monitoring method of a corrugated board production control system of the present invention;

[0056] Figure 2 It is a structural schematic diagram of the cardboard area to be monitored of the present invention;

[0057] Figure 3 is a structural schematic diagram of a first abnormality recognition model of the present invention;

[0058] Figure 4 is a structural schematic diagram of a second abnormality recognition model of the present invention;

[0059] Figure 5 is a schematic structural diagram of the first interlayer region of the present invention;

[0060] Figure 6 A schematic diagram for constructing a first coordinate system of the present invention;

[0061] Figure 7 It is a structural schematic diagram of an abnormality monitoring system of a corrugated board production control system of the present invention;

[0062] Figure 2 100, first cross section; 101, first panel; 102, second panel; 103, first interlayer region; 104, second interlayer region; Figure 5 103, a first interlayer region; 105, a first corrugated paper shape; Figure 6 106, x-axis of the first coordinate system; 107, y-axis of the first coordinate system; 108, origin of the first coordinate system. DETAILED DESCRIPTION

[0063] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0064] In order to better monitor and warn of abnormal situations in the production process of corrugated paperboard, the present invention proposes an abnormality monitoring method and system for a corrugated paperboard production control system.

[0065] Embodiment 1

[0066] The present invention proposes an abnormality monitoring method for a corrugated board production control system, the process of which is as follows: Figure 1 As shown, including:

[0067] S10. Obtain the production process and production cycle of the corrugated board production system, determine the abnormal monitoring position according to the production process, and determine the abnormal monitoring cycle according to the production cycle;

[0068] The process of determining the cardboard area to be monitored based on the abnormal monitoring position and abnormal monitoring cycle is as follows:

[0069] The abnormal monitoring cycle includes an abnormal monitoring time and an abnormal monitoring duration; at the abnormal monitoring time, the corrugated cardboard passing through the abnormal monitoring position is marked to obtain a first cross section, and after the abnormal monitoring duration has passed, the corrugated cardboard passing through the abnormal monitoring position is marked to obtain a second cross section;

[0070] The corrugated cardboard area between the first cross section and the cross section is used as the cardboard area to be monitored; the cardboard area to be monitored includes a first panel, a second panel, a first interlayer area, and a second interlayer area.

[0071] The cardboard area to be monitored is obtained as Figure 2 As shown, 100 represents a first section, 101 represents a first panel, 102 represents a second panel, 103 represents a first sandwich region, and 104 represents a second sandwich region. The first section is perpendicular to the plane where the first panel, the second panel and the first sandwich region are located.

[0072] The present invention determines an abnormal monitoring position according to a production process, determines an abnormal monitoring period according to a production cycle, and the abnormal monitoring period includes an abnormal monitoring time and an abnormal monitoring duration; determines a cardboard area to be monitored according to the position to be monitored and the abnormal monitoring period, and obtains a first panel, a second panel, a first interlayer area, and a second interlayer area of ​​the cardboard area to be monitored, providing a basis for subsequent data acquisition and early warning.

[0073] S20. Determine a cardboard area to be monitored according to the abnormal monitoring position and the abnormal monitoring period; acquire data of the cardboard area to be monitored to obtain a first data set and a second data set; the first data set includes a first panel image, a first panel infrared image, a second panel image, and a second panel infrared image; the second data set includes a first interlayer image and a second interlayer image.

[0074] S30. Construct a first anomaly recognition model to identify the first data set and obtain a first anomaly coefficient; the first anomaly recognition model includes an image anomaly recognition unit, a temperature anomaly recognition unit and a first anomaly fusion unit.

[0075] The first abnormality recognition model is constructed based on a convolutional neural network model, and its structure is as follows: Figure 3 As shown, it includes an image anomaly recognition unit, a temperature anomaly recognition unit and a first anomaly fusion unit;

[0076] The image abnormality recognition unit recognizes the first panel image and the second panel image by color features, texture features and shape features, and determines the corresponding first image abnormal area and second image abnormal area; the first image abnormal area is the abnormal area in the first panel image, and the second image abnormal area is the abnormal area in the second panel image;

[0077] The temperature anomaly recognition unit recognizes the first panel infrared image and the second panel infrared image by temperature distribution to obtain corresponding first temperature anomaly area and second temperature anomaly area; the first temperature anomaly area is an abnormal area in the first panel infrared image, and the second temperature anomaly area is an abnormal area in the second panel infrared image;

[0078] The first abnormal fusion unit determines a first abnormal coefficient according to the first image abnormal area, the second image abnormal area, the first temperature abnormal area, and the second temperature abnormal area.

[0079] The calculation formula of the first abnormal coefficient is:

[0080]

[0081] represents the first anomaly coefficient; Indicates The area of ​​the abnormal region in the image; Indicates The area of ​​the temperature anomaly region; represents the image abnormality weight; represents the temperature anomaly weight; Indicates The area of ​​the panel.

[0082] The present invention constructs a first abnormality recognition model, including an image abnormality recognition unit, a temperature abnormality recognition unit and a first abnormality fusion unit; the image abnormality area of ​​the panel is identified by the image abnormality recognition unit, and the temperature abnormality area of ​​the panel is identified by the temperature abnormality recognition unit. The first abnormality coefficient is obtained according to the identification of the image abnormality area and the temperature abnormality area, so as to accurately identify the panel abnormality in the monitored area.

[0083] S40. Construct a second anomaly recognition model to recognize the second data set and obtain a second anomaly coefficient; the second anomaly recognition model includes a first interlayer recognition unit, a second interlayer recognition unit, an adhesion point recognition unit and a second anomaly fusion unit.

[0084] The second abnormality recognition model is constructed based on a convolutional neural network model, and its structure is as follows: Figure 4 As shown, it includes a first interlayer identification unit, a second interlayer identification unit, an adhesion point identification unit, and a second abnormal fusion unit;

[0085] The first interlayer recognition unit recognizes the corrugated paper shape in the first interlayer image and the second interlayer image to obtain a first corrugated paper abnormality coefficient and a second corrugated paper abnormality coefficient; the first interlayer abnormality coefficient is calculated according to the first corrugated paper abnormality coefficient and the second corrugated paper abnormality coefficient;

[0086] The recognition process of the first interlayer recognition unit is:

[0087] Obtain a standard corrugated paper shape of corrugated paper; identify the corrugated paper shape in the first sandwich image to obtain a first corrugated paper shape; Figure 5 As shown in the figure, 103 represents the first interlayer area, and 105 represents the first corrugated paper shape; the corrugated paper shape in the second interlayer image is identified to obtain the second corrugated paper shape; according to the difference between the first corrugated paper shape and the standard corrugated paper shape, the first corrugated paper anomaly coefficient is obtained; according to the difference between the second corrugated paper shape and the standard corrugated paper shape, the second corrugated paper anomaly coefficient is obtained; the first interlayer anomaly coefficient is calculated based on the first corrugated paper anomaly coefficient and the second corrugated paper anomaly coefficient.

[0088] The calculation formula of the first interlayer anomaly coefficient is:

[0089]

[0090] in, represents the first interlayer anomaly coefficient; It indicates the abnormal coefficient of the first corrugated paper; Indicates the abnormal coefficient of the second corrugated paper.

[0091] The present invention identifies a first interlayer image and a second interlayer image to obtain a first corrugated paper shape and a second corrugated paper shape; obtains a first corrugated paper anomaly coefficient and a second corrugated paper anomaly coefficient according to the difference between the first corrugated paper shape and the second corrugated paper shape and a standard corrugated paper shape, and further calculates to obtain a first interlayer coefficient; and accurately identifies the abnormal situation of the corrugated paper shape in the interlayer image.

[0092] The second interlayer recognition unit recognizes the difference between the shapes of the corrugated paper in the first interlayer image and the second interlayer image to obtain a second interlayer abnormality coefficient;

[0093] The calculation formula of the second interlayer anomaly coefficient is:

[0094]

[0095] in, represents the anomaly coefficient of the second interlayer; Represents the similarity measurement function; Data representing a first corrugated paper shape; Data representing the second corrugated paper shape.

[0096] The bonding point identification unit identifies the bonding points between the corrugated paper and the panel, and includes a bonding degree identification layer and a bonding position identification layer;

[0097] The bonding degree identification layer identifies the bonding length of the bonding point between the corrugated paper and the panel, and obtains the bonding degree coefficient through the bonding length and the standard bonding length;

[0098] The bonding position identification layer is used to obtain the coordinates of the midpoint of the bonding length as the bonding point coordinates; the bonding position coefficient is obtained by the difference between the bonding point coordinates and the standard coordinates, and the bonding point coordinates are obtained by the first coordinate system and the second coordinate system constructed based on the first interlayer image and the second interlayer image respectively; the standard coordinates are obtained by the standard corrugated paper shape and the first coordinate system and the second coordinate system.

[0099] The construction process of the first coordinate system is as follows Figure 6 As shown: obtain a first sandwich image; use the projection of the first panel in the first sandwich image as the x-axis, as shown Figure 6 As shown in 106; the projection of the first section in the first sandwich image is used as the y-axis, as shown in Figure 6 As shown in 107; the intersection of the projection of the first panel in the first sandwich image and the projection of the first section in the first sandwich image is used as the origin of the first coordinate system, as shown in Figure 6 As shown in Figure 108.

[0100] The bonding abnormality coefficient is obtained by identifying the bonding degree coefficient and the bonding position coefficient; the calculation formula of the bonding abnormality coefficient is:

[0101]

[0102]

[0103]

[0104] in, represents the bonding anomaly coefficient; Indicates bonding point The adhesion coefficient; Indicates bonding point Bond length; Indicates standard bonding length; represents the first bonding weight; Indicates bonding point The bonding position coefficient; Indicates bonding point Distance from standard bonding point; represents the distance threshold; represents the second bonding weight.

[0105] The present invention identifies the bonding point between the corrugated paper and the panel in the first interlayer image and the second interlayer image to obtain the bonding length; obtains the midpoint of the bonding length as the bonding point coordinates; obtains the bonding degree coefficient according to the bonding length and the standard bonding length, and obtains the bonding position coefficient according to the difference between the bonding point coordinates and the standard coordinates; obtains the bonding abnormality coefficient according to the bonding degree coefficient and the bonding position coefficient; and accurately identifies the bonding condition of the corrugated paperboard.

[0106] The second abnormal fusion unit calculates the second abnormal coefficient according to the first interlayer abnormal coefficient, the second interlayer abnormal coefficient and the bonding abnormal coefficient.

[0107] The calculation formula of the second abnormal coefficient is:

[0108]

[0109] in, represents the second anomaly coefficient; Represents an exponential function with a natural constant as its base.

[0110] The present invention constructs a second abnormality recognition model, including a first interlayer abnormality recognition unit, a second interlayer abnormality recognition unit, an adhesive point recognition unit and a second abnormality fusion unit; the first interlayer abnormality recognition unit is used to recognize the abnormality of the shape of corrugated paper in the interlayer image, the second interlayer abnormality recognition unit is used to recognize the difference degree of the corrugated paper shape between the first interlayer image and the second interlayer image, and the adhesive point recognition unit is used to recognize the abnormality of the adhesive point between the corrugated paper and the panel; a second abnormality coefficient is obtained by comprehensively measuring the abnormal situation; the abnormal situation of the corrugated paper is accurately recognized by the second abnormality coefficient.

[0111] S50. Calculate a third abnormality coefficient of the cardboard area to be monitored based on the first abnormality coefficient and the second abnormality coefficient.

[0112] S60. Obtain multiple third abnormality coefficients of the same abnormal monitoring position and the same production cycle, calculate the comprehensive abnormality coefficient, and issue an early warning based on the comprehensive abnormality threshold.

[0113] The process of calculating the comprehensive abnormality coefficient according to the first abnormality coefficient and the second abnormality coefficient is as follows:

[0114] The first abnormality coefficient and the second abnormality coefficient of the same production cycle at the same abnormal monitoring position are collected; the third abnormality coefficient is calculated based on the corresponding first abnormality coefficient and the second abnormality coefficient, and the third abnormality coefficient sequence is obtained by arranging them in chronological order; in the third abnormality coefficient sequence, the third abnormality coefficient is weighted by the change rate of the third abnormality coefficient, and the abnormal warning coefficient is calculated; an abnormal threshold is set to judge the abnormal warning coefficient, and when the abnormal threshold is exceeded, a warning message is issued.

[0115] The calculation formula of the comprehensive anomaly coefficient is:

[0116]

[0117]

[0118] in, Indicates Comprehensive abnormality coefficient of an abnormal monitoring cycle; Indicates The third anomaly coefficient of an anomaly monitoring cycle; represents the first anomaly weight; Indicates The first abnormality coefficient of an abnormal monitoring cycle; represents the second anomaly weight; Indicates The second abnormality coefficient of an abnormal monitoring cycle.

[0119] The present invention obtains a third abnormal coefficient by measuring the first abnormal coefficient and the second abnormal coefficient, and obtains a comprehensive abnormal coefficient by weighting the third abnormal coefficient according to the change rate of the third abnormal coefficient; a threshold value is judged according to the comprehensive abnormal coefficient, and an abnormal warning is issued according to the judgment result, so as to accurately reflect the abnormal situation of corrugated paper production.

[0120] Figure 2 The figure shows three layers of corrugated paper, namely, a first panel, a second panel and a layer of corrugated paper; for other specifications of corrugated paper, the outermost panels are regarded as the first panel and the second panel, and the area between the first panel and the second panel is regarded as the interlayer area, and the method described in the present invention can also be used to identify abnormalities and conduct production warnings.

[0121] The present invention determines a cardboard area to be monitored through an abnormal monitoring position and an abnormal monitoring cycle of a corrugated cardboard production system; monitors the cardboard area to be monitored to obtain a first data set and a second data set; constructs a first abnormal recognition model to identify the first data set to obtain a first abnormal coefficient; constructs a second abnormal recognition model to identify the second data set to obtain a second abnormal coefficient; calculates a third abnormal coefficient of the cardboard area to be monitored according to the first abnormal coefficient and the second abnormal coefficient; obtains multiple third abnormal coefficients of the same abnormal monitoring position and the same production cycle, calculates a comprehensive abnormal coefficient, and issues an early warning according to a comprehensive abnormal threshold, thereby realizing accurate identification and early warning of abnormal conditions in the cardboard area to be monitored.

[0122] Embodiment 2

[0123] The present invention also proposes an abnormality monitoring system for a corrugated board production control system, such as Figure 7 As shown, including:

[0124] The process determination module obtains the production process and production cycle of the corrugated cardboard production system, determines the abnormal monitoring position according to the production process, and determines the abnormal monitoring cycle according to the production cycle.

[0125] The data acquisition module determines the cardboard area to be monitored according to the abnormal monitoring position and the abnormal monitoring period; acquires data on the cardboard area to be monitored to obtain a first data set and a second data set; the first data set includes a first panel image, a first panel infrared image, a second panel image and a second panel infrared image; the second data set includes a first interlayer image and a second interlayer image.

[0126] The process of determining the cardboard area to be monitored based on the abnormal monitoring position and abnormal monitoring cycle is as follows:

[0127] The abnormal monitoring cycle includes an abnormal monitoring time and an abnormal monitoring duration; at the abnormal monitoring time, the corrugated cardboard passing through the abnormal monitoring position is marked to obtain a first cross section, and after the abnormal monitoring duration has passed, the corrugated cardboard passing through the abnormal monitoring position is marked to obtain a second cross section;

[0128] The corrugated cardboard area between the first cross section and the second cross section is used as the cardboard area to be monitored;

[0129] The paperboard area to be monitored includes a first panel, a second panel, a first interlayer area and a second interlayer area.

[0130] A first recognition module constructs a first anomaly recognition model to recognize the first data set and obtain a first anomaly coefficient; the first anomaly recognition model includes an image anomaly recognition unit, a temperature anomaly recognition unit and a first anomaly fusion unit;

[0131] The first anomaly recognition model includes an image anomaly recognition unit, a temperature anomaly recognition unit and a first anomaly fusion unit;

[0132] The image abnormality recognition unit recognizes the first panel image and the second panel image by color features, texture features and shape features, and determines the corresponding first image abnormal area and second image abnormal area;

[0133] The temperature anomaly recognition unit recognizes the first panel infrared image and the second panel infrared image by temperature distribution to obtain corresponding first temperature anomaly area and second temperature anomaly area;

[0134] The first abnormal fusion unit determines a first abnormal coefficient according to the first image abnormal area, the second image abnormal area, the first temperature abnormal area, and the second temperature abnormal area.

[0135] A second recognition module constructs a second anomaly recognition model to recognize the second data set to obtain a second anomaly coefficient; the second anomaly recognition model includes a first interlayer recognition unit, a second interlayer recognition unit, a bonding point recognition unit, and a second anomaly fusion unit;

[0136] The second anomaly recognition model includes a first interlayer recognition unit, a second interlayer recognition unit, an adhesion point recognition unit, and a second anomaly fusion unit;

[0137] The first interlayer recognition unit recognizes the corrugated paper shape in the first interlayer image and the second interlayer image to obtain a first corrugated paper abnormality coefficient and a second corrugated paper abnormality coefficient; the first interlayer abnormality coefficient is calculated according to the first corrugated paper abnormality coefficient and the second corrugated paper abnormality coefficient;

[0138] The second interlayer recognition unit recognizes the difference between the shapes of the corrugated paper in the first interlayer image and the second interlayer image to obtain a second interlayer abnormality coefficient;

[0139] The bonding point identification unit identifies the bonding point between the corrugated paper and the panel to obtain a bonding degree coefficient and a bonding position coefficient; the bonding abnormality coefficient is obtained by identifying the bonding degree coefficient and the bonding position coefficient;

[0140] The second abnormal fusion unit calculates the second abnormal coefficient according to the first interlayer abnormal coefficient, the second interlayer abnormal coefficient and the bonding abnormal coefficient.

[0141] An abnormality identification module, which calculates a third abnormality coefficient according to the first abnormality coefficient and the second abnormality coefficient;

[0142] The abnormal warning module obtains multiple third abnormal coefficients of the same abnormal monitoring position and the same production cycle, calculates the comprehensive abnormal coefficient, and issues an early warning based on the comprehensive abnormal threshold.

[0143] The process of calculating the comprehensive abnormality coefficient according to the first abnormality coefficient and the second abnormality coefficient is as follows:

[0144] The first abnormality coefficient and the second abnormality coefficient of the same production cycle at the same abnormal monitoring position are collected; the third abnormality coefficient is calculated based on the corresponding first abnormality coefficient and the second abnormality coefficient, and the third abnormality coefficient sequence is obtained by arranging them in chronological order; in the third abnormality coefficient sequence, the third abnormality coefficient is weighted by the change rate of the third abnormality coefficient to calculate the comprehensive abnormality coefficient; a comprehensive abnormality threshold is set to judge the comprehensive abnormality coefficient, and when the comprehensive abnormality threshold is exceeded, an early warning message is issued.

[0145] The present invention determines a cardboard area to be monitored through an abnormal monitoring position and an abnormal monitoring cycle of a corrugated cardboard production system; monitors the cardboard area to be monitored to obtain a first data set and a second data set; constructs a first abnormal recognition model to identify the first data set to obtain a first abnormal coefficient; constructs a second abnormal recognition model to identify the second data set to obtain a second abnormal coefficient; calculates a third abnormal coefficient of the cardboard area to be monitored according to the first abnormal coefficient and the second abnormal coefficient; obtains multiple third abnormal coefficients of the same abnormal monitoring position and the same production cycle, calculates a comprehensive abnormal coefficient, and issues an early warning according to a comprehensive abnormal threshold, thereby realizing accurate identification and early warning of abnormal conditions in the cardboard area to be monitored.

[0146] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for monitoring abnormality of a corrugated board production control system, characterized in that: include: S10. Obtain the production process and production cycle of the corrugated board production system, determine the abnormal monitoring position according to the production process, and determine the abnormal monitoring cycle according to the production cycle; S20. Determine the cardboard area to be monitored according to the abnormal monitoring position and the abnormal monitoring period; Acquiring data from the cardboard area to be monitored to obtain a first data set and a second data set; the first data set includes a first panel image, a first panel infrared image, a second panel image, and a second panel infrared image; the second data set includes a first interlayer image and a second interlayer image; The process of determining the cardboard area to be monitored based on the abnormal monitoring position and abnormal monitoring cycle is as follows: The abnormal monitoring cycle includes an abnormal monitoring time and an abnormal monitoring duration; at the abnormal monitoring time, the corrugated cardboard passing through the abnormal monitoring position is marked to obtain a first cross section, and after the abnormal monitoring duration has passed, the corrugated cardboard passing through the abnormal monitoring position is marked to obtain a second cross section; The corrugated cardboard area between the first cross section and the second cross section is used as the cardboard area to be monitored; The paperboard area to be monitored includes a first panel, a second panel, a first interlayer area, and a second interlayer area; S30. Construct a first anomaly recognition model to recognize the first data set and obtain a first anomaly coefficient; the first anomaly recognition model includes an image anomaly recognition unit, a temperature anomaly recognition unit and a first anomaly fusion unit; The image abnormality recognition unit recognizes the first panel image and the second panel image by color features, texture features and shape features, and determines the corresponding first image abnormal area and second image abnormal area; The temperature anomaly recognition unit recognizes the first panel infrared image and the second panel infrared image by temperature distribution to obtain corresponding first temperature anomaly area and second temperature anomaly area; The first abnormal fusion unit determines a first abnormal coefficient according to the first image abnormal area, the second image abnormal area, the first temperature abnormal area and the second temperature abnormal area; S40. Construct a second anomaly recognition model to recognize the second data set to obtain a second anomaly coefficient; the second anomaly recognition model includes a first interlayer recognition unit, a second interlayer recognition unit, a bonding point recognition unit, and a second anomaly fusion unit; The first interlayer recognition unit recognizes the corrugated paper shapes in the first interlayer image and the second interlayer image to obtain a first corrugated paper abnormality coefficient and a second corrugated paper abnormality coefficient; the first interlayer abnormality coefficient is calculated based on the first corrugated paper abnormality coefficient and the second corrugated paper abnormality coefficient; The second interlayer recognition unit recognizes the difference between the shapes of the corrugated paper in the first interlayer image and the second interlayer image to obtain a second interlayer abnormality coefficient; The bonding point identification unit identifies the bonding point between the corrugated paper and the panel to obtain a bonding degree coefficient and a bonding position coefficient; the bonding abnormality coefficient is obtained by identifying the bonding degree coefficient and the bonding position coefficient; The second abnormal fusion unit calculates and obtains a second abnormal coefficient according to the first interlayer abnormal coefficient, the second interlayer abnormal coefficient and the bonding abnormal coefficient; S50. Calculate the third abnormality coefficient of the cardboard area to be monitored based on the first abnormality coefficient and the second abnormality coefficient; S60. Obtain multiple third abnormality coefficients of the same abnormal monitoring position and the same production cycle, calculate the comprehensive abnormality coefficient, and issue an early warning based on the comprehensive abnormality threshold.

2. The abnormality monitoring method of a corrugated board production control system according to claim 1, characterized in that: The recognition process of the first interlayer recognition unit is: Acquire a standard corrugated paper shape of corrugated paper; recognize the corrugated paper shape in the first sandwich image to acquire a first corrugated paper shape; recognize the corrugated paper shape in the second sandwich image to acquire a second corrugated paper shape; Obtaining a first corrugated paper anomaly coefficient according to a difference between the first corrugated paper shape and the standard corrugated paper shape; Obtaining a second corrugated paper anomaly coefficient according to a difference between the second corrugated paper shape and the standard corrugated paper shape; The first interlayer abnormal coefficient is calculated according to the first corrugated paper abnormal coefficient and the second corrugated paper abnormal coefficient.

3. The abnormality monitoring method for a corrugated board production control system according to claim 1, characterized in that: The bonding point identification unit includes a bonding degree identification layer and a bonding position identification layer; The bonding degree identification layer is used to identify the bonding length of the bonding point between the corrugated paper and the panel, and obtain the bonding degree coefficient through the bonding length and the standard bonding length; the bonding position identification layer is used to obtain the coordinates of the midpoint of the bonding length as the bonding point coordinates; the bonding position coefficient is obtained by the difference between the bonding point coordinates and the standard coordinates; the bonding point coordinates are obtained by the first coordinate system and the second coordinate system constructed based on the first interlayer image and the second interlayer image respectively.

4. The abnormality monitoring method for a corrugated board production control system according to claim 1, characterized in that: The process of calculating the comprehensive abnormality coefficient according to the first abnormality coefficient and the second abnormality coefficient is as follows: Collect and obtain the first abnormal coefficient and the second abnormal coefficient of the same production cycle at the same abnormal monitoring position; A third abnormal coefficient is calculated based on the corresponding first abnormal coefficient and second abnormal coefficient, and a third abnormal coefficient sequence is obtained by arranging them in chronological order; in the third abnormal coefficient sequence, the third abnormal coefficient is weighted by the change rate of the third abnormal coefficient, and a comprehensive abnormal coefficient is calculated; a comprehensive abnormal threshold is set to judge the comprehensive abnormal coefficient, and when the comprehensive abnormal threshold is exceeded, a warning message is issued.

5. An abnormality monitoring system for a corrugated board production control system, characterized in that: The abnormality monitoring method for implementing a corrugated board production control system according to any one of claims 1 to 4 comprises: A process determination module, which obtains the production process and production cycle of the corrugated board production system, determines the abnormality monitoring position according to the production process, and determines the abnormality monitoring cycle according to the production cycle; A data acquisition module, which determines a cardboard area to be monitored according to the abnormal monitoring position and the abnormal monitoring period; acquires data on the cardboard area to be monitored to obtain a first data set and a second data set; the first data set includes a first panel image, a first panel infrared image, a second panel image, and a second panel infrared image; the second data set includes a first interlayer image and a second interlayer image; The process of determining the cardboard area to be monitored based on the abnormal monitoring position and abnormal monitoring cycle is as follows: The abnormal monitoring cycle includes an abnormal monitoring time and an abnormal monitoring duration; at the abnormal monitoring time, the corrugated cardboard passing through the abnormal monitoring position is marked to obtain a first cross section, and after the abnormal monitoring duration has passed, the corrugated cardboard passing through the abnormal monitoring position is marked to obtain a second cross section; The corrugated cardboard area between the first cross section and the second cross section is used as the cardboard area to be monitored; the cardboard area to be monitored includes a first panel, a second panel, a first interlayer area, and a second interlayer area; A first recognition module constructs a first anomaly recognition model to recognize the first data set and obtain a first anomaly coefficient; the first anomaly recognition model includes an image anomaly recognition unit, a temperature anomaly recognition unit and a first anomaly fusion unit; The image abnormality recognition unit recognizes the first panel image and the second panel image by color features, texture features and shape features, and determines the corresponding first image abnormal area and second image abnormal area; The temperature anomaly recognition unit recognizes the first panel infrared image and the second panel infrared image by temperature distribution to obtain corresponding first temperature anomaly area and second temperature anomaly area; The first abnormal fusion unit determines a first abnormal coefficient according to the first image abnormal area, the second image abnormal area, the first temperature abnormal area and the second temperature abnormal area; A second recognition module constructs a second anomaly recognition model to recognize the second data set to obtain a second anomaly coefficient; the second anomaly recognition model includes a first interlayer recognition unit, a second interlayer recognition unit, a bonding point recognition unit, and a second anomaly fusion unit; The first interlayer recognition unit recognizes the corrugated paper shapes in the first interlayer image and the second interlayer image to obtain a first corrugated paper abnormality coefficient and a second corrugated paper abnormality coefficient; the first interlayer abnormality coefficient is calculated based on the first corrugated paper abnormality coefficient and the second corrugated paper abnormality coefficient; The second interlayer recognition unit recognizes the difference between the shapes of the corrugated paper in the first interlayer image and the second interlayer image to obtain a second interlayer abnormality coefficient; The bonding point identification unit identifies the bonding point between the corrugated paper and the panel to obtain a bonding degree coefficient and a bonding position coefficient; the bonding abnormality coefficient is obtained by identifying the bonding degree coefficient and the bonding position coefficient; The second abnormal fusion unit calculates and obtains a second abnormal coefficient according to the first interlayer abnormal coefficient, the second interlayer abnormal coefficient and the bonding abnormal coefficient; An abnormality identification module, which calculates a third abnormality coefficient according to the first abnormality coefficient and the second abnormality coefficient; The abnormal warning module obtains multiple third abnormal coefficients of the same abnormal monitoring position and the same production cycle, calculates the comprehensive abnormal coefficient, and issues an early warning based on the comprehensive abnormal threshold.

Images