A defect detection avoidance method based on full data collection of gypsum board production

Abstract

The application discloses a kind of based on gypsum board production full data acquisition's defect detection avoidance method, comprising the following steps: time series analysis of the adjustment amount of gypsum board production equipment adjustment item is carried out, the associated relationship between gypsum board production equipment adjustment item is extracted, and associated relationship is abstracted as the associated network between gypsum board production equipment adjustment item;And according to the number of gypsum board real-time production equipment adjustment item, the production condition of gypsum board production equipment is set in real time.This application carries out learning training to the characterization data of gypsum board defect item and the adjustment amount of gypsum board production equipment adjustment item to obtain defect adjustment calculation model, and the associated relationship between gypsum board production equipment adjustment item is extracted by time series analysis of the adjustment amount of gypsum board production equipment adjustment item, and associated relationship is abstracted as the associated network between gypsum board production equipment adjustment item, realize the full detection and full avoidance of potential defect of gypsum board, improve the initiative of defect detection and avoidance.

Description

Technical Field

[0001] This invention relates to the field of defect detection technology, specifically to a defect detection and avoidance method based on full data acquisition in gypsum board production. Background Technology

[0002] During the production, forming, cutting, and internal transportation of paper-faced gypsum board, various problems such as board breakage and damaged corners may occur. Problematic boards are mainly identified by visual inspection by on-site personnel. Sometimes, due to being busy with other tasks, individual products with external defects may not be detected in time, allowing boards to enter the stack, which is not conducive to product quality control. Furthermore, it requires on-site personnel to use forklifts to remove the entire stack of boards for quality selection and re-stacking, which is a large workload.

[0003] Therefore, an intelligent automatic defect monitoring and avoidance method is needed to monitor and avoid defects in the surface shape, temperature and humidity of gypsum board. Existing intelligent automatic defect monitoring methods generally first detect and avoid defects on one surface of the gypsum board, then flip the gypsum board over and monitor and avoid defects on the other surface. This detection method still uses passive defect detection and avoidance, which requires detecting and avoiding defects one by one, making it difficult to predict and avoid potential defects. Summary of the Invention

[0004] The purpose of this invention is to provide a defect detection and avoidance method based on full data acquisition in gypsum board production, in order to solve the technical problem that the existing technology uses passive defect detection and avoidance, which requires each gypsum board defect to be detected and avoided individually, making it difficult to predictively monitor and avoid potential defects.

[0005] To solve the above-mentioned technical problems, the present invention specifically provides the following technical solution:

[0006] A defect detection and avoidance method based on full data acquisition in gypsum board production includes the following steps:

[0007] Step S1: Match the gypsum board defects with the gypsum board production equipment adjustment items accordingly;

[0008] Step S2: Based on the targeted matching relationship between the gypsum board defect items and the adjustment items of the gypsum board production equipment in Step S1, the defect adjustment calculation model is obtained by learning and training the characterization data of the gypsum board defect items and the adjustment amount of the gypsum board production equipment adjustment items in the full data collection log of gypsum board production.

[0009] Step S3: In the full data collection log of gypsum board production, perform time-series analysis on the adjustment amount of the gypsum board production equipment adjustment items to extract the associated relationships between the gypsum board production equipment adjustment items, and abstract the associated relationships into an associated network between the gypsum board production equipment adjustment items.

[0010] Step S4: Extract the characterization data of real-time gypsum board defects from the real-time gypsum board production data obtained by real-time monitoring, and input the characterization data of real-time gypsum board defects into the defect adjustment calculation model to obtain the adjustment amount of the real-time gypsum board production equipment adjustment item.

[0011] Step S5: Count the number of real-time production equipment adjustment items for gypsum board, and determine the real-time production conditions of the gypsum board production equipment based on the number of real-time production equipment adjustment items, so as to achieve full detection and full avoidance of potential defects in gypsum board.

[0012] As a preferred embodiment of the present invention, the step of specifically matching gypsum board defects with gypsum board production equipment adjustment items includes:

[0013] The defects of the gypsum board include: gypsum board width, gypsum board vertical edge angle, gypsum board thickness, gypsum board wet weight, gypsum board dry weight, gypsum board temperature after drying, and gypsum board humidity after drying.

[0014] The adjustments to the gypsum board production equipment include: gypsum board forming blade distance, gypsum board forming blade angle, gypsum board forming height, dry gypsum board feed rate, wet gypsum board feed rate, and gypsum board dryer gate temperature.

[0015] The adjustment of the distance and angle of the gypsum board forming blades is specifically designed to avoid defects in the width and vertical edge angle of the gypsum board.

[0016] The adjustment of the height of the gypsum board molding is specifically designed to avoid defects in the thickness of the gypsum board;

[0017] The adjustment of the dry material feed amount and wet material feed amount of gypsum board is specifically designed to avoid the defects of the single weight of wet gypsum board and the single weight of dry gypsum board.

[0018] The adjustment of the gate temperature of the gypsum board dryer is specifically designed to avoid defects in the temperature and humidity of the gypsum board after drying.

[0019] As a preferred embodiment of the present invention, the defect adjustment calculation model obtained by learning and training the characterization data of gypsum board defects and the adjustment amount of gypsum board production equipment adjustment items includes:

[0020] In the full data collection log of gypsum board production, the characterization data of gypsum board defects and the adjustment amount of gypsum board production equipment adjustment items are extracted.

[0021] The characterization data of gypsum board defects are used as input to the BP neural network, and the adjustment amount of the gypsum board production equipment is used as output to the BP neural network.

[0022] A defect adjustment measurement model is obtained by convolutional learning of the input and output terms of a BP neural network.

[0023] The model expression for the defect adjustment measurement model is as follows:

[0024] Wr = BP(Ds);

[0025] In the formula, Wr is the adjustment amount of the gypsum board production equipment adjustment item r, Ds is the characterization data of the gypsum board defect item s, BP is the BP neural network, r is the identifier of the gypsum board production equipment adjustment item, and s is the identifier of the gypsum board defect item.

[0026] The complete data for gypsum board production consists of a combination of characterizing data such as gypsum board width, gypsum board vertical edge angle, gypsum board thickness, wet gypsum board weight, dry gypsum board weight, gypsum board temperature after drying, and gypsum board humidity after drying.

[0027] The gypsum board production full data collection log is a log consisting of gypsum board production full data collected in chronological order;

[0028] The extraction of characterization data for the defects in the gypsum board includes:

[0029] In the full data collection log of gypsum board production, the characteristic data of gypsum board width, vertical edge angle, thickness, wet weight, dry weight, temperature after drying, and humidity after drying at each time point are compared with the standard data of gypsum board width, vertical edge angle, thickness, wet weight, dry weight, temperature after drying, and humidity after drying. The corresponding items with similarity between the characteristic data and the standard data at each time point less than the allowable threshold are regarded as gypsum board defects at each time point.

[0030] In the full data collection log of gypsum board production, characterization data of gypsum board defects at each time point were extracted;

[0031] The extraction of the adjustment amount for the gypsum board production equipment adjustment items includes:

[0032] In the full data collection log of gypsum board production, the targeted matching relationship is used to determine the gypsum board production equipment adjustment items that are targetedly matched with the gypsum board defect items at each time point.

[0033] In the full data collection log of gypsum board production, the adjustment amount of the gypsum board production equipment adjustment item that is specifically matched with the gypsum board defect item at each time point is extracted.

[0034] As a preferred embodiment of the present invention, the step of performing time-series analysis on the adjustment items of gypsum board production equipment to extract the associated relationships between the adjustment items includes:

[0035] The adjustment amounts of each gypsum board production equipment adjustment item are sorted according to time sequence to obtain the time sequence of adjustment amounts of the gypsum board production equipment adjustment items.

[0036] The time series of adjustment amounts for each gypsum board production equipment adjustment item are analyzed by sequence similarity analysis to quantify the co-occurrence rate, which characterizes the co-occurrence relationship between each gypsum board production equipment adjustment item. The co-occurrence relationship is a characterization of the degree of mutual influence between the gypsum board production equipment adjustment items.

[0037] The quantitative formula for the co-occurrence rate is:

[0038] H ij =|W i,T -W j,T |;

[0039] In the formula, H ij W represents the co-occurrence rate between adjustment item i and adjustment item j of gypsum board production equipment. i,T For the adjustment amount time sequence of adjustment item i of gypsum board production equipment, W j,T For the adjustment amount time sequence of adjustment item j of gypsum board production equipment, |W i,T -W j,T |for W i,T and W j,T The Euclidean distance, i, j are the identifiers of the adjustment items of the gypsum board production equipment, and T is the time sequence identifier.

[0040] As a preferred embodiment of the present invention, the abstraction of the symbiotic relationship into a symbiotic network among the adjustment items of gypsum board production equipment includes:

[0041] The adjustment items of gypsum board production equipment are abstracted as topological nodes, and topological edges are set between topological nodes to make topological connections.

[0042] The co-occurrence rate between the adjustment items of gypsum board production equipment is used as the edge weight of the topological edge between the corresponding topological nodes, and is abstracted as the co-occurrence network.

[0043] As a preferred embodiment of the present invention, the step of determining the real-time production status setting of the gypsum board production equipment based on the number of real-time adjustment items of the gypsum board production equipment includes:

[0044] If the number of adjustment items for the gypsum board production equipment in real time is 0, then the current production condition settings of the gypsum board production equipment will be maintained.

[0045] If the number of adjustment items for the real-time gypsum board production equipment is 1, then the adjustment amount of the adjustment item for the real-time gypsum board production equipment is mapped to the associated network to obtain a single real-time associated network. The adjustment amount of the real-time associated adjustment item for the gypsum board production equipment is obtained in the single real-time associated network. Based on the adjustment amount of the real-time adjustment item for the gypsum board production equipment and the adjustment amount of the real-time associated adjustment item for the gypsum board production equipment, the production conditions of the gypsum board production equipment are set in real time to achieve full detection and full avoidance of potential defects in gypsum board.

[0046] If the number of adjustment items for gypsum board production equipment is greater than 1, the adjustment amount of each real-time adjustment item for gypsum board production equipment is sequentially mapped to the associated network to obtain multiple real-time associated networks. The multiple real-time associated networks are then merged to obtain the adjustment amount of the real-time associated adjustment item for gypsum board production equipment. Based on the adjustment amount of the real-time adjustment item for gypsum board production equipment and the adjustment amount of the real-time associated adjustment item for gypsum board production equipment, the production conditions of the gypsum board production equipment are set in real time to achieve full detection and full avoidance of potential defects in gypsum board.

[0047] As a preferred embodiment of the present invention, the step of mapping the adjustment amount of the real-time gypsum board production equipment adjustment item to the associated network to obtain a single real-time associated network includes:

[0048] The adjustment amount of the real-time gypsum board production equipment adjustment item is mapped to the node value of the corresponding topology node of the associated network, and the node value of the remaining topology node in the associated network is calculated one by one according to the edge weight of the topology node and the topology edge to obtain a single real-time associated network.

[0049] As a preferred embodiment of the present invention, obtaining the adjustment amount of the real-time associated adjustment item of the gypsum board production equipment in a single real-time associated network includes:

[0050] The topology nodes in a single real-time companion network, excluding the topology nodes corresponding to the real-time production equipment adjustment items for gypsum board, are taken as the real-time companion adjustment items for gypsum board production equipment.

[0051] The node value of the topology node corresponding to the real-time accompanying adjustment item of the gypsum board production equipment in a single real-time accompanying network is used as the adjustment amount of the real-time accompanying adjustment item of the gypsum board production equipment.

[0052] As a preferred embodiment of the present invention, the step of sequentially mapping the adjustment amount of each real-time gypsum board production equipment adjustment item to the associated network to obtain multiple real-time associated networks includes:

[0053] The adjustment amount of each real-time gypsum board production equipment adjustment item is mapped to the node value of the corresponding topology node of each associated network in turn, and the node value of the remaining topology node in each associated network is calculated one by one according to the edge weight of the topology node and the topology edge to obtain multiple real-time associated networks.

[0054] The node values ​​of multiple real-time companion networks are superimposed and fused to obtain a single fused real-time companion network.

[0055] As a preferred embodiment of the present invention, the step of fusing multiple real-time associated networks to obtain the adjustment amount of the real-time associated adjustment item for the gypsum board production equipment includes:

[0056] In a single fused real-time companion network, all topology nodes except those corresponding to the real-time gypsum board production equipment adjustment items are considered as real-time companion adjustment items for gypsum board production equipment.

[0057] The node value of the topology node corresponding to the real-time accompanying adjustment item of the gypsum board production equipment in a single fused real-time accompanying network is used as the adjustment amount of the real-time accompanying adjustment item of the gypsum board production equipment.

[0058] Compared with the prior art, the present invention has the following advantages:

[0059] This invention uses the full data collection log of gypsum board production to learn and train the characterization data of gypsum board defects and the adjustment amount of gypsum board production equipment adjustment items to obtain a defect adjustment calculation model. It also performs time series analysis on the adjustment amount of gypsum board production equipment adjustment items to extract the symbiotic relationship between gypsum board production equipment adjustment items, and abstracts the symbiotic relationship into a symbiotic network between gypsum board production equipment adjustment items. This enables full detection and full avoidance of potential defects in gypsum board, improving the initiative of defect detection and avoidance. Attached Figure Description

[0060] 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.

[0061] Figure 1 This is a flowchart of a defect detection and avoidance method provided in an embodiment of the present invention. Detailed Implementation

[0062] 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.

[0063] Existing intelligent automatic defect monitoring methods typically involve first detecting and avoiding defects on one surface of the gypsum board, then flipping the board over and performing the same detection and avoidance on the other surface. This method still employs passive defect detection and avoidance, requiring individual detection and avoidance of each defect, making it difficult to predictively detect and avoid potential defects. Therefore, this invention provides a defect detection and avoidance method based on full data acquisition from gypsum board production. It utilizes a defect adjustment calculation model based on full data acquisition from gypsum board production to determine adjustment items for gypsum board production equipment to avoid defects. Furthermore, based on a constructed associated network, it obtains adjustment items for all gypsum board production equipment to avoid potential defects, achieving full detection and avoidance of all potential defects in gypsum board.

[0064] like Figure 1 As shown, this invention provides a defect detection and avoidance method based on full data acquisition in gypsum board production, comprising the following steps:

[0065] Step S1: Match the gypsum board defects with the gypsum board production equipment adjustment items accordingly;

[0066] Step S2: Based on the targeted matching relationship between the gypsum board defect items and the adjustment items of the gypsum board production equipment in Step S1, the defect adjustment calculation model is obtained by learning and training the characterization data of the gypsum board defect items and the adjustment amount of the gypsum board production equipment adjustment items in the full data collection log of gypsum board production.

[0067] Step S3: In the full data collection log of gypsum board production, perform time-series analysis on the adjustment amount of the gypsum board production equipment adjustment items to extract the associated relationships between the gypsum board production equipment adjustment items, and abstract the associated relationships into an associated network between the gypsum board production equipment adjustment items.

[0068] Step S4: Extract the characterization data of real-time gypsum board defects from the real-time gypsum board production data obtained by real-time monitoring, and input the characterization data of real-time gypsum board defects into the defect adjustment calculation model to obtain the adjustment amount of the real-time gypsum board production equipment adjustment item.

[0069] Step S5: Count the number of real-time production equipment adjustment items for gypsum board, and determine the real-time production conditions of the gypsum board production equipment based on the number of real-time production equipment adjustment items, so as to achieve full detection and full avoidance of potential defects in gypsum board.

[0070] This invention constructs a defect adjustment calculation model that can calculate the adjustment amount of gypsum board production equipment adjustment items based on the characterization data of gypsum board defects. The adjustment amount of the gypsum board production equipment adjustment items is used to adjust the characterization data of gypsum board defects to standard data. For example, if the characterization data of gypsum board width and gypsum board vertical edge angle deviates from the standard data, the corresponding adjustment amount of the gypsum board forming knife distance and gypsum board forming knife angle can restore the gypsum board width and gypsum board vertical edge angle from the characterization data to the standard data, thereby avoiding defects in gypsum board width and gypsum board vertical edge angle. Defects in gypsum board thickness, wet gypsum board weight, dry gypsum board weight, gypsum board temperature after drying, and gypsum board humidity after drying are the same as those in gypsum board width and gypsum board vertical edge angle, and will not be elaborated here. Therefore, the defect adjustment calculation model achieves targeted matching calculation of the adjustment amount of gypsum board production equipment adjustment items, improves the degree of automation in calculation, and thus improves the efficiency of gypsum board defect avoidance.

[0071] The defects in gypsum board are matched with the adjustments made to the gypsum board production equipment, including:

[0072] Defects in gypsum board include: gypsum board width, gypsum board vertical edge angle, gypsum board thickness, wet gypsum board weight, dry gypsum board weight, gypsum board temperature after drying, and gypsum board humidity after drying.

[0073] Adjustments to gypsum board production equipment include: distance between gypsum board forming blades, angle of gypsum board forming blades, height of gypsum board forming plate, dry gypsum board feed rate, wet gypsum board feed rate, and gate temperature of gypsum board dryer.

[0074] Among them, the adjustment of the distance and angle of the gypsum board forming blade is specifically designed to avoid the defects of the gypsum board width and the vertical edge angle.

[0075] Adjusting the height of the gypsum board molding board specifically avoids defects in gypsum board thickness;

[0076] Adjustments to the dry and wet gypsum board material quantities are specifically designed to avoid the defects in the single weight of wet and dry gypsum board.

[0077] Adjusting the gate temperature of the gypsum board dryer is specifically designed to avoid defects in temperature and humidity after gypsum board drying.

[0078] A defect adjustment calculation model is obtained by learning and training the characterization data of gypsum board defects and the adjustment amounts of gypsum board production equipment adjustment items, including:

[0079] In the full data collection log of gypsum board production, the characterization data of gypsum board defects and the adjustment amount of gypsum board production equipment adjustment items are extracted.

[0080] The characterization data of gypsum board defects are used as input to the BP neural network, and the adjustment amount of gypsum board production equipment is used as output to the BP neural network.

[0081] A defect adjustment measurement model is obtained by convolutional learning of the input and output terms of a BP neural network.

[0082] The model expression for the defect adjustment and measurement model is:

[0083] Wr = BP(Ds);

[0084] In the formula, Wr is the adjustment amount of the gypsum board production equipment adjustment item r, Ds is the characterization data of the gypsum board defect item s, BP is the BP neural network, r is the identifier of the gypsum board production equipment adjustment item, and s is the identifier of the gypsum board defect item.

[0085] The complete data for gypsum board production consists of a combination of characterizing data such as gypsum board width, gypsum board vertical edge angle, gypsum board thickness, wet gypsum board weight, dry gypsum board weight, gypsum board temperature after drying, and gypsum board humidity after drying.

[0086] The gypsum board production full data collection log is a log consisting of gypsum board production full data collected in chronological order;

[0087] Extraction of characterization data for gypsum board defects, including:

[0088] In the full data collection log of gypsum board production, the characteristic data of gypsum board width, vertical edge angle, thickness, wet weight, dry weight, temperature after drying, and humidity after drying at each time point are compared with the standard data of gypsum board width, vertical edge angle, thickness, wet weight, dry weight, temperature after drying, and humidity after drying. The corresponding items with similarity between the characteristic data and the standard data at each time point less than the allowable threshold are regarded as gypsum board defects at each time point.

[0089] In the full data collection log of gypsum board production, characterization data of gypsum board defects at each time point were extracted;

[0090] The extraction of adjustment amounts for gypsum board production equipment adjustment items includes:

[0091] In the full data collection log of gypsum board production, the gypsum board defect items at each time point are used to determine the gypsum board production equipment adjustment items that are specifically matched with the gypsum board defect items at each time point using targeted matching relationships.

[0092] In the full data collection log of gypsum board production, the adjustment amount of the gypsum board production equipment adjustment item that is specifically matched with the gypsum board defect item at each time point is extracted.

[0093] The characterization data of real-time defects in gypsum board are extracted from the real-time full data of gypsum board production obtained by real-time monitoring. The characterization data of real-time defects in gypsum board are then input into the defect adjustment calculation model to obtain the adjustment amount of the real-time production equipment adjustment item of gypsum board.

[0094] Extraction of characterization data for real-time defects in gypsum board, including:

[0095] In the real-time collected full data of gypsum board production, the real-time characterization data of gypsum board width, gypsum board vertical edge angle, gypsum board thickness, wet gypsum board weight, dry gypsum board weight, gypsum board temperature after drying, and gypsum board humidity after drying are compared with the standard data of gypsum board width, gypsum board vertical edge angle, gypsum board thickness, wet gypsum board weight, dry gypsum board weight, gypsum board temperature after drying, and gypsum board humidity after drying. The corresponding items with the similarity between the real-time characterization data and the standard data less than the allowable threshold are regarded as real-time defects of gypsum board.

[0096] Characteristic data of real-time defects in gypsum board are extracted from the real-time collected full data of gypsum board production.

[0097] Extraction of adjustment amounts for real-time gypsum board production equipment adjustment items, including:

[0098] In the real-time collected full data of gypsum board production, the real-time defect items of gypsum board are used to determine the real-time production equipment adjustment items that are specifically matched with the gypsum board defect items.

[0099] From the real-time collected full data of gypsum board production, the adjustment amount of the real-time production equipment adjustment item for gypsum board that is specifically matched with the real-time defect items of gypsum board is extracted.

[0100] This invention mines the degree of mutual influence between production equipment adjustment items. Since production equipment adjustment items and gypsum board defect items have a targeted matching relationship, mining the degree of mutual influence between production equipment adjustment items is also mining the degree of mutual influence between gypsum board defect items. Reflecting on the production process, mining the influence relationship between gypsum board defects allows for the early detection of other potential gypsum board defects that a particular defect might cause. That is, if gypsum board defect A and another gypsum board defect B have a mutual influence relationship, then after detecting the occurrence of gypsum board defect A, it can be directly known that gypsum board defect B will occur. Therefore, this invention mines the degree of mutual influence between production equipment adjustment items to construct a symbiotic network, which can be used for proactively and predictively perceiving and avoiding potential gypsum board defects, as detailed below:

[0101] A time-series analysis of the adjustment items for gypsum board production equipment was conducted to extract the associated relationships among these adjustment items, including:

[0102] The adjustment amounts of each gypsum board production equipment adjustment item are sorted according to time sequence to obtain the time sequence of adjustment amounts of the gypsum board production equipment adjustment items.

[0103] The time series of adjustment amounts for each gypsum board production equipment adjustment item are analyzed by sequence similarity analysis to quantify the co-occurrence rate, which represents the co-occurrence relationship between each gypsum board production equipment adjustment item. The co-occurrence relationship is a representation of the degree of mutual influence between the gypsum board production equipment adjustment items.

[0104] The quantitative formula for the co-occurrence rate is:

[0105] H ij =|W i,T -W j,T |;

[0106] In the formula, H ij W represents the co-occurrence rate between adjustment item i and adjustment item j of gypsum board production equipment. i,T For the adjustment amount time sequence of adjustment item i of gypsum board production equipment, W j,T For the adjustment amount time sequence of adjustment item j of gypsum board production equipment, |W i,T -W j,T |for W i,T and W j,T The Euclidean distance, i, j are the identifiers of the adjustment items of the gypsum board production equipment, and T is the time sequence identifier.

[0107] By using the time series sequence of adjustment amounts for gypsum board production equipment adjustments, sequence similarity analysis was performed to quantify the co-occurrence rate, which characterizes the symbiotic relationship between various gypsum board production equipment adjustment items. The higher the similarity of the time series sequence of adjustment amounts among gypsum board production equipment adjustment items, the higher the co-occurrence rate between them. Reflecting on the production process, this indicates that when adjusting production equipment to avoid gypsum board defect A, adjustments to production equipment to avoid gypsum board defect B are usually accompanied by adjustments to production equipment to avoid gypsum board defect B. Furthermore, the more consistent the adjustment rhythm between the corresponding production equipment for avoiding gypsum board defect A and the corresponding production equipment for avoiding gypsum board defect B, the stronger the correlation between gypsum board defect B and gypsum board defect B. The more consistent the rhythm of the occurrence of gypsum board defects A, the higher the degree of mutual influence between gypsum board defects B and A, potentially forming a symbiotic relationship. That is, the occurrence of gypsum board defect A will definitely lead to the occurrence of gypsum board defect B. Therefore, this invention uses the time series of adjustment amounts of gypsum board production equipment adjustment items to perform sequence similarity analysis to quantify the symbiotic relationship between various gypsum board production equipment adjustment items. This fully mines the data information between production equipment adjustment items, thereby obtaining the symbiotic relationship between various gypsum board production equipment adjustment items, and ultimately achieving proactive and predictive perception of potential gypsum board defects, as well as proactive and predictive avoidance of potential gypsum board defects.

[0108] This invention constructs a symbiotic network that topologically visualizes the symbiotic relationships between adjustment items in gypsum board production equipment. It fully visualizes these relationships and, based on the adjustment amounts of real-time gypsum board production equipment adjustment items that avoid detected gypsum board defects, obtains the adjustment amounts of real-time symbiotic adjustment items that are potential gypsum board defects relative to the detected defects. This achieves full avoidance of all potential gypsum board defects, improving the comprehensiveness of defect avoidance. Furthermore, the adjustment amounts of real-time symbiotic adjustment items are specifically matched with potential gypsum board defects; therefore, full avoidance of all potential gypsum board defects is equivalent to full detection of all potential gypsum board defects, improving the comprehensiveness of defect detection.

[0109] The symbiotic relationship is abstracted as a symbiotic network among the adjustment items of gypsum board production equipment, including:

[0110] The adjustment items of gypsum board production equipment are abstracted as topological nodes, and topological edges are set between topological nodes to make topological connections.

[0111] The co-occurrence rate between the adjustment items of gypsum board production equipment is used as the edge weight of the topological edge between the corresponding topological nodes, and it is abstracted as a co-occurrence network.

[0112] The real-time production conditions of the gypsum board production equipment are determined based on the number of real-time adjustment items, including:

[0113] If the number of adjustment items for the gypsum board production equipment in real time is 0, then the current production condition settings of the gypsum board production equipment will be maintained.

[0114] If the number of adjustment items for the real-time gypsum board production equipment is 1, then the adjustment amount of the adjustment item for the real-time gypsum board production equipment is mapped to the associated network to obtain a single real-time associated network. The adjustment amount of the real-time associated adjustment item for the gypsum board production equipment is obtained in the single real-time associated network. Based on the adjustment amount of the adjustment item for the real-time gypsum board production equipment and the adjustment amount of the real-time associated adjustment item for the gypsum board production equipment, the production conditions of the gypsum board production equipment are set in real time to achieve full detection and full avoidance of potential defects in gypsum board.

[0115] If the number of adjustment items for gypsum board production equipment is greater than 1, the adjustment amount of each real-time adjustment item for gypsum board production equipment is sequentially mapped to the associated network to obtain multiple real-time associated networks. These multiple real-time associated networks are then merged to obtain the adjustment amount of the real-time associated adjustment item for gypsum board production equipment. Based on the adjustment amount of the real-time adjustment item for gypsum board production equipment and the adjustment amount of the real-time associated adjustment item for gypsum board production equipment, the production conditions of the gypsum board production equipment are set in real time to achieve full detection and full avoidance of potential defects in gypsum board.

[0116] Mapping the adjustment amounts of the real-time gypsum board production equipment adjustment items to the associated network yields a single real-time associated network, including:

[0117] The adjustment amount of the real-time gypsum board production equipment adjustment item is mapped to the node value of the corresponding topology node of the associated network, and the node value of the remaining topology node in the associated network is calculated one by one according to the edge weight of the topology node and the topology edge to obtain a single real-time associated network.

[0118] The adjustment amounts of the real-time associated adjustment items for gypsum board production equipment are obtained in a single real-time associated network, including:

[0119] The topology nodes in a single real-time companion network, excluding the topology nodes corresponding to the real-time production equipment adjustment items for gypsum board, are taken as the real-time companion adjustment items for gypsum board production equipment.

[0120] The node value of the topology node corresponding to the real-time accompanying adjustment item of the gypsum board production equipment in a single real-time accompanying network is used as the adjustment amount of the real-time accompanying adjustment item of the gypsum board production equipment.

[0121] If multiple gypsum board defects are detected simultaneously, a companion network is mapped for each defect, and the mapped companion networks are fused to generate a simultaneous combined adjustment, thus avoiding all potential defects at once, improving defect avoidance efficiency, and reducing avoidance interference, as detailed below:

[0122] By sequentially mapping the adjustment amount of each real-time gypsum board production equipment adjustment item to the associated network, multiple real-time associated networks are obtained, including:

[0123] The adjustment amount of each real-time gypsum board production equipment adjustment item is mapped to the node value of the corresponding topology node of each associated network in turn, and the node value of the remaining topology node in each associated network is calculated one by one according to the edge weight of the topology node and the topology edge to obtain multiple real-time associated networks.

[0124] The node values ​​of multiple real-time companion networks are superimposed and fused to obtain a single fused real-time companion network.

[0125] The adjustment amount of the real-time associated adjustment item for gypsum board production equipment is obtained by fusing multiple real-time associated networks, including:

[0126] In a single fused real-time companion network, all topology nodes except those corresponding to the real-time gypsum board production equipment adjustment items are considered as real-time companion adjustment items for gypsum board production equipment.

[0127] The node value of the topology node corresponding to the real-time accompanying adjustment item of the gypsum board production equipment in a single fused real-time accompanying network is used as the adjustment amount of the real-time accompanying adjustment item of the gypsum board production equipment.

[0128] This invention uses the full data collection log of gypsum board production to learn and train the characterization data of gypsum board defects and the adjustment amount of gypsum board production equipment adjustment items to obtain a defect adjustment calculation model. It also performs time series analysis on the adjustment amount of gypsum board production equipment adjustment items to extract the symbiotic relationship between gypsum board production equipment adjustment items, and abstracts the symbiotic relationship into a symbiotic network between gypsum board production equipment adjustment items. This enables full detection and full avoidance of potential defects in gypsum board, improving the initiative of defect detection and avoidance.

[0129] 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 defect detection and avoidance method based on full data acquisition in gypsum board production, characterized in that: Includes the following steps: Step S1: Match the gypsum board defects with the gypsum board production equipment adjustment items accordingly; Step S2: Based on the targeted matching relationship between the gypsum board defect items and the adjustment items of the gypsum board production equipment in Step S1, the defect adjustment calculation model is obtained by learning and training the characterization data of the gypsum board defect items and the adjustment amount of the gypsum board production equipment adjustment items in the full data collection log of gypsum board production. Step S3: In the full data collection log of gypsum board production, perform time-series analysis on the adjustment amount of the gypsum board production equipment adjustment items to extract the associated relationships between the gypsum board production equipment adjustment items, and abstract the associated relationships into an associated network between the gypsum board production equipment adjustment items. Step S4: Extract the characterization data of real-time gypsum board defects from the real-time gypsum board production data obtained by real-time monitoring, and input the characterization data of real-time gypsum board defects into the defect adjustment calculation model to obtain the adjustment amount of the real-time gypsum board production equipment adjustment item. Step S5: Count the number of real-time production equipment adjustment items for gypsum board, and determine the real-time production conditions of the gypsum board production equipment based on the number of real-time production equipment adjustment items, so as to achieve full detection and full avoidance of potential defects in gypsum board. A time-series analysis of the adjustment items for gypsum board production equipment was conducted to extract the associated relationships among these adjustment items, including: The adjustment amounts of each gypsum board production equipment adjustment item are sorted according to time sequence to obtain the time sequence of adjustment amounts of the gypsum board production equipment adjustment items. The time series of adjustment amounts for each gypsum board production equipment adjustment item are analyzed by sequence similarity analysis to quantify the co-occurrence rate, which represents the co-occurrence relationship between each gypsum board production equipment adjustment item. The co-occurrence relationship is a representation of the degree of mutual influence between the gypsum board production equipment adjustment items. The quantitative formula for the co-occurrence rate is: ； In the formula, H ij W represents the co-occurrence rate between adjustment item i and adjustment item j of gypsum board production equipment. i,T For the adjustment amount time sequence of adjustment item i of gypsum board production equipment, W j,T For the adjustment amount time sequence of adjustment item j of gypsum board production equipment, |W i,T -W j,T |for W i,T and W j,T The Euclidean distance, i, j are the identifiers of the adjustment items of the gypsum board production equipment, and T is the time sequence identifier; The symbiotic relationship is abstracted as a symbiotic network among the adjustment items of gypsum board production equipment, including: The adjustment items of gypsum board production equipment are abstracted as topological nodes, and topological edges are set between topological nodes to make topological connections. The co-occurrence rate between the adjustment items of gypsum board production equipment is used as the edge weight of the topological edge between the corresponding topological nodes, and it is abstracted as a co-occurrence network. The real-time production conditions of the gypsum board production equipment are determined based on the number of real-time adjustment items, including: If the number of adjustment items for the gypsum board production equipment in real time is 0, then the current production condition settings of the gypsum board production equipment will be maintained. If the number of adjustment items for the real-time gypsum board production equipment is 1, then the adjustment amount of the adjustment item for the real-time gypsum board production equipment is mapped to the associated network to obtain a single real-time associated network. The adjustment amount of the real-time associated adjustment item for the gypsum board production equipment is obtained in the single real-time associated network. Based on the adjustment amount of the adjustment item for the real-time gypsum board production equipment and the adjustment amount of the real-time associated adjustment item for the gypsum board production equipment, the production conditions of the gypsum board production equipment are set in real time to achieve full detection and full avoidance of potential defects in gypsum board. If the number of adjustment items for gypsum board production equipment is greater than 1, the adjustment amount of each real-time adjustment item for gypsum board production equipment is sequentially mapped to the associated network to obtain multiple real-time associated networks. These multiple real-time associated networks are then merged to obtain the adjustment amount of the real-time associated adjustment item for gypsum board production equipment. Based on the adjustment amount of the real-time adjustment item for gypsum board production equipment and the adjustment amount of the real-time associated adjustment item for gypsum board production equipment, the production conditions of the gypsum board production equipment are set in real time to achieve full detection and full avoidance of potential defects in gypsum board.

2. The defect detection and avoidance method based on full data acquisition in gypsum board production according to claim 1, characterized in that: The specific matching of gypsum board defects with gypsum board production equipment adjustment items includes: The defects of the gypsum board include: gypsum board width, gypsum board vertical edge angle, gypsum board thickness, gypsum board wet weight, gypsum board dry weight, gypsum board temperature after drying, and gypsum board humidity after drying. The adjustments to the gypsum board production equipment include: gypsum board forming blade distance, gypsum board forming blade angle, gypsum board forming height, dry gypsum board feed rate, wet gypsum board feed rate, and gypsum board dryer gate temperature. The adjustment of the distance and angle of the gypsum board forming blades is specifically designed to avoid defects in the width and vertical edge angle of the gypsum board. The adjustment of the height of the gypsum board molding is specifically designed to avoid defects in the thickness of the gypsum board; The adjustment of the dry material feed amount and wet material feed amount of gypsum board is specifically designed to avoid the defects of the single weight of wet gypsum board and the single weight of dry gypsum board. The adjustment of the gate temperature of the gypsum board dryer is specifically designed to avoid defects in the temperature and humidity of the gypsum board after drying.

3. The defect detection and avoidance method based on full data acquisition in gypsum board production according to claim 2, characterized in that: The defect adjustment calculation model is obtained by learning and training the characterization data of gypsum board defects and the adjustment amount of gypsum board production equipment adjustment items, including: In the full data collection log of gypsum board production, the characterization data of gypsum board defects and the adjustment amount of gypsum board production equipment adjustment items are extracted. The characterization data of gypsum board defects are used as input to the BP neural network, and the adjustment amount of the gypsum board production equipment is used as output to the BP neural network. A defect adjustment measurement model is obtained by convolutional learning of the input and output terms of a BP neural network. The model expression for the defect adjustment measurement model is as follows: Wr=BP(Ds); In the formula, Wr is the adjustment amount of the gypsum board production equipment adjustment item r, Ds is the characterization data of the gypsum board defect item s, BP is the BP neural network, r is the identifier of the gypsum board production equipment adjustment item, and s is the identifier of the gypsum board defect item. The complete data for gypsum board production consists of a combination of characterizing data such as gypsum board width, gypsum board vertical edge angle, gypsum board thickness, wet gypsum board weight, dry gypsum board weight, gypsum board temperature after drying, and gypsum board humidity after drying. The gypsum board production full data collection log is a log consisting of gypsum board production full data collected in chronological order; The extraction of characterization data for the defects in the gypsum board includes: In the full data collection log of gypsum board production, the characteristic data of gypsum board width, vertical edge angle, thickness, wet weight, dry weight, temperature after drying, and humidity after drying at each time point are compared with the standard data of gypsum board width, vertical edge angle, thickness, wet weight, dry weight, temperature after drying, and humidity after drying. The corresponding items with similarity between the characteristic data and the standard data at each time point less than the allowable threshold are regarded as gypsum board defects at each time point. In the full data collection log of gypsum board production, characterization data of gypsum board defects at each time point were extracted; The extraction of the adjustment amount for the adjustment items of the gypsum board production equipment includes: In the full data collection log of gypsum board production, the targeted matching relationship is used to determine the gypsum board production equipment adjustment items that are targetedly matched with the gypsum board defect items at each time point. In the full data collection log of gypsum board production, the adjustment amount of the gypsum board production equipment adjustment item that is specifically matched with the gypsum board defect item at each time point is extracted.

4. The defect detection and avoidance method based on full data acquisition in gypsum board production according to claim 3, characterized in that, The process of mapping the adjustment amount of the real-time gypsum board production equipment adjustment item to the associated network to obtain a single real-time associated network includes: The adjustment amount of the real-time gypsum board production equipment adjustment item is mapped to the node value of the corresponding topology node of the associated network, and the node value of the remaining topology node in the associated network is calculated one by one according to the edge weight of the topology node and the topology edge to obtain a single real-time associated network.

5. The defect detection and avoidance method based on full data acquisition in gypsum board production according to claim 4, characterized in that, The adjustment amount of the real-time associated adjustment item of the gypsum board production equipment obtained in a single real-time associated network includes: The topology nodes in a single real-time companion network, excluding the topology nodes corresponding to the real-time production equipment adjustment items for gypsum board, are taken as the real-time companion adjustment items for gypsum board production equipment. The node value of the topology node corresponding to the real-time accompanying adjustment item of the gypsum board production equipment in a single real-time accompanying network is used as the adjustment amount of the real-time accompanying adjustment item of the gypsum board production equipment.

6. The defect detection and avoidance method based on full data acquisition in gypsum board production according to claim 5, characterized in that, The process of sequentially mapping the adjustment amount of each real-time gypsum board production equipment adjustment item to the associated network results in multiple real-time associated networks, including: The adjustment amount of each real-time gypsum board production equipment adjustment item is mapped to the node value of the corresponding topology node of each associated network in turn, and the node value of the remaining topology node in each associated network is calculated one by one according to the edge weight of the topology node and the topology edge to obtain multiple real-time associated networks. The node values ​​of multiple real-time companion networks are superimposed and fused to obtain a single fused real-time companion network.

7. The defect detection and avoidance method based on full data acquisition in gypsum board production according to claim 6, characterized in that, The adjustment amount of the real-time associated adjustment item for gypsum board production equipment obtained by fusing multiple real-time associated networks includes: In a single fused real-time companion network, all topology nodes except those corresponding to the real-time gypsum board production equipment adjustment items are considered as real-time companion adjustment items for gypsum board production equipment. The node value of the topology node corresponding to the real-time accompanying adjustment item of the gypsum board production equipment in a single fused real-time accompanying network is used as the adjustment amount of the real-time accompanying adjustment item of the gypsum board production equipment.

Images