CN-116820048-B - Defect detection avoiding method based on gypsum board production full data acquisition

CN116820048BCN 116820048 BCN116820048 BCN 116820048BCN-116820048-B

Abstract

The invention discloses a defect detection avoidance method based on full data acquisition of gypsum board production, which comprises the following steps of extracting the associated relation among gypsum board production equipment adjustment items through time sequence analysis of adjustment quantity of the gypsum board production equipment adjustment items, abstracting the associated relation into an associated network among the gypsum board production equipment adjustment items, and determining the real-time setting of production working conditions of gypsum board production equipment according to the number of the gypsum board real-time production equipment adjustment items. According to the invention, the characterization data of the gypsum board defect items and the adjustment quantity of the adjustment items of the gypsum board production equipment are learned and trained to obtain the defect adjustment measuring and calculating model, the time sequence analysis of the adjustment quantity of the adjustment items of the gypsum board production equipment is carried out to extract the associated relation among the adjustment items of the gypsum board production equipment, and the associated relation is abstracted into an associated network among the adjustment items of the gypsum board production equipment, so that the full detection and the full avoidance of the potential defects of the gypsum board are realized, and the initiative of defect detection and avoidance is improved.

Inventors

YANG XIAODONG
FENG XIANLIANG
WU JIANJIANG
DONG LEI

Assignees

中建材创新科技研究院有限公司
北新集团建材股份有限公司
中国建材集团有限公司

Dates

Publication Date: 20260512
Application Date: 20230628

Claims (7)

1. A defect detection and avoidance method based on gypsum board production full data acquisition is characterized by comprising the following steps: Step S1, pertinently matching a gypsum board defect item with a gypsum board production equipment adjustment item; S2, based on the pertinence matching relation between the gypsum board defect items and the adjustment items of the gypsum board production equipment in the S1, learning and training the characterization data of the gypsum board defect items and the adjustment amount of the adjustment items of the gypsum board production equipment in a full data acquisition log of gypsum board production to obtain a defect adjustment measuring and calculating model; s3, in a full data acquisition log of gypsum board production, carrying out time sequence analysis of adjustment quantity on adjustment items of gypsum board production equipment to extract associated relations among the adjustment items of the gypsum board production equipment, and abstracting the associated relations into associated networks among the adjustment items of the gypsum board production equipment; S4, extracting characterization data of real-time defect items of the gypsum board from real-time gypsum board production total data obtained through real-time monitoring, and inputting the characterization data of the real-time defect items of the gypsum board into a defect adjustment measuring and calculating model to obtain adjustment quantity of adjustment items of real-time production equipment of the gypsum board; S5, counting the number of adjustment items of the real-time gypsum board production equipment, and determining the real-time setting of the production working conditions of the gypsum board production equipment according to the number of adjustment items of the real-time gypsum board production equipment so as to realize the full detection and full avoidance of potential defects of the gypsum board; Extracting associated relationships among gypsum board production equipment adjustment items by time sequence analysis of adjustment amounts of the gypsum board production equipment adjustment items, comprising: sequencing the adjustment amount of each gypsum board production equipment adjustment item according to the time sequence to obtain an adjustment amount time sequence of the gypsum board production equipment adjustment item; carrying out sequence similarity analysis on the time sequence of the adjustment amount of each gypsum board production equipment adjustment item to quantify the association rate representing the association relation among the gypsum board production equipment adjustment items, wherein the association relation is the degree representation of the mutual influence among the gypsum board production equipment adjustment items; the accompanying rate quantization formula is: ; Wherein H ij is the associated rate between the gypsum board production equipment adjustment item i and the gypsum board production equipment adjustment item j, W i,T is the adjustment amount time sequence of the gypsum board production equipment adjustment item i, W j,T is the adjustment amount time sequence of the gypsum board production equipment adjustment item j, |W i,T -W j,T | is the Euclidean distance of W i,T and W j,T , i, j is the identifier of the gypsum board production equipment adjustment item, and T is the time sequence identifier; abstracting the associated relationship into an associated network among adjustment items of gypsum board production equipment, comprising: abstracting adjustment items of gypsum board production equipment into topological nodes, and setting topological edges among the topological nodes to perform topological connection; the associated rate among adjustment items of gypsum board production equipment is used as the edge weight of the topological edge among the corresponding topological nodes so as to be abstracted into an associated network; Determining a real-time setting of production conditions of the gypsum board production facility according to the number of adjustment items of the gypsum board real-time production facility, comprising: If the number of adjustment items of the real-time gypsum board production equipment is 0, maintaining the current production working condition setting of the gypsum board production equipment; If the number of the adjustment items of the real-time gypsum board production equipment is 1, mapping the adjustment quantity of the adjustment items of the real-time gypsum board production equipment into an associated network to obtain a single real-time associated network, obtaining the adjustment quantity of the real-time associated adjustment items of the gypsum board production equipment in the single real-time associated network, and setting the production working condition of the gypsum board production equipment in real time according to the adjustment quantity of the adjustment items of the real-time gypsum board production equipment and the adjustment quantity of the real-time associated adjustment items of the gypsum board production equipment so as to realize the full detection and the full avoidance of potential defects of the gypsum board; If the number of the adjustment items of the gypsum board production equipment is larger than 1, mapping the adjustment quantity of the adjustment items of each gypsum board real-time production equipment into the associated network to obtain a plurality of real-time associated networks, fusing the plurality of real-time associated networks to obtain the adjustment quantity of the adjustment items of the gypsum board production equipment, and setting the production working conditions of the gypsum board production equipment in real time according to the adjustment quantity of the adjustment items of the gypsum board real-time production equipment and the adjustment quantity of the adjustment items of the real-time associated adjustment items of the gypsum board production equipment so as to realize full detection and full avoidance of potential defects of the gypsum board.
2. The method for avoiding defect detection based on full data acquisition in gypsum board production according to claim 1, wherein the step of pertinently matching the gypsum board defect item with the adjustment item of gypsum board production equipment comprises the steps of: the defect items of the gypsum board comprise the width of the gypsum board, the vertical edge angle of the gypsum board, the thickness of the gypsum board, the single wet gypsum board, the single dry gypsum board, the temperature after the gypsum board is dried and the humidity after the gypsum board is dried; the adjusting items of the gypsum board production equipment comprise a gypsum board forming cutter distance, a gypsum board forming cutter angle, a gypsum board forming board height, a gypsum board dry material blanking amount, a gypsum board wet material blanking amount and a gypsum board dryer flashboard temperature; the adjustment of the distance between the gypsum board forming cutters and the angle of the gypsum board forming cutters specifically avoids the defects of the width of the gypsum board and the vertical edge angle of the gypsum board; The height adjustment of the gypsum board forming plate specifically avoids the defect of the thickness of the gypsum board; the adjustment of the blanking amount of the dry gypsum board and the blanking amount of the wet gypsum board pertinently avoids the defects of single wet gypsum board and single dry gypsum board; the temperature of the flashboard of the gypsum board dryer is adjusted to pertinently avoid the defects of the temperature and humidity of the gypsum board after being dried.
3. The method for avoiding defect detection based on full data acquisition in gypsum board production according to claim 2, wherein the learning and training of the characterization data of the gypsum board defect item and the adjustment amount of the adjustment item of gypsum board production equipment to obtain the defect adjustment measuring and calculating model comprises the following steps: extracting characterization data of a gypsum board defect item and adjustment quantity of adjustment items of gypsum board production equipment from a full data acquisition log of gypsum board production; taking the characterization data of the defect term of the gypsum board as an input term of the BP neural network, and taking the adjustment quantity of the gypsum board production equipment as an output term of the BP neural network; Performing convolution learning on an input item and an output item of the BP neural network by using the BP neural network to obtain a defect adjustment measuring and calculating model; The model expression of the defect adjustment measuring and calculating model is as follows: Wr=BP(Ds); Wherein Wr is the adjustment amount of an adjustment item r of gypsum board production equipment, ds is characterization data of a gypsum board defect item s, BP is BP neural network, r is an identifier of the adjustment item of gypsum board production equipment, and s is an identifier of the gypsum board defect item; The full data of gypsum board production is composed of the combination of the characterization data of the width of the gypsum board, the vertical edge angle of the gypsum board, the thickness of the gypsum board, the wet single gypsum board, the dry single gypsum board, the temperature after the gypsum board is dried and the humidity after the gypsum board is dried; The gypsum board production full data acquisition log is a log formed by acquiring gypsum board production full data according to time sequence; The extraction of the characterization data of the defect item of the gypsum board comprises the following steps: In a full data acquisition log of gypsum board production, comparing the corresponding item similarity of the characterization data of the gypsum board width, the gypsum board vertical angle, the gypsum board thickness, the gypsum board wet board single, the gypsum board dry board single, the gypsum board post-drying temperature and the gypsum board post-drying humidity at each time sequence with the standard data of the gypsum board width, the gypsum board vertical angle, the gypsum board thickness, the gypsum board wet board single, the gypsum board dry board single, the gypsum board post-drying temperature and the gypsum board post-drying humidity, and taking the corresponding item of the characterization data at each time sequence with the standard data with the similarity smaller than an allowable threshold value as a gypsum board defect item at each time sequence; Extracting characterization data of gypsum board defect items at each time sequence from a full data acquisition log of gypsum board production; extraction of adjustment amounts of adjustment items of the gypsum board production apparatus, comprising: In the full data acquisition log of gypsum board production, determining gypsum board production equipment adjustment items which are pertinently matched with the gypsum board defect items at each time sequence by using the pertinence matching relation; and extracting the adjustment quantity of adjustment items of gypsum board production equipment, which are pertinently matched with the defect items of the gypsum board, at each time sequence in the full data acquisition log of gypsum board production.
4. A method for avoiding defect detection based on full data acquisition of gypsum board production according to claim 3, wherein mapping the adjustment amount of the adjustment item of the gypsum board real-time production equipment into the associated network to obtain a single real-time associated network comprises: And mapping the adjustment quantity of the adjustment items of the gypsum board real-time production equipment into node values of corresponding topological nodes of the associated network, and calculating the node values of the rest topological nodes in the associated network one by one according to the edge weights of the topological nodes and the topological edges to obtain a single real-time associated network.
5. The method for avoiding defect detection based on full data acquisition for gypsum board production according to claim 4, wherein the obtaining the adjustment amount of real-time associated adjustment items of gypsum board production equipment in a single real-time associated network comprises: Taking topological nodes except for the topological nodes corresponding to the adjustment items of the real-time gypsum board production equipment in the single real-time associated network as real-time associated adjustment items of the gypsum board production equipment; and taking the node value of the topological node corresponding to the real-time associated adjustment item of the gypsum board production equipment in the single real-time associated network as the adjustment quantity of the real-time associated adjustment item of the gypsum board production equipment.
6. The method for avoiding defect detection based on full data collection for gypsum board production according to claim 5, wherein the sequentially mapping the adjustment amount of each adjustment item of real-time gypsum board production equipment to the associated network to obtain a plurality of real-time associated networks comprises: The adjustment amount of the adjustment item of each gypsum board real-time production equipment is mapped into the node value of the corresponding topological node of each associated network in sequence, the node value of the rest topological nodes in each associated network is calculated one by one according to the edge weights of the topological nodes and the topological edges, and a plurality of real-time associated networks are obtained; And superposing and fusing the node values of the plurality of real-time associated networks to obtain a single fused real-time associated network.
7. The method for avoiding defect detection based on full data acquisition for gypsum board production according to claim 6, wherein the merging of the plurality of real-time associated networks to obtain the adjustment amount of the real-time associated adjustment item of the gypsum board production equipment comprises: Taking all topological nodes except for all topological nodes corresponding to the adjustment items of the gypsum board real-time production equipment in the single fusion real-time associated network as real-time associated adjustment items of the gypsum board production equipment; And taking the node value of the topological node corresponding to the real-time associated adjustment item of the gypsum board production equipment in the single fusion real-time associated network as the adjustment quantity of the real-time associated adjustment item of the gypsum board production equipment.

Description

Defect detection avoiding method based on gypsum board production full data acquisition Technical Field The invention relates to the technical field of defect detection, in particular to a defect detection avoidance method based on gypsum board production full data acquisition. Background The gypsum board has the advantages that various problems such as board damage and board damage angle can occur in the processes of production forming, cutting off, internal transportation of a dryer and the like, the problem board is mainly judged through naked eye observation of post personnel, and sometimes, the problem board is in danger of entering stacking due to the fact that other work busy in the post does not find out external defects of individual products in time, so that quality control of the products is not facilitated, and the post is required to select and stack the whole stacking board by using forklift forks, so that the workload is large. Therefore, an intelligent automatic defect monitoring and defect avoidance mode is needed to monitor and avoid defects on the surface shape and the temperature and humidity of the gypsum board, the existing intelligent automatic defect monitoring mode generally carries out defect detection and defect avoidance on one surface of the gypsum board, then the gypsum board is turned over, and then defect monitoring and defect avoidance are carried out on the other surface of the gypsum board. Disclosure of Invention The invention aims to provide a defect detection and avoidance method based on full data acquisition in gypsum board production, which aims to solve the technical problems that passive defect detection and defect avoidance are adopted in the prior art, gypsum board defects need to be detected one by one and avoided one by one, and the potential defects are difficult to predictively monitor and avoid. In order to solve the technical problems, the invention specifically provides the following technical scheme: a defect detection avoiding method based on gypsum board production full data acquisition comprises the following steps: Step S1, pertinently matching a gypsum board defect item with a gypsum board production equipment adjustment item; S2, based on the pertinence matching relation between the gypsum board defect items and the adjustment items of the gypsum board production equipment in the S1, learning and training the characterization data of the gypsum board defect items and the adjustment amount of the adjustment items of the gypsum board production equipment in a full data acquisition log of gypsum board production to obtain a defect adjustment measuring and calculating model; s3, in a full data acquisition log of gypsum board production, carrying out time sequence analysis of adjustment quantity on adjustment items of gypsum board production equipment to extract associated relations among the adjustment items of the gypsum board production equipment, and abstracting the associated relations into associated networks among the adjustment items of the gypsum board production equipment; S4, extracting characterization data of real-time defect items of the gypsum board from real-time gypsum board production total data obtained through real-time monitoring, and inputting the characterization data of the real-time defect items of the gypsum board into a defect adjustment measuring and calculating model to obtain adjustment quantity of adjustment items of real-time production equipment of the gypsum board; And S5, counting the number of adjustment items of the real-time gypsum board production equipment, and determining the real-time setting of the production working conditions of the gypsum board production equipment according to the number of adjustment items of the real-time gypsum board production equipment so as to realize the full detection and full avoidance of the potential defects of the gypsum board. As a preferred aspect of the present invention, the matching the defective gypsum board item with the adjustment item of the gypsum board production apparatus includes: the defect items of the gypsum board comprise the width of the gypsum board, the vertical edge angle of the gypsum board, the thickness of the gypsum board, the single wet gypsum board, the single dry gypsum board, the temperature after the gypsum board is dried and the humidity after the gypsum board is dried; the adjusting items of the gypsum board production equipment comprise a gypsum board forming cutter distance, a gypsum board forming cutter angle, a gypsum board forming board height, a gypsum board dry material blanking amount, a gypsum board wet material blanking amount and a gypsum board dryer flashboard temperature; the adjustment of the distance between the gypsum board forming cutters and the angle of the gypsum board forming cutters specifically avoids the defects of the width of the gypsum board and the vertical edge angle of the gypsum board; The height adjustment of the gyp