CN-122016917-A - Internal flaw detection method in high-temperature printing and dyeing process of heavy fabric

CN122016917ACN 122016917 ACN122016917 ACN 122016917ACN-122016917-A

Abstract

The invention discloses an internal flaw detection method in a high-temperature printing and dyeing process of heavy fabrics, which relates to the field of printing and dyeing and is characterized by comprising the following steps of S1, uniformly mixing a specific auxiliary agent into a dye liquor in the printing and dyeing process; S2, establishing a standard database through experiments in advance according to different fabric types, tissue structures and used dye-auxiliary agent systems, S3, after finishing printing and dyeing, removing the fabric from a dye vat, performing padding treatment on the fabric, then smoothly conveying the fabric into a constant-temperature cavity for detection, generating an actually measured temperature drop curve, and S4, comparing the actually measured temperature drop curve acquired by each local detection unit in the step S3 with a standard temperature drop curve model corresponding to the database in the step S2. According to the invention, the difference between the dyed area of the fabric and the undyed area of the fabric in the cooling process is amplified by adding the auxiliary agent, so that the dyeing flaws in the deep part of the fabric can be nondestructively judged through thermal imaging.

Inventors

HOU JIANGHUI
ZHANG JIANYONG
REN LIFENG

Assignees

浙江稽山印染有限公司

Dates

Publication Date: 20260512
Application Date: 20251223

Claims (10)

1. The internal flaw detection method in the high-temperature printing and dyeing process of the heavy fabric is characterized by comprising the following steps of: S1, establishing a standard database in advance through experiments aiming at different fabric types, tissue structures and used dye-auxiliary agent systems, wherein the database stores a standard temperature drop curve model of a fabric sample successfully dyed in a specific constant-temperature cooling environment in a defect-free state, and the standard temperature drop curve model represents the corresponding relation between the unit area temperature of the fabric and time in the process of reducing the temperature from a fixed high temperature to the ambient temperature; S2, uniformly mixing a specific auxiliary agent into the dye liquor in the printing and dyeing process, wherein the auxiliary agent is formed by compounding glycerol and fatty alcohol polyoxyethylene ether; S3, after finishing printing and dyeing, removing the fabric from the dye vat, performing padding treatment on the fabric, and then smoothly conveying the fabric into a constant-temperature cavity for detection; the method comprises the steps of carrying out global scanning monitoring on a tiled fabric by using a non-contact infrared thermal imaging sensor array arranged in a cavity, logically dividing the surface of the fabric into a plurality of local detection units with equal areas, synchronously collecting temperature data of each unit in a cooling process in real time, and generating an actually measured temperature drop curve corresponding to each unit; S4, comparing the measured temperature drop curve acquired by each local detection unit in the step S3 with a standard temperature drop curve model corresponding to the database in the step S1, identifying local units with significant differences between the measured curve and the standard curve based on a preset difference threshold criterion, wherein the significant differences comprise temperature reduction rate differences, temperature value differences at specific time points or integral curve morphology differences, judging the units with significant differences as areas with printing and dyeing flaws, and outputting position information of the units.
2. The method for detecting internal flaws in a high-temperature printing process for heavy fabrics according to claim 1, wherein in the step S4, the difference threshold criterion is dynamically adjusted by an adaptive algorithm, and the adaptive algorithm performs the steps of: a) After the fabric enters a constant temperature environment and begins to cool, calculating standard deviation (sigma) or variation Coefficient (CV) of all effective local detection unit temperature values in a current frame or a specific time period, and taking the unit temperature values as an index U for representing the cooling overall uniformity of the fabric; b) Setting one or more reference temperature measuring points which are not in contact with the fabric in a constant-temperature environment cavity, monitoring the stability of the environment temperature, calculating the fluctuation amplitude of the range (R) or the moving average line (MA) of the temperature value of the reference temperature measuring points in the same time period, and taking the temperature value of the temperature measuring points as an environment fluctuation index E; c) Establishing a predefined dynamic threshold mapping table, wherein the dynamic threshold mapping table defines the corresponding relation between the combination of U and E and the final difference threshold T; d) And (4) applying the calculated real-time dynamic threshold T to a flaw identification algorithm in the step (S4) for judging whether the difference between the actual measurement curve and the standard curve is obvious or not.
3. The method for detecting internal flaws in a high-temperature printing and dyeing process of heavy fabrics according to claim 2, wherein in the step S1, the standard database takes all measured temperature drop curves corresponding to local units which are judged to be flaw-free by the algorithm of the step S4 for each batch of detected fabrics as a potential effective data set, and the process sets a data quality filtering threshold value, and only when the overall uniformity index U of the batch of fabrics is superior to a set standard and the environment fluctuation index E is in a stable range, the automatic input flow of the batch of data is triggered.
4. The method for detecting internal flaws in the high-temperature printing and dyeing process of heavy fabrics according to claim 3, wherein the standard database is provided with a time stamp when data are input, the standard database periodically starts a model updating task, all or latest data in a data set are used as a training set by the task, a machine learning algorithm for generating a standard temperature drop curve model is retrained, and therefore new thermodynamic characteristics reflected by recent normal production are integrated into the latest model, and progressive optimization and iteration of the model are achieved.
5. The method for detecting internal flaws in a high-temperature printing and dyeing process of heavy fabrics according to claim 4, wherein the standard database evaluates the performance of each updated new model and applies the new model to an isolated historical data set for verification, and if the false alarm rate of the new model is significantly higher than that of an old model, the new model automatically rolls back to a stable model of a previous version, thereby ensuring the continuity and reliability of production detection.
6. The method for detecting internal flaws in the high-temperature printing and dyeing process of heavy fabrics according to claim 2, wherein in the step S3, the space coordinate information of each local unit is synchronously collected, in the step S4, after the abnormal units are identified, the abnormal units are spatially clustered by adopting an image processing algorithm based on connected domain analysis, and the abnormal areas which are continuously distributed in space are combined into the same flaw area for marking and outputting, so that the sporadic false detection is eliminated, and the sizes and the shapes of flaws are accurately described.
7. The method for detecting internal flaws in a high-temperature printing and dyeing process of heavy fabrics according to claim 6, wherein the step S4 is performed to output flaw position information and a confidence score related to the judgment result, and the confidence score integrates the factors such as the significance degree of curve difference, the continuity with peripheral units, and historical detection data.
8. The method for detecting internal flaws in high-temperature printing and dyeing process of heavy fabrics according to claim 1, wherein in the step S2, the addition amount of glycerol is 1.5-3% of the weight of the dye liquor, and the addition amount of fatty alcohol-polyoxyethylene ether is 0.2-0.5% of the weight of the dye liquor.
9. The method for detecting internal flaws in a high-temperature printing process for heavy fabrics according to claim 1, wherein in S3, the initial temperature of the fabric is at least 30 ℃ higher than the ambient temperature of the chamber when the fabric is transported to the chamber, so as to ensure that a sufficiently significant temperature drop process is obtained for analysis.
10. The method for detecting internal flaws in high-temperature printing and dyeing of heavy fabrics according to claim 9, wherein in S3, two adjacent batches of fabrics transported into the chamber are separated by at least 3 minutes to ensure complete resetting of the temperature of the chamber.

Description

Internal flaw detection method in high-temperature printing and dyeing process of heavy fabric Technical Field The invention relates to the field of printing and dyeing, in particular to an internal flaw detection method in the high-temperature printing and dyeing process of heavy fabrics. Background In the textile printing and dyeing industry field, heavy fabrics are often hidden in the fabrics due to compact fabric structure and large thickness, internal flaws such as uneven dye permeation, partial undyed or abnormal combination of the inside of the fibers generated in the printing and dyeing process, at present, common flaw detection techniques are mostly based on image recognition or spectrum analysis, and the methods are more effective for recognizing surface designs, stains or obvious chromatic aberration, but cannot be used for internal permeation conditions, deep-level problems and the like, and the hidden flaws can have the problems of white exposure after stretching, light transmittance difference and the like in the subsequent use process. There is therefore a need to propose a new solution to this problem. Disclosure of Invention Aiming at the defects existing in the prior art, the invention aims to provide an internal flaw detection method in the high-temperature printing and dyeing process of heavy fabrics. The technical aim of the invention is realized by the following technical scheme that the internal flaw detection method in the high-temperature printing and dyeing process of the heavy fabric comprises the following steps: S1, establishing a standard database in advance through experiments aiming at different fabric types, tissue structures and used dye-auxiliary agent systems, wherein the database stores a standard temperature drop curve model of a fabric sample successfully dyed in a specific constant-temperature cooling environment in a flawless state, and the standard temperature drop curve model represents the corresponding relation between the temperature of a unit area of the fabric and time in the process of reducing the temperature from a fixed high temperature to the ambient temperature; S2, uniformly mixing a specific auxiliary agent into the dye liquor in the printing and dyeing process, wherein the auxiliary agent is formed by compounding glycerol and fatty alcohol-polyoxyethylene ether; S3, after finishing printing and dyeing, removing the fabric from the dye vat, performing padding treatment on the fabric, and then smoothly conveying the fabric into a constant-temperature cavity for detection; the method comprises the steps of carrying out global scanning monitoring on a tiled fabric by using a non-contact infrared thermal imaging sensor array arranged in a cavity, logically dividing the surface of the fabric into a plurality of local detection units with equal areas, synchronously collecting temperature data of each unit in a cooling process in real time, and generating an actually measured temperature drop curve corresponding to each unit; S4, comparing the measured temperature drop curve acquired by each local detection unit in the step S3 with a standard temperature drop curve model corresponding to the database in the step S1, identifying local units with significant differences between the measured curve and the standard curve based on a preset difference threshold criterion, wherein the significant differences comprise temperature reduction rate differences, temperature value differences at specific time points or integral curve morphology differences, judging the units with significant differences as areas with printing and dyeing flaws, and outputting position information of the units. The invention is further arranged that in the S4, the difference threshold criterion is dynamically adjusted by an adaptive algorithm, and the adaptive algorithm executes the following steps: a) After the fabric enters a constant temperature environment and begins to cool, calculating a standard deviation (sigma) or a variation Coefficient (CV) of temperature values of all effective local detection units in a current frame or a specific time period, and taking the value as an index U for representing the overall uniformity of cooling of the fabric. B) One or more reference temperature measuring points which are not contacted with the fabric are arranged in the constant temperature environment cavity, the stability of the environment temperature is monitored, the fluctuation amplitude of the range (R) or the moving average line (MA) of the temperature value of the reference temperature measuring points in the same time period is calculated, and the value is used as an environment fluctuation index E. C) A predefined dynamic threshold mapping table is established, which defines the correspondence between the combination of U and E and the final difference threshold T. D) And (4) applying the calculated real-time dynamic threshold T to a flaw identification algorithm in the step (S4) for