CN-121978956-A - Textile dyeing control system for spinning

CN121978956ACN 121978956 ACN121978956 ACN 121978956ACN-121978956-A

Abstract

The invention relates to the technical field of textile printing and dyeing, and particularly discloses a textile dyeing control system for textile, which comprises a collaborative collection module, a benchmark matching module, a deviation compensation module and an evaluation evolution module, wherein a multi-sensor collaborative collection unit is constructed, key material fingerprints and key process variables are collected, a benchmark process track is set for a current dyeing batch by matching similar optimal batches in a historical database based on the key material fingerprints, a lightweight real-time digital twin body is constructed and adopted, a dyeing real-time state track is formed by simulation prediction based on the intercepted current actual measurement track, track synchronous comparison and systematic deviation diagnosis are carried out with the benchmark process track, if systematic deviation is judged, a dynamic differential compensation process is triggered to generate an optimal control strategy, if dyeing batch is finished, dyeing quality is evaluated, if the dyeing batch is qualified, the dyeing batch is classified into the historical database, and a contribution degree analysis process is started based on the optimal control strategy to update the historical database.

Inventors

Zheng Guancheng
CHEN QIWEI
FAN JIANLING
ZHOU XIAOPING

Assignees

浙江齐越新材料有限公司

Dates

Publication Date: 20260505
Application Date: 20260210

Claims (10)

1. A textile dyeing control system for spinning is characterized by comprising the following modules: the collaborative collection module is used for constructing a multi-sensing collaborative collection unit and collecting key material fingerprints and key process variables; the reference matching module is used for matching similar optimal batches in the historical database based on the key material fingerprints and setting a reference process track for the current dyeing batch; the deviation compensation module is used for constructing and adopting a lightweight real-time digital twin body, carrying out simulation prediction based on the intercepted current actual measurement track to form a dyeing real-time state track, carrying out track synchronous comparison and systematic deviation diagnosis with a reference process track, and triggering a dynamic differential compensation process to generate an optimal control strategy if systematic deviation is judged to occur; And the evaluation evolution module is used for evaluating the dyeing quality if the dyeing batch is completed, classifying the dyeing quality into a historical database if the dyeing batch is qualified, and starting a contribution analysis flow based on an optimal control strategy to update the historical database.
2. The dyeing control system for textile fabrics according to claim 1, wherein the reference process track is obtained by the following steps: the historical database stores dyeing data records of dyeing batches with qualified quality evaluation, wherein the dyeing data records comprise key material fingerprints, process variable tracks and dyeing quality evaluation indexes of the dyeing batches; calculating the similarity score between the current key material fingerprint and each key material fingerprint in the historical database by adopting a weighted mahalanobis distance algorithm, classifying the dyeing batches with the similarity score reaching or exceeding a preset similarity threshold as high-similarity candidate sets, marking the dyeing batch with the smallest final color difference delta E value in the dyeing quality evaluation index of each dyeing batch in the high-similarity candidate sets as a similar optimal batch, and setting the corresponding process variable track as the reference process track of the current dyeing batch.
3. The dyeing control system for textile fabrics according to claim 2, wherein the similarity score is calculated by the following steps: Z-score standardized preprocessing is carried out on key material fingerprints of a current dyeing batch and key material fingerprints recorded by all dyeing data in a historical database, and based on a weighted Markov distance algorithm, the Markov distance between the current key material fingerprint and each key material fingerprint in the historical database is calculated by combining an inverse matrix of a covariance matrix obtained by calculating all key material fingerprints in the historical database, and the similarity score between 0 and 1 is mapped through a Gaussian kernel function.
4. The dyeing control system for textile fabrics according to claim 1, wherein the real-time dyeing state track is obtained by the following steps: And continuously inputting the current actual measurement track into a lightweight real-time digital twin body taking an embedded reduced dynamics model as a calculation core, wherein the reduced dynamics model is a transfer function set obtained by training by a system identification method based on the process variable track in a historical database, and carrying out forward rolling simulation prediction by taking the current actual measurement track as an initial condition, outputting a predicted evolution track of each key process variable in a short time domain in the future, and jointly forming a dyeing real-time state track by the current actual measurement track and the predicted evolution track.
5. A textile dyeing control system according to claim 4, characterized in that the systematic deviation diagnosis is as follows: And (3) carrying out nonlinear alignment on the dyeing real-time state track and the reference process track by adopting a dynamic time warping algorithm, calculating the instantaneous deviation of corresponding values on the dyeing real-time state track and the reference process track and the accumulated deviation integral and the change trend of the instantaneous deviation in a sliding time window on the dyeing real-time state track aiming at each key process variable, judging that the key process variable is subjected to trend deviation when any key process variable meets a preset trend deviation condition rule, and judging that the key process variable is subjected to trend deviation if two or more key process variables with strong coupling relation are disclosed based on a reduced order dynamics model and are subjected to trend deviation simultaneously, and judging that the systematic deviation occurs when the deviation direction is consistent with the disclosed strong coupling relation logic.
6. The textile fabric dyeing control system of claim 5, wherein the trend deviation condition rules include an amplitude condition and a trend condition: The amplitude condition is that the absolute value of the instantaneous deviation continuously exceeds the corresponding preset static allowable deviation for a preset deviation duration, the trend condition is that the signs of the change trend are consistent on a continuous preset sliding time window, and the absolute value of the accumulated deviation integral exceeds a preset dynamic deviation threshold.
7. The textile dyeing control system according to claim 1, wherein the optimal control strategy is generated by: If systematic deviation occurs, recording the systematic deviation state as an initial condition, performing forward rolling simulation in a lightweight real-time digital twin body under a plurality of future tentative control strategies which are generated based on a quadratic programming algorithm and aim at main regulating equipment, predicting evolution paths of key process variables under different control strategies and comprehensive deviation values of the evolution paths and the reference process tracks, constructing a multi-objective optimization problem, and solving the multi-objective optimization problem by adopting a quadratic programming algorithm under the condition that physical limits and process safety constraints of the main regulating equipment are met, so as to obtain an optimal control strategy in an optimization time domain.
8. A textile dyeing control system as set forth in claim 7, wherein the multi-objective optimization problem comprises a primary objective function and a secondary objective function, the primary objective function being designed to minimize the predicted integrated deviation value, the secondary objective function being designed to constrain the amplitude and frequency of the motion of the backbone regulating device.
9. The textile dyeing control system according to claim 8, wherein the comprehensive deviation value is calculated by: Calculating the difference between the predicted value of each moment in the optimization time domain and the corresponding value of the reference process track for each key process variable, normalizing based on a preset normalization factor to obtain normalized deviation, giving a quality weight coefficient to the normalized deviation of each key process variable by adopting a analytic hierarchy process to calculate a weighted deviation square, and introducing a time discount factor to accumulate and sum the weighted deviation squares of each moment of each key process variable along the time axis of the optimization time domain to obtain a comprehensive deviation value.
10. The textile dyeing control system according to claim 1, wherein the history database is updated by: And constructing qualified key material fingerprints, process variable tracks and dyeing quality assessment indexes of the current dyeing batch into dyeing data records, storing the dyeing data records in a historical database, comparing the process variable tracks under the intervention of the optimal control strategy of the current dyeing batch with reference process tracks by adopting a SHAP value analysis algorithm, quantifying contribution degree scores of control actions of all trunk regulating and controlling equipment on the dyeing quality assessment indexes, associating the calculated contribution degree scores with corresponding control actions, and updating metadata fields of the corresponding dyeing data records in the historical database.

Description

Textile dyeing control system for spinning Technical Field The invention relates to the technical field of textile printing and dyeing, in particular to a textile dyeing control system for textile. Background In the textile printing and dyeing industry, the dyeing process is a core link for determining the final appearance quality and the wearability of textiles, and the process control level is directly related to key quality indexes such as color difference, uniformity, color fastness and the like of products. Traditional dyeing control is highly dependent on experience of operators, and is adjusted by setting a fixed temperature and time curve and assisting manual sampling comparison. In the method, when the interference of fabric raw material difference, dye batch fluctuation, equipment state change and the like is faced, response is delayed, adjustment is rough, quality problems such as batch-to-batch chromatic aberration and cylinder difference are easily caused, and repair and energy and raw material waste are caused. In order to solve the above problems, an online monitoring and automatic control system is generally introduced, and the real-time acquisition of process parameters and the automatic control of a partial loop are realized by installing sensors such as temperature, pH, flow and the like, however, such a system is mostly limited to the monitoring and adjustment of single or few isolated variables, and lacks of overall cognition and collaborative optimization of the internal multivariable strong coupling and nonlinear dynamic characteristics in the dyeing process. When multiple parameters deviate at the same time, the system can only alarm, and the root cause is difficult to automatically diagnose and a comprehensive regulation strategy is provided. Further analysis finds that the prior art scheme has common perception and decision fracture, the collected data is not deeply related to the essential factors for determining the dyeing result, the setting of the control reference is still based on a fixed formula or rough classification, and the individuation and accurate initial setting can not be carried out according to the actual state of each dye cylinder and the specific characteristics of the fabric. On the other hand, the control and prediction are disjoint, the advanced prediction and rolling optimization capability based on a process dynamic model is lacked, the predictive intervention cannot be performed before the quality deviation actually occurs, experience and knowledge are solidified, and the intelligent iterative evolution of the system is difficult. In view of the above problems, the present invention provides a textile dyeing control system for textile. Disclosure of Invention The invention aims to provide a textile fabric dyeing control system for spinning, which solves the background problem. The invention aims at realizing the technical scheme that the textile dyeing control system for spinning comprises the following modules: the collaborative collection module is used for constructing a multi-sensing collaborative collection unit and collecting key material fingerprints and key process variables; the reference matching module is used for matching similar optimal batches in the historical database based on the key material fingerprints and setting a reference process track for the current dyeing batch; the deviation compensation module is used for constructing and adopting a lightweight real-time digital twin body, carrying out simulation prediction based on the intercepted current actual measurement track to form a dyeing real-time state track, carrying out track synchronous comparison and systematic deviation diagnosis with a reference process track, and triggering a dynamic differential compensation process to generate an optimal control strategy if systematic deviation is judged to occur; The evaluation evolution module is used for evaluating the dyeing quality if the dyeing batch is completed, classifying the dyeing quality into a historical database if the dyeing batch is qualified, and starting a contribution analysis flow based on an optimal control strategy to update the historical database; Further, the reference process track is obtained by the following steps: the historical database stores dyeing data records of dyeing batches with qualified quality evaluation, wherein the dyeing data records comprise key material fingerprints, process variable tracks and dyeing quality evaluation indexes of the dyeing batches; Calculating the similarity score between the current key material fingerprint and each key material fingerprint in the historical database by adopting a weighted Markov distance algorithm, classifying the dyeing batches with the similarity score reaching or exceeding a preset similarity threshold as high-similarity candidate sets, marking the dyeing batch with the smallest final color difference delta E value in the dyeing quality evaluation index of