A dynamic adaptive optimization system and method for grey fabric dyeing process

Abstract

The present application relates to the technical field of dyeing process optimization, and particularly relates to a dynamic self-adaptive optimization system and method for grey cloth dyeing process, which specifically comprises a process dynamic management center, a process data module, a processing and feature extraction module, a process diagnosis module, an early diagnosis module, a dynamic adjustment module, a dynamic verification module and a back-end visual module; through collaborative analysis of actual dyeing rate and dyeing synchronism index, many dominant states are accurately identified, time series linear regression analysis is introduced to early warn potential risks of level dyeing, extension from dominant problem identification to implicit risk prediction is realized, accuracy of process state judgment is ensured, through early diagnosis analysis, potential problems such as expected rate abnormality and expected level dyeing deterioration are identified in advance, and through a precise matching mode of diagnosis result-adjustment scheme-index tracking, fine and quantitative regulation and control of process parameters are realized, and adjustment efficiency and effect are improved.

Description

Technical Field

[0001] This invention relates to the field of dyeing process optimization technology, and in particular to a dynamic adaptive optimization system and method for dyeing fabric. Background Technology

[0002] In the textile dyeing industry, the level of process control directly determines the stability of product quality and production efficiency. Greige fabric dyeing is a key process in textile production, and its technical quality directly affects the color consistency, color fastness, and hand feel of the final product. Traditional greige fabric dyeing processes typically rely on pre-set process formulas (such as temperature, time, auxiliary agent concentration, pH value, etc.) and the experience of operators for control, which presents many pain points that urgently need to be addressed. These are specifically manifested as follows: In terms of condition diagnosis, there is a lack of systematic indicator system. The focus is only on the final dyeing result, while neglecting the trend analysis of process parameters. It is difficult to identify potential risks such as delayed dyeing, excessive dyeing, and deterioration of evenness in advance, often falling into the dilemma of "discovery after the fact and passive remediation". In terms of process adjustment, the direction and extent of adjustment are mostly determined by the subjective experience of operators, lacking quantitative basis. The scientific nature and pertinence of the adjustment are insufficient, and the situation of "the more you adjust, the more it deviates" is easy to occur, resulting in increased fabric loss. Moreover, there is no perfect closed-loop verification mechanism. Whether the process status is stable after adjustment depends entirely on experience. Quality fluctuations occur repeatedly, reducing production efficiency.

[0003] To address the aforementioned technical shortcomings, a solution is proposed. Summary of the Invention

[0004] The purpose of this invention is to provide a dynamic adaptive optimization system and method for fabric dyeing process. It is a full-process design of data collection, diagnosis, prediction, control and verification. On the one hand, it improves the uniformity and pass rate of dyed products and enhances product quality competitiveness through precise control and stable monitoring. At the same time, it shortens the process adjustment cycle, reduces the time for post-process remediation and improves production efficiency. On the other hand, it reduces fabric loss and waste of auxiliary materials such as leveling agents through pre-control, and achieves effective control of production costs, thereby solving the above-mentioned technical defects.

[0005] The objective of this invention can be achieved through the following technical solution: a dynamic adaptive optimization system for fabric dyeing process, comprising a process dynamic management center, a process data module, a processing and feature extraction module, a process diagnosis module, an early diagnosis module, a dynamic adjustment module, a dynamic verification module, and a back-end visualization module; The process data module is used to retrieve the baseline process curve from the preset standard process library according to the fabric type and target color number, and obtain key parameters in real time. The processing and feature extraction module is used to preprocess key parameters to obtain standardized parameters, and at the same time extract key feature parameters from the standardized parameters, and send the key feature parameters to the process dynamic management center for storage. The process diagnostic module is used to acquire indicators and diagnose the status of key characteristic parameters, and obtain a list of diagnostic results. The early diagnosis module is used to collect historical data over a recent period of time, and to perform early prediction and diagnosis of indicators based on the historical data, as well as fusion output analysis, and output a list of diagnosis results or early diagnosis results. The dynamic adjustment module is used to perform dynamic adaptive adjustment analysis on the list of diagnostic results and early diagnostic results, and output adaptive adjustment decisions. The dynamic verification module is used to perform dynamic verification analysis on the core indicators after the acquisition, adaptive adjustment decision-making and adjustment are completed, and output the instruction of process status stabilization or deviation.

[0006] Preferably, the analysis process of the process diagnostic module is as follows: In the main staining stage, based on the extracted key feature parameters, two key process indicators are obtained: actual staining rate and staining synchronicity index. The calculated actual dyeing rate and dyeing synchronicity index are compared with the target dyeing rate range [SRmin, SRmax] and synchronicity threshold preset in the current process stage to diagnose whether the current process state deviates from the expectation, and obtain the dyeing lag state, dyeing over-excitation state, uniformity deterioration alarm state, uniformity potential risk warning state and uniformity improvement results. A list of diagnostic results is constructed based on the states of delayed staining, excessive staining, alarm state of deteriorated uniformity, early warning state of potential risk to uniformity, and state of improved uniformity.

[0007] Preferably, the actual staining rate analysis process is as follows: Dyeing rate v1 = (current average fabric color depth value minus the average fabric color depth value of the previous diagnostic cycle) divided by diagnostic cycle duration; Dyeing rate v2 = (relative concentration of dye liquor in the previous diagnostic cycle minus the current relative concentration of dye liquor) divided by the calculated diagnostic cycle. Obtain the preset weight coefficients corresponding to the dyeing rate v1 and the dyeing rate v2, and calculate the actual dyeing rate V based on the dyeing rate v1 × the corresponding preset weight coefficient + the dyeing rate v2 × the corresponding preset weight coefficient. Dyeing synchronicity index = fabric color depth distribution dispersion ÷ (α × current average fabric color depth value + β), where α is a preset adjustment coefficient, 0 < α < 1, and β is a preset adjustment coefficient, 0 < β < 1.

[0008] Preferably, the analysis process of the early diagnosis module is as follows: Using the most recent historical data as the basis for prediction, and employing a pre-defined prediction model, we make single-step predictions for the actual dyeing rate and the dyeing synchronicity index after a future ΔT time (prediction duration). Based on the pre-set prediction model, two predicted values ​​are obtained: predicted dyeing rate and predicted synchronicity index. Early diagnostic analysis is performed on the predicted dyeing rate and predicted synchronicity index to obtain expected rate anomalies (abnormality direction is too high or too low), expected deterioration of uniformity, and confirmation and aggravation of uniformity risk. Early diagnostic results are constructed based on anticipated rate anomalies (expected too high / expected too low), anticipated homogenization deterioration, and homogenization risk confirmation and exacerbation.

[0009] Preferably, based on the fusion analysis of the diagnostic result list and the early diagnostic results, if only the diagnostic result list is available, the diagnostic result list is output; if the early diagnostic results exist, the early diagnostic results are output first.

[0010] Preferably, the analysis process of the dynamic adjustment module is as follows: T1: Based on the list of diagnostic results and early diagnostic results, if the staining is delayed or the expected rate is abnormal (expected too low), the temperature adjustment range and the heating rate adjustment range are obtained. If the predicted dyeing rate still fails to meet the target after adjusting the temperature and heating rate, adjust the pH value as follows: pH adjustment range = preset pH sensitivity coefficient × (minimum target dyeing rate SRmin - current dyeing rate). T2: If the dyeing is overexcited / the expected rate is abnormal (expected too high), the adjustment range of the heating rate is obtained; in the segmented feeding process, the proportion of the dye replenishment interval extension or the proportion of the single feeding amount reduction is obtained. T3: If it is a leveling degradation alarm state / expected leveling degradation, the circulation flow rate adjustment range, stirring rate adjustment range, and leveling agent replenishment amount are obtained. T4: If the risk of uneven dyeing is confirmed and aggravated, then on the basis of T2 above, additional heating compensation will be turned on to make the temperature difference between different areas in the dyeing machine ≤ the preset temperature value. T5: If it is a potential risk warning state for uniformity, then obtain the circulation flow rate increase ratio. If the current temperature change rate exceeds the preset change rate threshold, then reduce the current temperature change rate to below the preset change rate threshold. Adaptive adjustment decisions are constructed based on T1 to T5.

[0011] Preferably, the analysis process of the dynamic verification module is as follows: After the adaptive adjustment decision is completed, the start time is recorded immediately, and the sampling interval is calculated based on that time. Sampling is performed once at a preset interval to obtain the core indicators of each sample (actual dyeing rate, dyeing synchronicity index, and color depth distribution dispersion). The core indicators are then processed to determine whether they are stable or deviating from the standard. If the process is deemed stable after three consecutive tests, it is considered to be in a stable state, and sampling is stopped.

[0012] The beneficial effects of this invention are as follows: (1) This invention extracts core feature parameters through standardized preprocessing and feature extraction, providing high-quality and standardized data support for subsequent diagnosis and control. Through the synergistic analysis of actual dyeing rate and dyeing synchronicity index, it accurately identifies explicit states such as dyeing delay, dyeing over-excitation, and deterioration of uniformity. At the same time, it introduces time series linear regression analysis to capture the trend change of dyeing synchronicity index, provide early warning of potential risks to uniformity, and realize the extension from explicit problem identification to implicit risk prediction, ensuring the accuracy of current process status judgment.

[0013] (2) This invention also identifies potential problems such as expected rate abnormalities and expected deterioration of dyeing level in advance through early diagnostic analysis. In particular, it confirms the trend of the early warning potential risk of dyeing level, forming a risk control link of prediction-verification-upgrade, effectively reducing the loss of greige fabric caused by process deviation and improving the stability of dyeing quality. Through the precise matching mode of diagnostic results-adjustment scheme, it avoids the experience-based and blind nature of traditional process adjustment, realizes the fine and quantitative control of process parameters, improves the adjustment efficiency and effect, and continuously tracks the core indicators. Based on the changes in the indicators, it dynamically optimizes the control strategy and further reduces the risk of subsequent quality fluctuations. Attached Figure Description

[0014] The invention will now be further described with reference to the accompanying drawings; Figure 1 This is a flowchart of the system of the present invention; Figure 2 This is a reference diagram of the method of the present invention; Figure 3 This is a reference diagram for adaptive adjustment decision analysis in this invention. Detailed Implementation

[0015] 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.

[0016] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments; Example 1: Please refer to Figures 1 to 3 As shown, the present invention is a dynamic adaptive optimization system for fabric dyeing process, including a process dynamic management center, a process data module, a processing and feature extraction module, a process diagnosis module, an early diagnosis module, a dynamic adjustment module, a dynamic verification module, and a back-end visualization module; The process dynamic management center has a one-way communication connection with both the process data module and the process diagnosis module. The process data module has a one-way communication connection with the processing and feature extraction module. The processing and feature extraction module has a one-way communication connection with the process dynamic management center. The process diagnosis module has a one-way communication connection with both the early diagnosis module and the dynamic adjustment module. The early diagnosis module has a one-way communication connection with both the dynamic adjustment module and the back-end visualization module. The dynamic adjustment module has a one-way communication connection with the dynamic verification module. The dynamic verification module has a one-way communication connection with the back-end visualization module. The process data module is used to retrieve the baseline process curve from the preset standard process library according to the fabric type and target color number, and to obtain the following key parameters in real time: fabric color parameters, dye liquor state parameters, and process environment parameters. Fabric color parameters are obtained by using an online colorimeter to measure the reflectance or color depth of the fabric at a specific wavelength. Dye liquor state parameters are obtained by using an online spectrometer or concentration sensor to measure the relative concentration of characteristic dyes in the dye liquor. Process environment parameters include liquid temperature, pH value, conductivity, and circulation flow rate. The processing and feature extraction module is used to preprocess key parameters to obtain standardized parameters, and at the same time extract key feature parameters from the standardized parameters, and send the key feature parameters to the process dynamic management center for storage. Key parameters within the diagnostic period are collected and preprocessed to obtain standardized parameters. Preprocessing includes noise reduction and standardization. Key feature parameters are extracted from the standardized parameters, including the current average color depth of the fabric, the current relative concentration of the dye liquor, the dye liquor temperature, the temperature change rate, and the dispersion of the color depth distribution of the fabric. Among them, the dispersion of fabric color depth distribution: by comparing the color depth values ​​of different monitoring points (such as the inlet, outlet, and intermediate sampling points) in the dyeing machine at the same time, the standard deviation is calculated as a direct measure of the uniformity of fabric color. The process diagnostic module is used to acquire indicators and diagnose the status of key characteristic parameters, and to obtain a list of diagnostic results, which specifically includes: During the main staining phase (the period between the start and end of staining), two key process metrics are obtained based on the extracted key feature parameters: Based on the fabric color depth, the dyeing rate v1 = (current average fabric color depth value minus the average fabric color depth value of the previous diagnostic cycle) divided by the diagnostic cycle duration. This method directly reflects the accumulation rate of dye on the fiber. Based on the consumption of dye liquor, the dyeing rate v2 = (relative concentration of dye liquor in the previous diagnostic cycle minus the current relative concentration of dye liquor) divided by the diagnostic cycle. This method reflects the rate at which dye is absorbed from the dye liquor. Obtain the preset weight coefficients corresponding to the dyeing rate v1 and the dyeing rate v2, and calculate the actual dyeing rate V based on the dyeing rate v1 × the corresponding preset weight coefficient + the dyeing rate v2 × the corresponding preset weight coefficient. Dyeing Synchronization Index = Fabric Color Depth Distribution Dispersion ÷ (α × Current Average Fabric Color Depth + β), where α is a preset adjustment coefficient (0 < α < 1) and β is a preset adjustment coefficient (0 < β < 1) to prevent the denominator from being zero. This dyeing synchronization index not only reflects the absolute degree of color unevenness but also correlates it with the current average dyeing depth. In the early stages of dyeing, when the average color depth is small, this index will be amplified, thereby increasing the system's sensitivity to the risk of uneven dyeing in the early stages. The calculated actual dyeing rate and dyeing synchronicity index are compared with the preset target dyeing rate range [SRmin, SRmax] and synchronicity threshold for the current process stage to diagnose whether the current process state deviates from expectations: If the actual dyeing rate is less than SRmin and the duration exceeds the preset duration threshold, it is determined to be a dyeing lag state. If the actual dyeing rate is greater than SRmax and the duration exceeds the preset duration threshold, it is determined to be an over-dyeing state. If the up-dyeing synchronicity index exceeds the synchronicity threshold, it is judged as an alarm state of deteriorated uniformity. If the up-dyeing synchronicity index does not exceed the synchronicity threshold, a univariate linear regression analysis is performed on the time series data points of the up-dyeing synchronicity index to obtain the slope of the regression line. This slope is the trend change rate. At the same time, a positive warning slope threshold and a negative improvement slope threshold are set. If the trend change rate is greater than the warning slope threshold, it indicates that the up-dyeing synchronicity index is in a continuous and obvious upward trend, and is judged as a potential risk warning state of uniform dyeing. If the trend change rate is less than the improvement slope threshold, it indicates that the up-dyeing synchronicity index is in a continuous and obvious downward trend, and is judged as a state of uniform dyeing improvement. A list of diagnostic results is constructed based on the states of delayed staining, excessive staining, alarm state of deteriorated uniformity, early warning state of potential risk to uniformity, and state of improved uniformity.

[0017] Example 2: The early diagnosis module is used to collect historical data over a recent period and perform early prediction, diagnosis, and fusion output analysis based on the historical data. Specifically, it includes: The most recent historical data period is used as the basis for prediction. The most recent period is a preset time period to facilitate the collection of historical data. Using a preset prediction model, the actual dyeing rate and the dyeing synchronicity index are predicted step by step after a future ΔT time (ΔT prediction time). The ΔT is set as a reasonable process adjustment lead time, usually 1 to 3 minutes. Two predicted values ​​are obtained based on the pre-set prediction model: the predicted staining rate and the predicted synchronicity index. Early diagnostic analysis of predicted staining rate and predicted synchronicity index: If the predicted dyeing rate will exceed the target dyeing rate range [SRmin, SRmax] corresponding to the current process stage after ΔT time, it is judged as an expected rate anomaly. If it exceeds SRmax, the direction of the anomaly is that the expected rate is too high. If it is lower than SRmin, the direction of the anomaly is that the expected rate is too low. If the predicted synchronicity index will exceed the synchronicity index safety threshold after ΔT time, it is judged as an expected deterioration of homogeneity. If the diagnosis results list indicates a potential risk of homogeneity, and the predictive synchronicity index confirms that the upward trend will continue to rise within ΔT time, it is determined that the risk of homogeneity has been confirmed and exacerbated. Early diagnostic results are constructed based on anticipated rate anomalies (over-anticipated / under-anticipated), anticipated homogenization deterioration, and the identification and exacerbation of homogenization risk. Based on the fusion analysis of the diagnostic result list and early diagnostic results, if only the diagnostic result list is available, the diagnostic result list will be output and the backend visualization module will respond to and display it immediately. If there are early diagnostic results, the early diagnostic results will be output first and the backend visualization module will respond to and display them immediately, because they represent problems that are about to occur and require early intervention. That is, potential problems can be discovered in advance through the prediction layer, and they can be elevated to the system's primary focus through the "prediction priority" principle, so that the control system can intervene before the quality problems actually occur and eliminate the problems in their infancy. The dynamic adjustment module is used to perform dynamic adaptive adjustment analysis of the diagnostic result list and early diagnostic results, specifically including: T1: Based on the list of diagnostic results and early diagnostic results, if the staining is delayed or the expected rate is abnormal (expected too low), the temperature adjustment range and the heating rate adjustment range are obtained. Temperature adjustment range = preset hysteresis state sensitivity coefficient × (minimum target dyeing rate SRmin - current dyeing rate); The adjustment range of the heating rate = preset heating rate sensitivity coefficient × (minimum target dyeing rate SRmin - current dyeing rate). For example, if the preset heating rate sensitivity coefficient is set to 0.8, the heating rate after a single adjustment shall not be lower than 0.5℃ / min and not higher than 5℃ / min. If the predicted dyeing rate still does not meet the target after adjusting the temperature and heating rate, the pH value should be adjusted. The pH value adjustment range = preset pH value sensitivity coefficient × (minimum target dyeing rate SRmin - current dyeing rate). If the preset pH value sensitivity coefficient is set to 0.3, the single adjustment shall not exceed 0.5 units, and the constraint range is 4-12. T2: If the dyeing is in an overexcitation state / expected rate is abnormal (expected rate is too high), the heating rate adjustment range is obtained. The heating rate adjustment range = preset overexcitation state sensitivity coefficient × (current dyeing rate - maximum target dyeing rate SRmax). If the preset overexcitation state sensitivity coefficient is set to 1.0, the heating rate after adjustment will not be lower than 0.5℃ / minute. In the segmented feeding process, the dye replenishment interval extension ratio is obtained. The dye replenishment interval extension ratio = preset ratio coefficient × (current dyeing rate - maximum target dyeing rate SRmax) ÷ maximum target dyeing rate SRmax. The single extension does not exceed 5 minutes, or the single replenishment amount reduction ratio = preset percentage × (current dyeing rate - maximum target dyeing rate SRmax) ÷ maximum target dyeing rate SRmax, to ensure that the dye concentration decreases steadily. T3: If it is a leveling degradation alarm state / expected leveling degradation, then the following values ​​are obtained: circulation flow rate adjustment range = preset flow rate sensitivity coefficient × (current synchronization index - synchronization threshold), stirring rate adjustment range = preset stirring sensitivity coefficient × (current synchronization index - synchronization threshold), and leveling agent replenishment amount = preset leveling agent sensitivity coefficient × (current synchronization index - synchronization threshold) × dye liquor volume. For example: the preset flow sensitivity coefficient is set to 10%, which means that for every 0.1 that the synchronization index exceeds the threshold, the flow rate increases by 10%, with a maximum not exceeding 110% of the equipment's rated flow rate; The preset stirring sensitivity coefficient is set to 15 rpm. After a single adjustment, it should not be lower than the equipment's minimum stirring rate, nor higher than 105% of the rated rate. The preset leveling agent sensitivity coefficient is set to 0.2 g / L, and the replenishment amount should not exceed the maximum safe dose (e.g., 2 g / L). T4: If the risk of uneven dyeing is confirmed and aggravated, then on the basis of T2 above, additional heating compensation will be turned on to make the temperature difference between different areas in the dyeing machine ≤ the preset temperature value. T5: If it is a potential risk warning state for uniform dyeing, then obtain the circulation flow rate increase ratio. Circulation flow rate increase ratio = preset ratio factor × (synchronicity index trend change rate ÷ warning slope threshold) to ensure that the flow rate increase does not affect the stability of the equipment. If the current temperature change rate exceeds the preset change rate threshold, then reduce the current temperature change rate to below the preset change rate threshold to avoid temperature fluctuations causing uneven dyeing. Based on T1 to T5, an adaptive adjustment decision is constructed. The backend visual module is used to respond to the adaptive adjustment decision and display it immediately. At the same time, the adaptive adjustment decision is executed to realize dynamic reference adjustment. The dynamic verification module is used to perform dynamic verification analysis on the core indicators after the adaptive adjustment decision-making process is completed. Specifically, it includes: After the adaptive adjustment decision is completed, the start time is recorded immediately, and the sampling interval is calculated based on that time. Sampling is performed once at a preset time interval to obtain the core indicators of each sample (actual dyeing rate, dyeing synchronicity index, color depth distribution dispersion, etc.). The core indicators are then processed for judgment. If all core indicators are within the preset target range, the system is considered to be stable. If any core indicator deviates from the preset target range, a deviation instruction is generated. The backend visual module responds to the deviation instruction and immediately performs the preset warning operation corresponding to the deviation instruction, so that the operation and management personnel can make targeted adjustments. For example, if the indicator still deviates from the target range, the strategy needs to be adjusted based on the following: If the indicators have improved in the direction of the target (such as the dyeing rate gradually increasing after the dyeing hysteresis is adjusted), maintain the original adjustment parameters, shorten the sampling interval to 1-2 minutes, and continue to track; If the indicators do not improve or worsen (e.g., the synchronicity index still rises after leveling adjustment), a second adaptive adjustment is immediately triggered, the adjustment range is recalculated (e.g., increasing circulation flow rate, adding more leveling agent), and the sampling frequency is increased simultaneously. Anomaly warning: If a sudden change in the index is detected (such as a sudden increase in the dispersion of color depth distribution), dynamic sampling should be suspended, and the operating status of the equipment (such as stirrer failure, abnormal temperature sensor) and the execution of adjustment instructions should be checked. Monitoring can be resumed only after non-process factors have been ruled out. If the process is deemed stable after three consecutive tests, it is considered to be in a stable state, and sampling is stopped.

[0018] Example 3: This invention also proposes a dynamic adaptive optimization method for fabric dyeing processes, comprising the following steps: Step 1: Real-time data acquisition and preprocessing: Based on the fabric type and target color number, retrieve the baseline process curve from the standard process library and acquire key parameters in the dyeing process in real time; Step 2: Key Feature Parameter Extraction: Preprocess the key parameters to obtain standardized parameters, and extract the key feature parameters from the standardized parameters. Step 3: Calculation of process indicators and current diagnosis: In the main staining stage, based on the key feature parameters, two key process indicators are calculated and compared with the preset target staining rate range [SRmin, SRmax] and synchronicity threshold of the current process stage to generate a list of diagnostic results. Step 4: Early Risk Prediction: Using a pre-set prediction model, the predicted staining rate and predicted synchronicity index are obtained. Based on the predicted staining rate and predicted synchronicity index, early diagnostic analysis is performed to generate early diagnostic results. Step 5: Decision Fusion and Adaptive Adjustment: Fusion of the diagnostic result list and the early diagnostic results, prioritizing the processing of the early diagnostic results, and generating adaptive adjustment decisions based on the fused diagnostic status by invoking the corresponding adjustment strategy; Step Six: Post-Adjustment Verification and Feedback: After executing the adaptive adjustment decision, initiate periodic sampling verification.

[0019] In summary, standardized preprocessing and feature extraction are used to extract core feature parameters, providing high-quality, standardized data support for subsequent diagnosis and control. Through the synergistic analysis of actual dyeing rate and dyeing synchronicity index, explicit states such as dyeing lag, excessive dyeing, and deterioration of evenness are accurately identified. Simultaneously, time-series linear regression analysis is introduced to capture the trend changes of the dyeing synchronicity index, providing early warning of potential evenness risks. This extends the assessment from explicit problem identification to implicit risk prediction, ensuring the accuracy of current process status judgment. Early diagnostic analysis identifies potential problems such as expected rate anomalies and expected evenness deterioration, especially confirming the trends of already warned potential evenness risks, forming a prediction-verification-upgrade risk management chain. This effectively reduces fabric loss due to process deviations and improves the stability of dyeing quality. Furthermore, the precise matching mode of diagnostic results and adjustment plans avoids the experience-based and blind nature of traditional process adjustments, achieving refined and quantitative control of process parameters, improving adjustment efficiency and effectiveness. Continuous tracking of core indicators and dynamic optimization of control strategies based on indicator changes further reduce the risk of subsequent quality fluctuations.

[0020] The threshold is set for comparative analysis of results to determine whether they are good or bad. The value of the threshold is determined by a combination of large-scale model analysis of sample data and human experience. It can also be adjusted appropriately based on seasonal or common-sense influencing factors. The size of the coefficient is a specific value obtained by quantifying each parameter to facilitate subsequent comparison. The size of the coefficient depends on the amount of sample data and the corresponding operating coefficient initially set by those skilled in the art for each set of sample data; as long as it does not affect the proportional relationship between the parameter and the quantified value.

[0021] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A dynamic adaptive optimization system for fabric dyeing process, characterized in that, It includes a process dynamic management center, a process data module, a processing and feature extraction module, a process diagnosis module, an early diagnosis module, a dynamic adjustment module, a dynamic verification module, and a back-end visualization module; The process data module is used to retrieve the baseline process curve from the preset standard process library according to the fabric type and target color number, and obtain key parameters in real time. The processing and feature extraction module is used to preprocess key parameters to obtain standardized parameters, and at the same time extract key feature parameters from the standardized parameters, and send the key feature parameters to the process dynamic management center for storage. The process diagnostic module is used to acquire indicators and diagnose the status of key characteristic parameters, and obtain a list of diagnostic results. The early diagnosis module is used to collect historical data over a recent period of time, and to perform early prediction and diagnosis of indicators based on the historical data, as well as fusion output analysis, and output a list of diagnosis results or early diagnosis results. The dynamic adjustment module is used to perform dynamic adaptive adjustment analysis on the list of diagnostic results and early diagnostic results, and output adaptive adjustment decisions. The dynamic verification module is used to perform dynamic verification analysis on the core indicators after the acquisition, adaptive adjustment decision-making and adjustment are completed, and output the instruction of process status stabilization or deviation.

2. The dynamic adaptive optimization system for fabric dyeing process according to claim 1, characterized in that, The analysis process of the process diagnostic module is as follows: In the main staining stage, based on the extracted key feature parameters, two key process indicators are obtained: actual staining rate and staining synchronicity index. The calculated actual dyeing rate and dyeing synchronicity index are compared with the target dyeing rate range [SRmin, SRmax] and synchronicity threshold preset in the current process stage to diagnose whether the current process state deviates from the expectation, and obtain the dyeing lag state, dyeing over-excitation state, uniformity deterioration alarm state, uniformity potential risk warning state and uniformity improvement results. A list of diagnostic results is constructed based on the states of delayed staining, excessive staining, alarm state of deteriorated uniformity, early warning state of potential risk to uniformity, and state of improved uniformity.

3. The dynamic adaptive optimization system for fabric dyeing process according to claim 2, characterized in that, The actual staining rate analysis process is as follows: Dyeing rate v1 = (current average fabric color depth value minus the average fabric color depth value of the previous diagnostic cycle) divided by diagnostic cycle duration; Dyeing rate v2 = (relative concentration of dye liquor in the previous diagnostic cycle minus the current relative concentration of dye liquor) divided by the calculated diagnostic cycle. Obtain the preset weight coefficients corresponding to the dyeing rate v1 and the dyeing rate v2, and calculate the actual dyeing rate V based on the dyeing rate v1 × the corresponding preset weight coefficient + the dyeing rate v2 × the corresponding preset weight coefficient. Dyeing synchronicity index = fabric color depth distribution dispersion ÷ (α × current average fabric color depth value + β), where α is a preset adjustment coefficient, 0 < α < 1, and β is a preset adjustment coefficient, 0 < β < 1.

4. The dynamic adaptive optimization system for fabric dyeing process according to claim 1, characterized in that, The analysis process of the early diagnosis module is as follows: Using the most recent historical data as the basis for prediction, and employing a pre-defined prediction model, we make single-step predictions for the actual dyeing rate and the dyeing synchronicity index after a future ΔT time (prediction duration). Based on the pre-set prediction model, two predicted values ​​are obtained: predicted dyeing rate and predicted synchronicity index. Early diagnostic analysis is performed on the predicted dyeing rate and predicted synchronicity index to obtain expected rate anomalies (abnormality direction is too high or too low), expected deterioration of uniformity, and confirmation and aggravation of uniformity risk. Early diagnostic results are constructed based on anticipated rate anomalies (expected too high / expected too low), anticipated homogenization deterioration, and homogenization risk confirmation and exacerbation.

5. The dynamic adaptive optimization system for fabric dyeing process according to claim 4, characterized in that, Based on the fusion analysis of the diagnostic result list and the early diagnostic results, if only the diagnostic result list is available, the diagnostic result list will be output; if the early diagnostic results exist, the early diagnostic results will be output first.

6. The dynamic adaptive optimization system for fabric dyeing process according to claim 1, characterized in that, The analysis process of the dynamic adjustment module is as follows: T1: Based on the list of diagnostic results and early diagnostic results, if the staining is delayed or the expected rate is abnormal (expected too low), the temperature adjustment range and the heating rate adjustment range are obtained. If the predicted dyeing rate still fails to meet the target after adjusting the temperature and heating rate, adjust the pH value as follows: pH adjustment range = preset pH sensitivity coefficient × (minimum target dyeing rate SRmin - current dyeing rate). T2: If the dyeing is overexcited / the expected rate is abnormal (expected too high), the adjustment range of the heating rate is obtained; in the segmented feeding process, the proportion of the dye replenishment interval extension or the proportion of the single feeding amount reduction is obtained. T3: If it is a leveling degradation alarm state / expected leveling degradation, the circulation flow rate adjustment range, stirring rate adjustment range, and leveling agent replenishment amount are obtained. T4: If the risk of uneven dyeing is confirmed and aggravated, then on the basis of T2 above, additional heating compensation will be turned on to make the temperature difference between different areas in the dyeing machine ≤ the preset temperature value. T5: If it is a potential risk warning state for uniformity, then obtain the circulation flow rate increase ratio. If the current temperature change rate exceeds the preset change rate threshold, then reduce the current temperature change rate to below the preset change rate threshold. Adaptive adjustment decisions are constructed based on T1 to T5.

7. The dynamic adaptive optimization system for fabric dyeing process according to claim 1, characterized in that, The analysis process of the dynamic verification module is as follows: After the adaptive adjustment decision is completed, the start time is recorded immediately, and the sampling interval is calculated based on that time. Sampling is performed once at a preset time interval to obtain the core indicators of each sample (actual dyeing rate, dyeing synchronicity index, color depth distribution dispersion), and the core indicators are processed to obtain the normal stable or deviation instructions. If the process is deemed stable after three consecutive tests, it is considered to be in a stable state, and sampling is stopped.

8. A dynamic adaptive optimization method for a fabric dyeing process, wherein the method is applied to a dynamic adaptive optimization system for a fabric dyeing process as described in any one of claims 1-7, characterized in that, Includes the following steps: Step 1: Real-time data acquisition and preprocessing; Step 2: Extraction of key feature parameters; Step 3: Calculation of process indicators and current diagnosis; Step 4: Early risk prediction; Step 5: Decision fusion and adaptive adjustment; Step 6: Post-adjustment verification and feedback.

Images