A method, system and medium for detecting chemical residues in fabrics

Abstract

The present invention relates to the field of electromagnetic detection technology, and in particular to a method, system and medium for detecting chemical residues in fabrics. The method comprises the following steps: obtaining fabric production sensor data and integrating fabric chemical processing information to obtain fabric processing chemical data; obtaining fabric electromagnetic detection data and screening abnormal electromagnetic patterns to obtain abnormal electromagnetic reflection pattern data; performing fabric residue fingerprint matching on abnormal electromagnetic reflection pattern data based on fabric processing chemical data to obtain fabric residue fingerprint matching data; constructing a chemical residue prediction model based on fabric processing chemical data and fabric residue fingerprint matching data; obtaining real-time fabric production data, and performing fabric chemical residue risk assessment on real-time fabric production data through a chemical residue prediction model, thereby obtaining a real-time fabric production chemical residue risk assessment report. The present invention can effectively identify low-concentration residues.

Description

Technical Field

[0001] The present invention relates to the field of electromagnetic detection technology, and in particular to a method, system and medium for detecting chemical residues in fabrics. Background Art

[0002] With the rapid development of the textile industry and the increasing variety of fabric products, the chemical treatment of fabrics and the chemicals used in the production process are becoming increasingly diverse. In order to improve the performance of fabrics, such as antibacterial, waterproof, fireproof, and UV-resistant, manufacturers often use a series of chemicals for treatment. These chemicals include dyes, finishing agents, flame retardants, waterproofing agents, antibacterial agents, etc. However, many chemicals have potential harmful effects and may pose a threat to human health and the environment. Therefore, the detection of chemical residues in fabrics has become a focus of attention for the textile industry and consumers. However, although the existing textile chemical residue detection technology has met the detection needs to a certain extent, traditional detection methods still have some significant defects. Traditional detection methods such as gas chromatography-mass spectrometry (GC-MS) and high performance liquid chromatography (HPLC) can provide more accurate detection results in some cases, but their sensitivity and selectivity are usually limited, making detection difficult. Many low-concentration chemicals or chemicals with similar structures may not be effectively distinguished and accurately measured, which makes it difficult for these methods to meet the high requirements when multiple chemicals are present at the same time. Summary of the invention

[0003] Based on this, it is necessary for the present invention to provide a method, system and medium for detecting chemical residues in textiles to solve at least one of the above technical problems.

[0004] To achieve the above object, a method for detecting chemical residues in textiles comprises the following steps:

[0005] Step S1: Acquire fabric production sensor data, and perform fabric chemical processing flow structure modeling on the fabric production sensor data, thereby obtaining a fabric chemical processing flow structure model; integrate fabric chemical processing information based on the fabric chemical processing flow structure model, thereby obtaining fabric processing chemical data;

[0006] Step S2: Acquire fabric electromagnetic detection data, and perform fabric electromagnetic reflection pattern recognition based on the fabric electromagnetic detection data, thereby obtaining fabric electromagnetic reflection pattern data; perform abnormal electromagnetic pattern screening based on the fabric electromagnetic reflection pattern data, thereby obtaining abnormal electromagnetic reflection pattern data;

[0007] Step S3: constructing a chemical molecular fingerprint library based on the textile processing chemical data, and performing fabric residual fingerprint matching on the abnormal electromagnetic reflection pattern data according to the chemical molecular fingerprint library, thereby obtaining fabric residual fingerprint matching data;

[0008] Step S4: Calculate the chemical residue concentration factor according to the textile processing chemical data and the textile residue fingerprint matching data, thereby obtaining the chemical residue concentration factor; and construct a chemical residue prediction model based on the chemical residue concentration factor;

[0009] Step S5: acquiring real-time fabric production data, and performing fabric chemical residue risk assessment on the real-time fabric production data through a chemical residue prediction model, thereby obtaining a real-time fabric production chemical residue risk assessment report.

[0010] The present invention can fully grasp all chemical substances involved in the production process and their treatment methods by acquiring real-time sensor data in fabric production and establishing a chemical processing flow structure model. This helps to analyze the use and residual trend of chemicals and provide accurate data support for subsequent detection and evaluation. The constructed chemical processing flow structure model can help manufacturers better control the processing of fabrics, thereby reducing unnecessary chemical use and waste, and reducing potential chemical residue risks. By integrating fabric chemical processing information, the use of chemicals in each production link can be efficiently tracked, which helps to implement precise quality control and supervision. Through electromagnetic detection means (such as microwaves or radio frequencies), electromagnetic reflection patterns related to chemical residues can be obtained without destroying or complex pre-treatment of the fabric. This method not only improves the efficiency of detection, but also reduces the complexity of sample processing. Electromagnetic reflection pattern recognition can sensitively capture the presence of trace chemicals in fabrics, especially for the detection of low-concentration residues, which makes up for the lack of sensitivity of traditional chemical detection methods. Abnormal screening based on electromagnetic reflection patterns helps to identify abnormal chemical residues or potential sources of pollution, thereby providing early warning for further analysis and intervention. By constructing a chemical molecular fingerprint library, various chemical residues that may exist in fabrics can be accurately identified and matched. This molecular fingerprint identification method can efficiently and accurately distinguish various different chemical components, and even chemical substances with similar structures can be distinguished. Compared with traditional chemical detection methods, molecular fingerprint matching can significantly improve the selectivity and accuracy of detection, especially in complex situations where multiple chemicals exist at the same time, and can accurately detect and identify the residues of each chemical substance. The chemical molecular fingerprint library makes it possible to quickly screen a large number of samples, greatly improving the detection efficiency and avoiding the shortcomings of traditional methods that require repeated analysis and cumbersome operations. By calculating the chemical residue concentration factor, the residue of each chemical substance in the fabric can be quantified. This not only helps to identify potential health and environmental risks, but also provides guidance for the control and optimization of chemicals in the production process. The constructed chemical residue prediction model can predict the concentration of residual chemicals in fabrics based on real-time data and identify potential problems in advance. This prediction capability helps manufacturers adjust processes in a timely manner to reduce the risk of excessive chemical residues. Using the chemical residue concentration factor, the production process can get optimization suggestions, such as reducing the use of certain chemicals or changing the order of use, to ensure the chemical safety of fabrics. By conducting real-time risk assessment of chemical residues in the production of fabrics, the safety of fabrics can be monitored in real time during the production process. This means that if the residue concentration exceeds the standard or other abnormal conditions occur, an immediate warning can be issued to prevent substandard products from entering the market. Through real-time risk assessment, strict quality control can be implemented at every stage of fabric production.The risk assessment report based on real-time data and prediction models provides a scientific basis for production decisions, helping managers to adjust production processes and control strategies in a timely manner to reduce the risk of chemical residues and ensure product quality. In summary, through the implementation of this series of steps, accurate, rapid and low-cost monitoring and evaluation of chemical residues in the textile production process can be achieved, solving the problems of insufficient sensitivity and poor selectivity of traditional detection methods, and effectively identifying low-concentration residues and similar chemicals.

[0011] Optionally, step S1 specifically includes:

[0012] Step S11: acquiring fabric production sensor data, and performing chemical sensor data extraction on the fabric production sensor data, thereby obtaining fabric production chemical sensor data;

[0013] Step S12: extracting chemical sensing amplitude fluctuation characteristics based on the chemical sensing data of fabric production, thereby obtaining fabric chemical sensing fluctuation data;

[0014] Step S13: dividing the chemical sensing fluctuation data of the fabric into sensing fluctuation time series, thereby obtaining chemical sensing fluctuation time series data, and performing fluctuation stage dynamic time regularization according to the chemical sensing fluctuation time series data, thereby obtaining chemical processing fluctuation stage data;

[0015] Step S14: performing fabric chemical processing flow structure modeling based on the chemical processing fluctuation stage data, thereby obtaining a fabric chemical processing flow structure model;

[0016] Step S15: Integrate the textile chemical processing information based on the textile chemical processing flow structure model, so as to obtain textile processing chemical data.

[0017] The present invention can realize real-time monitoring of the chemical processing of textiles by collecting sensor data in the process of textile production. This is important for discovering potential problems, optimizing process flow, etc. Chemical sensor data extraction can convert complex raw data into meaningful information that can be directly applied, helping subsequent analysis and decision-making. This process can filter out noise and improve the accuracy and reliability of data. Chemical sensor amplitude fluctuation feature extraction can help identify subtle fluctuations in chemical reactions, such as changes in parameters such as temperature, pressure, and chemical concentration. These features are crucial to understanding and optimizing chemical processing. By extracting amplitude fluctuation features, abnormal fluctuations occurring in the production process, such as overreaction or incomplete reaction, can be effectively identified, thereby providing data support for adjusting the process or early warning system. Time series division of fluctuation data helps to understand the progress and changes of chemical reactions in different time periods and helps to study the reaction characteristics of different stages. Time series analysis can reveal the potential laws in the production process. Using the dynamic time warping (DTW) algorithm to dynamically adjust the fluctuation stage can eliminate the time difference of fluctuations in different stages and ensure the consistency of data and the accuracy of comparison. This is of great significance for optimizing the production cycle and improving production accuracy. Analyzing the chemical reaction process in stages helps to adjust the process parameters of each stage in a targeted manner and improve the overall production efficiency. By establishing a structural model of the textile chemical processing process, the interrelationships, key nodes and fluctuation effects of each link in the chemical processing process can be clearly depicted. This is directly helpful for optimizing the production process, reducing waste and improving resource utilization. The process structure model can provide a theoretical basis for the adjustment of the production process, help analyze and identify the key factors affecting production efficiency and quality, and thus make precise process improvements. By integrating information such as the use of chemical substances, reaction processes, and fluctuation data at each stage, a comprehensive view of the processing process can be formed. This integration process helps to grasp every detail of the production process as a whole. After information integration, it can provide data support for subsequent decision-making, including optimizing material use, adjusting reaction conditions, and predicting future fluctuation trends. Through data-driven decision-making processes, the level of intelligent and automated production can be significantly improved. Information integration can help establish a complete production history archive and improve the traceability of production. This is of great significance for quality control and compliance inspection.

[0018] Optionally, step S15 is specifically:

[0019] Step S151: extracting the time structure features of the chemical processing flow of the textile product from the structure model of the chemical processing flow, thereby obtaining the time data of the chemical processing flow;

[0020] Step S152: integrating the process environment sensing characteristics of the textile chemical process structure model according to the textile production chemical sensing data, thereby obtaining the process environment sensing data;

[0021] Step S153: obtaining a chemical substance characteristic database, and identifying chemical substances in the chemical process flow of the textile based on the chemical process time data, the process environment sensor data, and the chemical process fluctuation stage data, thereby obtaining chemical substance data in the chemical process flow of the textile;

[0022] Step S154: integrating the information of textile processing reaction chemicals into the chemical substance characteristic database based on the textile chemical processing flow reaction chemical data, thereby obtaining textile processing chemical data.

[0023] By extracting the structural characteristics of the processing flow time, the present invention can accurately understand the time consumption, bottlenecks and influencing factors of each link. This helps to identify which steps may affect the overall production efficiency, and then optimize the time, reduce the production cycle, and improve production efficiency. The extraction of time data provides a solid foundation for subsequent analysis, and can be continuously optimized and adjusted in a data-driven manner to improve the controllability of the entire chemical processing process. Through real-time monitoring of time data, abnormal fluctuations or delays in the processing process can be discovered immediately, so that remedial measures can be taken quickly to reduce unexpected stagnation in the production process. Environmental parameters (such as temperature, humidity, chemical concentration, etc.) in the processing process have an important impact on the processing quality of fabrics. Integrating these environmental sensor data can fully understand the actual conditions of the chemical processing process, thereby further optimizing the processing parameters and ensuring the consistency of fabric quality. The integration of environmental parameter data can help identify and predict potential problems in the production process, such as excessive temperature or low chemical concentration, and make timely adjustments to prevent the production of defective products. Based on environmental sensor data, the system can adjust processing conditions (such as temperature, pressure, etc.) in real time to ensure that the processing flow is carried out under optimal conditions, improve the stability of the production process and product quality. By calling the chemical characteristic database, the system can accurately identify the reaction chemicals that may be produced at a specific processing stage, helping to determine whether the chemical reaction is as expected. This is crucial for controlling the chemical processing quality of fabrics and can effectively prevent quality problems caused by incomplete or abnormal reactants. Identifying abnormal substances in chemical reactions helps to discover problems that may cause quality fluctuations in advance, so that corresponding measures can be taken in advance to avoid production defects. By identifying reaction chemicals, the use of chemicals can be optimized, resource waste can be reduced, and the quality of fabrics can be ensured not to be affected by excessive or insufficient chemicals. Integrating the reaction chemical information in the chemical processing flow of fabrics into the chemical characteristic database can continuously enrich the content of the database, making it more targeted and comprehensive. This provides a more accurate reference for subsequent production processes. The dynamic update of the database not only helps to optimize the existing production process, but also improves and perfects the chemical processing process through the reverse feedback mechanism in the process of continuous accumulation, and improves the overall intelligent level of production. The integrated data provides a basis for subsequent reaction chemical analysis, optimization and decision-making, and can help the production system make more accurate control and adjustment, thereby reducing errors and improving production stability and fabric quality.

[0024] Optionally, step S2 specifically includes:

[0025] Step S21: acquiring fabric electromagnetic detection data, and performing noise reduction processing on the fabric electromagnetic detection data, thereby obtaining fabric electromagnetic noise reduction data;

[0026] Step S22: dividing the fabric electromagnetic noise reduction data into fabric space grids, thereby obtaining fabric grid electromagnetic detection data;

[0027] Step S23: extracting fabric mesh electromagnetic reflection intensity features from the fabric mesh electromagnetic detection data to obtain fabric mesh electromagnetic reflection intensity data, and performing reflection intensity grid division according to the fabric mesh reflection intensity data to obtain high-frequency reflection intensity grid data and low-frequency reflection intensity grid data;

[0028] Step S24: performing high-frequency reflection intensity grid reflection pattern recognition on the electromagnetic detection data of the fabric grid according to the high-frequency reflection intensity grid data, thereby obtaining the high-frequency reflection pattern data of the fabric grid; performing low-frequency reflection intensity grid reflection pattern recognition on the electromagnetic detection data of the fabric grid according to the low-frequency reflection intensity grid data, thereby obtaining the low-frequency reflection pattern data of the fabric grid;

[0029] Step S25: performing fabric grid spatial merging on the fabric grid high-frequency reflection pattern data and the fabric grid low-frequency reflection pattern data, thereby obtaining fabric electromagnetic reflection pattern data;

[0030] Step S26: performing abnormal electromagnetic pattern screening based on the electromagnetic reflection pattern data of the fabric, thereby obtaining abnormal electromagnetic reflection pattern data.

[0031] In the present invention, the electromagnetic detection data of fabrics is often affected by factors such as environmental noise and electromagnetic interference. Noise reduction processing can effectively remove these irrelevant signals, making the original data clearer and more accurate. The data after noise reduction processing can significantly improve the accuracy of subsequent feature extraction and pattern recognition, reduce errors, and improve the reliability of the overall detection system. By gridding the data, complex two-dimensional or three-dimensional fabric electromagnetic data can be converted into more easily processed discrete spatial data. In this way, the electromagnetic characteristics of fabrics at different spatial positions can be analyzed more conveniently. Refined grid division helps to accurately capture the electromagnetic change characteristics of local fabrics, thereby improving the ability to recognize details. Electromagnetic reflection intensity is an important indicator for analyzing the electromagnetic characteristics of fabrics. By extracting the reflection intensity characteristics, the reflection behavior of fabrics at different frequencies can be obtained. This helps to understand the electromagnetic response characteristics of fabrics at different frequencies. By gridding the reflection intensity, the high-frequency and low-frequency reflection intensities are separated, which can help to better understand the differences in electromagnetic performance of fabrics in different frequency ranges, and lay the foundation for subsequent pattern recognition and reflection pattern analysis. Identifying electromagnetic reflection patterns at different frequencies helps to analyze the physical properties of fabrics in different electromagnetic bands. For example, high-frequency signals are usually used to identify the surface characteristics of fabrics, while low-frequency signals are more helpful in detecting the internal structure of fabrics. Processing high-frequency and low-frequency data separately can capture the response patterns of fabrics in different electromagnetic bands, further improving the accuracy and reliability of reflection pattern recognition. Spatially merging high-frequency and low-frequency reflection pattern data helps to understand the electromagnetic characteristics of fabrics as a whole, thereby achieving more comprehensive and multi-dimensional detection results. The merged data can provide a more comprehensive and intuitive evaluation result for the electromagnetic characteristics of fabrics, and help identify more complex electromagnetic reflection patterns. By screening out abnormal electromagnetic reflection patterns, it can help identify defects, quality problems or other abnormal conditions in fabrics (such as fabric damage, structural changes, etc.), providing an important basis for quality control and fault detection. By detecting abnormal electromagnetic patterns, potential problems of fabrics can be discovered earlier and corrected in time, thereby improving the production quality of fabrics.

[0032] Optionally, step S26 is specifically:

[0033] Step S261: performing electromagnetic reflection pattern clustering according to the electromagnetic reflection pattern data of the fabric, thereby obtaining electromagnetic reflection pattern clustering data;

[0034] Step S262: performing clustered reflection pattern grid distribution statistics based on the electromagnetic reflection pattern clustering data, thereby obtaining clustered reflection pattern grid distribution data;

[0035] Step S263: screening the clustered reflection pattern grid distribution data for the reflection pattern with high difference in local reflection intensity of the fabric, thereby obtaining first abnormal electromagnetic reflection pattern data;

[0036] Step S264: performing distribution discrete reflection pattern screening on the clustered reflection pattern grid distribution data, thereby obtaining second abnormal electromagnetic reflection pattern data;

[0037] Step S265: performing an abnormal electromagnetic reflection pattern intersection operation on the first abnormal electromagnetic reflection pattern data and the second abnormal electromagnetic reflection pattern data, so as to obtain abnormal electromagnetic reflection pattern data.

[0038] The present invention can classify complex reflection data by clustering electromagnetic reflection patterns, so that the original electromagnetic reflection pattern data can be compressed and organized, thereby facilitating subsequent processing and analysis. The clustering operation helps to identify some unusual reflection patterns, which may be caused by fabric defects, variations or external factors, thereby providing a basis for subsequent abnormal screening and analysis. Clustering reduces the complexity of the data, so that the subsequent analysis steps can be focused on specific reflection pattern categories, thereby improving work efficiency. By performing grid distribution statistics on the clustered reflection patterns, the spatial distribution law of the reflection patterns on the surface of the fabric can be revealed. This helps to understand the differences in reflection characteristics in different regions and find possible local defects or abnormal areas. Grid distribution statistics help to accurately locate those areas that show significant reflection changes in space, providing a basis for subsequent quality inspection and problem location. Grid distribution data can provide accurate numerical information, which is convenient for further quantitative analysis, such as differences in reflection intensity, uniformity of reflection patterns, etc. By screening areas with large differences in local reflection intensity, abnormalities in the reflection intensity of the fabric surface can be effectively found. These abnormalities may be caused by local defects or external interference generated during the weaving process of the fabric. Screening for areas with large differences in reflection intensity helps to focus on analyzing quality issues in these areas and avoid unnecessary comprehensive testing of the entire fabric. By screening areas with large dispersion of reflection pattern distribution, some parts of the fabric surface with extremely uneven reflection patterns can be revealed. These areas are manifestations of unstable fabric quality and uneven fabric structure. Distributed discrete reflection patterns usually mean that there is obvious unevenness in the reflection characteristics of the fabric surface, which may be related to defects in the production process, material inconsistency or processing deviation. Through this screening step, these uneven areas can be discovered in time. By performing an intersection operation on two types of abnormal patterns (high intensity difference and discrete distribution), the accuracy of abnormal pattern recognition can be improved. Only when the reflection pattern meets both abnormal conditions at the same time will it be considered a true abnormality, which helps to reduce false detection and missed detection. The intersection operation combines different abnormal characteristics, making the final screened abnormal pattern more comprehensive and reliable, and can better reflect the comprehensive quality status of the fabric. Through this multi-level and multi-dimensional screening method, quality problems can be discovered more accurately, providing a basis for subsequent quality improvement measures.

[0039] Optionally, step S3 specifically includes:

[0040] Step S31: extracting processing chemical substance characteristics from a chemical substance characteristic database based on the textile processing chemical substance data, thereby obtaining processing chemical substance characteristic data;

[0041] Step S32: constructing a chemical molecular fingerprint library according to the characteristic data of the processed chemical substances;

[0042] Step S33: performing principal component analysis on the abnormal electromagnetic reflection pattern data to obtain an abnormal reflection feature vector;

[0043] Step S34: Calculate the Tanimoto coefficient of the abnormal reflection feature vector and the chemical molecule fingerprint library to obtain the fingerprint Tanimoto coefficient;

[0044] Step S35: performing residual chemical molecular fingerprint matching on the abnormal electromagnetic reflection pattern data according to the fingerprint Tanimoto coefficient, thereby obtaining fabric residual fingerprint matching data.

[0045] The present invention can accurately identify the chemical substances used in the fabric processing process by extracting the characteristic data of the processed chemical substances. This is of great significance for the environmental protection, sustainability assessment and compliance with relevant environmental protection standards of the fabric. By extracting the characteristics of chemical substances, data from different sources can be systematically summarized and integrated, laying a good data foundation for subsequent analysis. By constructing a molecular fingerprint library of chemical substances, different chemical substances can be described in a standardized manner. This fingerprint library provides a unified standard for subsequent material identification, matching and classification, and can improve the accuracy and efficiency of matching. The construction of a fingerprint library can store and process large-scale chemical substance data in a simplified manner, avoiding manual comparison of each data, thereby improving work efficiency. The comparison method based on the molecular fingerprint of chemical substances has higher speed and computational efficiency than the traditional chemical structure comparison method, and can quickly identify the chemical components in the fabric. Principal component analysis (PCA) can effectively reduce the dimension of the data, extract the feature vector that contributes most to the abnormal electromagnetic reflection pattern, thereby simplifying the subsequent analysis process. This helps to improve computational efficiency and reduce the complexity of data processing. Through principal component analysis, abnormal parts in the electromagnetic reflection pattern can be found, helping to identify potential problems in the fabric, such as uneven fabric surface, chemical residues, etc. The Tanimoto coefficient is a common method for measuring the similarity of two fingerprints. By calculating the Tanimoto coefficient, the similarity between the abnormal reflection feature vector and the molecular fingerprint in the chemical fingerprint library can be quantified, thereby helping to determine the residual chemicals or pollutants in the fabric. The Tanimoto coefficient calculation can efficiently compare a large number of molecular fingerprints, especially for large-scale database comparison. This is very important for application scenarios that require rapid screening. By calculating the Tanimoto coefficient, the accuracy of fingerprint matching can be improved, thereby improving the correspondence between abnormal reflection pattern data and chemical substances, helping to better analyze the quality and safety of fabrics. By matching the abnormal reflection pattern data with the chemical fingerprint library, it is possible to accurately identify whether there are residual chemicals in the fabric. This process can help discover potential environmental problems in the production process. This matching technology makes the assessment of fabric quality more accurate and can detect chemical residues in fabrics in real time during the production process, thereby preventing consumers from being exposed to harmful chemicals.

[0046] Optionally, step S4 is specifically:

[0047] Step S41: Calculating the residual concentration of fabric chemicals based on the fabric residual fingerprint matching data, thereby obtaining fabric chemical residual concentration data;

[0048] Step S42: extracting the usage characteristics of textile processing chemicals from the textile processing chemical data, thereby obtaining textile processing chemical usage data;

[0049] Step S43: performing textile chemical processing correlation on the processing process environment sensor data and the textile processing chemical dosage data, thereby obtaining a textile chemical processing parameter group;

[0050] Step S44: Calculating the chemical residue concentration factor according to the fabric chemical residue concentration data and the fabric chemical processing parameter group, thereby obtaining the chemical residue concentration factor;

[0051] Step S45: constructing a chemical residue prediction model based on the chemical residue concentration factor and the textile chemical processing parameter group.

[0052] The present invention can effectively identify and quantify the residual chemicals on the fabric through fingerprint matching data, thereby providing accurate data support for subsequent processing. This helps to control the residual amount of chemicals and ensure that the fabric meets safety and quality standards. It can detect harmful residual chemicals on the fabric in a timely manner to ensure the safety of the final product, especially in sensitive products that may affect human skin (such as baby products, underwear, etc.), which plays an important role. By extracting the usage characteristics of processing chemicals, it can help production managers understand the consumption of each chemical, so as to more accurately manage the use of raw materials and avoid waste. By combining environmental sensor data (such as temperature, humidity, pressure, etc.) in the processing flow with chemical usage data, the impact of environmental factors on the use of chemicals can be better understood, thereby optimizing the environmental conditions in the processing process. The relationship between different environmental conditions and chemical use can be effectively discovered, thereby achieving accurate control of the processing process, avoiding unnecessary production fluctuations, and improving overall production stability. By calculating the chemical residual concentration factor, a quantitative indicator can be provided for the chemical residues in the fabric processing process, which helps to identify the impact of different processing parameters on the residual concentration and assess risks in advance. By calculating the concentration factor, fabric safety issues (such as allergies, skin irritation, etc.) caused by improper processing or excessive use of chemicals can be effectively prevented. By building a chemical residue prediction model, it is possible to predict in advance the residue problems that may occur under different processing conditions, so that preventive measures can be taken. This can significantly improve product quality and production efficiency. By continuously optimizing the prediction model, the amount of chemical used, processing temperature, humidity and other conditions can be adjusted more accurately to maximize the final quality of the fabric while reducing waste and environmental pollution.

[0053] Optionally, step S5 specifically includes:

[0054] Step S51: acquiring real-time production fabric data, and extracting production fabric material features from the real-time production fabric data, thereby obtaining real-time production fabric material data;

[0055] Step S52: acquiring real-time chemical processing environment data, and performing fabric chemical residue prediction on the real-time production fabric material data and the real-time chemical processing environment data through a chemical residue prediction model, thereby obtaining fabric chemical residue prediction data;

[0056] Step S53: performing statistics of chemicals with high residual amounts according to the textile chemical residue prediction data, thereby obtaining data of chemicals with high residual amounts;

[0057] Step S54: Calculating the chemical residue percentage of textile processing chemical usage data and high-value residual chemical data, thereby obtaining textile chemical residue percentage data;

[0058] Step S55: Performing a residue risk assessment based on the textile chemical residue percentage data, thereby obtaining a real-time textile chemical residue risk assessment report.

[0059] The present invention can monitor key features of the fabric production process, such as fiber type, fabric structure, etc., in real time by acquiring real-time fabric production data. This provides an accurate data basis for subsequent chemical treatment, residue prediction, etc. The extraction of fabric material features can identify different requirements of different textiles for chemical treatment, thereby improving production efficiency, avoiding excessive or insufficient chemical treatment, and reducing resource waste. Material feature extraction helps to discover potential quality problems in the production process, such as uneven textile fibers, unqualified fabric structure, etc., and then make adjustments in the early stage. Based on the combination of chemical processing environment data (such as temperature, humidity, chemical concentration, etc.) and fabric material data, the chemical residue of the fabric can be accurately predicted through the prediction model to reduce the error of manual judgment. Through the chemical residue prediction data, possible chemical residue problems can be discovered in time during the production process, providing a basis for subsequent processing adjustments, thereby avoiding the outflow of unqualified products. The use of the prediction model can be optimized and adjusted according to the environmental data in the production process to ensure that the chemical residue is within a safe range and improve production efficiency and product quality. Statistics on high-value residual chemical substances can help identify chemical substances with greater potential risks, and can more accurately control their usage to avoid affecting the quality and safety of fabrics. By counting and analyzing high-value residual chemical substances, it can help manufacturers adjust the amount and use of chemicals to reduce the generation of harmful residues from the source. By calculating the chemical residue percentage, the residual proportion of each chemical substance in the fabric can be quantified, and then the safety and applicability of the fabric can be evaluated in detail. It provides a quantitative analysis method to make the chemical residue situation in the fabric production process more transparent and manageable. Enterprises can optimize and adjust according to the residue percentage data to ensure that the product quality meets the standards. Risk assessment based on chemical residue percentage data can provide real-time risk warnings for the production process, timely discover potential quality problems or safety hazards, and avoid excessive residues of harmful substances. Risk assessment reports can help production managers and quality control personnel make decisions quickly, adjust production parameters or conduct subsequent tests, so as to effectively control the chemical residues of fabrics.

[0060] Optionally, the present invention further provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the method for detecting chemical residues in fabrics as described above.

[0061] Optionally, the present invention further provides a textile chemical residue detection system, which is used to perform the textile chemical residue detection method as described above, and the textile chemical residue detection system comprises:

[0062] A processing information integration module is used to obtain fabric production sensor data, and to perform fabric chemical processing flow structure modeling on the fabric production sensor data, thereby obtaining a fabric chemical processing flow structure model; and to perform fabric chemical processing information integration based on the fabric chemical processing flow structure model, thereby obtaining fabric processing chemical data;

[0063] The electromagnetic reflection pattern recognition module is used to obtain electromagnetic detection data of fabrics, and perform electromagnetic reflection pattern recognition of fabrics according to the electromagnetic detection data of fabrics, so as to obtain electromagnetic reflection pattern data of fabrics; perform abnormal electromagnetic pattern screening based on the electromagnetic reflection pattern data of fabrics, so as to obtain abnormal electromagnetic reflection pattern data;

[0064] A fabric residual fingerprint matching module is used to construct a chemical molecular fingerprint library based on textile processing chemical data, and to perform fabric residual fingerprint matching on abnormal electromagnetic reflection pattern data according to the chemical molecular fingerprint library, thereby obtaining fabric residual fingerprint matching data;

[0065] A residual concentration factor calculation module is used to calculate the chemical residual concentration factor according to the textile processing chemical data and the textile residual fingerprint matching data, so as to obtain the chemical residual concentration factor; and to construct a chemical residue prediction model based on the chemical residual concentration factor;

[0066] The residual risk assessment module is used to obtain real-time fabric production data and conduct fabric chemical residue risk assessment on the real-time fabric production data through a chemical residue prediction model, thereby obtaining a real-time fabric production chemical residue risk assessment report.

[0067] The textile chemical residue detection system of the present invention can realize any textile chemical residue detection method of the present invention, and is used to combine the operation and signal transmission medium between various modules to complete the textile chemical residue detection method. The internal modules of the system cooperate with each other to effectively identify low-concentration residues. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0069] Figure 1 Schematic diagram of the steps of the method for detecting chemical residues in textiles of the present invention;

[0070] Figure 2 Detailed step flow diagram of step S1 in the present invention;

[0071] Figure 3 Detailed step flow diagram of step S2 in the present invention;

[0072] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0073] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0074] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0075] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0076] To achieve this, please refer to Figures 1 to 3 The present invention provides a method for detecting chemical residues in fabrics, the method comprising the following steps:

[0077] Step S1: Acquire fabric production sensor data, and perform fabric chemical processing flow structure modeling on the fabric production sensor data, thereby obtaining a fabric chemical processing flow structure model; integrate fabric chemical processing information based on the fabric chemical processing flow structure model, thereby obtaining fabric processing chemical data;

[0078] In this embodiment, real-time data of fabric production is obtained by installing sensors at different locations of the production line. These sensors include temperature sensors, humidity sensors, pH sensors, flow sensors, etc., which are used to monitor the physical and chemical changes of fabrics during chemical processing. Next, these sensor data are used to establish a fabric chemical processing process structure model. For example, these sensor data are associated with specific chemical treatment processes (such as dyeing, bleaching, etc.) through machine learning algorithms (such as decision trees, support vector machines, etc.) to form a complete fabric processing process model. In this model, based on the time series of sensor data, the environment and parameter conditions of each step of chemical treatment can be accurately described, and the concentration and type of the chemical substances used can be inferred.

[0079] Step S2: Acquire fabric electromagnetic detection data, and perform fabric electromagnetic reflection pattern recognition based on the fabric electromagnetic detection data, thereby obtaining fabric electromagnetic reflection pattern data; perform abnormal electromagnetic pattern screening based on the fabric electromagnetic reflection pattern data, thereby obtaining abnormal electromagnetic reflection pattern data;

[0080] In this embodiment, during the fabric processing, an electromagnetic sensor is installed to detect the fabric in real time. The sensor obtains electromagnetic reflection data of the fabric based on the electromagnetic reflection characteristics of the fabric surface (such as microwave and infrared spectrum). Then, the electromagnetic reflection data is analyzed using image processing and pattern recognition algorithms (such as convolutional neural network CNN) to identify the reflection patterns of different fabrics, such as dyed fabrics, untreated fabrics, and fabrics with chemical residues. Based on these pattern recognitions, the data is further processed to screen out abnormal electromagnetic reflection patterns that do not meet the standards.

[0081] Step S3: constructing a chemical molecular fingerprint library based on the textile processing chemical data, and performing fabric residual fingerprint matching on the abnormal electromagnetic reflection pattern data according to the chemical molecular fingerprint library, thereby obtaining fabric residual fingerprint matching data;

[0082] In this embodiment, a chemical molecular fingerprint library is constructed using the molecular fingerprint data of chemical substances. This fingerprint library includes the molecular structure characteristics (such as molecular weight, functional group, etc.) of different chemical substances and their associated characteristics with the fabric reflection pattern. Then, the acquired abnormal electromagnetic reflection pattern data is matched with the chemical substance fingerprint in the molecular fingerprint library. Fingerprint matching can be performed by a similarity algorithm (such as cosine similarity or Euclidean distance) to determine whether a specific chemical substance remains on the fabric and to infer its possible concentration.

[0083] Step S4: Calculate the chemical residue concentration factor according to the textile processing chemical data and the textile residue fingerprint matching data, thereby obtaining the chemical residue concentration factor; and construct a chemical residue prediction model based on the chemical residue concentration factor;

[0084] In this embodiment, based on the obtained fabric residual fingerprint matching data, combined with the amount of chemicals used in the fabric processing process (such as dye concentration, chemical treatment dosage, etc.), the chemical residual concentration factor is calculated. Regression analysis (such as linear regression, random forest regression, etc.) is used to associate the residual concentration with processing parameters (such as temperature, time, etc.) to construct a chemical residual concentration prediction model. This model can predict the residual concentration of chemical substances on fabrics according to different processing conditions.

[0085] Step S5: acquiring real-time fabric production data, and performing fabric chemical residue risk assessment on the real-time fabric production data through a chemical residue prediction model, thereby obtaining a real-time fabric production chemical residue risk assessment report.

[0086] In this embodiment, risk assessment is performed on each batch of fabrics produced based on the chemical residue prediction model. Real-time production data (such as temperature, humidity, chemical concentration, etc.) obtained by sensors is input into the chemical residue prediction model to calculate the chemical residue risk of the fabric in real time. If the residue concentration exceeds the set safety standard (such as national standards or internal enterprise standards), a risk assessment report will be generated and an alarm will be issued to alert the relevant operators. The report includes the expected residual concentration of each chemical substance, possible safety risks, and prevention and control measures that need to be taken.

[0087] Optionally, step S1 specifically includes:

[0088] Step S11: acquiring fabric production sensor data, and performing chemical sensor data extraction on the fabric production sensor data, thereby obtaining fabric production chemical sensor data;

[0089] In this embodiment, chemical reaction data on the surface of the fabric can be collected in real time through chemical sensors (such as gas sensors, temperature sensors, humidity sensors, etc.) installed on the production line. For example, in the dyeing process, the sensor can monitor the change in the concentration of the gas released when the fabric reacts with the dye, and extract relevant chemical signals through these sensor data, such as the pH value of the dye, the reaction temperature, the concentration of the reaction gas, etc., which constitute the chemical sensing data of the fabric production. These sensing data can be uploaded, stored and subsequently processed through data acquisition equipment. The acquisition frequency of the sensing data can be set to once per second or once per minute to ensure real-time monitoring of the production process.

[0090] Step S12: extracting chemical sensing amplitude fluctuation characteristics based on the chemical sensing data of fabric production, thereby obtaining fabric chemical sensing fluctuation data;

[0091] In this embodiment, by analyzing the obtained chemical sensing data, the amplitude fluctuation characteristics can be extracted. Specifically, through algorithms such as fast Fourier transform (FFT) or wavelet transform, the time series fluctuation of the sensing data is analyzed in the frequency domain and time domain, and the amplitude, frequency and change trend of the fluctuation are extracted. Taking the gas sensor data in the dyeing process as an example, during the dyeing process, the chemical reaction on the surface of the fabric will cause the gas concentration to fluctuate over time. These fluctuations can reflect the different stages of the dye reaction. By calculating the amplitude of these fluctuations, the chemical fluctuation characteristic data in the fabric production process can be obtained. These characteristic data can be further applied to quality control and process optimization.

[0092] Step S13: dividing the chemical sensing fluctuation data of the fabric into sensing fluctuation time series, thereby obtaining chemical sensing fluctuation time series data, and performing fluctuation stage dynamic time regularization according to the chemical sensing fluctuation time series data, thereby obtaining chemical processing fluctuation stage data;

[0093] In this embodiment, after the chemical sensing fluctuation data is obtained, the data can be divided by a time series analysis algorithm. First, the overall sensing fluctuation data is divided into different time periods or fluctuation stages according to the changes in the predetermined time window or fluctuation characteristics. For example, in the dyeing process, the sensing data can be divided into a pretreatment stage, a reaction stage, and a cleaning stage according to the change in the rate of the chemical reaction. Subsequently, the dynamic time warping (DTW) algorithm is used to calibrate these fluctuation stages and adjust the time axis of the fluctuation data so that the fluctuation characteristics of the same stage can be compared and analyzed on the same time scale. This process can improve the monitoring accuracy of each stage and the identification of fluctuation laws between stages, providing a basis for optimizing process parameters.

[0094] Step S14: performing fabric chemical processing flow structure modeling based on the chemical processing fluctuation stage data, thereby obtaining a fabric chemical processing flow structure model;

[0095] In this embodiment, the chemical processing fluctuation stage data obtained can be used to model the chemical processing flow of fabrics using data modeling techniques (such as system modeling, causal reasoning model or machine learning algorithm). For example, the various stages of chemical processing of fabrics and their interrelationships can be described by establishing a graph model. Assuming that the dyeing process includes several stages such as pretreatment, dyeing, and cleaning, the chemical sensing fluctuation data of each stage can be used as the input data of the stage. By constructing a weighted directed graph, the input-output relationship of each stage is connected to form a structural model of the chemical processing flow of fabrics. This model can help analyze and predict how chemical fluctuations affect the processing effect of fabrics under different conditions, and thus provide decision support for process optimization.

[0096] Step S15: Integrate the textile chemical processing information based on the textile chemical processing flow structure model, so as to obtain textile processing chemical data.

[0097] In this embodiment, based on the constructed fabric chemical processing flow structure model, the chemical reaction data of each stage can be integrated to generate fabric processing chemical data. For example, in the dyeing process, the chemical substances (such as dyes, auxiliaries, acid-base regulators, etc.) involved in each stage will affect the color, quality and reaction speed of the fabric. By integrating the chemical substance information, usage, reaction conditions and other data of each stage in the model, a complete fabric processing chemical data set can be generated. These data can not only reflect the types and quantities of chemicals used in each stage, but also reflect the effects of different chemicals on the fabric processing effect. These data can provide important support for subsequent quality control, environmental monitoring and production efficiency improvement.

[0098] Optionally, step S15 is specifically:

[0099] Step S151: extracting the time structure features of the chemical processing flow of the textile product from the structure model of the chemical processing flow, thereby obtaining the time data of the chemical processing flow;

[0100] In this embodiment, the model constructed in the pre-step includes the time parameters of each processing stage. By collecting historical production data, the time structure characteristics of each stage are extracted, including the start time, end time and stage duration of each chemical treatment stage. Specifically, the time points of each link of fabric processing are recorded in real time by intelligent sensors or automated production systems. For example, in the dyeing stage, the sensor records the time when the fabric enters the dyeing tank and the time when it is taken out, and the time difference of the dyeing process is calculated. Next, the time data of the processing flow is processed using data analysis methods (such as time series analysis) to identify the time characteristics of different processing stages, such as the time fluctuations of each link such as dyeing, drying, and shaping, the average processing time and its relationship with environmental factors (such as temperature and humidity).

[0101] Step S152: integrating the process environment sensing characteristics of the textile chemical process structure model according to the textile production chemical sensing data, thereby obtaining the process environment sensing data;

[0102] In this embodiment, it relies on the environmental sensor data provided during the fabric production process, such as temperature, humidity, gas concentration, pH value, dye concentration and other chemical environmental information. Taking the dyeing process as an example, a temperature sensor and a pH sensor are configured in the dyeing tank to monitor the temperature and pH changes of the dye solution in real time. When the fabric passes through the dyeing tank, the sensor data will be collected and integrated with the processing time data based on the time series characteristics, thereby forming a complete processing flow environmental feature data set.

[0103] Step S153: obtaining a chemical substance characteristic database, and identifying chemical substances in the chemical process flow of the textile based on the chemical process time data, the process environment sensor data, and the chemical process fluctuation stage data, thereby obtaining chemical substance data in the chemical process flow of the textile;

[0104] In this embodiment, a chemical substance characteristic database is introduced by consulting relevant materials and expert experience, and the database contains a large amount of reaction characteristic data of chemical substances (such as dyes, chemical additives, cleaning agents, etc.). These data include reaction temperature, reaction time, chemical substance concentration, reaction rate, etc. During the fabric processing process, especially in the chemical processing stages such as dyeing, printing, and finishing, the fabric reacts with the chemical substances, resulting in changes in physical and chemical properties. By comparing and analyzing the extracted time data, environmental data, and reaction data in the chemical substance characteristic database, the chemical substances involved in the reaction process can be accurately identified. For example, during the dyeing process, it is possible to identify whether a chemical reaction of a certain dye has occurred based on the reaction temperature and time data, and then determine whether the ideal color effect has been achieved. In addition, by combining the fluctuation stage data (such as abnormal temperature fluctuations during the processing process), the reaction of the chemical substances during the processing process can be further analyzed, and early warning of possible quality problems can be given.

[0105] Step S154: integrating the information of textile processing reaction chemicals into the chemical substance characteristic database based on the textile chemical processing flow reaction chemical data, thereby obtaining textile processing chemical data.

[0106] In this embodiment, the identified textile chemical processing process reaction chemical data is used to extract the information such as the dosage change, concentration change, reaction temperature, reaction time, chemical properties, etc. of the chemical substance reaction involved in the reaction chemical in the chemical substance characteristic database to obtain textile processing chemical data.

[0107] Optionally, step S2 specifically includes:

[0108] Step S21: acquiring fabric electromagnetic detection data, and performing noise reduction processing on the fabric electromagnetic detection data, thereby obtaining fabric electromagnetic noise reduction data;

[0109] In this embodiment, electromagnetic wave detection equipment (such as a microwave reflectometer or an antenna array) is used to scan the surface of the fabric with electromagnetic waves to collect electromagnetic reflection data of the fabric. The collected raw data are usually affected by noise, environmental factors (such as temperature, humidity) and measurement errors. Therefore, these data must be denoised. For example, wavelet transform or Kalman filter algorithm can be used to remove high-frequency noise, and the main features of the original signal are retained by smoothing the data. After denoising, the electromagnetic noise reduction data of the fabric obtained can effectively reflect the electromagnetic properties of the fabric.

[0110] Step S22: dividing the fabric electromagnetic noise reduction data into fabric space grids, thereby obtaining fabric grid electromagnetic detection data;

[0111] In this embodiment, the electromagnetic data after noise reduction is divided into several grids according to the actual size of the fabric and the measurement accuracy. For example, assume that the actual size of the fabric is 50×50 cm and the resolution of the measuring device is 5 mm. By dividing the fabric into 10×10 grids, each grid has a side length of 5 cm, electromagnetic detection is performed on each grid area, and the electromagnetic reflection signal of the corresponding area is collected. The data in each grid will be used as a separate electromagnetic detection data point. After grid division, the electromagnetic detection data of the fabric grid can be obtained, which is specifically represented by the electromagnetic reflection intensity and phase information of each grid.

[0112] Step S23: extracting fabric mesh electromagnetic reflection intensity features from the fabric mesh electromagnetic detection data to obtain fabric mesh electromagnetic reflection intensity data, and performing reflection intensity grid division according to the fabric mesh reflection intensity data to obtain high-frequency reflection intensity grid data and low-frequency reflection intensity grid data;

[0113] In this embodiment, the characteristics of the reflection intensity are extracted from the electromagnetic reflection data of each grid. For example, the frequency spectrum characteristics of the signal can be extracted by Fourier transform to analyze the high-frequency and low-frequency components. The high-frequency signal represents the surface details of the fabric or the high-density area of ​​the fabric, while the low-frequency signal is related to the large-scale structural characteristics of the fabric. By setting a frequency threshold, the high-frequency reflection intensity and the low-frequency reflection intensity are distinguished. Finally, the obtained high-frequency reflection intensity grid data and low-frequency reflection intensity grid data can be used for subsequent pattern recognition and feature analysis respectively.

[0114] Step S24: performing high-frequency reflection intensity grid reflection pattern recognition on the electromagnetic detection data of the fabric grid according to the high-frequency reflection intensity grid data, thereby obtaining the high-frequency reflection pattern data of the fabric grid; performing low-frequency reflection intensity grid reflection pattern recognition on the electromagnetic detection data of the fabric grid according to the low-frequency reflection intensity grid data, thereby obtaining the low-frequency reflection pattern data of the fabric grid;

[0115] In this embodiment, machine learning or pattern recognition algorithms (such as support vector machines (SVM) or neural networks) are used to perform pattern recognition on high-frequency reflection intensity grid data and low-frequency reflection intensity grid data. Specifically, the K-means clustering algorithm or hierarchical clustering algorithm is first used to classify the grid data to identify different reflection patterns. For example, the high-frequency reflection pattern is related to the microstructure or surface morphology changes of the fabric, while the low-frequency reflection pattern is related to the macrostructure and thickness changes of the fabric. Through this pattern recognition, the high-frequency reflection pattern data and low-frequency reflection pattern data of the fabric grid are finally obtained.

[0116] Step S25: performing fabric grid spatial merging on the fabric grid high-frequency reflection pattern data and the fabric grid low-frequency reflection pattern data, thereby obtaining fabric electromagnetic reflection pattern data;

[0117] In this embodiment, the high-frequency and low-frequency reflection pattern data are spatially merged. This can be done by weighted averaging or splicing. For example, for each grid point, the weighted sum of the high-frequency reflection pattern data and the low-frequency reflection pattern data is calculated to obtain a comprehensive electromagnetic reflection pattern data. In this merging process, weighting can be performed according to the contribution of each mode to ensure that the merged data can fully reflect the comprehensive electromagnetic properties of the fabric. The electromagnetic reflection pattern data of the fabric finally obtained contains the electromagnetic reflection characteristics of the fabric at different frequencies, which can provide data support for subsequent abnormal pattern detection.

[0118] Step S26: performing abnormal electromagnetic pattern screening based on the electromagnetic reflection pattern data of the fabric, thereby obtaining abnormal electromagnetic reflection pattern data.

[0119] In this embodiment, an abnormal detection algorithm (such as an isolation forest or a Gaussian mixture model) is used to analyze the electromagnetic reflection pattern data of the fabric to screen out abnormal electromagnetic reflection patterns. These abnormal patterns may correspond to defective areas or irregular physical properties of the fabric. For example, by analyzing the statistical characteristics of the reflection pattern data (such as mean, standard deviation, etc.), abnormal patterns that are significantly different from most of the data can be identified, indicating that the fabric has electromagnetic reflection anomalies in certain areas. Finally, the quality problems or potential defects of the fabric can be further analyzed by screening out abnormal electromagnetic reflection pattern data, such as wrinkles, damage or other abnormal structures on the surface of the fabric.

[0120] Optionally, step S26 is specifically:

[0121] Step S261: performing electromagnetic reflection pattern clustering according to the electromagnetic reflection pattern data of the fabric, thereby obtaining electromagnetic reflection pattern clustering data;

[0122] In this embodiment, a clustering algorithm (such as K-means, DBSCAN or spectral clustering) is applied to classify the electromagnetic reflection pattern data of the fabric. The input of the clustering algorithm is the characteristic values ​​of the reflection pattern such as the electromagnetic reflection intensity, reflection angle and frequency of each fabric. By setting the number of clusters (such as k value or distance threshold), the algorithm divides the electromagnetic reflection pattern data into multiple clusters, each cluster representing a similar reflection pattern. Finally, the clustering data of the electromagnetic reflection pattern is obtained for subsequent analysis.

[0123] Step S262: performing clustered reflection pattern grid distribution statistics based on the electromagnetic reflection pattern clustering data, thereby obtaining clustered reflection pattern grid distribution data;

[0124] In this embodiment, based on the clustering results, grid distribution statistics are performed on the electromagnetic reflection patterns of different clusters. Specifically, the spatial distribution of the reflection pattern is discretized using a two-dimensional grid or a three-dimensional grid. For example, the surface of the fabric is divided into a plurality of small grids, each of which represents a certain reflection pattern area. By counting the number of clustered reflection patterns, intensity distribution and other data in each grid, the mean, standard deviation or distribution range of each grid can be further calculated. In this way, the distribution data of the clustered reflection pattern of each grid can be obtained, and the electromagnetic reflection characteristics of the fabric can be analyzed and understood in more detail.

[0125] Step S263: screening the clustered reflection pattern grid distribution data for the reflection pattern with high difference in local reflection intensity of the fabric, thereby obtaining first abnormal electromagnetic reflection pattern data;

[0126] In this embodiment, the obtained clustered reflection pattern grid distribution data is screened to find out the area with large local reflection intensity difference. Specifically, by calculating the reflection intensity difference between adjacent grids, if the reflection intensity of a grid is significantly higher than that of the surrounding grids (for example, the difference is greater than a predetermined threshold), the area is considered to belong to the abnormal reflection mode. This can be done by setting a threshold for the reflection intensity difference (for example, a difference greater than a certain percentage) and evaluating it using statistical methods (such as standard deviation, mean difference, etc.). Finally, the first abnormal electromagnetic reflection pattern data corresponding to the grid with a large reflection intensity difference is screened out, and these data will be used for further analysis.

[0127] Step S264: performing distribution discrete reflection pattern screening on the clustered reflection pattern grid distribution data, thereby obtaining second abnormal electromagnetic reflection pattern data;

[0128] In this embodiment, based on the clustered reflection pattern grid distribution data, distribution discreteness analysis is used to screen out reflection pattern data with higher discreteness. Specifically, the distribution discreteness of the reflection pattern in each grid can be calculated, for example, by calculating the variance or entropy value of the reflection intensity or reflection angle in each grid. If the reflection pattern of a certain grid has a large discreteness (for example, the reflection pattern varies greatly at different frequencies or angles), the grid can be determined to be an abnormal reflection pattern. A threshold value of discreteness is set (for example, the variance is greater than a set value or the entropy value is greater than a predetermined value), and discrete reflection patterns are screened out. Finally, the second abnormal electromagnetic reflection pattern data is obtained, which can be used for further comparison or combination with other data.

[0129] Step S265: performing an abnormal electromagnetic reflection pattern intersection operation on the first abnormal electromagnetic reflection pattern data and the second abnormal electromagnetic reflection pattern data, so as to obtain abnormal electromagnetic reflection pattern data.

[0130] In this embodiment, the two parts of abnormal data are compared through intersection operation to screen out the common abnormal patterns. Specifically, the first and second abnormal patterns can be compared using set operations (such as intersection) to find the reflection patterns that meet the conditions of high reflection intensity difference and discrete distribution. These common abnormal electromagnetic reflection patterns usually indicate that there are abnormal areas or defects in the electromagnetic reflection characteristics of the fabric surface. In this step, the intersection conditions can be set, such as data points that meet the requirements of intensity difference and discreteness at the same time, and finally comprehensive abnormal electromagnetic reflection pattern data can be obtained for further analysis or fault diagnosis.

[0131] Optionally, step S3 specifically includes:

[0132] Step S31: extracting processing chemical substance characteristics from a chemical substance characteristic database based on the textile processing chemical substance data, thereby obtaining processing chemical substance characteristic data;

[0133] In this embodiment, based on the textile processing chemical data, the chemical characteristics of the processing chemical (such as chemical reactivity, molecular polarity, acidity and alkalinity, etc.) in the chemical characteristic database are extracted to obtain the processing chemical characteristic data. The extracted chemical characteristics must include but are not limited to molecular weight, chemical structure, polarity, hydrophilicity, lipophilicity, chemical reactivity, etc.

[0134] Step S32: constructing a chemical molecular fingerprint library according to the characteristic data of the processed chemical substances;

[0135] In this embodiment, a chemical substance molecular fingerprint library is constructed based on molecular fingerprint technology (such as ECFP, MACCS and other algorithms) using the extracted processing chemical substance characteristic data. In specific implementation, the molecular structure of each chemical substance is first converted into a binary vector through a molecular fingerprint algorithm, and each bit in the vector represents whether the chemical substance has a certain structure or chemical feature. For example, for a specific chemical substance (such as a dye molecule), a binary fingerprint vector with a length of 512 can be generated, in which each bit identifies the presence or absence of a certain structural unit. By encoding a large number of chemical substances in this way, a library containing multiple chemical molecular fingerprints is finally established.

[0136] Step S33: performing principal component analysis on the abnormal electromagnetic reflection pattern data to obtain an abnormal reflection feature vector;

[0137] In this embodiment, the abnormal electromagnetic reflection pattern data is subjected to dimensionality reduction processing by principal component analysis (PCA) in order to extract effective feature vectors. The PCA algorithm is applied to process these high-dimensional data of the abnormal electromagnetic reflection pattern data, and the main reflection features are extracted by removing redundant information and noise. The reflection feature vector finally obtained can reflect the abnormal electromagnetic reflection characteristics after the fabric surface interacts with the processed chemical substances.

[0138] Step S34: Calculate the Tanimoto coefficient of the abnormal reflection feature vector and the chemical molecule fingerprint library to obtain the fingerprint Tanimoto coefficient;

[0139] In this embodiment, the extracted abnormal reflection feature vector is compared with the established chemical molecular fingerprint library. The similarity of the two fingerprints is calculated by the Tanimoto coefficient to evaluate the relationship between the reflection feature and the chemical substance. The Tanimoto coefficient is calculated by comparing the intersection and union size of two sets, and the formula is: \text {Tanimoto coefficient}=\frac{|A \cap B|}{|A \cup B|}; wherein A and B are the sets of reflection features and molecular fingerprints, respectively. The closer the calculation result is to 1, the higher the similarity between the two. Through this calculation, it can be determined whether the abnormal reflection pattern matches a specific chemical substance or its characteristics.

[0140] Step S35: performing residual chemical molecular fingerprint matching on the abnormal electromagnetic reflection pattern data according to the fingerprint Tanimoto coefficient, thereby obtaining fabric residual fingerprint matching data.

[0141] In this embodiment, the obtained Tanimoto coefficient is used to perform residual matching of chemical molecular fingerprints on abnormal electromagnetic reflection pattern data. For example, assuming that a certain batch of fabrics uses a certain chemical dye during the dyeing process, through the previous reflection data analysis and molecular fingerprint comparison, it is found that it has a high similarity with a certain chemical substance. Through the residual fingerprint matching technology, it can be further confirmed whether these chemicals still exist on the fabric surface, or whether chemical reactions or conversions have occurred. Ultimately, the obtained fabric residual fingerprint matching data can help determine the chemical residues in the processing process and ensure the safety of the fabric and compliance with environmental standards.

[0142] Optionally, step S4 is specifically:

[0143] Step S41: Calculating the residual concentration of fabric chemicals based on the fabric residual fingerprint matching data, thereby obtaining fabric chemical residual concentration data;

[0144] In this embodiment, during the fabric production process, the textile may adsorb some chemical substances (such as dyes, finishing agents, etc.). Surface analysis technology (such as mass spectrometry or high performance liquid chromatography) is used to quantitatively detect chemical residues on the fabric. The concentration of chemical residues on the fabric surface is calculated by comparing and analyzing different chemical components. For example, if some harmful chemical substances, such as dyes or waterproofing agents, remain on the fabric surface, high performance liquid chromatography (HPLC) can be used for analysis, and the concentration can be determined by the standard curve method, and finally the chemical residue concentration data can be obtained.

[0145] Step S42: extracting the usage characteristics of textile processing chemicals from the textile processing chemical data, thereby obtaining textile processing chemical usage data;

[0146] In this embodiment, in the production process of fabrics, the use of all chemical treatment agents involved, including dyes, finishing agents, waterproofing agents, etc., is recorded and analyzed. The chemical materials in each fabric processing process contained in the fabric processing chemical data are extracted. For example, in the fabric dyeing process, the type, amount and ratio of the dye are recorded, and the key parameters related to the fabric chemical treatment, such as the consumption of chemical substances, operating temperature and time, etc., are extracted using big data analysis methods.

[0147] Step S43: performing textile chemical processing correlation on the processing process environment sensor data and the textile processing chemical dosage data, thereby obtaining a textile chemical processing parameter group;

[0148] In this embodiment, environmental factors (such as temperature, humidity, air pressure, etc.) have a significant impact on the effect and residual concentration of chemicals. The purpose is to determine the impact of environmental conditions on the chemical processing effect during the processing process by combining environmental sensor data and textile chemical dosage data. For example, suppose that in a certain coating process, the temperature sensor and humidity sensor respectively record a temperature of 70°C and a humidity of 60%, while the chemical dosage is 300g of waterproofing agent and 200g of softener. Through the correlation analysis of these environmental data and dosage data, a statistical model (such as regression analysis or machine learning algorithm) is used to establish the association rules between environmental conditions and chemical dosage, and finally a set of textile chemical processing parameters, such as the chemical concentration distribution under the optimal temperature and humidity conditions, are obtained, which provides support for the subsequent calculation and optimization of residual concentration factors.

[0149] Step S44: Calculating the chemical residue concentration factor according to the fabric chemical residue concentration data and the fabric chemical processing parameter group, thereby obtaining the chemical residue concentration factor;

[0150] In this embodiment, the weights (factors) of various factors affecting the residual concentration of fabric chemicals are calculated. First, the obtained fabric chemical residual concentration data are combined with the obtained fabric chemical processing parameter group, and a mathematical model is used for analysis. For example, a multivariate linear regression model or a machine learning algorithm such as a support vector machine (SVM) is used to determine the relationship between the fabric residual concentration and the processing environment parameters (such as temperature, humidity) and the amount of chemicals used. Hypothesis analysis shows that temperature has a greater impact on the residual concentration of a certain chemical substance, humidity has a relatively small impact, and the impact of the amount of chemicals is moderate. Based on these analyses, the weight of each factor (such as temperature, humidity, and chemical amount) on the residual concentration of the chemical is finally calculated to form a chemical residual concentration factor.

[0151] Step S45: constructing a chemical residue prediction model based on the chemical residue concentration factor and the textile chemical processing parameter group.

[0152] In this embodiment, a chemical residue prediction model is constructed based on the chemical residue concentration factor and the fabric chemical processing parameter group obtained by the above calculation. The model outputs the expected residual concentration by inputting environmental conditions (such as processing temperature, humidity, processing time, etc.) and chemical dosage data. For example, a machine learning algorithm (such as a decision tree, a neural network or an XGBoost model) is used to train the data set to establish a prediction relationship for the chemical residue concentration. Assuming that the training data includes fabric samples using different amounts of chemicals at different temperatures and humidities, after model training, new chemical dosages and environmental parameters are input to predict the residual concentration. This prediction model can not only help optimize the production process, but also provide a scientific basis for quality control to ensure that the product meets environmental protection and safety standards.

[0153] Optionally, step S5 specifically includes:

[0154] Step S51: acquiring real-time production fabric data, and extracting production fabric material features from the real-time production fabric data, thereby obtaining real-time production fabric material data;

[0155] In this embodiment, in a real-time production environment, the production data of the fabric is collected through sensor systems and data acquisition equipment, such as optical fiber sensors, cameras, temperature and humidity sensors, etc. These data include the weight, density, fiber type, thickness, tensile strength, etc. of the fabric. In the data processing stage, image recognition technology and material analysis models, such as deep learning algorithms (convolutional neural networks), are used to extract the material characteristics of the fabric. For example, the texture of the fabric surface is analyzed using image recognition technology, and the material type of the fabric (such as cotton, linen, silk, chemical fiber, etc.) and its related mechanical properties data are extracted in combination with the temperature and humidity data collected by the sensor. These data ultimately constitute "real-time production fabric material data" for subsequent analysis.

[0156] Step S52: acquiring real-time chemical processing environment data, and performing fabric chemical residue prediction on the real-time production fabric material data and the real-time chemical processing environment data through a chemical residue prediction model, thereby obtaining fabric chemical residue prediction data;

[0157] In this embodiment, chemical processing environment data including chemical concentration, reaction temperature, humidity, harmful gases in the air, etc. are monitored in real time through chemical sensors, gas detectors, etc. At the same time, the chemical residue prediction model is combined with real-time fabric material data (such as fiber hygroscopicity, fabric chemical resistance, etc.) to predict possible chemical residues. For example, when the fabric is exposed to certain chemicals during the dyeing process, the model will predict whether the dye or other chemicals will leave residues in the fabric based on the input fabric material and chemical environment data, including the type and concentration of the residual chemicals.

[0158] Step S53: performing statistics of chemicals with high residual amounts according to the textile chemical residue prediction data, thereby obtaining data of chemicals with high residual amounts;

[0159] In this embodiment, different types of chemicals are counted and analyzed, with particular attention paid to chemicals with high residual amounts. For example, chemicals such as dyes and bleaches whose residual amounts exceed the prescribed limit can be identified based on the prediction results. By setting a threshold (e.g., dye residual amounts greater than 0.5 mg / cm² are high residual amounts), those chemicals with high residual amounts will be screened out and counted. For each chemical, its residual amount in different batches of fabrics is counted, and a "high residual amount chemical data" report is generated. This report lists the name, residual amount, residual ratio, and comparison data with the standard limit of each chemical to facilitate judgment of whether it exceeds the standard. For example, it is found that a batch of fabrics has a high amount of dye residuals during the dyeing process, which exceeds the prescribed safety standard (e.g., 1 mg / cm²). These high residual chemicals will be treated and improved as key issues.

[0160] Step S54: Calculating the chemical residue percentage of textile processing chemical usage data and high-value residual chemical data, thereby obtaining textile chemical residue percentage data;

[0161] In this embodiment, the ratio calculation is performed on the processing chemical usage data of the fabric (such as the input amount of dyes, bleaching agents, auxiliaries, etc.) and the obtained high-value residual chemical data. This can be calculated by the following formula: Chemical Residue Percentage = Ratio of High-value Residue of Total Chemical Usage to Chemical Residue. For example, if 10 kg of dye is input during the dyeing process of a batch of fabrics, but the dye residue is 1 kg, the dye residue ratio of the batch of fabrics is 10%. Through this calculation, the contribution of different chemicals to the final residue of the fabric can be evaluated, which helps optimize the process and reduce unnecessary chemical use.

[0162] Step S55: Performing a residue risk assessment based on the textile chemical residue percentage data, thereby obtaining a real-time textile chemical residue risk assessment report.

[0163] In this embodiment, the risk of chemical residues is assessed based on the data of the percentage of chemical residues in the fabric using a risk assessment model (such as AHP method, fuzzy comprehensive evaluation method, hierarchical analysis method, etc.). The assessment model will take into account multiple factors, such as the toxicity of the chemical, the residual concentration, the control status of the processing environment, etc., and ultimately derive the overall chemical residue risk level of the fabric. For example, when the percentage of the residue of certain chemical substances exceeds a predetermined safety threshold, an early warning report will be issued, indicating that the risk of the batch of fabrics is high and requires additional inspection or treatment. Finally, a real-time production fabric chemical residue risk assessment report is generated, which will list in detail the risk level of each chemical substance, possible health effects, recommended treatment measures, etc., to provide a basis for production decisions.

[0164] Optionally, the present invention further provides a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the method for detecting chemical residues in fabrics as described above.

[0165] Optionally, the present invention further provides a textile chemical residue detection system, which is used to perform the textile chemical residue detection method as described above, and the textile chemical residue detection system comprises:

[0166] A processing information integration module is used to obtain fabric production sensor data, and to perform fabric chemical processing flow structure modeling on the fabric production sensor data, thereby obtaining a fabric chemical processing flow structure model; and to perform fabric chemical processing information integration based on the fabric chemical processing flow structure model, thereby obtaining fabric processing chemical data;

[0167] The electromagnetic reflection pattern recognition module is used to obtain electromagnetic detection data of fabrics, and perform electromagnetic reflection pattern recognition of fabrics according to the electromagnetic detection data of fabrics, so as to obtain electromagnetic reflection pattern data of fabrics; perform abnormal electromagnetic pattern screening based on the electromagnetic reflection pattern data of fabrics, so as to obtain abnormal electromagnetic reflection pattern data;

[0168] A fabric residual fingerprint matching module is used to construct a chemical molecular fingerprint library based on textile processing chemical data, and to perform fabric residual fingerprint matching on abnormal electromagnetic reflection pattern data according to the chemical molecular fingerprint library, thereby obtaining fabric residual fingerprint matching data;

[0169] A residual concentration factor calculation module is used to calculate the chemical residual concentration factor according to the textile processing chemical data and the textile residual fingerprint matching data, so as to obtain the chemical residual concentration factor; and to construct a chemical residue prediction model based on the chemical residual concentration factor;

[0170] The residual risk assessment module is used to obtain real-time fabric production data and conduct fabric chemical residue risk assessment on the real-time fabric production data through a chemical residue prediction model, thereby obtaining a real-time fabric production chemical residue risk assessment report.

[0171] The textile chemical residue detection system of the present invention can realize any textile chemical residue detection method of the present invention, and is used to combine the operation and signal transmission medium between various modules to complete the textile chemical residue detection method. The internal modules of the system cooperate with each other to effectively identify low-concentration residues.

[0172] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0173] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for detecting chemical residues in fabrics, characterized in that: The following steps are involved: Step S1: Acquire fabric production sensor data, and perform fabric chemical processing flow structure modeling on the fabric production sensor data, thereby obtaining a fabric chemical processing flow structure model; integrate fabric chemical processing information based on the fabric chemical processing flow structure model, thereby obtaining fabric processing chemical data; Step S2: Acquire fabric electromagnetic detection data, and perform fabric electromagnetic reflection pattern recognition based on the fabric electromagnetic detection data, thereby obtaining fabric electromagnetic reflection pattern data; perform abnormal electromagnetic pattern screening based on the fabric electromagnetic reflection pattern data, thereby obtaining abnormal electromagnetic reflection pattern data; Step S3: constructing a chemical molecular fingerprint library based on the textile processing chemical data, and performing fabric residual fingerprint matching on the abnormal electromagnetic reflection pattern data according to the chemical molecular fingerprint library, thereby obtaining fabric residual fingerprint matching data; Step S4: Calculate the chemical residue concentration factor according to the textile processing chemical data and the textile residue fingerprint matching data, thereby obtaining the chemical residue concentration factor; and construct a chemical residue prediction model based on the chemical residue concentration factor; Step S5: acquiring real-time fabric production data, and performing fabric chemical residue risk assessment on the real-time fabric production data through a chemical residue prediction model, thereby obtaining a real-time fabric production chemical residue risk assessment report.

2. The method for detecting chemical residues in fabrics according to claim 1, characterized in that: Step S1 is specifically as follows: Step S11: acquiring fabric production sensor data, and performing chemical sensor data extraction on the fabric production sensor data, thereby obtaining fabric production chemical sensor data; Step S12: extracting chemical sensing amplitude fluctuation characteristics based on the chemical sensing data of fabric production, thereby obtaining fabric chemical sensing fluctuation data; Step S13: dividing the chemical sensing fluctuation data of the fabric into sensing fluctuation time series, thereby obtaining chemical sensing fluctuation time series data, and performing fluctuation stage dynamic time regularization according to the chemical sensing fluctuation time series data, thereby obtaining chemical processing fluctuation stage data; Step S14: performing fabric chemical processing flow structure modeling based on the chemical processing fluctuation stage data, thereby obtaining a fabric chemical processing flow structure model; Step S15: Integrate the textile chemical processing information based on the textile chemical processing flow structure model, so as to obtain textile processing chemical data.

3. The method for detecting chemical residues in fabrics according to claim 2, characterized in that: Step S15 is specifically as follows: Step S151: extracting the time structure features of the chemical processing flow of the textile product from the structure model of the chemical processing flow, thereby obtaining the time data of the chemical processing flow; Step S152: integrating the process environment sensing characteristics of the textile chemical process structure model according to the textile production chemical sensing data, thereby obtaining the process environment sensing data; Step S153: obtaining a chemical substance characteristic database, and identifying chemical substances in the chemical process flow of the textile based on the chemical process time data, the process environment sensor data, and the chemical process fluctuation stage data, thereby obtaining chemical substance data in the chemical process flow of the textile; Step S154: integrating the information of textile processing reaction chemicals into the chemical substance characteristic database based on the textile chemical processing flow reaction chemical data, thereby obtaining textile processing chemical data.

4. The method for detecting chemical residues in textiles according to claim 1, characterized in that: Step S2 is specifically as follows: Step S21: acquiring fabric electromagnetic detection data, and performing noise reduction processing on the fabric electromagnetic detection data, thereby obtaining fabric electromagnetic noise reduction data; Step S22: dividing the fabric electromagnetic noise reduction data into fabric space grids, thereby obtaining fabric grid electromagnetic detection data; Step S23: extracting fabric mesh electromagnetic reflection intensity features from the fabric mesh electromagnetic detection data to obtain fabric mesh electromagnetic reflection intensity data, and performing reflection intensity grid division according to the fabric mesh reflection intensity data to obtain high-frequency reflection intensity grid data and low-frequency reflection intensity grid data; Step S24: performing high-frequency reflection intensity grid reflection pattern recognition on the fabric grid electromagnetic detection data according to the high-frequency reflection intensity grid data, thereby obtaining the fabric grid high-frequency reflection pattern data; According to the low-frequency reflection intensity grid data, the electromagnetic detection data of the fabric grid is subjected to low-frequency reflection intensity grid reflection pattern recognition, thereby obtaining the low-frequency reflection pattern data of the fabric grid; Step S25: performing fabric grid spatial merging on the fabric grid high-frequency reflection pattern data and the fabric grid low-frequency reflection pattern data, thereby obtaining fabric electromagnetic reflection pattern data; Step S26: performing abnormal electromagnetic pattern screening based on the electromagnetic reflection pattern data of the fabric, thereby obtaining abnormal electromagnetic reflection pattern data.

5. The method for detecting chemical residues in textiles according to claim 4, characterized in that: Step S26 is specifically as follows: Step S261: performing electromagnetic reflection pattern clustering according to the electromagnetic reflection pattern data of the fabric, thereby obtaining electromagnetic reflection pattern clustering data; Step S262: performing clustered reflection pattern grid distribution statistics based on the electromagnetic reflection pattern clustering data, thereby obtaining clustered reflection pattern grid distribution data; Step S263: screening the clustered reflection pattern grid distribution data for the reflection pattern with high difference in local reflection intensity of the fabric, thereby obtaining first abnormal electromagnetic reflection pattern data; Step S264: performing distribution discrete reflection pattern screening on the clustered reflection pattern grid distribution data, thereby obtaining second abnormal electromagnetic reflection pattern data; Step S265: performing an abnormal electromagnetic reflection pattern intersection operation on the first abnormal electromagnetic reflection pattern data and the second abnormal electromagnetic reflection pattern data, so as to obtain abnormal electromagnetic reflection pattern data.

6. The method for detecting chemical residues in fabrics according to claim 1, characterized in that: Step S3 is specifically as follows: Step S31: extracting processing chemical substance characteristics from a chemical substance characteristic database based on the textile processing chemical substance data, thereby obtaining processing chemical substance characteristic data; Step S32: constructing a chemical molecular fingerprint library according to the characteristic data of the processed chemical substances; Step S33: performing principal component analysis on the abnormal electromagnetic reflection pattern data to obtain an abnormal reflection feature vector; Step S34: Calculate the Tanimoto coefficient of the abnormal reflection feature vector and the chemical molecule fingerprint library to obtain the fingerprint Tanimoto coefficient; Step S35: performing residual chemical molecular fingerprint matching on the abnormal electromagnetic reflection pattern data according to the fingerprint Tanimoto coefficient, thereby obtaining fabric residual fingerprint matching data.

7. The method for detecting chemical residues in fabrics according to claim 1, characterized in that: Step S4 is specifically as follows: Step S41: Calculating the residual concentration of fabric chemicals based on the fabric residual fingerprint matching data, thereby obtaining fabric chemical residual concentration data; Step S42: extracting the usage characteristics of textile processing chemicals from the textile processing chemical data, thereby obtaining textile processing chemical usage data; Step S43: performing textile chemical processing correlation on the processing process environment sensor data and the textile processing chemical dosage data, thereby obtaining a textile chemical processing parameter group; Step S44: Calculating the chemical residue concentration factor according to the fabric chemical residue concentration data and the fabric chemical processing parameter group, thereby obtaining the chemical residue concentration factor; Step S45: constructing a chemical residue prediction model based on the chemical residue concentration factor and the textile chemical processing parameter group.

8. The method for detecting chemical residues in fabrics according to claim 1, characterized in that: Step S5 is specifically as follows: Step S51: acquiring real-time production fabric data, and extracting production fabric material features from the real-time production fabric data, thereby obtaining real-time production fabric material data; Step S52: acquiring real-time chemical processing environment data, and performing fabric chemical residue prediction on the real-time production fabric material data and the real-time chemical processing environment data through a chemical residue prediction model, thereby obtaining fabric chemical residue prediction data; Step S53: performing statistics of chemicals with high residual amounts according to the textile chemical residue prediction data, thereby obtaining data of chemicals with high residual amounts; Step S54: Calculating the chemical residue percentage of textile processing chemical usage data and high-value residual chemical data, thereby obtaining textile chemical residue percentage data; Step S55: Performing a residue risk assessment based on the textile chemical residue percentage data, thereby obtaining a real-time textile chemical residue risk assessment report.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for detecting chemical residues in textiles as claimed in any one of claims 1 to 8 is implemented.

10. A detection system for chemical residues in textiles, characterized in that: The method for detecting chemical residues in textiles according to claim 1 is used to perform the method, the detection system for chemical residues in textiles comprising: A processing information integration module is used to obtain fabric production sensor data, and to perform fabric chemical processing flow structure modeling on the fabric production sensor data, thereby obtaining a fabric chemical processing flow structure model; and to perform fabric chemical processing information integration based on the fabric chemical processing flow structure model, thereby obtaining fabric processing chemical data; The electromagnetic reflection pattern recognition module is used to obtain electromagnetic detection data of fabrics, and perform electromagnetic reflection pattern recognition of fabrics according to the electromagnetic detection data of fabrics, so as to obtain electromagnetic reflection pattern data of fabrics; perform abnormal electromagnetic pattern screening based on the electromagnetic reflection pattern data of fabrics, so as to obtain abnormal electromagnetic reflection pattern data; A fabric residual fingerprint matching module is used to construct a chemical molecular fingerprint library based on textile processing chemical data, and to perform fabric residual fingerprint matching on abnormal electromagnetic reflection pattern data according to the chemical molecular fingerprint library, thereby obtaining fabric residual fingerprint matching data; A residual concentration factor calculation module is used to calculate the chemical residual concentration factor according to the textile processing chemical data and the textile residual fingerprint matching data, so as to obtain the chemical residual concentration factor; and to construct a chemical residue prediction model based on the chemical residual concentration factor; The residual risk assessment module is used to obtain real-time fabric production data and conduct fabric chemical residue risk assessment on the real-time fabric production data through a chemical residue prediction model, thereby obtaining a real-time fabric production chemical residue risk assessment report.

Images