Method and system for monitoring the entire recycling process of fluorine-containing rubber waste

Abstract

The application discloses a recycling whole-process monitoring method and system for fluorine-containing rubber waste, and relates to the technical field of rubber waste recycling, which comprises the following steps: performing multi-modal sensing analysis on the fluorine-containing rubber waste to generate a set of waste characteristic parameters; constructing a recycling process database to obtain real-time recycling state data; performing online detection according to the real-time recycling state data to extract quality abnormality early warning signals; establishing a recycling whole-process data chain for dynamic monitoring to generate recycling correction parameters, updating the recycling whole-process data chain through the recycling correction parameters, and obtaining a recycling whole-process optimized data chain. The application solves the technical problems of low recycling efficiency and unstable product quality caused by the lagging recycling process monitoring and inaccurate control of fluorine-containing rubber waste in the prior art, achieves real-time monitoring and accurate control of the recycling whole process of fluorine-containing rubber waste, and improves the recycling efficiency and product quality.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of rubber waste recycling, in particular to a recycling whole-process monitoring method and system for fluorine-containing rubber waste. BACKGROUND

[0002] With the rapid development of industry, fluorine-containing rubber is widely used in many fields, and the resulting fluorine-containing rubber waste is also increasing. If fluorine-containing rubber waste is not properly treated, it will not only cause resource waste, but also may cause environmental pollution problems, so its recycling is of great significance. However, there are many problems in the current fluorine-containing rubber waste recycling process, the traditional recycling monitoring method relies on manual experience judgment, lacks accurate analysis of waste characteristics, and it is difficult to realize real-time and dynamic monitoring of the whole process, resulting in low recycling efficiency and unstable product quality, which cannot meet the requirements of modern green environmental protection and efficient production.

[0003] The prior art has the technical problems of low recycling efficiency and unstable product quality caused by lagging monitoring of the fluorine-containing rubber waste recycling process and inaccurate regulation. SUMMARY

[0004] The present application provides a recycling whole-process monitoring method and system for fluorine-containing rubber waste, which is used to solve the technical problems of low recycling efficiency and unstable product quality caused by lagging monitoring of the fluorine-containing rubber waste recycling process and inaccurate regulation in the prior art.

[0005] In view of the above problems, the present application provides a recycling whole-process monitoring method and system for fluorine-containing rubber waste.

[0006] In a first aspect of the present application, a recycling whole-process monitoring method for fluorine-containing rubber waste is provided, the method comprising:

[0007] Multi-modal sensing analysis is performed on the fluorine-containing rubber waste to generate a set of waste characteristic parameters; a recycling process database is constructed, and based on the recycling process database, recycling records are made according to the set of waste characteristic parameters to obtain real-time recycling state data; online detection is performed according to the real-time recycling state data, a product quality evaluation matrix is constructed according to a detection data set, and a quality abnormality early warning signal is extracted; a recycling whole-process data chain is established, dynamic monitoring is performed according to the mapping of the quality abnormality early warning signal to the recycling whole-process data chain, a recycling correction parameter is generated, the recycling whole-process data chain is updated by the recycling correction parameter, and a recycling whole-process optimization data chain is obtained.

[0008] In a possible implementation, the fluororubber waste is detected by a multi-modal sensor group to obtain a multi-modal sensor data set; the multi-modal sensor data set is spatio-temporally aligned, principal component analysis is performed on the multi-modal initial alignment data set to obtain a multi-modal alignment data set; confidence evaluation is performed based on the multi-modal alignment data set, and the multi-modal alignment data set is excluded according to the evaluation result to generate a multi-modal data set; composite feature analysis is performed based on the multi-modal data set to generate the waste feature parameter set.

[0009] In a possible implementation, a recycling process database is constructed, and the recycling process database includes a material characteristic sub-database, a process parameter sub-database, and an environmental protection standard sub-database; the waste feature parameter set is dynamically matched with the material characteristic sub-database, the process parameter sub-database, and the environmental protection standard sub-database to generate target recycling process data; the device control instruction based on the target recycling process data is activated to record the recycling time sequence to obtain recycling record data flow; the recycling record data flow is encrypted and stored to generate a recycling process traceability chain, and the real-time recycling state data is extracted according to the recycling process traceability chain.

[0010] In a possible implementation, the molecular structure spectral features of the fluororubber waste are extracted based on the waste feature parameter set, and a fluororubber material fingerprint library is constructed based on the molecular structure spectral features; a material degradation degree evaluation is performed based on the fluororubber material fingerprint library to generate a degradation degree score; the thermal stability grade of the fluororubber waste is calculated according to the degradation degree score, and the material characteristics are integrated based on the thermal stability grade and the fluororubber material fingerprint library to construct the material characteristic sub-database.

[0011] In a possible implementation, historical recycling data logs are called, and the historical recycling data logs include a plurality of recycling process parameters and a plurality of waste recycling parameters, and the plurality of recycling process parameters and the plurality of waste recycling parameters have a corresponding relationship; the plurality of recycling process parameters and the plurality of waste recycling parameters are associated to construct a parameter association matrix; process control is performed by using digital twinning according to the parameter association matrix to generate a virtual process parameter set, and the data blind area is supplemented based on the virtual process parameter set to construct the process parameter sub-database.

[0012] In a possible implementation, the fluororubber waste is detected in real time according to the real-time recycling state data to obtain a detection data set; a multi-dimensional quality evaluation index is constructed based on the environmental protection standard sub-library, the detection data set is mapped to the multi-dimensional quality evaluation index to generate a real-time quality parameter vector; dynamic weight distribution is performed based on the real-time quality parameter vector to construct a product quality evaluation matrix; matrix outlier evaluation is performed on the product quality evaluation matrix to obtain a plurality of outlier data points, abnormality recognition is performed based on the plurality of outlier data points, and a quality abnormality feature vector is extracted according to an abnormal mode; the historical abnormality database is called, the quality abnormality feature vector is matched with the historical abnormality database to generate a hierarchical early warning signal, and the hierarchical early warning signal is added to the quality abnormality early warning signal.

[0013] In a possible implementation, the product quality evaluation matrix is subjected to dynamic matrix decomposition to obtain a decomposition result, and the decomposition result includes an outlier residual matrix; multi-dimensional spatial outlier degree calculation is performed on the outlier residual matrix to extract a plurality of outlier data points; the plurality of outlier data points are mapped to a time-space coordinate system of a recycling process flow to construct an outlier time-space correlation graph; an outlier mode feature is extracted by traversing the outlier time-space correlation graph for analysis to generate the quality abnormality feature vector.

[0014] In a possible implementation, the quality abnormality early warning signal is mapped to the recycling full-process data chain to construct a time-space correlation graph; positioning and identification are performed according to the time-space correlation graph to determine a plurality of abnormal source position labels; multi-dimensional influence analysis of abnormal sources is performed by traversing the plurality of abnormal source position labels to generate a recycling correction parameter; the recycling correction parameter is traced back to the recycling full-process data chain for online learning to generate a data learning result; closed-loop verification testing is performed on the recycling full-process data chain according to the data learning result to construct the recycling full-process optimization data chain.

[0015] In a possible implementation, a plurality of industrial Internet of Things nodes are deployed based on a recycling process flow to perform real-time collection to obtain equipment operation parameters; process parameter correlation analysis is performed according to the recycling process flow to construct a process parameter correlation network, and a process parameter coupling relationship is identified based on the process parameter correlation network; the equipment operation parameters are subjected to synchronous simulation according to the process parameter coupling relationship to construct the recycling full-process data chain.

[0016] In a second aspect of the present application, a recycling full-process monitoring system for fluororubber waste is provided, and the system comprises:

[0017] The multi-modal sensing analysis module is used for multi-modal sensing analysis of the fluorine-containing rubber waste to generate a waste characteristic parameter set; the recycling record module is used for constructing a recycling process database, performing recycling record based on the recycling process database according to the waste characteristic parameter set, and obtaining real-time recycling state data; the abnormal early warning signal extraction module is used for online detection according to the real-time recycling state data, constructing a product quality evaluation matrix according to a detection data set, and extracting a quality abnormal early warning signal; and the dynamic monitoring module is used for establishing a recycling whole-process data chain, performing dynamic monitoring according to the quality abnormal early warning signal mapped to the recycling whole-process data chain, generating a recycling correction parameter, updating the recycling whole-process data chain through the recycling correction parameter, and obtaining a recycling whole-process optimization data chain.

[0018] The one or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0019] The multi-modal sensing analysis module is used for multi-modal sensing analysis of the fluorine-containing rubber waste to generate a waste characteristic parameter set; the recycling record module is used for constructing a recycling process database, performing recycling record based on the recycling process database according to the waste characteristic parameter set, and obtaining real-time recycling state data; the abnormal early warning signal extraction module is used for online detection according to the real-time recycling state data, constructing a product quality evaluation matrix according to a detection data set, and extracting a quality abnormal early warning signal; and the dynamic monitoring module is used for establishing a recycling whole-process data chain, performing dynamic monitoring according to the quality abnormal early warning signal mapped to the recycling whole-process data chain, generating a recycling correction parameter, updating the recycling whole-process data chain through the recycling correction parameter, and obtaining a recycling whole-process optimization data chain. The technical effect of realizing real-time monitoring and precise regulation of the fluorine-containing rubber waste recycling whole process and improving recycling efficiency and product quality is achieved. BRIEF DESCRIPTION OF DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0021] Figure 1 A recycling whole-process monitoring method flowchart for fluorine-containing rubber waste is provided for the embodiments of the present application.

[0022] Figure 2 A recycling whole-process monitoring system structure diagram for fluorine-containing rubber waste is provided for the embodiments of the present application.

[0023] Legend of the drawings: multi-modal sensing analysis module 10, recycling record module 20, abnormal early warning signal extraction module 30, dynamic monitoring module 40. DETAILED DESCRIPTION

[0024] The application provides a full-process monitoring method and system for fluorine-containing rubber waste recovery, which is used to solve the technical problems of low recovery efficiency and unstable product quality caused by lagging recovery process monitoring and inaccurate regulation in the prior art.

[0025] The technical solutions in the embodiments of the application will be clearly and completely described below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by a person of ordinary skill in the art without creative work belong to the protection scope of the application.

[0026] Embodiment one, as shown in the application, a full-process monitoring method for fluorine-containing rubber waste recovery is provided, which comprises the following steps: Figure 1

[0027] Step S100: multi-modal sensing analysis is performed on the fluorine-containing rubber waste to generate a waste characteristic parameter set.

[0028] Specifically, the chemical composition (near-infrared spectrometer, X-ray fluorescence spectrometer), three-dimensional morphology (three-dimensional vision system) and moisture content (microwave moisture sensor) of the fluorine-containing rubber waste are detected by the multi-modal sensor group composed of a near-infrared spectrometer, a three-dimensional vision system, an X-ray fluorescence spectrometer and a microwave moisture sensor, to obtain a multi-modal sensing data set containing spectral feature data, three-dimensional image data, element composition data and moisture content data; the data set is time and space aligned to unify the time and space reference, and the principal component analysis is used to remove redundant information to obtain a multi-modal aligned data set; the aligned data set is subjected to confidence evaluation to remove low-confidence abnormal data to generate a multi-modal data set; finally, through composite feature analysis, key information such as molecular structure spectral features, element composition ratios, three-dimensional size parameters and moisture contents is extracted from the multi-modal data to generate a waste characteristic parameter set representing the characteristics of the fluorine-containing rubber waste.

[0029] Step S200: a recovery process database is constructed, and real-time recovery state data is obtained by recording recovery according to the waste characteristic parameter set based on the recovery process database.

[0030] ​Specifically, a recycling process database is constructed, which includes a material property sub-database, a process parameter sub-database, and an environmental standard sub-database. The material property sub-database is formed by extracting the spectral features of the molecular structure of waste materials to construct a material fingerprint library and evaluating degradation degree and thermal stability grade. The process parameter sub-database is generated by constructing a parameter correlation matrix based on historical recycling data logs and supplementing data blind spots through digital twinning. The environmental standard sub-database includes information such as pollutant emission standards. The characteristic parameter set of the waste material is dynamically matched with the three sub-databases to generate target recycling process data, activate equipment control instructions, record the recycling sequence to form a recycling record data stream, and generate a recycling process traceability chain by encrypting the data stream. Real-time recycling state data is extracted from the traceability chain.

[0031] Step S300: Online detection is performed according to the real-time recycling state data, a product quality evaluation matrix is constructed based on the detection data set, and a quality abnormality early warning signal is extracted.

[0032] Specifically, according to the real-time recycling state data, the recycling process of fluorine-containing rubber waste is detected online, and a detection data set containing indicators such as component purity and physical properties is obtained. Based on the environmental standard sub-database, evaluation indicators covering environmental protection and quality dimensions are constructed, and the detection data is mapped to generate a real-time quality parameter vector. Through a dynamic weight distribution algorithm, each parameter is assigned a weight, and a product quality evaluation matrix is constructed. The matrix is dynamically decomposed, the multi-dimensional spatial outlying degree of the outlying residual matrix is calculated, the outlying data points are mapped to the time-space coordinate system of the recycling process, an outlying space-time correlation map is constructed, the outlying mode features in the map are analyzed, and a quality abnormality feature vector is generated. The vector is matched with the historical abnormality database to generate a hierarchical warning signal and integrated into a quality abnormality early warning signal.

[0033] Step S400: Establish a recycling full-process data chain, dynamically monitor the recycling full-process data chain according to the mapping of the quality abnormality early warning signal, generate recycling correction parameters, update the recycling full-process data chain through the recycling correction parameters, and obtain a recycling full-process optimization data chain.

[0034] Specifically, based on the recycling process deployment of industrial Internet of Things nodes, real-time collection of equipment operation parameters (such as temperature, pressure, speed, etc.), construction of process parameter correlation network through process parameter correlation analysis, and identification of coupling relationship between parameters. The equipment operation parameters are simulated synchronously according to the coupling relationship, and the data flow chain covering the whole recycling process, i.e. the recycling whole process data chain, is constructed. When receiving the quality abnormal early warning signal, it is mapped to the data chain, a space-time correlation graph is constructed, and the abnormal source position label is located. Multi-dimensional influence analysis (such as influence on product quality, process efficiency and environmental protection index) is carried out on each abnormal source, and recycling correction parameters including process parameter adjustment and equipment control strategy optimization are generated. The correction parameters are traced back to the data chain for online learning, and the correction effect is verified through closed-loop verification test, and finally the optimized recycling whole process data chain is generated, realizing dynamic optimization and continuous monitoring of the recycling process.

[0035] In one possible implementation manner, step S100 further includes:

[0036] Step S110: The fluorine-containing rubber waste is detected by a multi-modal sensor group to obtain a multi-modal sensing data set.

[0037] Step S120: The multi-modal sensing data set is spatio-temporally aligned, and principal component analysis is performed on the multi-modal initial alignment data set to obtain a multi-modal alignment data set.

[0038] Step S130: Confidence evaluation is performed based on the multi-modal alignment data set, and the multi-modal alignment data set is excluded according to the evaluation result to generate a multi-modal data set.

[0039] Step S140: Compound feature analysis is performed based on the multi-modal data set to generate the waste feature parameter set.

[0040] Specifically, the fluororubber waste is detected by a multi-modal sensor group composed of a near-infrared spectrometer, a three-dimensional vision system, an X-ray fluorescence spectrometer and a microwave moisture sensor: the near-infrared spectrometer scans the surface of the waste at multiple wavelengths to obtain fluororubber molecular structure characteristic data for analyzing its chemical composition and molecular chain structure; the three-dimensional vision system constructs a three-dimensional point cloud model of the waste by optical imaging technology to calculate geometric parameters such as volume density, particle size and surface roughness of the waste; the X-ray fluorescence spectrometer scans the cross-section of the waste to generate a distribution map of heavy metal pollutants (such as iron, copper, zinc, etc.) to identify the types and contents of impurity elements; the microwave moisture sensor uses microwave reflection principle to non-contact measure the internal moisture content of the waste to obtain dynamic moisture gradient data at different positions of the waste. Through synchronous data collection of multiple sensors, a multi-modal sensor data set containing molecular structure characteristics, three-dimensional geometric parameters, element distribution information and moisture gradient data is formed to realize multi-dimensional and high-precision detection of the physical and chemical properties of fluororubber waste.

[0041] For the multi-modal sensor data set of fluororubber waste collected by the multi-modal sensor group including near-infrared spectrometer, three-dimensional vision system, X-ray fluorescence spectrometer and microwave moisture sensor, firstly, the time and space alignment processing is carried out, the time stamp and spatial coordinates of each sensor data are unified (such as taking the coordinate system of the waste treatment equipment as the reference), the time and space deviation caused by the difference in sampling time sequence and physical position of the sensors is eliminated, and the multi-modal initial alignment data set is formed; then principal component analysis algorithm is used for dimension reduction processing of the multi-modal initial alignment data set, high-dimensional data is mapped to low-dimensional feature space through linear transformation, redundant information (such as repeated background noise signals, non-key feature parameters) is removed, and principal components representing the core characteristics of the waste are retained, so that the multi-modal alignment data set with reduced dimension and independent features is obtained, laying a foundation for subsequent data fusion and feature analysis.

[0042] For the multi-modal alignment data set after time and space alignment and principal component analysis, a confidence evaluation method combining statistical analysis and machine learning is used: first, the Z-score algorithm is used to calculate the standard score of each dimension data to identify outliers deviating from the mean by more than 3 times the standard deviation; at the same time, the isolation forest algorithm is used to detect outliers in high-dimensional space, a binary tree is constructed to isolate each data point, and the path length of isolation is calculated to evaluate the outlier probability. Set a confidence threshold (such as 95%), mark the low-confidence data points with Z-score absolute value > 3 or outlier probability > 0.9 as outliers, automatically remove them through the data cleaning module, and retain the valid data with high confidence to finally generate a multi-modal data set with pure structure, ensuring the accuracy of subsequent feature analysis.

[0043] Based on the multi-modal data set, Fourier transform is adopted to perform frequency domain analysis on near-infrared spectrum data to extract molecular structure parameters such as C-F bond vibration characteristic peaks and characteristic absorption spectrum of fluorine-containing rubber molecules; three-dimensional point cloud model is combined with morphological algorithm to calculate three-dimensional geometric parameters such as volume density, average particle size and surface roughness of the waste; element quantitative analysis is performed on X-ray fluorescence spectrum data by using partial least squares method to determine the content of main elements such as fluorine, carbon and hydrogen and the concentration distribution of heavy metal impurities such as iron and copper; based on the time domain reflection data of the microwave moisture sensor, a moisture diffusion model is constructed to obtain internal moisture content and gradient distribution parameters of the waste; finally, the above multi-dimensional feature parameters are normalized and integrated to generate a waste feature parameter set containing chemical composition, physical form, impurity content and moisture state.

[0044] In one possible implementation manner, step S200 further includes:

[0045] Step S210: constructing a recycling process database containing a material property sub-database, a process parameter sub-database and an environmental protection standard sub-database.

[0046] Step S220: dynamically matching the waste feature parameter set with the material property sub-database, the process parameter sub-database and the environmental protection standard sub-database respectively to generate target recycling process data.

[0047] Step S230: based on the target recycling process data, activating device control instructions according to a recycling time sequence to record and obtain recycling record data stream.

[0048] Step S240: encrypting and storing the recycling record data stream to generate a recycling process traceability chain, and extracting the real-time recycling state data according to the recycling process traceability chain.

[0049] Specifically, a recycling process database containing a material property sub-database, a process parameter sub-database and an environmental protection standard sub-database is constructed. The material property sub-database stores fluorine-containing rubber types (such as fluorine rubber 23, fluorine rubber 26, etc.), molecular weight distribution (number average molecular weight and weight average molecular weight data measured by gel permeation chromatography), and impurity characteristic data including heavy metal content and foreign fiber mixing ratio, etc.; the process parameter sub-database records historical pyrolysis temperature curves including different stage heating rates, constant temperature interval temperature values, catalyst ratio and equipment control parameters (rotation speed of crushing equipment, screen mesh aperture of screening equipment); the environmental protection standard sub-database integrates compliance indicators such as fluorine emission limit value and heavy metal residue threshold value specified by the state and industry to form a stereoscopic data support system covering material attributes, process execution and environmental compliance.

[0050] The waste material characteristic parameter set, including molecular structure spectral characteristics, three-dimensional geometric parameters, element composition ratio, moisture content, etc., is dynamically matched with three sub-databases of the recycling process database respectively: through the spectral characteristic matching algorithm, the molecular structure data of the waste material such as C-F bond vibration frequency and characteristic absorption peak position are calculated with the cosine similarity degree with the standard spectrum of materials such as fluororubber 23 and fluororubber 26 in the material characteristic sub-database, and the material species with a matching degree > 90% are screened out, combined with the molecular weight distribution and impurity characteristics, the cracking temperature interval of the corresponding material (such as fluororubber 26 corresponding to 300-350°C) is retrieved; based on the physical parameters such as the bulk density and particle size of the waste material, the pyrolysis temperature curve in the process parameter sub-database is matched through the parameter correlation matrix (regression model trained by historical data), for example, when the average particle size of the waste material is > 5mm, the curve with a heating rate of 10°C / min and a constant temperature of 320°C is automatically associated, and according to the moisture content (> 15%), the amount of vulcanization accelerator is increased in proportion (for every 1% increase in moisture, the catalyst ratio is increased by 0.5%); the concentration of heavy metals such as lead and cadmium in the waste material and the potential release amount of fluoride are input into the compliance verification model of the environmental protection standard sub-database, if the lead content is > 80ppm, the control instruction of increasing the purification efficiency of the waste gas treatment equipment to 95% is triggered; finally, through the rule engine matching and model calculation of multi-dimensional data, the target recycling process data containing material degradation process parameters (cracking temperature 320°C), equipment control parameters (crushing rotation speed 1200r / min) and environmental threshold (fluoride emission real-time monitoring frequency increased to 1 time per minute) are integrated and generated.

[0051] Based on the target recycling process data (such as cracking temperature 320°C, crushing rotation speed 1200r / min, catalyst ratio adjustment coefficient, etc.), the control instructions are transmitted in real time to the controllers of the devices (such as the crusher unit, the cracking reaction kettle, and the catalyst automatic proportioning system) through the industrial internet of things platform, triggering the devices to perform operations according to the preset recycling sequence (pretreatment, crushing, cracking, and purification); the sensors deployed on the devices are used to collect running parameters (actual temperature, rotation speed, catalyst injection amount, etc.) in real time, and the data are packaged in time stamp order through the edge computing gateway, forming a structured data stream containing device ID, parameter name, real-time value, and execution time; the message queue is used to realize real-time transmission and buffering of data, ensuring the continuity and integrity of the recycling record data stream, and providing original data support for subsequent traceability chain construction and real-time state analysis.

[0052] Using blockchain technology or hash encryption algorithms, information such as equipment operating parameters, process execution time, and control commands in the recycling record data stream is encrypted. Data blocks are linked chronologically through a distributed ledger or chain data structure to generate an immutable recycling process traceability chain. Smart contracts or data parsing engines are used to traverse the timestamps, equipment IDs, parameter sequences, and other information in the traceability chain to extract key status data of the current recycling stage in real time (such as real-time temperature of the cracking reactor, operating speed of the crushing equipment, and actual amount of catalyst injected). This data is then compared with preset thresholds in the target recycling process data to form real-time recycling status data reflecting process execution deviations. This data is used for quality anomaly analysis and early warning signal generation in the subsequent online detection module.

[0053] In one possible implementation, step S210 further includes:

[0054] Step S211: Extract the molecular structure spectral features of fluorinated rubber waste based on the waste feature parameter set, and construct a fingerprint library of fluorinated rubber materials based on the molecular structure spectral features.

[0055] Step S212: Based on the fluorinated rubber material fingerprint library, evaluate the material degradation degree and generate a degradation degree score.

[0056] Step S213: Calculate the thermal stability level of the fluororubber waste based on the degradation score, and integrate the material properties based on the thermal stability level and the fluororubber material fingerprint library to construct the material property sub-library.

[0057] Specifically, near-infrared spectroscopic data of fluorinated rubber molecular structure (such as CF bond vibration frequency, characteristic absorption peak position and intensity) collected by a near-infrared spectrometer are extracted from the waste characteristic parameters. After standardization and normalization, baseline correction and noise interference are eliminated, a digital fingerprint template containing characteristic spectra of different types of fluorinated rubber is established to construct a fingerprint library of fluorinated rubber materials, thereby achieving accurate identification of the chemical composition of waste.

[0058] Based on the fingerprint database of fluororubber materials, the degree of molecular chain breakage and aging degradation of the material are evaluated by comparing the differences between the spectral characteristics of the waste and the standard fingerprint template (such as the degree of characteristic peak shift and the number of impurity peaks) and combining the functional group attenuation rate analysis of Fourier transform infrared spectroscopy. The degradation degree score of 0 to 100 is generated by using fuzzy comprehensive evaluation method or hierarchical analysis method to quantify the recyclability level of the waste.

[0059] According to the degradation score, the thermal stability of the fluorine-containing rubber waste is divided into different grades (such as 1st grade for high stability, 2nd grade for medium stability, and 3rd grade for low stability) through a preset threshold rule, for example, setting the degradation score ≥ 80 points as the 1st grade, 60-79 points as the 2nd grade, and < 60 points as the 3rd grade; then, the thermal stability grade is associated and integrated with the molecular structure spectral characteristics, degradation score and other data in the fingerprint library of the fluorine-containing rubber material, to form a multi-dimensional data set containing material types, spectral fingerprints, degradation degrees and thermal stability grades, which is stored in a structured manner by using a database management system, to build a material characteristic sub-database, and to provide material attribute basic data support for subsequent dynamic matching of recycling process parameters.

[0060] In one possible implementation, step S210 further includes:

[0061] Step S214: retrieving a historical recycling data log, the historical recycling data log containing a plurality of recycling process parameters and a plurality of waste recycling parameters, the plurality of recycling process parameters and the plurality of waste recycling parameters having a corresponding relationship.

[0062] Step S215: associating the plurality of recycling process parameters with the plurality of waste recycling parameters to construct a parameter association matrix.

[0063] Step S216: using digital twinning to control the process according to the parameter association matrix to generate a virtual process parameter set, supplementing a data blind area based on the virtual process parameter set, and constructing the process parameter sub-database.

[0064] Specifically, a historical recycling data log is retrieved from the recycling process database, which stores a plurality of data generated in the past fluorine-containing rubber waste recycling process, wherein the recycling process parameters include pyrolysis temperature, catalyst ratio, crushing equipment speed, cracking time, etc., and the waste recycling parameters include waste type degradation score, molecular structure spectral characteristics, impurity content (such as heavy metal concentration) and moisture content, etc.; after cleaning abnormal values and missing values by data preprocessing technology, a structured parameter corresponding relationship data set is formed, for example, fluorine-containing rubber 26 waste (degradation score 75 points, moisture content 12%) corresponds to pyrolysis temperature 320℃, catalyst ratio 5%, crushing speed 1500r / min, etc., clearly showing the actual corresponding relationship between the process parameters and the waste parameters, and providing a reliable data basis for subsequent parameter association analysis.

[0065] The correlation coefficients of the recycling process parameters (such as pyrolysis temperature, catalyst ratio, and crushing speed) and the waste recycling parameters (such as degradation score, impurity content, and moisture content) in the historical recycling data log are calculated by the Pearson correlation coefficient algorithm, and a threshold (such as |p|≥0.6) is set to screen strong correlation parameter pairs; the Apriori algorithm is used to mine the frequent item sets between the parameters to generate association rules (such as when the pyrolysis temperature is >300°C and the catalyst ratio is >4%, the degradation score is >70 points with a confidence of 85%); the above association results are structured and processed to construct an m*n-dimensional parameter association matrix, where the rows represent the process parameters and the columns represent the waste parameters, and the matrix elements are the correlation coefficients or confidence values of the parameter pairs. The correlation strength can be visualized through a heat map to form an association matrix that intuitively reflects the coupling relationship between parameters, providing a basis for the association model of digital twin simulation.

[0066] Based on the constructed parameter association matrix, a virtual simulation model of the fluororubber waste recycling process is created using digital twin technology, and the parameter association rules in the matrix are converted into constraint conditions for the simulation model; by inputting the key parameter combinations in the historical recycling data (such as the process parameter intervals corresponding to different waste types), various process scenarios are simulated in a virtual environment to generate a virtual process parameter set that covers the data blind area; the virtual parameter set is fused with the historical measured parameters to supplement the process parameter intervals not covered by the historical data (such as catalyst ratio under extreme humidity conditions), and a database management system is used to store the parameter association matrix, historical parameters, virtual parameters, and simulation verification results to construct a process parameter sub-library containing process parameter association relationships, parameter intervals, and simulation data, providing comprehensive data support for real-time recycling process parameter matching.

[0067] In one possible implementation, step S300 further includes:

[0068] Step S310: Real-time recycling online detection of the fluororubber waste according to the real-time recycling state data to obtain the detection data set.

[0069] Step S320: Construction of multi-dimensional quality evaluation indexes based on the environmental protection standard sub-library, mapping of the detection data set to the multi-dimensional quality evaluation indexes, and generation of a real-time quality parameter vector.

[0070] Step S330: Dynamic weight allocation based on the real-time quality parameter vector to construct a product quality evaluation matrix.

[0071] Step S340: Matrix outlier evaluation of the product quality evaluation matrix to obtain a plurality of outlier data points, abnormality identification based on the plurality of outlier data points, and extraction of a quality abnormality feature vector according to the abnormality mode.

[0072] Step S350: retrieve the historical abnormality database, match the quality abnormality feature vector with the historical abnormality database, generate a hierarchical early warning signal, and add the hierarchical early warning signal to the quality abnormality early warning signal.

[0073] Specifically, according to real-time recycling state data, the physical and chemical parameters of fluororubber waste at the stages of crushing, cracking, and purification are dynamically collected by deploying online detection equipment (such as near-infrared spectrometers, X-ray fluorescence spectrometers, microwave moisture sensors, and gas analyzers) at key processes in the recycling production line. The near-infrared spectrometer scans the surface of the waste in real time to obtain molecular structure spectral data to monitor the degree of material degradation; the X-ray fluorescence spectrometer analyzes the element distribution of the cross-section of the waste to track changes in heavy metal impurities; the microwave moisture sensor non-contact measures the internal moisture content fluctuations; and the gas analyzer monitors the emission concentration of fluorides and volatile organic compounds in real time during pyrolysis. The data collected by each device is preprocessed (denoising and format unification) by an edge computing unit, and then integrated to form a structured detection data set containing time stamps, process nodes, parameter types, and numerical values, providing real-time data support for subsequent quality evaluation and abnormality early warning.

[0074] From the environmental standard sub-library, extract compliance indicators such as fluoride emission limits, heavy metal residue thresholds, and energy consumption standards, and combine them with the physical performance requirements of the recycled products (such as particle size uniformity ≥ 90% and pyrolysis residue rate ≤ 5%) to build a multi-dimensional quality evaluation index system that includes chemical pollution, physical performance, and environmental compliance. The fluoride ion concentration, heavy metal content, moisture content, and equipment energy consumption parameters in the detection data set are mapped to the corresponding evaluation indicators through data normalization processing (such as min-max standardization), and normalized values in the 0-1 interval are generated according to the indicator thresholds to form a real-time quality parameter vector (such as [0.85, 0.92, 0.78] corresponding to the compliance scores of fluoride emissions, heavy metal residues, and energy consumption), which directly reflects the quality status of the recycled products and the degree of compliance with environmental standards.

[0075] A weight regulator based on fuzzy logic is adopted to assign real-time weights to multi-dimensional quality evaluation indicators according to the recycling stages (crushing, cracking, purification) and waste thermal stability grades. For example, when a waste thermal stability grade of 3 is detected, the weight of the temperature control parameter in the pyrolysis stage is automatically increased to 35%, the weight of the fluoride emission parameter is adjusted to 30%, the weight of the heavy metal residue parameter is set to 25%, and the weight of the energy consumption parameter is 10%. The real-time quality parameter vector is multiplied by the weight matrix through matrix operation to generate a two-dimensional evaluation matrix containing each process node and quality indicator. For example, the matrix elements of the cracking stage are [temperature compliance score 0.92 x 0.35, fluoride emission score 0.88 x 0.3, heavy metal residue score 0.95 x 0.25, and energy consumption score 0.80 x 0.1], realizing quantitative evaluation and visual display of quality risks at each link.

[0076] Non-negative matrix factorization is used to reduce the dimensionality of the product quality evaluation matrix, separating the basic matrix representing the normal mode and the outlier residual matrix containing abnormal information. By calculating the local outlier factor of each data point in the residual matrix, a threshold (such as outlier degree > 3σ) is set to extract outlier data points that significantly deviate from the mean. These outlier points are mapped to the space-time coordinate system of the recycling process (such as the time axis and the process node axis), and an outlier space-time correlation map is constructed. Through graph traversal algorithm, abnormal trajectories formed by continuous outlier points (such as cracking temperature continuously deviating from the threshold and fluoride emission exceeding the standard in a certain time period) are identified. Based on the trajectory characteristics (duration, parameter fluctuation amplitude, number of related indicators), the quality abnormal feature vector containing the abnormal type, severity, and occurrence node is extracted, such as [abnormal type-temperature out of control, severity-high, occurrence node-cracking furnace, duration-45 minutes].

[0077] Access the historical anomaly database, which stores the feature vectors of various abnormal events (such as fluoride leakage, catalyst ratio anomaly, equipment overheating, etc.) recorded in past recycling processes and their corresponding grading standards (I for emergency, II for warning, III for prompt); cosine similarity algorithm is used to match the real-time extracted quality abnormal feature vector with the feature vectors in the historical database, and the matching degree score is calculated; according to the preset matching degree threshold (such as matching degree ≥ 85% for I, 60%-84% for II, and < 60% for III), a graded warning signal is generated, for example, if a historical cracking temperature out of control event is matched with a matching degree of 92%, a I-level red warning is triggered; finally, the graded warning signal and the real-time abnormal feature are integrated to form a complete quality abnormal warning signal containing the risk level, abnormal type, and occurrence location, which is pushed to the system monitoring interface and related responsible persons.

[0078] In one possible implementation, step S340 further includes:

[0079] Step S341: Dynamic matrix decomposition is performed on the product quality evaluation matrix to obtain a decomposition result, and the decomposition result includes an outlier residual matrix.

[0080] Step S342: Multidimensional spatial outlying degree calculation is performed on the outlier residual matrix to extract a plurality of outlying data points.

[0081] Step S343: The plurality of outlying data points are mapped to a time-space coordinate system of a recycling process to construct an outlying time-space correlation graph.

[0082] Step S344: Outlying mode features are extracted from the outlying time-space correlation graph for analysis to generate the quality anomaly feature vector.

[0083] Specifically, dynamic modal decomposition is adopted to orthogonally decompose the product quality evaluation matrix. By setting a sliding time window (such as 5 minutes) and a process stage marker (crushing / lysis / purification) as dynamic adjustment factors, the original matrix is decomposed into a stable base matrix and an outlier residual matrix: the stable base matrix extracts core feature vectors under normal working conditions through non-negative matrix decomposition, representing the correlation mode of each quality parameter under standard process (such as the benchmark mapping relationship between temperature and emission concentration in the lysis stage); the outlier residual matrix is obtained by calculating the difference between the original matrix and the stable base matrix, and the element values thereof are standardized to reflect the abnormal degree of parameter fluctuation (such as the standard deviation of the deviation of the measured value of fluorinated emissions from the benchmark value). This process realizes real-time decomposition through GPU accelerated calculation, ensuring that a single batch matrix processing is completed within 200 ms, providing real-time data support for subsequent anomaly detection.

[0084] For the outlier residual matrix, a local outlier factor algorithm is used for multidimensional spatial outlying quantification. First, the dimensional parameters (such as fluorinated concentration, temperature, energy consumption, etc.) in the matrix are standardized to eliminate dimensional differences; then a local neighborhood is constructed with a sliding window (such as 10 consecutive sampling periods), the distance density ratio of each data point to other points in the neighborhood is calculated, and by setting a threshold, such as LOF>1.5, data points significantly deviating from the local neighborhood are screened out. For example, if the LOF value of the fluorinated emissions residual value in a lysis stage is 2.1, which exceeds the preset threshold, it is determined as an outlying data point. Finally, a plurality of outlying data points are extracted by this method, providing accurate coordinates for anomaly positioning.

[0085] The extracted multiple outlier data points are mapped to a two-dimensional space-time coordinate system with time axis as horizontal coordinate and process node as vertical coordinate according to their corresponding collection timestamps and process node positions (such as equipment numbers in the stages of crushing, cracking and purification). Each outlier data point is marked with a different color and is associated with its corresponding parameter type (such as fluoride concentration, equipment speed) and deviation degree. Through a graph database, an outlier space-time correlation graph is constructed, nodes represent outlier data points, and edges represent the time sequence relationship or process logic correlation between data points, which intuitively presents the continuity of the anomaly in the time dimension (such as exceeding the standard for three consecutive periods) and the propagation path in the process space (such as the abnormal temperature of the cracking furnace leading to the increase of impurity rate in the purification section), providing visual support for multi-dimensional anomaly analysis.

[0086] Using depth-first search, the outlier space-time correlation graph is traversed to identify abnormal trajectories and multi-parameter correlation patterns composed of consecutive outlier data points in the graph. For example, the co-occurrence pattern of fluoride emission exceeding the standard and temperature parameter anomaly in the cracking process within a certain period of time is extracted, and the lag relationship (such as the temperature rising for 15 minutes before the fluoride concentration starts to exceed the standard) and the causal relationship (such as high temperature leading to incomplete rubber cracking) of the two in the time axis and the process logic are analyzed. At the same time, the duration of the abnormal trajectory, the number of affected process nodes, the types of parameters involved (chemical indicators / physical indicators) and the degree of deviation threshold (such as residual value reaching 2.5σ) are counted. Based on the above characteristics, a multi-dimensional quality anomaly feature vector containing abnormal type, occurrence position, time characteristics, parameter correlation and severity level is generated, providing structured data support for subsequent anomaly matching and early warning.

[0087] In one possible implementation manner, step S400 further includes:

[0088] Step S410: mapping the quality anomaly early warning signal to the recycling full-process data chain to construct a space-time correlation graph.

[0089] Step S420: positioning and identifying according to the space-time correlation graph to determine a plurality of abnormal source position labels.

[0090] Step S430: traversing the plurality of abnormal source position labels to perform multi-dimensional impact analysis on the abnormal sources to generate recycling correction parameters.

[0091] Step S440: tracing the recycling correction parameters back to the recycling full-process data chain for online learning to generate a data learning result.

[0092] Step S450: performing closed-loop verification testing on the recycling full-process data chain according to the data learning result to construct the recycling full-process optimization data chain.

[0093] Specifically, the key information carried in the quality anomaly early warning signal, such as the abnormal type (e.g., excessive fluoride emission), the occurrence timestamp, and the process node, is accurately matched with the device operation parameters (e.g., real-time temperature, rotation speed) recorded in the recycling full-process data chain and the process correlation network in the time and space dimensions; a two-dimensional time-space correlation graph is constructed with the time axis as the horizontal coordinate and the process node (crushing / cracking / purification) as the vertical coordinate, each abnormal point in the graph is marked with a diamond icon, and the deviation amplitude of the measured value of the abnormal parameter from the standard threshold (e.g., 120% excessive fluoride concentration) is distinguished by different colors (e.g., red represents level I early warning, and yellow represents level II early warning); the process flowchart is embedded in the edge of the graph to intuitively show the propagation path of the anomaly in the physical space (device location) and time sequence, for example, the temperature anomaly of a certain cracking furnace is represented as red marker points in the time-space graph for three consecutive sampling periods and extends along the time axis to the right, and is associated to the impurity rate anomaly point in the subsequent purification process through a dashed arrow, forming a complete abnormal time-space correlation visualization model.

[0094] Based on the constructed time-space correlation graph, node positioning algorithms (e.g., centrality analysis, shortest path search) in graph theory are used for traceability analysis of abnormal data points. First, identify the earliest outlier data point in the graph as the initial abnormal source candidate, and judge the starting node of the abnormal propagation based on the logical relationship between processes in the process flowchart (e.g., material flow direction of crushing, cracking, and purification); then determine the key nodes that have the greatest impact on abnormal propagation by calculating the betweenness centrality of each node, for example, the betweenness centrality value of a certain cracking furnace temperature sensor node is the highest, indicating that it is the core hub of abnormal propagation. Finally, these key nodes are marked with red triangles in the graph to generate multiple abnormal source location labels containing device number (e.g., cracking furnace A-03), process stage (cracking process), and abnormal type (temperature sensor drift), achieving accurate mapping from time-space distribution to physical entity.

[0095] For each abnormal source location label, the parameter correlation matrix of the process parameter sub-library in the recycling process database is called to analyze the coupling relationship between the abnormal source and the upstream and downstream process parameters (e.g., correlation coefficient of cracking temperature and fluoride emission concentration), and the digital twin model is used to simulate the propagation path of the anomaly in the process chain and the degree of influence on the subsequent process; at the same time, the historical operation data (e.g., temperature fluctuation curve, motor load change) of the abnormal source device are extracted from the device operation parameter sub-library of the database, the influence of the anomaly on the device life or operation stability is predicted through the ARIMA algorithm, and calibration, maintenance, or replacement instructions are generated in combination with the device maintenance manual; in addition, the pollutant emission limits (e.g., gaseous fluoride ≤9 mg / m 3), and the compliance risk level, to generate environmental compensation measures (such as increasing the operation power of the waste gas treatment equipment by 10%). Finally, the multi-dimensional analysis results such as process parameter adjustment suggestions (such as reducing the pyrolysis temperature by 5°C), equipment maintenance operation instructions, and environmental protection control measures are integrated into structured recycling correction parameters, which are pushed to the control nodes of the production line in real time through the industrial Internet of Things interface for execution.

[0096] The recycling correction parameters are traced back to the recycling full-process data chain for online learning using an artificial neural network algorithm. Specifically, the process parameters (such as cracking temperature and fluoride emission concentration) in the data chain, the equipment state (such as sensor calibration records), and the correction parameters (such as temperature adjustment amount and maintenance instructions) are used as input layer neurons, and the quality index optimization target (such as fluoride emission compliance rate and product purity) is used as output layer neurons to construct a multi-layer perception model. The hidden layer weights are adjusted through the back propagation algorithm, for example, the correction parameter of reducing the pyrolysis temperature by 5°C is associated with the feedback result of reducing the fluoride concentration by 15% in the historical data, the mean square error is calculated, and the model parameters are iteratively optimized. During the learning process, the nonlinear mapping relationship between the correction parameters and the process state (such as the S-shaped curve characteristics of the temperature adjustment amplitude and the emission concentration change) is automatically extracted, and the data learning results including the parameter correlation weight matrix and the prediction error confidence interval are generated, for example, the output is a quantitative rule that when the cracking temperature is ≥305°C and the sensor drift value is >0.8°C, the temperature reduction amplitude increases by 1°C, and the fluoride concentration decreases by 6% with a probability of 6%, realizing the adaptive learning and prediction ability improvement of the recycling process model.

[0097] Based on the data learning results, a closed-loop verification test is started in the recycling full-process data chain: the optimized process parameters (such as the adjusted pyrolysis temperature threshold) and equipment control rules (such as the sensor calibration period is shortened to 4 hours) are synchronized to the digital twin model, the process running state of 30 consecutive production cycles is simulated, and the consistency of the predicted value and the historical measured data is compared (such as the fluoride emission concentration prediction error rate <5%). At the same time, the key indicators (such as the actual temperature fluctuation range of the cracking furnace and the impurity rate of the purification section) of the production line under the action of the correction parameters are collected in real time through the industrial Internet of Things nodes, and a verification data set is constructed; the significance difference between the verification data and the learning results is analyzed using hypothesis testing (such as t-test), if the measured values all fall within the prediction confidence interval (such as 95% confidence level), it is determined that the correction is effective, the optimized parameter combination (such as the updated value of the temperature-emission correlation matrix) is fixed to the process parameter sub-library of the data chain, and a recycling full-process optimization data chain containing dynamic coupling relationship is formed; if the verification fails, the relearning mechanism is triggered, and the machine learning model parameters are adjusted back to step S440 until the closed-loop optimization is completed.

[0098] In one possible implementation, step S450 further includes:

[0099] Step S451: Based on the recycling process, a plurality of industrial Internet of Things nodes are deployed for real-time collection to obtain equipment operation parameters.

[0100] Step S452: According to the recycling process, process parameter correlation analysis is performed, a process parameter correlation network is constructed, and a process parameter coupling relationship is identified based on the process parameter correlation network.

[0101] Step S453: The equipment operation parameters are synchronized and simulated according to the process parameter coupling relationship, and the recycling full-process data chain is constructed.

[0102] Specifically, industrial Internet of Things nodes are deployed on key process equipment (such as crusher units, cracking reaction furnaces, and purification towers) in the crushing, cracking, and purification processes of fluorine-containing rubber waste recycling, including temperature sensors, vibration sensors, and gas concentration detectors. Intelligent terminals such as real-time collection of equipment operation parameters (such as crushing motor speed, cracking furnace pressure, and fluorine compound concentration in the purification section) at a frequency of seconds. Each node transmits raw data to an edge computing unit through a wireless communication module (such as 5G, Wi-Fi), and after filtering and denoising, protocol conversion, a standardized set of equipment operation parameters (such as structured data containing timestamps, equipment numbers, parameter names, and values) is generated, and encrypted and stored in the equipment operation parameter sub-library of the recycling process database, providing real-time data support for full-process monitoring.

[0103] Table 1: Real-time collection data example of fluorine-containing rubber 26 waste recycling:

[0104] Table 1

[0105] Time stamp Process node Equipment number Collection parameter Measured value Standard threshold 2025 / 5 / 22 10:00 Crushing procedure B-001 Motor speed 1500 r / min 1450 ~ 1550 r / min B-001 Vibration amplitude 12.5 mm / s ≤ 15 mm / s 2025 / 5 / 22 10:05 Cracking procedure R-003 In-furnace temperature 320℃ 315~325℃ R-003 In-furnace pressure 101.3 kPa 100 ~ 103 kPa 2025 / 5 / 22 10:10 Purification procedure P-001 Fluoride concentration 8.2 mg / m 3 ]] ≤ 9 mg / m 3 ]] P-001 Product moisture content 0.80% ≤1.5% 2025 / 5 / 22 10:15 Total control system - Crushing procedure energy consumption 250 kWh -

[0106] According to the logical sequence of the fluorine-containing rubber waste recycling process (crushing-cracking-purification), the Apriori algorithm is used to mine the historical process parameters (such as crushing particle size, cracking temperature, and fluorine compound emission concentration) stored in the recycling process database, calculate the support and confidence between parameters (such as when the cracking temperature is greater than 300°C and the crushing particle size is less than 2mm, the confidence of fluorine compound emission exceeding the standard is 85%), and construct a process parameter correlation network containing parameter correlation edges and weights. By analyzing the connection density and betweenness centrality of the nodes in the network, strong coupling parameter pairs (such as the correlation coefficient of cracking temperature and catalyst activity reaches 0.92) and key hub parameters (such as the temperature parameter in the cracking stage as the core node connecting the crushing and purification processes) are identified, and the mutual influence path and dependence of each process parameter in the process are determined, providing a structured correlation model for subsequent synchronous simulation and abnormal tracing.

[0107] The device operation parameters (such as crushing motor speed, cracking furnace temperature, purification section pressure, etc.) collected in real time by the industrial Internet of Things nodes are input into the constructed process parameter correlation network, and dynamic synchronous simulation is performed based on the coupling relationship between the parameters (such as the quantified correlation that the fluorine emission concentration increases by 6% for every 5℃ increase in cracking temperature). A full-process virtual simulation model is constructed using digital twinning technology, the device operation parameters of each process are used as input variables, the model is driven to run through the parameter coupling relationship, and the cascading influence of parameter changes on upstream and downstream processes (such as a 15-minute increase in cracking time caused by a large crushing particle size, which in turn causes a 3% increase in impurity rate in the purification section) is simulated. In this process, the spatiotemporal data of device state, process logic, and quality indicators are integrated to construct a recovery full-process data chain covering the crushing, cracking, and purification processes. This data chain aligns the parameters of each process with the time axis, forming a dynamic digital mapping that includes parameter coupling relationships and abnormal propagation paths, providing real-time simulation support for full-process monitoring, abnormal positioning, and process optimization.

[0108] In Embodiment Two, based on the same inventive concept as the full-process monitoring method for fluorine-containing rubber waste in the preceding embodiments, as shown in Figure 2 The present application provides a full-process monitoring system for fluorine-containing rubber waste, and the system and method embodiments in the present application are based on the same inventive concept. The system comprises:

[0109] A multi-modal sensing analysis module 10 is configured to perform multi-modal sensing analysis on the fluorine-containing rubber waste to generate a set of waste characteristic parameters.

[0110] A recovery record module 20 is configured to construct a recovery process database and perform recovery record based on the recovery process database according to the set of waste characteristic parameters to obtain real-time recovery state data.

[0111] An abnormal early warning signal extraction module 30 is configured to perform online detection according to the real-time recovery state data, construct a product quality evaluation matrix according to the detection data set, and extract a quality abnormal early warning signal.

[0112] A dynamic monitoring module 40 is configured to establish a full-process data chain, perform dynamic monitoring according to the mapping of the quality abnormal early warning signal to the full-process data chain, generate a recovery correction parameter, update the full-process data chain through the recovery correction parameter, and obtain a full-process optimized data chain.

[0113] Further, the system is also used for the following functions:

[0114] The fluororubber waste is cooperatively detected by a multi-modal sensor group to obtain a multi-modal sensing data set; the multi-modal sensing data set is spatio-temporally aligned, principal component analysis is performed on the multi-modal initial alignment data set to obtain a multi-modal alignment data set; confidence evaluation is performed based on the multi-modal alignment data set, and the multi-modal alignment data set is excluded according to the evaluation result to generate a multi-modal data set; composite feature analysis is performed based on the multi-modal data set to generate the waste characteristic parameter set.

[0115] Further, the system is also used for the following functions:

[0116] A recycling process database is constructed, which includes a material characteristic sub-database, a process parameter sub-database, and an environmental protection standard sub-database; the waste characteristic parameter set is dynamically matched with the material characteristic sub-database, the process parameter sub-database, and the environmental protection standard sub-database to generate target recycling process data; device control instructions are activated based on the target recycling process data to record the recycling time sequence and obtain recycling record data stream; the recycling record data stream is encrypted and stored to generate a recycling process traceability chain, and the real-time recycling state data is extracted according to the recycling process traceability chain.

[0117] Further, the system is also used for the following functions:

[0118] The molecular structure spectral features of the fluororubber waste are extracted based on the waste characteristic parameter set, and a fluororubber material fingerprint library is constructed based on the molecular structure spectral features; material degradation degree evaluation is performed based on the fluororubber material fingerprint library to generate a degradation degree score; the thermal stability grade of the fluororubber waste is calculated according to the degradation degree score, and material characteristics are integrated based on the thermal stability grade and the fluororubber material fingerprint library to construct the material characteristic sub-database.

[0119] Further, the system is also used for the following functions:

[0120] The historical recycling data log is retrieved, which includes a plurality of recycling process parameters and a plurality of waste recycling parameters, and the plurality of recycling process parameters and the plurality of waste recycling parameters have a corresponding relationship; the plurality of recycling process parameters and the plurality of waste recycling parameters are associated to construct a parameter association matrix; process control is performed using digital twinning according to the parameter association matrix to generate a virtual process parameter set, and the data blind area is supplemented based on the virtual process parameter set to construct the process parameter sub-database.

[0121] Further, the system is also used for the following functions:

[0122] According to the real-time recycling state data, the fluorine-containing rubber waste is detected in real time, and a detection data set is obtained; a multi-dimensional quality evaluation index is constructed based on the environmental protection standard sub-library, the detection data set is mapped to the multi-dimensional quality evaluation index, and a real-time quality parameter vector is generated; dynamic weight distribution is performed based on the real-time quality parameter vector, and a product quality evaluation matrix is constructed; matrix outlier evaluation is performed on the product quality evaluation matrix, a plurality of outlier data points are obtained, abnormal identification is performed based on the plurality of outlier data points, and a quality abnormal feature vector is extracted according to an abnormal mode; the historical abnormal database is called, the quality abnormal feature vector is matched with the historical abnormal database, a hierarchical early warning signal is generated, and the hierarchical early warning signal is added to the quality abnormal early warning signal.

[0123] Further, the system is also used for the following functions:

[0124] The product quality evaluation matrix is dynamically decomposed, and a decomposition result is obtained, the decomposition result includes an outlier residual matrix; a plurality of outlier data points are extracted by performing multi-dimensional space outlier degree calculation on the outlier residual matrix; the plurality of outlier data points are mapped to a time-space coordinate system of a recycling process flow, and an outlier space-time correlation graph is constructed; an outlier mode feature is extracted by traversing the outlier space-time correlation graph for analysis, and the quality abnormal feature vector is generated.

[0125] Further, the system is also used for the following functions:

[0126] The quality abnormal early warning signal is mapped to the recycling whole-process data chain to construct a space-time correlation graph; a plurality of abnormal source position labels are determined by positioning and identifying according to the space-time correlation graph; a recycling correction parameter is generated by traversing the plurality of abnormal source position labels for multi-dimensional influence analysis of the abnormal source; the recycling correction parameter is traced back to the recycling whole-process data chain for online learning, and a data learning result is generated; a closed-loop verification test is performed on the recycling whole-process data chain according to the data learning result, and the recycling whole-process optimization data chain is constructed.

[0127] Further, the system is also used for the following functions:

[0128] Based on the recycling process flow, a plurality of industrial internet of things nodes are deployed for real-time collection, and equipment operation parameters are obtained; process parameter correlation analysis is performed according to the recycling process flow, a process parameter correlation network is constructed, and process parameter coupling relationships are identified based on the process parameter correlation network; the equipment operation parameters are synchronized and simulated according to the process parameter coupling relationships, and the recycling whole-process data chain is constructed.

[0129] It should be noted that the above-mentioned embodiment sequences of the present application are merely for description only, but not for representing the advantages and disadvantages of the embodiments. And the above-mentioned embodiments of the present specification have been described. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are also possible or can be advantageous.

[0130] The above only describes the preferred embodiments of the present application, and does not limit the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0131] The specification and drawings of the present application are merely exemplary descriptions of the present application, and are considered to cover any and all modifications, changes, combinations or equivalents within the scope of the present application. Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application belong to the scope of the present application and its equivalent technology, the present application is intended to include these modifications and variations.

Claims

1. A method for monitoring the entire recycling process of fluorine-containing rubber waste, characterized by, The method includes: Multimodal sensing analysis was performed on fluorinated rubber waste to generate a set of waste characteristic parameters; A recycling process database is constructed, and recycling records are made based on the waste characteristic parameter set according to the recycling process database to obtain real-time recycling status data. Online detection is performed based on the real-time recycling status data, and a product quality evaluation matrix is ​​constructed based on the detection dataset to extract quality anomaly warning signals. Establish a recycling end-to-end data chain, dynamically monitor the recycling end-to-end data chain according to the quality anomaly warning signal, generate recycling correction parameters, update the recycling end-to-end data chain through the recycling correction parameters, and obtain an optimized recycling end-to-end data chain; The method includes constructing a recycling process database, recording recycling data based on the database according to the waste characteristic parameter set, and obtaining real-time recycling status data. A recycling process database is constructed, which includes a material property sub-database, a process parameter sub-database, and an environmental protection standard sub-database. The waste characteristic parameter set is dynamically matched with the material property sub-library, the process parameter sub-library, and the environmental protection standard sub-library to generate target recycling process data; Based on the target recycling process data, the activated equipment control command is recorded according to the recycling sequence to obtain a recycling record data stream; The recycling record data stream is encrypted and stored to generate a recycling process traceability chain. The real-time recycling status data is extracted based on the recycling process traceability chain. Online detection is performed based on the real-time recovery status data. A product quality evaluation matrix is ​​constructed based on the detection dataset, and quality anomaly warning signals are extracted. The method includes: The fluororubber waste is subjected to real-time online detection of recycling based on the real-time recycling status data to obtain the detection dataset. Based on the environmental protection standard sub-library, a multidimensional quality evaluation index is constructed, and the detection dataset is mapped to the multidimensional quality evaluation index to generate a real-time quality parameter vector. Based on the real-time quality parameter vector, a dynamic weight allocation is performed to construct a product quality evaluation matrix; Matrix outlier evaluation is performed on the product quality evaluation matrix to obtain multiple outlier data points. Anomaly identification is performed based on the multiple outlier data points, and quality anomaly feature vectors are extracted according to the anomaly pattern. Retrieve the historical anomaly database, match the quality anomaly feature vector with the historical anomaly database, generate a graded early warning signal, and add the graded early warning signal to the quality anomaly early warning signal; The method involves dynamically monitoring the recycling process data chain by mapping the quality anomaly warning signal to the recycling process data chain, generating recycling correction parameters, and updating the recycling process data chain using these correction parameters to obtain an optimized recycling process data chain. The quality anomaly early warning signal is mapped to the entire recycling process data chain to construct a spatiotemporal correlation map; Based on the spatiotemporal correlation map, the location and identification of multiple anomaly sources are determined; The multiple anomaly source location labels are traversed to perform a multi-dimensional impact analysis on the anomaly sources, and recovery correction parameters are generated. The recovery correction parameter is traced back to the recovery whole-process data chain for online learning to generate a data learning result; According to the data learning result, a closed-loop verification test is performed on the recovery whole-process data chain to construct the recovery whole-process optimization data chain.

2. The method for monitoring the entire recycling process of fluororubber waste according to claim 1, wherein A multi-modal sensing analysis is performed on the fluororubber waste to generate a waste characteristic parameter set, and the method comprises: A multi-modal sensor group is used to cooperatively detect the fluororubber waste to obtain a multi-modal sensing data set; The multi-modal sensing data set is time-space aligned, and principal component analysis is performed on the multi-modal initial aligned data set to obtain a multi-modal aligned data set; Based on the multi-modal aligned data set, confidence evaluation is performed, and based on the evaluation result, the multi-modal aligned data set is excluded from the abnormal, to generate a multi-modal data set; Based on the multi-modal data set, composite feature analysis is performed to generate the waste characteristic parameter set.

3. The method for monitoring the entire recycling process of fluororubber waste according to claim 1, wherein The construction process of the material characteristic sub-library comprises: Based on the waste characteristic parameter set, the molecular structure spectral features of the fluororubber waste are extracted, and based on the molecular structure spectral features, a fluororubber material fingerprint library is constructed; Based on the fluororubber material fingerprint library, material degradation degree evaluation is performed to generate a degradation degree score; According to the degradation degree score, the thermal stability grade of the fluororubber waste is calculated, and based on the thermal stability grade, the material characteristics are integrated based on the fluororubber material fingerprint library to construct the material characteristic sub-library.

4. The method for monitoring the entire recycling process of fluororubber waste according to claim 1, wherein The construction process of the process parameter sub-library comprises: A historical recovery data log is retrieved, and the historical recovery data log contains a plurality of recovery process parameters and a plurality of waste recovery parameters, and the plurality of recovery process parameters and the plurality of waste recovery parameters have a corresponding relationship; According to the plurality of recovery process parameters and the plurality of waste recovery parameters, a parameter correlation matrix is constructed; According to the parameter correlation matrix, digital twinning is used for process control to generate a virtual process parameter set, and based on the virtual process parameter set, data blind areas are supplemented to construct the process parameter sub-library.

5. The method for monitoring the entire recycling process of fluororubber waste according to claim 1, wherein The product quality evaluation matrix is subjected to matrix outlier evaluation to obtain a plurality of outlier data points, based on the plurality of outlier data points, abnormality is identified, and according to the abnormal mode, a quality abnormal feature vector is extracted, and the method comprises: The product quality evaluation matrix is subjected to dynamic matrix decomposition to obtain a decomposition result, and the decomposition result contains an outlier residual matrix; The outlier residual matrix is subjected to multi-dimensional space outlier degree calculation to extract a plurality of outlier data points; The plurality of outlier data points are mapped to a time-space coordinate system of the recovery process flow to construct an outlier time-space correlation map; The outlier mode features are extracted by traversing the outlier time-space correlation map for analysis to generate the quality abnormal feature vector.

6. The method for monitoring a full recycling process of fluororubber waste according to claim 1, wherein The process of establishing a recovery whole-process data chain comprises: Based on the recovery process flow, a plurality of industrial internet of things nodes are deployed for real-time collection to obtain equipment operation parameters; According to the recovery process flow, process parameter correlation analysis is performed to construct a process parameter correlation network, and based on the process parameter correlation network, process parameter coupling relationships are identified; The device operation parameters are synchronously simulated according to the coupling relationship of the process parameters, and the recycling whole-process data chain is constructed.

7. A recovery full-process monitoring system for fluorine-containing rubber waste, characterized by, The system is used for implementing the recycling whole-process monitoring method for fluororubber waste according to any one of claims 1-6, and the system comprises: a multi-modal sensing analysis module for performing multi-modal sensing analysis on the fluororubber waste to generate a set of waste characteristic parameters; a recycling record module for constructing a recycling process database and performing recycling record based on the set of waste characteristic parameters according to the recycling process database to obtain real-time recycling state data; an abnormal early warning signal extraction module for performing online detection according to the real-time recycling state data, constructing a product quality evaluation matrix according to a detection data set, and extracting a quality abnormal early warning signal; a dynamic monitoring module for establishing a recycling whole-process data chain, performing dynamic monitoring according to the mapping of the quality abnormal early warning signal to the recycling whole-process data chain, generating a recycling correction parameter, updating the recycling whole-process data chain through the recycling correction parameter, and obtaining a recycling whole-process optimization data chain.

Images