CN114923869A - Method for identifying recycled plastic based on combination of spectroscopy, thermal analysis and data fusion strategy and chemometric method - Google Patents
Method for identifying recycled plastic based on combination of spectroscopy, thermal analysis and data fusion strategy and chemometric method Download PDFInfo
- Publication number
- CN114923869A CN114923869A CN202210513295.7A CN202210513295A CN114923869A CN 114923869 A CN114923869 A CN 114923869A CN 202210513295 A CN202210513295 A CN 202210513295A CN 114923869 A CN114923869 A CN 114923869A
- Authority
- CN
- China
- Prior art keywords
- data
- identification
- sample
- data fusion
- model
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/31—Investigating relative effect of material at wavelengths characteristic of specific elements or molecules, e.g. atomic absorption spectrometry
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N25/00—Investigating or analyzing materials by the use of thermal means
- G01N25/20—Investigating or analyzing materials by the use of thermal means by investigating the development of heat, i.e. calorimetry, e.g. by measuring specific heat, by measuring thermal conductivity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
- G06F18/251—Fusion techniques of input or preprocessed data
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02W—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO WASTEWATER TREATMENT OR WASTE MANAGEMENT
- Y02W30/00—Technologies for solid waste management
- Y02W30/50—Reuse, recycling or recovery technologies
- Y02W30/62—Plastics recycling; Rubber recycling
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Pathology (AREA)
- Immunology (AREA)
- General Health & Medical Sciences (AREA)
- Biochemistry (AREA)
- Analytical Chemistry (AREA)
- Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
- Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)
Abstract
The invention belongs to the technical field of high polymer materials and food packaging safety, and discloses a method for identifying recycled plastics based on a spectrometry method, a thermal analysis method and a data fusion strategy combined with a chemometrics method. The method comprises the steps of detecting a sample to be detected by using a spectrometry method and a thermal analysis method respectively, collecting and screening characteristic data obtained by using the two testing methods respectively, carrying out normalization processing on the two characteristic data to obtain a data fusion matrix, and identifying the data fusion matrix by using an identification model established by a chemometrics method to obtain an identification result. According to the invention, the characteristic data set is effectively extracted through the ultraviolet-visible spectrum and the differential scanning calorimetry, and is subjected to identification analysis, and the identification accuracy of 100% is realized by combining a data fusion strategy and a chemometrics method.
Description
Technical Field
The invention belongs to the technical field of high polymer materials and food packaging safety, and particularly relates to a method for identifying recycled plastics based on a method combining a spectrum method, a thermal analysis method and a data fusion strategy with a chemometrics method.
Background
The use of Recycled plastic materials (rPM) is one of the important components of recycling economy, and recycling plastic materials is also becoming one of the key solutions to the problem of plastic waste. The european EC 2023/2006 states that recycled plastics must meet the relevant safety standards in order to be used in the manufacture of food packaging, but there are some circumstances in which recycled plastics may flow into the food contact material supply chain without evaluation, and therefore, the identification of recycled plastics is of great importance in ensuring food safety and maintaining market order.
After the recycled plastic goes through a first life cycle and a recycling environment which are relatively complex, the polymer degradation products, monomer residual impurities, reaction process by-products, pollutants and other unintended additives introduced in the production process, additives, raw materials and other intentional additives are various, and the physical and chemical properties of the recycled plastic also change, so that characteristic differences inevitably exist between the recycled plastic and the virgin plastic, and the characteristic differences can be an effective means for identifying the recycled plastic. At present, methods for screening substances in recycled plastics and identifying the recycled plastics by means of instrument detection and analysis technologies, classification algorithms and the like exist, but the existing methods have the inevitable problem of false positive and false negative, cannot achieve 100% identification accuracy, and cannot provide sufficient guarantee for food safety.
Disclosure of Invention
In order to overcome the disadvantages and shortcomings of the prior art, the invention provides a method for identifying recycled plastics based on a combination of a spectroscopic method, a thermal analysis method and a data fusion strategy with a chemometric method.
It is another object of the present invention to provide the use of the above method for the identification of recycled plastics. The method of the invention can be applied to the identification of various recycled plastics and can accurately identify the recycled plastics.
The purpose of the invention is realized by the following scheme:
a method for identifying recycled plastics based on a spectroscopic method, a thermal analysis method and a data fusion strategy combined with a chemometrics method comprises the steps of detecting a sample to be detected by the spectroscopic method and the thermal analysis method respectively, collecting and screening characteristic data obtained by the two test methods respectively, carrying out normalization processing on the two characteristic data to obtain a data fusion matrix, and identifying the data fusion matrix by using an identification model of the chemometrics method to obtain an identification result.
Further, the formula of the normalization process is as follows:
where x represents the value in the data set, min (x) represents the minimum value in the series of data sets, and max (x) represents the maximum value in the series of data sets.
Further, the normalization process may be performed using matlab 2016R.
Further, the spectrum method is ultraviolet-visible spectrum method (UV-Vis).
Furthermore, the spectrum method detects the solution of the sample to be detected. For example, a sample to be tested is added into ethanol and subjected to ultrasonic heating treatment to obtain a sample solution. The sample to be tested is preferably ground to a powder and then used to prepare the sample solution.
Further, the thermal analysis method is Differential Scanning Calorimetry (DSC).
Further, the thermal analysis method detects a powdery sample to be detected.
Further, the establishment of the model: respectively measuring data of various primary plastic samples and recycled plastic samples by using a spectroscopy method and a thermal analysis method, respectively collecting and screening characteristic data obtained by two testing methods, carrying out normalization processing on the two characteristic data to obtain a data fusion matrix, and establishing an identification model by using a chemometrics method based on the matrix.
Further, the identification model is a Support Vector Machine (SVM) model or a Linear Discriminant Analysis (LDA) model.
Further, the screening may be performed by using an orthogonal partial least squares discriminant analysis (OPLS-DA) on the collected data to obtain the characteristic data.
Further, the criteria for the screening may be VIP > 1.
Furthermore, the data obtained by spectroscopy is preprocessed by smoothing and first derivative before being used for screening, such as by using the thermo INSIGHT 2 software.
Furthermore, the data obtained by the thermal analysis method is preprocessed, and the baseline is corrected to the same level and then is used for screening.
The sample to be tested, the primary plastic sample and the regenerated plastic sample which are the same or different can be respectively pretreated by the working procedures of crushing, cleaning, drying and the like before testing.
Further, the method specifically comprises the following steps:
(1) pretreating a sample to be detected to obtain a sample solution and sample powder;
(2) detecting the sample solution by using an ultraviolet-visible spectrum method;
(3) detecting the sample powder by adopting a differential scanning calorimetry method;
(4) collecting and processing data: collecting ultraviolet-visible spectrum data and DSC data, screening out characteristic data sets by respectively adopting orthogonal partial least squares discriminant analysis (OPLS-DA), and carrying out normalization processing on the two characteristic data sets to obtain a data fusion data set;
(5) and (3) identification: and identifying the data fusion data set by using a Support Vector Machine (SVM) model or a Linear Discriminant Analysis (LDA) model to obtain an identification result.
Furthermore, when the differential scanning calorimetry test is carried out, the sample to be tested can be directly used when the sample is powder, otherwise, the sample powder can be obtained by grinding the sample and then used for detection, for example, the sample powder can be ground by a high-throughput tissue grinder.
Furthermore, the sample solution can be obtained by adding a powdery sample into ethanol, carrying out ultrasonic heating treatment in a sealed state, and carrying out heat preservation at 30-70 ℃ for 8-72 h. The time of the heat treatment is preferably 0.5 to 4 hours.
In the sample solution, the mass-volume ratio of the sample to the ethanol is preferably 1:100-10:100, g/mL. Furthermore, the differential scanning calorimetry test is two times of heating detection, specifically, the temperature is firstly increased to 300 ℃ to eliminate the thermal history of the material, and after the temperature is reduced to room temperature, the temperature is increased to 300 ℃ again for detection.
The heating rates are respectively 5-15K/min, more preferably 10K/min, which are the same or different.
The invention provides a method for identifying regenerated PET (polyethylene terephthalate) based on ultraviolet-visible spectroscopy (UV-Vis), Differential Scanning Calorimetry (DSC) and a data fusion strategy combined with a chemometric method. The method effectively extracts the characteristic data set of the PET material through ultraviolet-visible spectrum and differential scanning calorimetry, performs identification analysis on the characteristic data set, and combines a data fusion strategy and a chemometrics method to realize 100% identification accuracy.
The invention also provides the application of the method in identification of the recycled plastics. The method of the invention can be applied to the identification of various recycled plastics and can accurately identify the recycled plastics. The method for identifying the recycled plastic is used for identifying the virgin plastic and the recycled plastic, combines a plurality of instrument detection and analysis technologies, a data fusion strategy and a chemometrics method, has the identification accuracy rate of 100 percent and more visual and stable identification result compared with a single instrument detection and analysis technology, and provides powerful guarantee for the identification work of the recycled plastic for food contact based on characteristic data.
Compared with the prior art, the invention has the following advantages and beneficial effects:
(1) the method uses a plurality of instrument detection and analysis technologies combined with a data fusion strategy and a chemometrics method to identify the primary/regenerated plastics, and has high identification accuracy rate which can reach 100 percent at most.
(2) The method has wide application range, is not limited to a certain fixed instrument detection and analysis technology or plastic types, and can be applied to various instrument detection and analysis technologies and plastic types.
(3) The method is also suitable for low-cost instruments and a small amount of samples, can realize accurate identification of the recycled plastics, saves the cost and reduces the identification threshold.
(4) The method disclosed by the invention fuses data through normalization processing, and can effectively improve the accuracy and the stability of the identification model.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.
FIG. 1 is a UV-Vis spectral curve of the sample in example 1.
FIG. 2 is a second temperature rise profile of DSC testing of the sample of example 1.
Fig. 3 shows SVM model discrimination results of hierarchical sampling of the data fusion data set in example 1.
Fig. 4 shows the discrimination results of a hierarchical sampling of the UV-Vis dataset.
FIG. 5 shows the results of RF model identification of a hierarchical sample of the data fusion data set in example 2.
Detailed Description
The present invention will be described in further detail with reference to examples, but the embodiments of the present invention are not limited thereto. The materials referred to in the following examples are commercially available without specific reference. The method is a conventional method unless otherwise specified. The dosage of each component is calculated by mass parts and volume parts, and the unit is g and mL.
In one embodiment, a method for identifying recycled plastics based on a spectroscopic method and a thermal analysis method combined with a chemometric method includes the steps of detecting a sample to be detected by the spectroscopic method and the thermal analysis method respectively, collecting and screening characteristic data obtained by the two test methods respectively, normalizing the two characteristic data to obtain a data fusion matrix, and identifying the data fusion matrix by using an identification model of the chemometric method to obtain an identification result.
In some preferred embodiments, the formula of the normalization process is as follows:
where x represents the value in the data set, min (x) represents the minimum value in the series of data sets, and max (x) represents the maximum value in the series of data sets.
In some preferred embodiments, the normalization process is performed using matlab 2016R.
In some preferred embodiments, the spectroscopy is ultraviolet-visible spectroscopy (UV-Vis).
In some preferred embodiments, the spectroscopy detects a solution of a sample to be tested. For example, a sample to be tested is added into ethanol and subjected to ultrasonic heating treatment to obtain a sample solution. The sample to be tested is preferably ground to a powder and then used to prepare the sample solution.
In some preferred embodiments, the thermal analysis method is Differential Scanning Calorimetry (DSC).
In some preferred embodiments, the thermal analysis detects a powdered sample to be tested.
In some preferred embodiments, the model is established by: respectively measuring data of various primary plastic samples and recycled plastic samples by using a spectroscopy method and a thermal analysis method, respectively collecting and screening characteristic data obtained by two testing methods, carrying out normalization processing on the two characteristic data to obtain a data fusion matrix, and establishing an identification model by using a chemometrics method based on the matrix.
In some preferred embodiments, the identification model is a Support Vector Machine (SVM) model or a Linear Discriminant Analysis (LDA) model.
In some preferred embodiments, the screening may be performed using orthogonal partial least squares discriminant analysis (OPLS-DA) on the collected data to obtain the characteristic data. Further, the criteria for the screening may be VIP > 1.
In some preferred embodiments, the data obtained by spectroscopy detection is subjected to smoothing and first derivative preprocessing before being used for screening. In one embodiment, the processing is performed using thermal INSIGHT 2 software.
In some preferred embodiments, the data from the thermographic testing is pre-processed, corrected to the same level as the baseline, and then used for screening.
The sample to be tested, the primary plastic sample and the regenerated plastic sample which are the same or different can be respectively pretreated by the working procedures of crushing, cleaning, drying and the like before testing.
In some preferred embodiments, the method of the invention comprises in particular the steps of:
(1) pretreating a sample to be detected to obtain a sample solution and sample powder;
(2) detecting the sample solution by using an ultraviolet-visible spectrum method;
(3) detecting the sample powder by adopting a differential scanning calorimetry method;
(4) data collection and processing: collecting ultraviolet-visible spectrum data and DSC data, respectively screening out characteristic data sets by adopting orthogonal partial least squares discriminant analysis (OPLS-DA), and carrying out normalization processing on the two characteristic data sets to obtain a data fusion data set;
(5) identification: and identifying the data fusion data set by using a Support Vector Machine (SVM) model or a Linear Discriminant Analysis (LDA) model to obtain an identification result.
In some preferred embodiments, the sample to be tested is a powder for direct use in performing the differential scanning calorimetry test, or the sample powder can be obtained by grinding the sample and then used for testing, e.g., grinding with a high-throughput tissue grinder.
In some preferred embodiments, the sample solution can be obtained by adding a powdered sample into ethanol, performing ultrasonic heat treatment in a sealed state, and keeping the temperature at 30-70 ℃ for 8-72 h. In one embodiment, the temperature is kept at 50 ℃ for 72 h; in another embodiment, the incubation is carried out at 60 ℃ for 48 h.
In some preferred embodiments, the time of the heat treatment is preferably 0.5 to 4 hours. In one embodiment, the time of the heat treatment is 0.5 h; in another embodiment, the time of the heat treatment is 4 h; in still another embodiment, the time of the heat treatment is 2 hours.
In some preferred embodiments, the mass-to-volume ratio of the sample to the ethanol in the sample solution is 1:100 to 10:100, g/mL. In one embodiment, the mass-to-volume ratio of the sample to the ethanol is 1:100, g/mL; in another embodiment, the mass to volume ratio of sample to ethanol is 4:100, g/mL; in yet another embodiment, the mass to volume ratio of sample to ethanol is 10:100, g/mL.
In some preferred embodiments, the differential scanning calorimetry test is two temperature rise tests, specifically, the temperature is first raised to 300 ℃ to eliminate the thermal history of the material, and after the temperature is lowered to room temperature, the temperature is raised to 300 ℃ again for testing.
In some preferred embodiments, the rate of temperature rise is 5-15K/min, respectively, which may be the same or different. In one embodiment, the rate of temperature rise is 5K/min; in another embodiment, the rate of temperature rise is 10K/min; in still another embodiment, the rate of temperature increase is 15K/min.
To more particularly represent the method of the invention, a method based on ultraviolet-visible spectroscopy (UV-Vis), Differential Scanning Calorimetry (DSC) and data fusion strategies combined with chemometric methods combined with the identification of recycled PET is provided in the examples below
Example 1: identification of recycled PET
(1) Sample preparation: the PET samples have 75 batches, wherein 41 batches of the recycled PET samples and 34 batches of the original PET samples have forms including bottle flakes, granules and powder, and the forms of the samples do not interfere with identification results through experimental verification.
(2) Sample pretreatment: grinding the 41 regenerated PET samples into powder by using a high-throughput tissue grinder, grinding for four minutes, and continuously grinding after radiating heat for two minutes every two minutes to obtain sample powder. Since the virgin PET is already in powder form, no separate grinding is performed. Then weighing 4 parts by mass of sample powder and 100 parts by volume of ethanol as a solvent, placing the sample powder and the ethanol in a container, sealing the container, heating the container for 1 hour by using an ultrasonic cleaner, and finally placing the container in an oven to keep the temperature for 48 hours at the temperature of 60 ℃ to obtain a sample solution.
(3) Detecting by using an ultraviolet-visible spectrum method: and (3) taking the supernatant of the sample solution treated in the step (2), measuring the spectrum of the supernatant by using an ultraviolet-visible spectrophotometer, controlling parameters in instrument operation and spectrum collection by using software thermo INSIGHT 2, setting the absorption range of the spectrum as 200-1000nm, setting the sampling interval as 1.0nm, setting the ordinate of the spectrogram as transmittance, and using ethanol as a blank control. And respectively detecting the 75 samples, and finally deriving and sorting all spectral curves, wherein the x axis is the wavelength, and the y axis is the transmittance.
The ultraviolet-visible spectrum of the sample is subjected to smoothing and first-order derivation pretreatment by using thermal INSIGHT 2 software, and a 75 (sample) × 201 (transmittance) matrix is formed by taking a 200-400nm wave band. And establishing an OPLS-DA model for the data set matrix by using SIMCA 14.1 software to obtain VIP values of 201 transmittance variables, screening wavelengths with the VIP value being more than 1 to form a new 75 (sample) x 85 (transmittance) matrix to form a characteristic data set matrix of the ultraviolet-visible spectrum for later use. FIG. 1(a) is a raw UV-Vis spectrum; FIG. 1(b) is a UV-Vis spectrum curve after smoothing and first derivative processing.
(4) Differential scanning calorimetry detection: about 10.0mg (to the nearest 0.1mg) of the powdery sample was weighed on an analytical balance, placed in a DSC special alumina crucible, and subjected to DSC measurement. The initial temperature is 30 ℃, the temperature is increased to 300 ℃ at the speed of 10K/min to eliminate the thermal history of the material, and finally the temperature is increased to 300 ℃ again at the speed of 10K/min after the temperature is cooled to the room temperature by air cooling. The standard substance tin was used to correct the two together before testing, the reference being an empty alumina crucible. And (3) respectively detecting the 75 samples, and finally obtaining two DSC temperature rise curves of each sample.
And (3) translating DSC curves with different initial positions to the same initial point by using excel, taking DSC data between 221-300 ℃ for the second temperature rise curve to form a 75 (sample) × 80(DSC power) matrix, establishing an OPLS-DA model for the data set matrix by using SIMCA 14.1 software, and screening wavelengths with VIP value of more than 1 to form a new 75 (sample) × 42 (transmittance) matrix to form a characteristic data set matrix of the DSC for later use. FIG. 2 shows the DSC second ramp.
(5) Data fusion: the data set of uv-vis spectra and DSC thermodynamic spectra were first normalized using matlab 2016R, with the formula for normalization as follows:
where x represents the value in the data set, min (x) represents the minimum value in the series of data sets, and max (x) represents the maximum value in the series of data sets.
And (3) normalizing and combining characteristic data sets of VIP & gt 1 in the ultraviolet-visible spectrum and the DSC thermodynamic spectrum to obtain a data matrix of 127 (characteristic value) × 75 (sample) under medium-level data fusion.
(6) And (3) identification: the SVM model is established on the matlab 2016R, and the SVM model is a machine learning identification method, so that a data set needs to be divided into a training set and a test set. By adopting three sampling modes of random sampling, layered sampling and system sampling, 30% of samples are extracted from 34 raw material samples and 41 regenerated material samples respectively to form a test set containing 25 samples, and the other 50 samples are used as a training set.
The overall identification results are shown in table 1 by comparing single uv-vis spectrum data without in-out normalization processing and single DSC data, wherein the identification results of the hierarchical sampling of the data fusion dataset of the method of the present invention are shown in fig. 3, the identification results of the hierarchical sampling of the single uv-vis spectrum data are shown in fig. 4, wherein 1 is defined to represent raw PET, 0 is defined to represent recycled PET, circles represent actual classification, crosses represent predicted classification, and when the actual classification is the same as the predicted classification, the circles and crosses will coincide on the graph.
TABLE 1 SVM model identification accuracy
Example 2: identification of recycled PET
The contents of step (1) to step (5) are the same as those of example 1.
(6) And (3) identification: the LDA model is established on the matlab 2016R, and similarly, 30% of samples are extracted from 34 raw material samples and 41 regenerated material samples respectively to form a test set containing 25 samples by adopting three sampling modes of random sampling, layered sampling and system sampling, and the rest 50 samples are used as training sets.
In which, the overall identification results are shown in table 2 and fig. 5 by comparing a Random Forest (RF) model and an Artificial Neural Network (ANN) model. Fig. 5 shows the RF model discrimination results.
TABLE 2 LDA model identification accuracy
According to the embodiment, the data obtained by a single UV-Vis and DSC test method are used for identifying the chemometrics method model, so that the condition of misjudgment cannot be avoided, the method for identifying the recycled plastic based on a plurality of instrument detection and analysis technologies and a data fusion strategy combined with the chemometrics method can accurately identify the recycled plastic, and the identification accuracy is up to 100%; the identification result is more visual and stable, and powerful guarantee is provided for the identification work of the recycled plastic for food contact based on the characteristic data.
The method is not limited to a certain fixed instrument detection and analysis technology or plastic types, and can be applied to various instrument detection and analysis technologies and plastic types; meanwhile, the method is still suitable for low-cost instruments and a small amount of samples, can realize accurate identification of the recycled plastics, saves the cost and reduces the identification threshold; the screening of characteristic data with larger contribution to identification can ensure the accuracy and the stability of the identification model while reducing the data volume.
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such changes, modifications, substitutions, combinations, and simplifications are intended to be included in the scope of the present invention.
Claims (10)
1. A method for identifying recycled plastics based on a spectroscopic method, a thermal analysis method and a data fusion strategy combined with a chemometrics method is characterized in that a sample to be detected is detected by the spectroscopic method and the thermal analysis method respectively, characteristic data obtained by the two test methods are collected and screened respectively, normalization processing is carried out on the two characteristic data to obtain a data fusion matrix, and an identification model of the chemometrics method is used for identifying the data fusion matrix to obtain an identification result.
2. The method of claim 1, wherein: the formula of the normalization process is as follows:
where x represents a value in the data set, min (x) represents a minimum value in the series of data sets, and max (x) represents a maximum value in the series of data sets.
3. The method of claim 1, wherein: the spectrum method is an ultraviolet-visible spectrum method.
4. The method of claim 1, wherein: the thermal analysis method is differential scanning calorimetry.
5. The method of claim 1, wherein: the identification model is a support vector machine model or a linear discriminant analysis model.
6. Method according to claim 1, characterized in that the establishment of the model: the method comprises the steps of measuring data of various primary plastic samples and recycled plastic samples by using a spectrometry method and a thermal analysis method respectively, collecting and screening characteristic data obtained by using two testing methods respectively, carrying out normalization processing on the two characteristic data to obtain a data fusion matrix, and establishing an identification model by using a chemometrics method based on the matrix.
7. The method of claim 1, wherein: and the screening refers to screening the collected data by adopting orthogonal partial least square discriminant analysis to obtain characteristic data.
8. The method of claim 7, wherein: the criteria for the screening is VIP > 1.
9. The method according to claim 1, characterized in that it comprises in particular the steps of:
(1) pretreating a sample to be detected to obtain a sample solution and sample powder;
(2) detecting the sample solution by using an ultraviolet-visible spectrum method;
(3) detecting the sample powder by adopting a differential scanning calorimetry method;
(4) data collection and processing: collecting ultraviolet-visible spectrum data and DSC data, screening out a characteristic data set by respectively adopting orthogonal partial least square discriminant analysis, and normalizing the two characteristic data sets to obtain a data fusion data set;
(5) identification: and identifying the data fusion data set by using a support vector machine model or a linear discriminant analysis model to obtain an identification result.
10. Use of the method of any one of claims 1 to 9 for the identification of recycled plastics.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210513295.7A CN114923869A (en) | 2022-05-12 | 2022-05-12 | Method for identifying recycled plastic based on combination of spectroscopy, thermal analysis and data fusion strategy and chemometric method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210513295.7A CN114923869A (en) | 2022-05-12 | 2022-05-12 | Method for identifying recycled plastic based on combination of spectroscopy, thermal analysis and data fusion strategy and chemometric method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114923869A true CN114923869A (en) | 2022-08-19 |
Family
ID=82807974
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210513295.7A Pending CN114923869A (en) | 2022-05-12 | 2022-05-12 | Method for identifying recycled plastic based on combination of spectroscopy, thermal analysis and data fusion strategy and chemometric method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114923869A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115616204A (en) * | 2022-12-21 | 2023-01-17 | 金发科技股份有限公司 | Method and system for identifying polyethylene terephthalate reclaimed materials |
| WO2024130804A1 (en) * | 2022-12-21 | 2024-06-27 | 国高材高分子材料产业创新中心有限公司 | Identification method for polyamide reclaimed material |
-
2022
- 2022-05-12 CN CN202210513295.7A patent/CN114923869A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115616204A (en) * | 2022-12-21 | 2023-01-17 | 金发科技股份有限公司 | Method and system for identifying polyethylene terephthalate reclaimed materials |
| WO2024130804A1 (en) * | 2022-12-21 | 2024-06-27 | 国高材高分子材料产业创新中心有限公司 | Identification method for polyamide reclaimed material |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114923869A (en) | Method for identifying recycled plastic based on combination of spectroscopy, thermal analysis and data fusion strategy and chemometric method | |
| Wang et al. | Improved PLS regression based on SVM classification for rapid analysis of coal properties by near-infrared reflectance spectroscopy | |
| Leggieri et al. | An electronic nose supported by an artificial neural network for the rapid detection of aflatoxin B1 and fumonisins in maize | |
| CN106404748B (en) | A kind of multiline combination laser induced breakdown spectroscopy cereal crops Production area recognition method | |
| Li et al. | Analysis and identification of rice adulteration using terahertz spectroscopy and pattern recognition algorithms | |
| Yang et al. | Rapid and Nondestructive On‐Site Classification Method for Consumer‐Grade Plastics Based on Portable NIR Spectrometer and Machine Learning | |
| Shi et al. | Optimization of electronic nose sensor array by genetic algorithms in Xihu-Longjing Tea quality analysis | |
| CN102564993A (en) | Method for identifying rice varieties by using Fourier transform infrared spectrum and application of method | |
| Guo et al. | Rapid authentication of edible bird's nest by FTIR spectroscopy combined with chemometrics | |
| Cámara-Martos et al. | Identification and quantification of lactic acid bacteria in a water-based matrix with near-infrared spectroscopy and multivariate regression modeling | |
| CN108760647A (en) | A kind of wheat content of molds line detecting method based on Vis/NIR technology | |
| Zou et al. | Support vector regression for determination of component of compound oxytetracycline powder on near-infrared spectroscopy | |
| Luqing et al. | Detection and quantification of sugar and glucose syrup in roasted green tea using near infrared spectroscopy | |
| Yu et al. | Identification of wine according to grape variety using near-infrared spectroscopy based on radial basis function neural networks and least-squares support vector machines | |
| Huang et al. | Qualitative discrimination of Chinese dianhong black tea grades based on a handheld spectroscopy system coupled with chemometrics | |
| CN105092579A (en) | Mango quality non-destructive testing device | |
| CN105699319A (en) | Near infrared spectrum quick detection method for total moisture of coal based on gaussian process | |
| Yang et al. | Analytical-performance improvement of laser-induced breakdown spectroscopy for the processing degree of wheat flour using a continuous wavelet transform | |
| CN114965973A (en) | Method for identifying recycled plastic based on instrument detection and analysis technology combined with multiple chemometrics methods and/or machine learning algorithm | |
| Song et al. | Rapid identification of adulterated rice based on data fusion of near-infrared spectroscopy and machine vision | |
| Chen et al. | Monitoring green tea fixation quality by intelligent sensors: comparison of image and spectral information | |
| CN103592256A (en) | Mid-infrared spectroscopic method for distinguishing normal edible vegetable oil from refined hogwash oil based on Fourier transform | |
| Li et al. | E-nose and colorimetric sensor array combining homologous data fusion strategy discriminating the roasting degree of large-leaf yellow tea | |
| Mohamed et al. | Food powders classification using handheld near-infrared spectroscopy and support vector machine | |
| CN113092403A (en) | Terahertz spectrum qualitative detection method for textile material of school uniform for students |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |