Construction method and application of polycarbonate reclaimed material identification model

Abstract

The invention belongs to the field of plastic classification and identification, and provides a construction method and application of a polycarbonate reclaimed material identification model. According to the method, the difference of various types of polycarbonate in volatile compounds is utilized, and the volatile compound characteristic database is utilized to construct the prediction model, so that accurate identification of polycarbonate with different sample sources, polycarbonate regenerated materials with different recovery sources and polycarbonate raw materials with different synthesis processes can be realized; compared with a common comparison method, the method has higher accuracy.

Description

Technical Field

[0001] The present invention belongs to the field of plastic classification and identification, and specifically relates to a method for constructing a polycarbonate recycled material identification model and its application. Background Art

[0002] Currently, about 99% of plastics come from petroleum-based sources. It is expected that by 2025, the global plastics industry will account for 16% of oil consumption and contribute 15% of carbon emissions. According to research reports, recycled plastics can significantly reduce the carbon emission value of products, and many developed countries advocate that manufacturers add recycled plastics to their products.

[0003] Polycarbonate is an excellent engineering plastic, with outstanding impact resistance and creep resistance, relatively high tensile strength, flexural strength, elongation and rigidity, good heat resistance and cold resistance, excellent electrical properties, low water absorption, good light transmittance, etc. In 2020, China's polycarbonate production capacity reached 1.79 million tons, accounting for about 30% of the global polycarbonate industry.

[0004] The application markets of polycarbonate mainly include fields such as electronic and electrical, sheets / films, vehicle lamps, vehicle windows, optics, household appliances, packaging, medical treatment, etc. During the use, recycling, and reprocessing of plastics, it is inevitable that pollutants, degradation products, organic impurities, oligomers, monomers, cleaning agents, fragrances, etc. remain in polycarbonate recycled plastic products. Secondly, in order to repair the use performance of recycled plastics to meet the application scenarios of the next life cycle, different modification aids may be added to polycarbonate recycled plastics, such as antioxidants, chain extenders, transesterification agents, crosslinking agents, etc. Traditional physical property detection methods have low applicability in distinguishing recycled plastics. Therefore, the accurate identification of polycarbonate recycled plastics is an urgent need in this field.

[0005] However, currently in the industry, the certification of recycled plastic products mostly relies on document review, and there are relatively few corresponding identification methods for whether recycled plastics are contained in products. Most of the existing identification technologies are identified through simple comparison, which is relatively crude and often has problems such as insufficient identification depth and low identification accuracy.

[0006] CN109870558 discloses a method for identifying polycarbonate plastic recycled materials. In this method, the peak appearance situation of a sample to be measured is determined by headspace gas chromatography. It is only determined that the sample to be measured is a recycled material based on the peak appearance time being greater than 14 minutes and the number of peaks being more than 7, which is prone to misjudgment and the accuracy of identification is relatively insufficient. Summary of the Invention

[0007] To solve the technical problem of insufficient accuracy in identifying recycled polycarbonate plastics in the prior art, the present invention provides a method for constructing an identification model for recycled polycarbonate materials and a method for identifying recycled polycarbonate materials.

[0008] The primary object of the present invention is to provide a method for constructing an identification model for recycled polycarbonate materials.

[0009] The secondary object of the present invention is to provide a method for identifying recycled polycarbonate materials.

[0010] The above objects of the present invention are achieved by the following technical solutions:

[0011] The present invention protects a method for constructing an identification model for recycled polycarbonate materials, including the following steps:

[0012] Step S1: Obtain polycarbonate samples, where the polycarbonate samples include recycled polycarbonate material samples from different recycling sources and virgin polycarbonate material samples from different synthesis processes. Mark classification labels for the polycarbonate samples, where the classification labels include at least one of sample source, recycling source, and synthesis process. The sample source includes virgin material and recycled material. The recycling sources include optical discs, bucket materials, car lights, black miscellaneous materials, small household appliances, and game console casings. The synthesis processes include the phosgene method and the non-phosgene method;

[0013] Step S2: Analyze the volatile compounds in the polycarbonate samples to establish a volatile compound characteristic database;

[0014] Among them, when the classification label includes the sample source, the volatile compound characteristic database includes at least one of dichloromethane, chlorobenzene, methyl methacrylate, etc. and at least one of n-heptane, styrene, benzene, phenol, etc.;

[0015] When the classification label includes the recycling source, the volatile compound characteristic database includes at least one of benzaldehyde, 1,4-dichlorobenzene, cyclohexanone, etc., at least one of 5,8-diethyldodecane, methyl acrylate, undecane, nonadecane, dodecane, etc., at least one of methyl isobutyrate, 3-methylideneheptane, ethylbenzene, 6-ethyl-2-methyl-decane, tetratetracontane, etc., at least one of 1-methoxy-2-propanol, 1-butanone, trans-1-butenyloxy-pentane, etc., and at least one of benzofuran, dichloromethane;

[0016] When the classification label includes the synthesis process, the volatile compound characteristic database includes at least one of carbon tetrachloride, acetone, m-xylene, p-tert-butylphenol, chlorobenzene, butyl acrylate, etc. and at least one of p-xylene, 2,6-bis(1,1-dimethylethyl)phenol, phenol, heptadecane, 1-butanol, nonadecane, etc.;

[0017] Step S3: Perform data preprocessing, data dimensionality reduction, and feature screening on the volatile compound feature database, train and optimize at least one machine learning model, and obtain at least one single prediction model.

[0018] The present invention provides a method for constructing a polycarbonate recycled material identification model. Polycarbonates from different sample sources, polycarbonates from different recycling sources, and polycarbonates from different synthesis processes each have differences in volatile compounds. This method utilizes the differences in volatile compounds among various types of polycarbonates, constructs a feature database based on volatile compounds by measuring volatile compounds, and constructs a prediction model using the volatile compound feature database, which can realize the identification of polycarbonates from different sample sources, polycarbonate recycled materials from different recycling sources, and polycarbonate virgin materials from different synthesis processes.

[0019] For identifying polycarbonates from different sample sources: The characteristic substances of polycarbonate virgin materials are dichloromethane, chlorobenzene, and methyl methacrylate, and the characteristic substances of polycarbonate recycled materials are n-heptane, styrene, benzene, and phenol. Therefore, the volatile compound feature database for identifying polycarbonates from different recycling sources should at least include at least one of dichloromethane, chlorobenzene, and methyl methacrylate and at least one of n-heptane, styrene, benzene, and phenol;

[0020] For identifying polycarbonates from different recycling sources: The characteristic substances of polycarbonate recycled materials from CD sources are benzaldehyde, 1,4-dichlorobenzene, and cyclohexanone; the characteristic substances of polycarbonate recycled materials from bucket sources are 5,8-diethyldodecane, methyl acrylate, undecane, nonadecane, and dodecane; the characteristic substances of polycarbonate recycled materials from black miscellaneous material sources are methyl isobutyrate, 3-methylideneheptane, ethylbenzene, 6-ethyl-2-methyl-decane, and tetratetracontane; the characteristic substances of polycarbonate recycled materials from headlight material sources are 1-methoxy-2-propanol, 1-butanone, and trans-1-butenyloxy-pentane; the characteristic substances of polycarbonate recycled materials from small household appliances and game console casings sources are benzofuran and dichloromethane. Therefore, the volatile compound feature database for identifying polycarbonates from different recycling sources should at least include at least one of benzaldehyde, 1,4-dichlorobenzene, and cyclohexanone, at least one of 5,8-diethyldodecane, methyl acrylate, undecane, nonadecane, and dodecane, at least one of methyl isobutyrate, 3-methylideneheptane, ethylbenzene, 6-ethyl-2-methyl-decane, and tetratetracontane, at least one of 1-methoxy-2-propanol, 1-butanone, and trans-1-butenyloxy-pentane, and at least one of benzofuran and dichloromethane, which can realize the identification of polycarbonate recycled materials from different recycling sources;

[0021] For identifying polycarbonates synthesized by different processes: The characteristic substances in the virgin materials of polycarbonates synthesized by the phosgene method are carbon tetrachloride, acetone, m-xylene, p-tert-butylphenol, chlorobenzene, and butyl acrylate. The characteristic substances in the virgin materials of polycarbonates synthesized by the non-phosgene method are p-xylene, 2,6-bis(1,1-dimethylethyl)phenol, phenol, heptadecane, 1-butanol, and nonadecane. Therefore, the characteristic database of volatile compounds for identifying polycarbonates synthesized by different processes should include at least one of carbon tetrachloride, acetone, m-xylene, p-tert-butylphenol, chlorobenzene, and butyl acrylate and at least one of p-xylene, 2,6-bis(1,1-dimethylethyl)phenol, phenol, heptadecane, 1-butanol, and nonadecane, enabling the identification of virgin materials of polycarbonates synthesized by different processes.

[0022] On the one hand, the types of volatile compounds in the characteristic database of volatile compounds are obtained through the analysis of a large number of polycarbonates, possessing both universality and accuracy. On the other hand, a prediction model is obtained by training the characteristic database of volatile compounds on a machine learning model, and using the prediction model for determination has higher accuracy compared to ordinary comparison methods.

[0023] The black miscellaneous materials described in the present invention are black recycled miscellaneous material fragments or particles.

[0024] Specifically, the methods used for the analysis of volatile compounds in step S2 include, but are not limited to, headspace-gas chromatography.

[0025] More specifically, the equilibrium temperature of the headspace-gas chromatography is 100 - 200 °C, and the equilibrium time is 200 - 400 min.

[0026] Specifically, the chromatographic column of the headspace-gas chromatography is selected from low-polarity columns or non-polarity columns, further selected from 100% methyl polysiloxane or 5% phenyl-95% methyl polysiloxane, including but not limited to Mega-5MS chromatographic columns.

[0027] Specifically, the temperature programming of the headspace-gas chromatography is as follows: The initial temperature is maintained at 30 - 50 °C for 4 - 6 min, then increased to 240 - 280 °C at a rate of 6 - 10 °C / min and maintained for 6 - 10 min.

[0028] Specifically, the flow rate of the headspace-gas chromatography is 1.0 - 1.5 mL / min.

[0029] Specifically, the inlet temperature of the headspace-gas chromatography is 220 - 280 °C.

[0030] Specifically, the split ratio of the headspace-gas chromatography is 1 - 10:1.

[0031] Specifically, the detectors of the headspace-gas chromatography method include but are not limited to mass spectrometry detectors, FID detectors, ECD detectors, TCD detectors, and FPD detectors.

[0032] More specifically, the temperature of the MSD transmission line of the mass spectrometry detector is 250-300 °C.

[0033] Specifically, the temperature of the ion source of the headspace-gas chromatography method is 200-250 °C.

[0034] Preferably, the construction method further includes the following steps:

[0035] Step S4: Assign weights to the prediction results of at least two single prediction models based on the optimal weighting method, and obtain a combined prediction model after training and optimization.

[0036] To further improve the discrimination accuracy, the present invention further constructs a combined prediction model based on the optimal weighting method.

[0037] Specifically, the machine learning model is selected from at least one of partial least squares discriminant analysis (PLS-DA), orthogonal partial least squares discriminant analysis (OPLS-DA), principal component analysis (PCA), decision tree (DT), support vector machine (SVM), random forest (RF), K-nearest neighbor (KNN), Bayesian network, artificial neural network (ANN), and AdaBoosting. Preferably, when the classification label includes the sample source, the machine learning model is a random forest, decision tree, or Adaboosting; when the classification label includes the synthesis process, the machine learning model includes a decision tree or Adaboosting; when the classification label includes the recycling source, the machine learning model includes a random forest or Adaboosting.

[0038] Specifically, the data preprocessing includes data summarization and / or data cleaning; the data dimensionality reduction uses the principal component analysis method. To reduce the complexity of the database for model training, testing, and optimization, the present invention uses the principal component analysis method to perform dimensionality reduction processing on the volatile feature database.

[0039] The present invention also protects a method for identifying recycled polycarbonate materials, including the following steps:

[0040] Step S1: Obtain a polycarbonate sample to be identified;

[0041] Step S2: Analyze the volatile compounds in the polycarbonate sample to be identified to obtain a dataset to be identified;

[0042] Step S3: Input the dataset to be identified into the above single prediction model to obtain the classification label of the polycarbonate sample to be identified.

[0043] Specifically, the method for analyzing volatile compounds described in step S2 includes, but is not limited to, headspace-gas chromatography.

[0044] More specifically, the equilibrium temperature of the headspace-gas chromatography is 100-200 °C, and the equilibrium time is 200-400 min.

[0045] Specifically, the chromatographic column of the headspace-gas chromatography is selected from low-polarity columns or non-polar columns, and further selected from 100% methyl polysiloxane or 5% phenyl-95% methyl polysiloxane, including but not limited to Mega-5MS chromatographic column.

[0046] Specifically, the temperature programming of the headspace-gas chromatography is as follows: the initial temperature is maintained at 30-50 °C for 4-6 min, and then increased to 240-280 °C at a rate of 6-10 °C / min and maintained for 6-10 min.

[0047] Specifically, the flow rate of the headspace-gas chromatography is 1.0-1.5 mL / min.

[0048] Specifically, the inlet temperature of the headspace-gas chromatography is 220-280 °C.

[0049] Specifically, the split ratio of the headspace-gas chromatography is 1-10:1.

[0050] Specifically, the detector of the headspace-gas chromatography includes, but is not limited to, mass spectrometry detector, FID detector, ECD detector, TCD detector, FPD detector.

[0051] More specifically, the MSD transfer line temperature of the mass spectrometry detector is 250-300 °C.

[0052] Specifically, the ion source temperature of the headspace-gas chromatography is 200-250 °C.

[0053] Compared with the existing technology, the beneficial effects of the present technical solution are as follows:

[0054] The present invention provides a method for constructing a polycarbonate recycled material identification model. By utilizing the differences in volatile compounds among various types of polycarbonates and constructing a prediction model using a volatile compound characteristic database, it is possible to identify polycarbonates from different sample sources, polycarbonate recycled materials from different recycling sources, and virgin polycarbonate materials from different synthesis processes.

[0055] The volatile characteristic substances in various types of polycarbonates in the present invention are obtained by analyzing a large number of polycarbonates, which has both universality and accuracy. And the determination using the prediction model has higher accuracy compared with ordinary comparison methods. Description of the Drawings

[0056] Figure 1 Confusion matrix results of multiple machine learning models for virgin materials and recycled materials in Example 3;

[0057] Figure 2 PLS-DA score plot (a), three-dimensional scatter score plot (b), loading plot (c), and biplot (d) of volatile substances for different synthesis processes of virgin polycarbonate in Example 4. Detailed Implementation Modes

[0058] The present method will be described below in conjunction with the embodiments.

[0059] Example 1 A method for constructing a polycarbonate recycled material identification model

[0060] The present example provides a method for constructing a polycarbonate recycled material identification model, including the following steps:

[0061] Step S1: Obtain polycarbonate samples, where the polycarbonate samples include polycarbonate recycled material samples from different recycling sources and polycarbonate virgin material samples from different synthesis processes. Mark classification labels for the polycarbonate samples, where the classification labels include sample source, recycling source, and synthesis process. The sample source includes virgin material and recycled material, the recycling source includes optical discs, water buckets, water bucket plates, car headlights, black miscellaneous materials, small household appliances, and game console casings, and the synthesis process includes the phosgene method and the non-phosgene method;

[0062] Step S2: Conduct an analysis of volatile compounds on the polycarbonate samples to establish a volatile compound characteristic database;

[0063] Among them, the analysis of volatile compounds uses the headspace-gas chromatography method, and the test method of the headspace-gas chromatography method is as follows:

[0064] Weigh 2 g (accurate to 0.001 g) of the sample and place it in a 20 mL headspace vial for on-machine testing. Seal it immediately after weighing to reduce the loss of volatile substances. The headspace vial is equilibrated at 150 °C for 300 min, and a sampling needle is used to aspirate 1000 μL of volatile substances into a headspace gas chromatography-mass spectrometry instrument (GCMS1000, China Hexin Mass Spectrometry Company) for analysis;

[0065] Chromatographic conditions: The chromatographic column is Mega-5ms, with dimensions of 30 m × 0.25 mm × 0.25 μm; carrier gas: helium, flow rate: 1.2 mL / min; injection port temperature 250 °C; column oven temperature program: initial temperature 40 °C, hold for 5 min, increase to 260 °C at a rate of 8 °C / min, and hold for 8 min; split ratio is 5:1;

[0066] Mass spectrometry conditions: MSD transfer line temperature: 280 °C; acquisition mode: full scan; scan mass range: 35 - 550 amu; ion source temperature 220 °C; solvent delay: 0.1 min;

[0067] The volatile compound database includes the compounds shown in Table 1:

[0068] Table 1 Volatile Compound Database

[0069]

[0070]

[0071] Continued Table 1 Volatile Compound Database

[0072]

[0073]

[0074] Continued Table 1 Volatile Compound Database

[0075]

[0076]

[0077]

[0078] Continued Table 1 Volatile Compound Database

[0079]

[0080]

[0081] Continued Table 1 Volatile Compound Database

[0082]

[0083]

[0084] Step S3: Perform data preprocessing, data dimensionality reduction, and feature screening on the volatile compound database, train and optimize at least one machine learning model, and obtain at least one single prediction model.

[0085] Example 2 A method for identifying recycled polycarbonate

[0086] This example provides a method for identifying recycled polycarbonate, including the following steps:

[0087] Step S1: Obtain the polycarbonate sample to be identified;

[0088] Step S2: Analyze the volatile compounds of the polycarbonate sample to be identified to obtain a dataset to be identified;

[0089] Step S3: Input the dataset to be identified into the above single-item prediction model to obtain the classification label of the polycarbonate sample to be identified;

[0090] Among them, the volatile compound analysis uses headspace-gas chromatography, and the test method of the headspace-gas chromatography is as follows:

[0091] Weigh 2 g (accurate to 0.001 g) of the sample and place it in a 20 mL headspace vial for on-machine testing. Seal it immediately after weighing to reduce the loss of volatile substances. The headspace vial is equilibrated at 150 °C for 300 min, and 1000 μL of volatile substances are aspirated by a sampling needle and injected into a gas chromatography-mass spectrometry (GCMS1000, Comix Mass Spectrometry Co., Ltd., China) for analysis;

[0092] Chromatographic conditions: The chromatographic column is Mega-5ms, with dimensions of 30 m × 0.25 mm × 0.25 μm; carrier gas: helium, flow rate: 1.2 mL / min; injection port temperature 250 °C; column oven temperature program: initial temperature 40 °C is maintained for 5 min, heated to 260 °C at a rate of 8 °C / min, and maintained for 8 min; split ratio is 5:1;

[0093] Mass spectrometry conditions: MSD transfer line temperature: 280 °C; acquisition mode: full scan; scan mass range: 35 - 550 amu; ion source temperature 220 °C; solvent delay: 0.1 min.

[0094] Example 3 A method for constructing a polycarbonate recycled material identification model

[0095] In this example, based on the construction method of Example 1, five single-item prediction models for the source of polycarbonate samples were constructed using machine learning models such as SVM, RF, KNN, DT, and adaboosting.

[0096] The verification results of the above five single-item prediction models are as Figure 1 shown in Table 2. According to the standard of VIP value > 1, characteristic substances are screened. Among them, the characteristic substances of virgin polycarbonate materials include dichloromethane, chlorobenzene, and methyl methacrylate, and the characteristic substances of polycarbonate recycled materials include n-heptane, styrene, benzene, and phenol.

[0097] Table 2 Evaluation results of various machine learning models for virgin materials and recycled materials

[0098]

[0099] For recycled polycarbonate plastics, the RF, DT, and adaboosting algorithms can achieve a 100% discrimination accuracy for the volatile component data in polycarbonate samples. Qualitative discrimination of recycled plastics can be effectively guided by combining chemical characterization analysis with machine learning algorithms.

[0100] Example 4: A method for constructing a discrimination model of recycled polycarbonate

[0101] Based on the construction method of Example 1, this example uses the PLS-DA machine learning model to construct a single-item prediction model for the synthesis process.

[0102] The verification results of the above single-item prediction model are as Figure 2 shown in Table 3. Characteristic substances are screened according to the standard of VIP value > 1. Among them, the characteristic substances in the virgin polycarbonate raw material synthesized by the phosgene method include carbon tetrachloride, acetone, m-xylene, p-tert-butylphenol, chlorobenzene, and butyl acrylate; the characteristic substances in the virgin polycarbonate raw material synthesized by the non-phosgene method include p-xylene, 2,6-bis(1,1-dimethylethyl)phenol, phenol, heptadecane, 1-butanol, and nonadecane.

[0103] Table 3: Evaluation results of various machine learning models for phosgene and non-phosgene PC

[0104]

[0105] As can be seen from Table 3, the single-item discrimination models constructed by the RF, DT, and adaboosting algorithms can accurately discriminate the virgin polycarbonate raw materials with different synthesis processes. Among them, the discrimination accuracy of the single-item discrimination models constructed by the DT and adaboosting algorithms can reach 100%.

[0106] Example 5: A method for constructing a discrimination model of recycled polycarbonate

[0107] Based on the construction method of Example 1, this example uses the PLS-DA machine learning model to construct a single-item prediction model for the recycling source.

[0108] The verification results of the above single prediction model are shown in Table 4. Feature substances were screened according to the criterion of VIP value > 1. Among them, the feature substances of polycarbonate recycled materials from optical discs include benzaldehyde, 1,4-dichlorobenzene, and cyclohexanone; the feature substances of polycarbonate recycled materials from water buckets include 5,8-diethyldodecane, methyl acrylate, undecane, nonadecane, and dodecane; the feature substances of polycarbonate recycled materials from black miscellaneous materials include methyl isobutyrate, 3-methylideneheptane, ethylbenzene, 6-ethyl-2-methyldecane, and tetratetracontane; the feature substances of polycarbonate recycled materials from headlight materials include 1-methoxy-2-propanol, 1-butanone, and trans-1-butenyloxy-pentane; the feature substances of polycarbonate recycled materials from small household appliances and game consoles include benzofuran and dichloromethane.

[0109] Table 4 Evaluation results of multiple machine learning models for recycled PC from different recycling sources

[0110]

[0111] As can be seen from Table 4, the single discrimination models constructed by RF, DT, and adaboosting algorithms can all accurately discriminate polycarbonate recycled materials from different recycling sources. Among them, the discrimination accuracy of the single discrimination models constructed by RF and adaboosting algorithms can reach 100%.

[0112] The specific embodiments described above further elaborate on the purpose, technical solutions, and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the protection scope of the present invention. In particular, it is pointed out that for those skilled in the art, any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for constructing a polycarbonate recycled material identification model, comprising the following steps: Step S1, obtaining polycarbonate samples, wherein the polycarbonate samples include polycarbonate recycled material samples from different recycling sources and polycarbonate virgin material samples from different synthesis processes, and marking the polycarbonate samples with classification labels, wherein the classification labels include at least one of the sample source, the recycling source, and the synthesis process, wherein the sample source includes virgin material and recycled material, the recycling source includes optical discs, bucket materials, car lights, black miscellaneous materials, small household appliances, and game console casings, and the synthesis process includes a phosgene method and a non-phosgene method; Step S2, analyzing the volatile compounds of the polycarbonate sample and establishing a volatile compound feature database; Wherein, when the classification label includes the sample source, the volatile compound feature database includes at least one of dichloromethane, chlorobenzene, methyl methacrylate and at least one of n-heptane, styrene, benzene, and phenol; When the classification label includes a recycled source, the volatile compound feature database includes at least one of benzaldehyde, 1,4-dichlorobenzene, and cyclohexanone, at least one of 5,8-diethyldodecane, 2-methyl acrylate, undecane, nonadecane, and dodecane, at least one of methyl isobutyrate, 3-methyleneheptane, ethylbenzene, 6-ethyl-2-methyl-decane, and tetradecane, at least one of 1-methoxy-2-propanol, 1-butanone, and trans-1-butenyloxy-pentane, and at least one of benzofuran and dichloromethane; When the classification label includes a synthesis process, the volatile compound feature database includes at least one of carbon tetrachloride, acetone, m-xylene, p-tert-butylphenol, chlorobenzene, 2-butyl acrylate and at least one of p-xylene, 2,6-bis(1,1-dimethylethyl)phenol, phenol, heptadecane, 1-butanol, and nonadecane; Step S3: performing data preprocessing, data dimension reduction and feature screening on the volatile compound feature database, training and optimizing at least one machine learning model, and obtaining at least one single prediction model.

2. The construction method according to claim 1, characterized in that: The method used for analyzing the volatile compounds in step S2 includes headspace-gas chromatography or solid phase microextraction-gas chromatography.

3. The construction method according to claim 2, characterized in that: The equilibrium temperature of the headspace-gas chromatography method is 100-200° C., and the equilibrium time is 200-400 minutes.

4. The construction method according to claim 2, characterized in that: The chromatographic column of the headspace-gas chromatography method is selected from a low-polarity column or a non-polar column.

5. The construction method according to claim 2, characterized in that: The temperature raising procedure of the headspace-gas chromatography method is as follows: the initial temperature is 30-50° C. and maintained for 4-6 minutes, then raised to 240-280° C. at a rate of 6-10° C. / min and maintained for 6-10 minutes.

6. The construction method according to claim 1, characterized in that: The construction method further comprises the following steps: Step S4: weight the prediction results of at least two single prediction models based on the optimal weighting method, and obtain a combined prediction model after training and optimization.

7. The construction method according to claim 1, characterized in that: The machine learning model is selected from at least one of PLS-DA, OPLS-DA, PCA, decision tree, support vector machine, random forest, K nearest neighbor, Bayesian network, artificial neural network, and AdaBoosting.

8. The construction method according to claim 1, characterized in that: The data preprocessing includes data aggregation and / or data cleaning.

9. The construction method according to claim 1, characterized in that: The data dimension reduction adopts principal component analysis method.

10. A method for identifying polycarbonate recycled materials, characterized in that: The following steps are involved: Step S1, obtaining a polycarbonate sample to be identified; Step S2, performing volatile compound analysis on the polycarbonate sample to be identified to obtain a data set to be identified; Step S3, inputting the data set to be identified into any one of the single-item prediction models of claims 1 to 9 to obtain a classification label of the polycarbonate sample to be identified.

Images