In-situ real-time determination method and system for dynamic change of fragrance in black tea drying process and application of in-situ real-time determination method and system

Abstract

The invention relates to an in-situ real-time determination method and equipment for dynamic change of aroma in a black tea drying process and application. The method comprises the following steps: acquiring a calibration sample, monitoring VOC (Volatile Organic Compounds) generated by the sample in the drying process on line, and screening to obtain a VOC marker combination associated with the aroma; constructing an aroma type discrimination model based on the VOC marker combination; obtaining a to-be-detected sample, drying the to-be-detected sample, and inputting VOC generated in real time based on the to-be-detected sample in the treatment process into the fragrance type discrimination model to obtain a corresponding fragrance type; the method is applied to distinguishing different stages of aroma type conversion in full fire drying of black tea. The equipment comprises heating and drying equipment for drying a sample in the equipment, an analyzer loaded with HPPI-TOFMS is communicated with the space of the heating and drying equipment through a stainless steel capillary tube, VOC in the drying process is monitored on line, and the corresponding fragrance is obtained. The method has the advantages that the response speed is high, the detection process is simple, pretreatment is not needed, and in-situ, real-time and online detection can be achieved.

Description

Technical Field

[0001] This invention relates to the technical field of testing or analyzing materials by measuring their chemical or physical properties, and particularly to an in-situ real-time method, system, and application for measuring the dynamic changes in aroma during the drying process of black tea. Background Technology

[0002] Tea, as one of the world's three major beverages, is loved by people all over the world. Based on different processing methods, tea can be divided into six different categories: green tea, black tea, oolong tea, yellow tea, dark tea, and white tea. Among them, black tea is a fully fermented tea and is the most consumed type of tea globally, wielding significant influence in the international market. In recent years, the market attention and demand for black tea have continued to rise.

[0003] In tea quality evaluation, aroma and taste are the most critical attributes, with aroma accounting for approximately 25% of the total sensory evaluation score, and even reaching 30% in broken black tea. Tea aroma characteristics are diverse; taking black tea as an example, common aromas include sweetness, floral and fruity notes, and caramel. Aroma formation is greatly influenced by processing techniques, with drying being crucial among the many processing steps. During the high-temperature drying process of black tea, the volatile compounds in the tea undergo profound and rapid transformations, continuously forming and converting through thermochemical reactions, accompanied by dynamic changes such as release, enhancement, and decay. In this process, the aroma profile evolves rapidly; grassy notes gradually disappear, while honey-sweet, floral and fruity, and caramel aromas gradually emerge, ultimately shaping the superior aroma quality of the finished tea under the combined action of aroma-active substances. Different drying methods, temperatures, and durations all significantly affect the changes in aroma substances in tea; for example, excessively high drying temperatures or prolonged drying times can easily cause an unpleasant, burnt flavor. Currently, judging the appropriate drying and aroma formation of black tea mainly relies on the tea master's smelling ability and human experience. Therefore, using modern analytical techniques to accurately identify the aroma transformation during the drying process of black tea in real time is of great value for precise control of the drying process and targeted shaping of aroma quality.

[0004] In existing technologies, the detection of aroma compounds in tea mainly relies on gas chromatography-mass spectrometry (GC-MS). This method is widely used in the analysis of volatile components. It utilizes solid-phase microextraction or solvent-assisted evaporation to enrich aroma molecules, which are then injected into the chromatographic system for analysis via thermal desorption. Although GC-MS technology is mature and highly accurate, its sample pretreatment process is relatively cumbersome, the detection cycle is long, and the analytical results are easily affected by the sample pretreatment method. Especially in the drying process, the high temperature conditions pose a significant challenge to GC-MS sample collection and make it difficult to accurately assess the dynamic transformation of aroma. More importantly, this technology currently cannot achieve real-time, online, in-situ detection, while the aroma of tea changes rapidly during the drying stage, making it difficult for GC-MS to achieve synchronous process monitoring and obtain corresponding dynamic spectral information in real time. Summary of the Invention

[0005] This invention addresses the problems existing in the prior art, providing an in-situ real-time determination method, system, and application for the dynamic changes in aroma during the drying process of black tea. It primarily utilizes HPPI-TOFMS technology, requiring no sample pretreatment, enabling real-time, online analysis and detection of multiple aroma substances during the black tea drying process, and accurately identifying dynamic aroma transitions. In recent years, online mass spectrometry based on soft ionization has developed rapidly. Among them, high-pressure photoionization-time-of-flight mass spectrometry (HPPI-TOF / MS), with its advantages of high resolution, rapid detection, wide coverage, and no sample pretreatment required, can achieve in-situ, real-time, and online monitoring of volatiles during tea drying. Simultaneously, soft ionization technology mainly generates molecular or quasi-molecular ions, facilitating qualitative and quantitative analysis of complex mixtures. By dynamically tracking key aroma-active substances during the drying process, a reliable basis can be provided for the accurate identification of aroma transitions.

[0006] The technical solution adopted in this invention is an in-situ real-time determination method for the dynamic changes of aroma in the drying process of black tea. The method obtains a calibration sample, monitors the VOCs (volatile organic compounds) generated by the sample during the drying process online, and screens to obtain a combination of VOC markers associated with the aroma.

[0007] Construct an aroma discrimination model based on the VOC marker combination;

[0008] A sample to be tested is obtained and dried. During the drying process, the VOC generated in real time by the sample is input into the aroma discrimination model to obtain the corresponding aroma.

[0009] Preferably, the combination of VOC markers associated with the aroma is geraniol, (E,E)-2,4-nonadienal, trans-2-hexenal, 1-octanol, phenylacetaldehyde, phenylethanol, β-linalool, trans-2-nonal, and trans-citral.

[0010] Preferably, the calibration sample is a black tea sample after initial roasting, with a moisture content of 20% to 25%.

[0011] Preferably, the process of screening and obtaining a combination of VOC markers associated with aroma type includes the following steps:

[0012] The calibration sample mentioned in S1.1 includes the analysis group. The calibration sample of the analysis group is dried at a preset experimental temperature. HPPI-TOFMS is used to monitor the VOC generated during the processing of the analysis group at a preset frequency f1 and the original mass spectrum is acquired in real time.

[0013] S1.2 Based on standards, perform compound characterization of VOCs detected by HPPI-TOFMS;

[0014] S1.3 preprocesses the raw mass spectrometry data, including simple noise filtering, mass calibration, and peak shape fitting (Gaussian fitting). Based on this, features are extracted to obtain the mass-to-charge ratio of VOC ions and their corresponding mass spectrometry response intensities.

[0015] S1.4 Establish a mixed screening mechanism to screen combinations of VOC markers corresponding to different aroma types.

[0016] Preferably, the calibration sample further includes a control group, wherein the calibration sample of the control group is dried at a preset experimental temperature, and the VOC generated during the treatment of the control group is used as a control sample at a preset frequency f2, and the expert evaluation data of the control sample is obtained accordingly to guide the mixed screening mechanism; f1>f2.

[0017] Preferably, in S1.4, the mixed screening mechanism performs statistical analysis based on a combination of data on moisture change rate, aroma attributes, and VOC relative intensity information (peak area), and selects VOC marker combinations corresponding to different aroma types according to a set threshold. This mixed screening mechanism is used to verify the consistency of aroma transformation patterns and to screen VOCs. In subsequent processing, the mixed screening mechanism is not needed again for the same product category or under the same processing conditions.

[0018] Preferably, the statistical analysis includes univariate analysis and multivariate analysis, and the multivariate analysis includes unsupervised analysis and supervised analysis; the unsupervised analysis combines the moisture change rate and aroma attributes to group the samples, and the supervised analysis is used to screen VOC markers that meet the conditions.

[0019] Preferably, a random forest model is constructed based on the VOC marker combinations associated with aroma obtained through screening as an aroma discrimination model, which is used to discriminate the aroma transformation during the drying process of black tea.

[0020] The application of an in-situ real-time measurement method for the dynamic changes in aroma during the drying process of black tea is used to distinguish different stages of aroma transformation during the final drying of black tea.

[0021] A measuring device employing the aforementioned in-situ real-time measurement method for the dynamic changes in aroma during the drying process of black tea includes:

[0022] A heating and drying device used to dry samples at a preset temperature;

[0023] An analyzer, equipped with HPPI-TOFMS, is spatially connected to the heating and drying equipment via a stainless steel capillary tube.

[0024] This invention relates to an in-situ real-time measurement method, equipment, and application for the dynamic changes in aroma during the drying process of black tea. The method involves acquiring a calibration sample, monitoring VOCs generated by the sample during the drying process online, and screening for VOC marker combinations associated with the aroma type. An aroma discrimination model based on these VOC marker combinations is constructed. A test sample is acquired and dried. During the drying process, the VOCs generated in real time by the test sample are input into the aroma discrimination model to obtain the corresponding aroma type. The method is applied to distinguish different stages of aroma transformation during the final drying of black tea. The equipment includes a heating and drying device for drying samples at a preset temperature. An analyzer equipped with HPPI-TOFMS is spatially connected to the heating and drying device via a stainless steel capillary tube to monitor VOCs during the drying process online and obtain the corresponding aroma type.

[0025] The beneficial effects of this invention are as follows:

[0026] (1) Using HPPI-TOFMS as a detection method for aroma substances in black tea has advantages such as in-situ, real-time, online, rapid, and preservation of initial aroma characteristics (without any pretreatment). It has high temporal resolution (collection frequency once per minute) and can capture instantaneous change patterns. Compared with other rapid identification methods, the identification capability of this invention reaches the compound level and can be applied to actual production, which is innovative and scientific.

[0027] (2) It weakens the subjective factors of tea masters using artificial senses to judge the aroma of black tea, making the results more objective, reliable and convincing;

[0028] (3) The random forest data model can quickly and accurately obtain the aroma classification results during the drying process of black tea, which helps to accurately control the drying process and shape the aroma quality in a targeted manner, and is applicable to a wider range of application scenarios. At the same time, it promotes the quality control of black tea drying from manual experience to a quantifiable digital model.

[0029] (4) The present invention has the advantages of fast response speed, simple detection process, no need for any sample pretreatment, and in-situ, real-time and online detection during the drying process of black tea. Attached Figure Description

[0030] Figure 1 This is a flowchart of the method of the present invention;

[0031] Figure 2 This is a schematic diagram of the device structure of the present invention;

[0032] Figure 3 This is a schematic diagram of sample grouping in this invention, showing (A) changes in water peak intensity during the drying process, (B) the quantitative sensory descriptive analysis (QDA) score results of aroma characteristics, and (C) the principal component analysis diagram of VOCs detected by HPPI-TOF / MS.

[0033] Figure 4 The diagram shows the PLS-DA model for VOCs detected by HPPI-TOF / MS during the drying process of black tea in this invention. (A) PLS-DA score diagram, (B) permutation test diagram.

[0034] Figure 5 In this invention, the random forest model prediction graphs are shown in (A) and (B) for the aroma components that contribute the most. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] like Figure 1 As shown, this invention relates to an in-situ real-time method for measuring the dynamic changes in aroma during the drying process of black tea. The method obtains a calibration sample, monitors the VOCs generated by the sample during the drying process online, and screens to obtain a combination of VOC markers associated with the aroma.

[0037] Construct an aroma discrimination model based on the VOC marker combination;

[0038] A sample to be tested is obtained and dried. During the drying process, the VOC generated in real time by the sample is input into the aroma discrimination model to obtain the corresponding aroma.

[0039] In this invention, aroma samples generated during the drying process of black tea are analyzed using HPPI-TOFMS technology to obtain raw mass spectra. The raw mass spectra are preprocessed, and differential VOC combinations are screened out using statistical analysis methods. A database of aroma compounds in the black tea drying process based on HPPI-TOFMS is established. These VOC combinations can be used to determine the aroma type in the subsequent black tea drying process, achieving rapid and accurate identification of aroma substances in tea during the drying process.

[0040] After processing and analysis, this invention proposes a combination of VOC markers associated with aroma type, namely geraniol (CAS 459-80-3), (E,E)-2,4-nonadienal (CAS 5910-87-2), trans-2-hexenal (CAS 6728-26-3), 1-octanol (CAS 111-87-5), phenylacetaldehyde (CAS 122-78-1), phenylethanol (CAS 60-12-8), β-linalool (CAS 78-70-6), trans-2-nonanal (CAS 18829-56-6), and trans-citral (CAS 5392-40-5).

[0041] The following is a detailed description with reference to specific embodiments.

[0042] The calibration sample is a black tea sample after initial roasting, with a moisture content of 20% to 25%.

[0043] In this embodiment, the raw material was fresh leaves of the Jin Xuan tea tree variety with one bud and two leaves, harvested in July 2024 at the Shengzhou Research and Experiment Base of the Chinese Academy of Agricultural Sciences. The calibration sample was obtained through the following treatment:

[0044] S0.1 Withering: Spread the fresh leaves evenly on the spreading rack to a thickness of 2-3cm. Wither at 28℃ and 70% relative humidity until the moisture content of the fresh leaves drops to about 64%. At this time, the color of the fresh leaves changes from bright green to dark green, the leaves are soft, and the stems do not break when broken.

[0045] S0.2 Kneading: Kneading is performed using a kneading machine for a total duration of 70 minutes, consisting of 20 minutes of dry kneading (without pressure), 15 minutes of light kneading, 20 minutes of medium kneading, 10 minutes of heavy kneading, and 5 minutes of light kneading. After processing, the leaf strip formation rate exceeds 90%, the strips are tightly bound, and the forming is good. After kneading, timely de-clumping is performed to prevent clumping.

[0046] S0.3 Fermentation: Place the kneaded leaves in a constant temperature and humidity fermentation chamber at 30℃ and 90% relative humidity for 3-4 hours, turning them manually every hour; after fermentation, the leaves turn yellowish-red.

[0047] S0.4 Initial drying: The tea leaves are initially dried using a hot air dryer at 110℃ for 10 minutes until the moisture content is about 20% to 25%. At this point, the leaves feel slightly prickly to the touch. After initial drying, the samples are collected, sealed, and stored in a cool, dry place at a low temperature for subsequent pre-set drying experiments.

[0048] In this invention, during the drying process at a preset experimental temperature, high-pressure photoionization time-of-flight mass spectrometry (HPPI-TOFMS) is used to detect volatiles in black tea in real time. It can achieve rapid analysis of trace volatile substances without sample pretreatment, and the detection sensitivity can reach the sub-ppbv level.

[0049] The process of identifying VOC biomarker combinations associated with aroma profiles includes the following steps:

[0050] The calibration sample mentioned in S1.1 includes the analysis group. The calibration sample of the analysis group is dried at a preset experimental temperature. HPPI-TOFMS is used to monitor the VOC generated during the processing of the analysis group at a preset frequency f1 online as the analysis sample, and the original mass spectrum is acquired in real time.

[0051] During implementation, the preset frequency f1 is to acquire the released VOC every 1 minute.

[0052] S1.2 Based on standards, perform compound characterization of VOCs detected by HPPI-TOFMS, including the following steps:

[0053] S1.2.1 Take the aroma substance standard solution and place it in a liquid chromatography vial, and then place it in a custom container connected to a stainless steel capillary tube;

[0054] S1.2.2 Using high-purity nitrogen (purity ≥99.999%) as the carrier gas, the volatiles purged from the standard were introduced into the mass spectrometer via a capillary tube to obtain the corresponding HPPI-TOFMS mass spectrometry standard data.

[0055] S1.2.3 A mass spectrometry library of aroma compounds of black tea was established by combining HPPI-TOFMS measured spectra, NIST database and information on key aroma compounds known in relevant literature reports;

[0056] S1.2.4 The original spectrum of the aroma sample is compared with that of the spectral library to complete the qualitative identification of the substance.

[0057] In this embodiment, a database of 38 common aroma compounds encountered during the drying process of black tea was established. Of these, 26 compounds were verified using standard references, enabling rapid and accurate identification of target compounds during the tea drying process. Table 1 shows the HPPI-TOFMS aroma substance identification table.

[0058] Table 1. Identification of aroma compounds in dried black tea by HPPI-TOFMS

[0059] No. Cat. Compound CAS Molecular formula Observedm / z IonAssignment Identification 1 alcohol benzyl alcohol 100-51-6 <![CDATA[C7H8O]]> 108.0568 <![CDATA[[M] + ]]> STD 2 alcohol 1-Octanol 111-87-5 <![CDATA[C8H 18 O]]> 112.1212 <![CDATA[[M-H2O] + ]]> STD 3 alcohol Phenylacetyl alcohol 60-12-8 <![CDATA[C8H 10 O]]> 122.0734 <![CDATA[[M] + ]]> STD 4 alcohol β-Linol 78-70-6 <![CDATA[C 10 H 18 O]]> 136.1255 <![CDATA[[M-H2O] + ]]> STD 5 alcohol Geraniol 106-24-1 <![CDATA[C 10 H 18 O]]> 154.134 <![CDATA[[M] + ]]> STD 6 alcohol Leaf alcohol 928-96-1 <![CDATA[C6H 12 O]]> 100.0862 <![CDATA[[M] + ]]> STD 7 alcohol 2-Pentylfuran 3777-69-3 <![CDATA[C9H 14 O]]> 138.1099 <![CDATA[[M] + ]]> STD 8 alcohol Citronellol 106-22-9 <![CDATA[C 10 H 20 O]]> 138.1379 <![CDATA[[M-H2O] + ]]> STD 9 alcohol 1-Octen-3-ol 3391-86-4 <![CDATA[C8H 16 O]]> 128.1139 <![CDATA[[M] + ]]> STD 10 aldehyde β-Citral 432-25-7 <![CDATA[C 10 H 16 O]]> 78.0472 <![CDATA[[C6H6] + ]]> STD 11 aldehyde (E,E)-2,4-Nonadienal 5910-87-2 <![CDATA[C9H 14 O]]> 96.0927 <![CDATA[[C7H 12 ] + ]]> STD 12 aldehyde trans-2-hexenal 6728-26-3 <![CDATA[C6H 10 O]]> 98.0726 <![CDATA[[M] + ]]> STD 13 aldehyde benzaldehyde 100-52-7 <![CDATA[C7H6O]]> 106.0465 <![CDATA[[M] + ]]> STD 14 aldehyde phenylacetaldehyde 122-78-1 <![CDATA[C8H8O]]> 120.0593 <![CDATA[[M] + ]]> STD 15 aldehyde trans-2-nonanal 18829-56-6 <![CDATA[C9H 16 O]]> 140.1169 <![CDATA[[M-H2O] + ]]> STD 16 aldehyde Nononal 124-19-6 <![CDATA[C9H 18 O]]> 124.1209 <![CDATA[[M-H2O] + ]]> STD 17 aldehyde trans-citral 5392-40-5 <![CDATA[C 10 H 16 O]]> 152.1167 <![CDATA[[M] + ]]> STD 18 aldehyde trans-2-,cis-6-nonadienal 557-48-2 <![CDATA[C9H 14 O]]> 65.0159 <![CDATA[[C4H1O2] + ]]> STD 19 aldehyde E-2-Octenal 2548-87-0 <![CDATA[C8H 14 O]]> 126.1036 <![CDATA[[M] + ]]> STD 20 acid Germ acid 459-80-3 <![CDATA[C 10 H 16 O2]]> 43.0181 <![CDATA[[C2H3O] + ]]> STD 21 acid Isovalerate 503-74-2 <![CDATA[C5H 10 O2]]> 102.0698 <![CDATA[[M] + ]]> STD 22 ketone (E)-β-ionone 79-77-6 <![CDATA[C 13 H 20 O]]> 192.1506 <![CDATA[[M] + ]]> STD 23 Heterocyclic 2-Ethylfuran 3208-16-0 <![CDATA[C6H8O]]> 96.0593 <![CDATA[[M] + ]]> STD 24 Heterocyclic Linalool Oxide II (Furan Form) 1365-19-1 <![CDATA[C 10 H 18 O2]]> 94.0763 <![CDATA[[C7H 10 ] + ]]> STD 25 ester 3-Hexenyl hexanoate 31501-11-8 <![CDATA[C 12 H 22 O2]]> 82.0781 <![CDATA[[C6H 10 ] + ]]> STD 26 ester Methyl salicylate 119-36-8 <![CDATA[C8H8O3]]> 152.0484 <![CDATA[[M] + ]]> STD 27 alcohol 1-Penten-3-ol 616-25-1 <![CDATA[C5H 10 O]]> 86.0715 <![CDATA[[M] + ]]> MS 28 alcohol trans-coniferol 32811-40-8 <![CDATA[C 10 H 12 O3]]> 180.0851 <![CDATA[[M] + ]]> MS 29 aldehyde trans-2-pentenal 1576-87-0 <![CDATA[C5H8O]]> 84.0558 <![CDATA[[M] + ]]> MS 30 aldehyde trans-2-trans-4-heptadecenal 2363-88-4 <![CDATA[C 10 H 16 O]]> 110.0738 <![CDATA[[M] + ]]> MS 31 acid trans-2-hexenoic acid 13419-69-7 <![CDATA[C6H 10 O2]]> 114.068 <![CDATA[[M] + ]]> MS 32 acid 3-Methylvalerate 105-43-1 <![CDATA[C6H 12 O2]]> 116.0836 <![CDATA[[M] + ]]> MS 33 acid nonanoic acid 112-05-0 <![CDATA[C9H 18 O2]]> 158.1228 <![CDATA[[M] + ]]> MS 34 Organic sulfur dimethyl sulfide 75-18-3 <![CDATA[C2H6S]]> 62.0185 <![CDATA[[M] + ]]> MS 35 ester 2-Ethylbutyrate Allyl 7493-69-8 <![CDATA[C9H 16 O2]]> 156.119 <![CDATA[[M] + ]]> MS 36 ester Geraniol formate 105-86-2 <![CDATA[C 11 H 18 O2]]> 182.1374 <![CDATA[[M] + ]]> MS 37 ester (Z)-3-hexenyl valerate 35852-46-1 <![CDATA[C 11 H 20 O2]]> 184.1507 <![CDATA[[M] + ]]> MS 38 ester γ-Dodecalactone 2305-05-7 <![CDATA[C 12 H 22 O2]]> 198.1655 <![CDATA[[M] + ]]> MS

[0060] S1.3 preprocesses the raw mass spectrometry data, including simple noise filtering, mass calibration, and peak shape fitting (Gaussian fitting). Based on this, feature extraction is performed to obtain the mass-to-charge ratio of VOC ions and their corresponding mass spectrometry response intensity.

[0061] The data is summed, normalized, and filtered to obtain a data table that can be used for statistical analysis.

[0062] S1.4 Establish a mixed screening mechanism to screen combinations of VOC markers corresponding to different aroma types.

[0063] The mixed screening mechanism uses statistical analysis based on data combinations of moisture change rate, aroma attributes and VOC relative intensity information to screen for VOC marker combinations corresponding to different aroma types according to a set threshold.

[0064] Specifically, the analysis of the rate of change in moisture content refers to using m / z = 18.01 (H2O) + The slope of the trend is divided into time periods. The rising period is defined as the slope being greater than 15% of the maximum slope, and a baseline intensity check is used to exclude early smoothing artifacts. The stable period is defined as the absolute value of the slope being less than 2% of the maximum slope. The falling period is defined as the slope being less than -10% of the maximum slope. The sampled data are classified into stable period I, rising period, stable period II, falling period and stable period III.

[0065] The analysis of aroma quality characteristics specifically involves classifying the aroma attributes of dry tea using intensity scores obtained from quantitative sensory descriptive analysis (QDA). Here, it is proposed that the calibration sample also includes a control group. The calibration sample of the control group is dried at a preset experimental temperature, and the VOCs generated during the treatment of the control group are used as control samples at a preset frequency f2. The expert evaluation data of the control samples are obtained accordingly to guide the mixed screening mechanism; f1 > f2.

[0066] During implementation, the preset frequency f2 is to acquire the released VOC every 4 minutes.

[0067] That is, sensory evaluation of the aroma characteristics of samples during the drying process of black tea is conducted. In this embodiment, an evaluation team composed of at least 5 reviewers with rich sensory experience conducts olfactory evaluation of control samples while HPPI-TOFMS is used to detect volatiles during the drying process of tea. All samples are coded with three-digit random numbers and provided to the reviewers in random order. Each sample is smelled at least 3 times, and the perceived aroma descriptive words are recorded. The characteristic aroma of each sample is determined by statistically analyzing the most frequently occurring descriptive words. Based on this, the aroma intensity is graded on a 5-point scale, where 0 indicates no such aroma, 1 indicates weak, 2 indicates slightly weak, 3 indicates medium, 4 indicates slightly strong, and 5 indicates strong. The reviewers score according to the characteristic aroma descriptive words, thereby obtaining supplementary sensory data on the aroma characteristics of black tea at different stages of the drying process.

[0068] Given that the sensory perception of aroma change points and the water peak dynamics inflection point are highly consistent, this embodiment uses objective water peak data to assist sensory perception data in order to reduce human subjective interference. Finally, a two-dimensional aroma grading standard based on moisture dynamics and sensory attributes was established: 0-3 min mild stage, 4-11 min sweet stage, 12-15 min sweet-caramel mixed stage, 15-20 min caramel stage, and >20 min high-heat stage.

[0069] Based on the above, the statistical analysis includes univariate analysis and multivariate analysis, the latter including unsupervised analysis and supervised analysis:

[0070] Unsupervised analysis employed principal component analysis (PCA), which reflects the original clustering and separation of VOCs. Samples were grouped based on moisture change rate and aroma attributes, such as... Figure 3 As shown;

[0071] like Figure 4 As shown, the supervised analysis employed orthogonal partial least squares (PLS-DA) discriminant analysis. It can be observed that during the drying process of black tea, the volatile compounds in the tea leaves continuously change, and samples of different aroma types exhibit corresponding discretization and clustering. In the PLS-DA model, R0... 2 Y=0.938, Q 2 =0.876, indicating that the model is reliable and has excellent predictive ability; further, 200 permutation tests were conducted, and its Q... 2 =-0.573 indicates that the PLS-DA model has not overfitted; it is used to screen for conditions that meet the criteria, such as relative intensity greater than the preset value, generally VIP>1; at the same time, univariate analysis is performed by importing the obtained data table containing the relative intensity information of all ions into SPSS, and performing nonparametric tests to evaluate its statistical significance.

[0072] VOC markers were selected by combining p-value < 0.05 and VIP value;

[0073] In this embodiment, nine substances with p<0.05 and VIP>1 were screened out, including geraniol, (E,E)-2,4-nonadienal, trans-2-hexenal, 1-octanol, phenylacetaldehyde, phenylethanol, β-linalool, trans-2-nonal, and trans-citral.

[0074] Furthermore, such as Figure 5 As shown, a random forest (RF) model was constructed based on the VOC biomarker combinations associated with aroma obtained through screening as an aroma discrimination model, which was used to discriminate the aroma transformation during the drying process of black tea.

[0075] Specifically, the raw peak area data (relative intensity) of the samples obtained from two parallel drying experiments were used as input, with 2 / 3 of the samples serving as the training set and 1 / 3 serving as the test set. The RF model constructed based on this achieved a discrimination accuracy of 100% on the test set, and the confusion matrix was used for visualization. Its out-of-bag (OOB) cross-validation score was 98.25%.

[0076] The results show that the model has good robustness and generalization ability, and there is no risk of overfitting.

[0077] By ranking the importance of each compound in the model, the results showed that β-linalool had the highest contribution, at 0.2017, indicating that it plays a key role in the aroma transformation during the drying process of black tea.

[0078] In practical applications, the VOCs generated by the sample during the drying process are identified by measuring the model online, and the aroma classification at each time point is obtained by screening the combination of VOC markers.

[0079] Overall, the nine differential compounds identified in this invention can effectively distinguish different stages of the black tea drying process, and their discriminative value for aroma changes is verified by a random forest model.

[0080] This invention also relates to the application of an in-situ real-time measurement method for the dynamic changes in aroma during the drying process of black tea, which can be used to distinguish different stages of aroma transformation during the final drying of black tea.

[0081] like Figure 2 As shown, the present invention also relates to a measuring device employing the aforementioned method for in-situ real-time determination of the dynamic changes in aroma during the drying process of black tea, comprising:

[0082] A heating and drying device 1 is used to dry the sample inside it at a preset temperature.

[0083] An analyzer 2, equipped with HPPI-TOFMS, is spatially connected to the heating and drying equipment 1 via a stainless steel capillary tube 3.

[0084] In this invention, 200g of black tea sample after initial roasting is evenly spread on a plate; the sample is placed in a heating and drying device 1 equipped with a heating tube and a built-in fan, such as a small household oven, and dried at a preset temperature. This device can simulate the hot air drying principle in actual production and ensure hot air circulation; in order to capture the changes in aroma substances during the drying process to the greatest extent, this embodiment sets the drying temperature to 130℃ and the drying time to 40min.

[0085] While the sample is placed into the heating and drying equipment 1, the stainless steel capillary tube 3 is inserted into the heating and drying equipment 1 through the heat insulation tube to form a relatively closed sampling space. At the same time, the other end is connected to the analyzer 2 loaded with HPPI-TOFMS to realize the direct acquisition of aroma signal.

[0086] Aroma signals were collected every minute during the drying process, resulting in 41 sets of continuous aroma data.

[0087] To ensure the stability and reproducibility of the data, the above experiment was repeated twice, and the results showed that continuous, rapid and real-time monitoring of aroma substances during the drying process of black tea can be achieved.

[0088] In practical applications, analyzer 2 is a mass spectrometry system (model PI-TOFMSR5020), including a high-voltage photoionization source (using a vacuum ultraviolet lamp), an ion transport module, and an orthogonal accelerated time-of-flight mass spectrometer. The key voltage parameters are: microchannel plate detector voltage 4000V, extraction voltage 18V, focusing voltage 16V, and cone voltage 10V. Under standard operating conditions, the mass resolution at m / z 108 is approximately 6000 (FWHM), and it is equipped with a 0.5m long field-free drift tube. The gas sample is introduced into the ion source through a 250μm inner diameter, 30cm long stainless steel capillary 3 at a flow rate of 50 mL / min, and one mass spectrum is acquired every minute to monitor the dynamic changes of aroma substances during the drying process.

[0089] Those skilled in the art will understand that the multi-level classification model and its applications mentioned in the embodiments of the present invention can be provided as a method, system, or computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0090] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0091] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0092] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0093] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0094] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for in-situ real-time determination of the dynamic changes in aroma during the drying process of black tea, characterized in that: The method obtains calibration samples, monitors VOCs generated by the samples during the drying process online, and screens for VOC biomarker combinations associated with aroma type; Construct an aroma discrimination model based on the VOC marker combination; A sample to be tested is obtained and dried. During the drying process, the VOC generated in real time by the sample is input into the aroma discrimination model to obtain the corresponding aroma.

2. The in-situ real-time measurement method for dynamic changes in aroma during the drying process of black tea according to claim 1, characterized in that: The VOC markers associated with the aroma are geraniol, (E,E)-2,4-nonadienal, trans-2-hexenal, 1-octanol, phenylacetaldehyde, phenylethanol, β-linalool, trans-2-nonanal, and trans-citral.

3. The in-situ real-time measurement method for dynamic changes in aroma during the drying process of black tea according to claim 1, characterized in that: The calibration sample is a black tea sample after initial roasting, with a moisture content of 20% to 25%.

4. The in-situ real-time measurement method for dynamic changes in aroma during the drying process of black tea according to claim 1, characterized in that: The process of identifying VOC biomarker combinations associated with aroma profiles includes the following steps: S1.1 The calibration samples include the analytical group. The calibration samples of the analytical group are dried at a preset experimental temperature and subjected to HPPI-TOFMS at a preset frequency. f 1. Online monitoring and analysis of VOCs generated during processing and real-time acquisition of raw mass spectra; S1.2 Based on standards, perform compound characterization of VOCs detected by HPPI-TOFMS; S1.3 preprocesses the raw mass spectrometry data, extracts features, and obtains the mass-to-charge ratio of VOC ions and their corresponding mass spectrometry response intensity. S1.4 Establish a mixed screening mechanism to screen combinations of VOC markers corresponding to different aroma types.

5. The in-situ real-time method for measuring the dynamic changes in aroma during the drying process of black tea according to claim 4, characterized in that: The calibration samples also include a control group, wherein the calibration samples of the control group are dried at a preset experimental temperature and at a preset frequency. f 2. VOCs generated during the treatment of the control group were used as control samples, and expert evaluation data of the control samples were obtained to guide the mixed screening mechanism. f 1> f 2.

6. The in-situ real-time determination method for dynamic changes in aroma during the drying process of black tea according to claim 4, characterized in that: In S1.4, the mixed screening mechanism performs statistical analysis based on the combination of data on moisture change rate, aroma attributes and VOC relative intensity information, and selects VOC marker combinations corresponding to different aroma types according to the set threshold.

7. The in-situ real-time method for measuring the dynamic changes in aroma during the drying process of black tea according to claim 6, characterized in that: The statistical analysis includes univariate analysis and multivariate analysis. The multivariate analysis includes unsupervised analysis and supervised analysis. The unsupervised analysis combines the moisture change rate and aroma attributes to group the samples, while the supervised analysis is used to screen VOC markers that meet the criteria.

8. The in-situ real-time method for measuring the dynamic changes in aroma during the drying process of black tea according to claim 1, characterized in that: A random forest model was constructed based on the VOC biomarker combinations associated with aroma obtained through screening, which was used as an aroma discrimination model to identify aroma changes during the drying process of black tea.

9. An application of the in-situ real-time measurement method for the dynamic changes in aroma during the drying process of black tea as described in any one of claims 1 to 8, characterized in that: It is used to distinguish the different stages of aroma transformation during the final drying of black tea.

10. A measuring device employing the in-situ real-time measurement method for the dynamic changes in aroma during the drying process of black tea as described in any one of claims 1 to 8, characterized in that: include: A heating and drying device used to dry samples at a preset temperature; An analyzer, equipped with HPPI-TOFMS, is spatially connected to the heating and drying equipment via a stainless steel capillary tube.

Images