Modeling method and detection method for near-infrared quantitative analysis of stevia rebaudiana components

Abstract

The invention belongs to the technical field of component detection, and particularly relates to a modeling method and a detection method for near-infrared quantitative analysis of stevia rebaudiana components. According to the method, stevioside, rebaudioside A, chlorogenic acid, cryptochlorogenic acid, neochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B or isochlorogenic acid C are extracted through a heating reflux synchronous extraction technology, and compared with ultrasonic extraction, the extraction efficiency of the heating reflux synchronous extraction technology for extracting stevia rebaudiana powder samples with the same mass is higher; the real content of the components in the stevia rebaudiana can be reflected. According to the method, a stable and accurate analysis model is established on the basis of the near infrared spectrum technology, then quantitative analysis on stevioside, rebaudioside A, chlorogenic acid, cryptochlorogenic acid, neochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B or isochlorogenic acid C in stevia rebaudiana is achieved, the detection speed is high, and the analysis precision is high. Compared with HPLC (High Performance Liquid Chromatography) detection, the detection method provided by the invention does not need to use a large amount of organic solvents and avoids environmental pollution.

Description

Technical Field

[0001] The present invention belongs to the technical field of component detection, and in particular relates to a modeling method and a detection method for near-infrared quantitative analysis of stevia components. Background Art

[0002] Stevia is a naturally sweet plant whose leaves are rich in natural products with significant physiological activity. The main functional components include steviol glycosides and chlorogenic acid compounds, specifically stevioside (STV, a diterpenoid compound), rebaudioside A (RA), chlorogenic acid (CGA), cryptochlorogenic acid (CCGA), neochlorogenic acid (NCGA, a phenolic compound), isochlorogenic acid A (ICAA), isochlorogenic acid B (ICAB) and isochlorogenic acid C (ICAC). The structures are shown in Formula I to Formula VIII, respectively:

[0003]

[0004] In order to control the quality of stevia, its main functional components are usually quantitatively analyzed. The commonly used analysis method is high performance liquid chromatography (HPLC), which has high accuracy and separation ability, but a long detection cycle.

[0005] Near-infrared spectroscopy (NIR) technology is a rapid detection method based on the principle of molecular vibration and spectral absorption. It has the advantages of being fast, non-destructive, requiring no sample pretreatment, and capable of simultaneous detection of multiple components. The NIR band mainly covers 4000-12800cm -1 , capable of capturing the absorption characteristics of functional groups such as CH, OH, and NH in organic matter. The use of near-infrared spectroscopy for the determination of steviol glycosides and chlorogenic acid has been reported, but only the total steviol glycosides or total chlorogenic acid have been measured. Summary of the Invention

[0006] The object of the present invention is to provide a modeling method and a detection method for near-infrared quantitative analysis of stevia components. The detection method provided by the present invention can realize the rapid and accurate detection of stevioside, rebaudioside A, chlorogenic acid, cryptochlorogenic acid, neochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B or isochlorogenic acid C in stevia.

[0007] In order to achieve the above object, the present invention provides the following technical solutions:

[0008] The present invention provides a modeling method for near-infrared quantitative analysis of stevia components, comprising the following steps:

[0009] (1) crushing and sieving the stevia sample in sequence to obtain a stevia powder sample;

[0010] (2) performing high performance liquid chromatography detection on one or more of stevioside, rebaudioside A, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B and isochlorogenic acid C in the stevia powder sample to obtain the content of the target component; the preparation method of the test solution for high performance liquid chromatography detection comprises the following steps: mixing the stevia powder sample with a first ethanol solution, performing a first extraction, and then performing a first solid-liquid separation to obtain a first solid phase and a first liquid phase; mixing the first solid phase with a second ethanol solution, performing a second extraction, and then performing a second solid-liquid separation to obtain a second solid phase and a second liquid phase; mixing the first liquid phase and the second liquid phase and removing the solvent to obtain an extract; mixing the extract with a chromatographic methanol aqueous solution to obtain a test solution;

[0011] (3) performing near-infrared spectroscopy analysis on the target component in the stevia powder sample to obtain near-infrared spectral data of the target component;

[0012] (4) Preprocessing the near-infrared spectrum data of the target component to select the characteristic wavelength corresponding to the target component; the characteristic wavelength of stevioside is 6248.5-7143.4 cm -1 , the dimension is 6, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of rebaudioside A is 6248.5~7143.4cm -1 and 7691.1~9091.2cm -1 , the dimension is 4, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of the new chlorogenic acid is 3999.8~5303.5cm -1 and 5936.1~6248.5cm -1 , the dimension is 5, and the second-order derivative is used to preprocess the near-infrared spectral data; the characteristic wavelength of chlorogenic acid is 3999.8~5303.5cm -1 , the dimension is 5, and the second-order derivative is used to preprocess the near-infrared spectral data; the characteristic wavelength of the cryptochlorogenic acid is 3999.8~5303.5cm -1 , the dimension is 7, and the near-infrared spectral data are preprocessed by eliminating the constant offset; the characteristic wavelength of the isochlorogenic acid B is 3999.8~5303.5cm -1 and 5623.7~6248.5cm -1 , the dimension is 4, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of the isochlorogenic acid A is 5623.7~6248.5cm -1 , the dimension is 4, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of the isochlorogenic acid C is 3999.8~5303.5cm -1and 5623.7~6248.5cm -1 , the dimension is 3, and the first-order derivative + vector normalization is used to preprocess the near-infrared spectral data;

[0013] (5) Based on the Quant 2 module of OPUS 7.0 software, a near-infrared quantitative calibration model of the target component of stevia was established. The characteristic wavelength of the target component determined in step (4) was used, and the partial least squares method was used to fit the near-infrared spectral data with the content of the target component to obtain a calibration model for near-infrared quantitative analysis of stevia components.

[0014] Preferably, the first ethanol solution is an ethanol-water solution; the volume content of the first ethanol solution is 30%; and the ratio of the mass of the stevia powder sample to the volume of the first ethanol solution is 1 g:8 mL.

[0015] Preferably, the temperature of the first extraction is 60° C. and the insulation time is 1 hour.

[0016] Preferably, the second ethanol solution is an ethanol-water solution; the volume content of the second ethanol solution is 30%; and the ratio of the mass of the stevia powder sample to the volume of the second ethanol solution is 1 g:6 mL.

[0017] Preferably, the temperature of the second extraction is 60° C. and the insulation time is 0.5 h.

[0018] Preferably, the solvent removal is performed by rotary evaporation; the temperature of the rotary evaporation is 30 to 50° C., and the insulation time is 10 to 40 minutes.

[0019] Preferably, the volume fraction of methanol in the chromatographic methanol aqueous solution is 10-70%; and the concentration of the extract in the test solution is 1-20 g / L.

[0020] Preferably, the chromatographic column used for the high performance liquid chromatography detection of stevioside is HPLC COLUMNNH2; the mobile phase of the chromatographic column includes acetonitrile and water; the volume ratio of acetonitrile to water is 80:20; the flow rate of the mobile phase is 1.2 mL / min; the detection wavelength of the high performance liquid chromatography detection is 210 nm, and the injection volume is 10 μL; the parameters of the high performance liquid chromatography detection of rebaudioside A are the same as those of stevioside; the chromatographic column used for the high performance liquid chromatography detection of the neochlorogenic acid is Shim-pack-GIST C18; the mobile phase of the chromatographic column includes acetic acid aqueous solution and acetonitrile; the volume content of acetic acid in the acetic acid aqueous solution is 1%; the high performance liquid chromatography detection of the neochlorogenic acid is gradient elution, and the volume ratio of the acetic acid aqueous solution and acetonitrile is shown in Table 1; the flow rate of the mobile phase is 1.0 mL / min; the detection wavelength of the high performance liquid chromatography detection is 330 nm, and the injection volume is 10 μL;

[0021] Table 1 Gradient elution parameters for HPLC detection of neochlorogenic acid

[0022] time Acetic acid aqueous solution Acetonitrile 0~15min 92～83％ 8～17％ 15-30 minutes 83～82％ 17～18％ 30-40 minutes 82～81％ 18～19％ 40-42 minutes 81～76％ 19～24％ 42-50 minutes 76～60％ 24～40％ 50-65 minutes 60～59.8％ 40～40.2％

[0023] The parameters for high performance liquid chromatography detection of the chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B and isochlorogenic acid C are the same as those for neochlorogenic acid.

[0024] Preferably, the near-infrared spectroscopy analysis adopts an integrating sphere diffuse reflectance method with a sampling range of 4000 to 10000 cm -1 , resolution of 8.0~64.0cm -1 , the number of scans is 16 to 64 times, and the number of acquisitions is 3 to 6 times.

[0025] The present invention also provides a method for near-infrared quantitative analysis of stevia components, comprising the following steps:

[0026] (1) crushing and sieving the stevia sample in sequence to obtain a stevia powder sample;

[0027] (2) performing high performance liquid chromatography detection on one or more of stevioside, rebaudioside A, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B and isochlorogenic acid C in the stevia powder sample to obtain the content of the target component; the preparation method of the test solution for high performance liquid chromatography detection comprises the following steps: mixing the stevia powder sample with a first ethanol solution, performing a first extraction, and then performing a first solid-liquid separation to obtain a first solid phase and a first liquid phase; mixing the first solid phase with a second ethanol solution, performing a second extraction, and then performing a second solid-liquid separation to obtain a second solid phase and a second liquid phase; mixing the first liquid phase and the second liquid phase and removing the solvent to obtain an extract; mixing the extract with a chromatographic methanol aqueous solution to obtain a test solution;

[0028] (3) performing near-infrared spectroscopy analysis on the target component in the stevia powder sample to obtain near-infrared spectral data of the target component;

[0029] (4) Preprocessing the near-infrared spectrum data of the target component to select the characteristic wavelength corresponding to the target component; the characteristic wavelength of stevioside is 6248.5-7143.4 cm -1 , the dimension is 6, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of rebaudioside A is 6248.5~7143.4cm -1 and 7691.1~9091.2cm -1 , the dimension is 4, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of the new chlorogenic acid is 3999.8~5303.5cm -1 and 5936.1~6248.5cm -1 , the dimension is 5, and the second-order derivative is used to preprocess the near-infrared spectral data; the characteristic wavelength of chlorogenic acid is 3999.8~5303.5cm -1 , the dimension is 5, and the second-order derivative is used to preprocess the near-infrared spectral data; the characteristic wavelength of the cryptochlorogenic acid is 3999.8~5303.5cm -1 , the dimension is 7, and the near-infrared spectral data are preprocessed by eliminating the constant offset; the characteristic wavelength of the isochlorogenic acid B is 3999.8~5303.5cm -1 and 5623.7~6248.5cm -1 , the dimension is 4, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of the isochlorogenic acid A is 5623.7~6248.5cm -1 , the dimension is 4, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of the isochlorogenic acid C is 3999.8~5303.5cm -1 and 5623.7~6248.5cm -1 , the dimension is 3, and the first-order derivative + vector normalization is used to preprocess the near-infrared spectral data;

[0030] (5) Based on the Quant 2 module of OPUS 7.0 software, a near-infrared quantitative calibration model of the target component of stevia was established. The characteristic wavelength of the target component determined in step (4) was used, and the partial least squares method was used to fit the near-infrared spectral data with the content of the target component to obtain a calibration model for near-infrared quantitative analysis of stevia components;

[0031] (6) The stevia to be tested is successively crushed and sieved to obtain a stevia powder to be tested, and a target component in the stevia powder to be tested is analyzed by near-infrared spectroscopy to obtain near-infrared spectral data of the target component. The component content of the stevia is obtained according to a calibration model for near-infrared quantitative analysis of stevia components.

[0032] The present invention provides a modeling method for near-infrared quantitative analysis of stevia components. The present invention uses a heated reflux synchronous extraction technique to extract stevioside, rebaudioside A, chlorogenic acid, cryptochlorogenic acid, neochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B, or isochlorogenic acid C. Compared to ultrasonic extraction, the heated reflux synchronous extraction technique has higher extraction efficiency for the same mass of stevia powder sample and can better reflect the true content of the above-mentioned components in stevia. Based on near-infrared spectroscopy technology, the present invention establishes a stable and accurate analytical model, thereby achieving quantitative analysis of stevioside, rebaudioside A, chlorogenic acid, cryptochlorogenic acid, neochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B, or isochlorogenic acid C in stevia, with rapid detection speed and high analytical accuracy.

[0033] The present invention provides a method for near-infrared quantitative analysis of stevia components. The present invention combines near-infrared spectroscopy with PLS modeling to achieve rapid and accurate detection of two steviol glycosides (stevioside and rebaudioside A) and six chlorogenic acid compounds (chlorogenic acid, cryptochlorogenic acid, neochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B, and isochlorogenic acid C) in stevia, with simple operation and low cost. The present invention can achieve rapid batch analysis through modeling and is suitable for various application scenarios such as online detection or on-site quality control; the detection method provided by the present invention can accurately identify component differences caused by different strains and different cultivation methods, providing a scientific basis and technical support for the establishment of production standard procedures for stevia raw materials; compared with HPLC detection, the detection method provided by the present invention does not require the use of large amounts of organic solvents, thus avoiding pollution to the environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0035] Figure 1 This is a flow chart of the modeling method for near-infrared quantitative analysis of stevia components of the present invention;

[0036] Figure 2 The HPLC chromatogram of stevioside and rebaudioside A of the present invention is as follows;

[0037] Figure 3is a high performance liquid chromatogram of chlorogenic acid, cryptochlorogenic acid, neochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B and isochlorogenic acid C of the present invention;

[0038] Figure 4 This is the near infrared spectrum of stevia of the present invention;

[0039] Figure 5 This is a cross-validation result diagram for stevioside of the present invention;

[0040] Figure 6 This is a cross-validation result diagram for rebaudioside A of the present invention;

[0041] Figure 7 This is a cross-validation result diagram of the present invention for new chlorogenic acid;

[0042] Figure 8 This is a cross-validation result diagram of chlorogenic acid according to the present invention;

[0043] Figure 9 This is a cross-validation result diagram of the present invention on cryptochlorogenic acid;

[0044] Figure 10 This is a cross-validation result diagram of the present invention on isochlorogenic acid B;

[0045] Figure 11 This is a cross-validation result diagram of the present invention for isochlorogenic acid A;

[0046] Figure 12 This is a cross-validation result diagram of the present invention for isochlorogenic acid C. DETAILED DESCRIPTION

[0047] The present invention provides a modeling method for near-infrared quantitative analysis of stevia components, comprising the following steps:

[0048] (1) crushing and sieving the stevia sample in sequence to obtain a stevia powder sample;

[0049] (2) performing high performance liquid chromatography on one or more of stevioside, rebaudioside A, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B, and isochlorogenic acid C in the stevia powder sample to obtain the content of the target component;

[0050] (3) performing near-infrared spectroscopy analysis on the target component in the stevia powder sample to obtain near-infrared spectral data of the target component;

[0051] (4) Preprocessing the near-infrared spectrum data of the target component to select the characteristic wavelength corresponding to the target component; the characteristic wavelength of stevioside is 6248.5-7143.4 cm -1, the dimension is 6, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of rebaudioside A is 6248.5~7143.4cm -1 and 7691.1~9091.2cm -1 , the dimension is 4, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of the new chlorogenic acid is 3999.8~5303.5cm -1 and 5936.1~6248.5cm -1 , the dimension is 5, and the second-order derivative is used to preprocess the near-infrared spectral data; the characteristic wavelength of chlorogenic acid is 3999.8~5303.5cm -1 , the dimension is 5, and the second-order derivative is used to preprocess the near-infrared spectral data; the characteristic wavelength of the cryptochlorogenic acid is 3999.8~5303.5cm -1 , the dimension is 7, and the near-infrared spectral data are preprocessed by eliminating the constant offset; the characteristic wavelength of the isochlorogenic acid B is 3999.8~5303.5cm -1 and 5623.7~6248.5cm -1 , the dimension is 4, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of the isochlorogenic acid A is 5623.7~6248.5cm -1 , the dimension is 4, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of the isochlorogenic acid C is 3999.8~5303.5cm -1 and 5623.7~6248.5cm -1 , the dimension is 3, and the first-order derivative + vector normalization is used to preprocess the near-infrared spectral data;

[0052] (5) Based on the Quant 2 module of OPUS 7.0 software, a near-infrared quantitative calibration model of the target component of stevia was established. The characteristic wavelength of the target component determined in step (4) was used, and the partial least squares method was used to fit the near-infrared spectral data with the content of the target component to obtain a calibration model for near-infrared quantitative analysis of stevia components.

[0053] The process of the modeling method for near infrared quantitative analysis of stevia components of the present invention is as follows: Figure 1 As shown. The present invention sequentially crushes and sieves a stevia sample to obtain a stevia powder sample. In the present invention, the stevia sample can be representative stevia; the sieve used for sieving can have an aperture of no greater than 40 mesh, specifically 40 mesh, 50 mesh, or 60 mesh. In a specific embodiment of the present invention, representative stevia collected from different farmers and plots of land by Xinjiang Huijia Biotechnology Co., Ltd. was used.

[0054] After obtaining the stevia powder sample, the present invention performs high performance liquid chromatography detection on one or more of stevioside, rebaudioside A, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B and isochlorogenic acid C in the stevia powder sample to obtain the content of the target component.

[0055] In the present invention, the preparation method of the reference solution for high performance liquid chromatography detection may include the following steps: taking a reference substance of stevioside, rebaudioside A, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B or isochlorogenic acid C and mixing them with a chromatographic methanol aqueous solution to obtain a reference substance solution; the volume fraction of methanol in the chromatographic methanol aqueous solution may be 10 to 70%, specifically 30%; the concentration of the reference substance in the reference substance solution may be 1 to 20 g / L, specifically 10 g / L.

[0056] In the present invention, the test solution for high performance liquid chromatography detection is prepared by a heating reflux synchronous extraction method, and stevioside, rebaudioside A, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B and isochlorogenic acid C are extracted simultaneously; the preparation method of the test solution may include the following steps: mixing the stevia powder sample with a first ethanol solution for a first extraction and then performing a first solid-liquid separation to obtain a first solid phase and a first liquid phase; mixing the first solid phase with a second ethanol solution for a second extraction and then performing a second solid-liquid separation to obtain a second solid phase and a second liquid phase; mixing the first liquid phase and the second liquid phase and removing the solvent to obtain an extract; mixing the extract with a chromatographic methanol aqueous solution to obtain a test solution.

[0057] In the present invention, the first ethanol solution may be an ethanol-water solution; the volume content of the first ethanol solution may be 30%; the ratio of the mass of the stevia powder sample to the volume of the first ethanol solution may be 1 g:8 mL; the temperature of the first extraction may be 60° C., and the insulation time may be 1 h; cooling may be further included before the first solid-liquid separation; the cooling may be natural cooling; the first solid-liquid separation may be suction filtration; and the pressure of the suction filtration may be -0.1 to -0.08 MPa.

[0058] In the present invention, the second ethanol solution may be an ethanol-water solution; the volume content of the second ethanol solution may be 30%; the ratio of the mass of the stevia powder sample to the volume of the second ethanol solution may be 1 g:6 mL; the temperature of the second extraction may be 60° C., and the insulation time may be 0.5 h; cooling may be further included before the second solid-liquid separation; the cooling may be natural cooling; the second solid-liquid separation may be suction filtration; and the pressure of the suction filtration may be -0.1 to -0.08 MPa.

[0059] In the present invention, the solvent removal can be rotary evaporation; the temperature of the rotary evaporation can be 30-50°C, specifically 40°C, and the insulation time can be 10-40 minutes, specifically 20 minutes or 30 minutes; the extract can be placed in a brown penicillin bottle and stored away from light.

[0060] In the present invention, the volume fraction of methanol in the chromatographic methanol aqueous solution can be 10-70%, specifically 30%; the extract concentration in the test solution can be 1-20 g / L, specifically 10 g / L.

[0061] In the present invention, the chromatographic column used for the high performance liquid chromatography detection of stevioside can be HPLCCOLUMN NH2; the column length of the chromatographic column can be 250 mm, the inner diameter can be 4.6 mm, the filler particle size can be 5 μm, and the column temperature can be 25-35°C, specifically 30°C; the mobile phase of the chromatographic column can include acetonitrile and water; the water can be pure water; the volume ratio of acetonitrile and water can be 80:20; the flow rate of the mobile phase can be 1.2 mL / min; the detection wavelength of the high performance liquid chromatography detection can be 210 nm, and the injection volume can be 10 μL.

[0062] In the present invention, the parameters for HPLC detection of rebaudioside A may be the same as those for stevioside, which will not be described in detail here.

[0063] In the present invention, the chromatographic column used for the high performance liquid chromatography detection of the neochlorogenic acid may be Shim-pack-GIST C18; the column length of the chromatographic column may be 250 mm, the inner diameter may be 4.6 mm, the filler particle size may be 5 μm, and the column temperature may be 25-35°C, specifically 30°C; the mobile phase of the chromatographic column may include aqueous acetic acid and acetonitrile; the volume content of acetic acid in the aqueous acetic acid may be 1%; the high performance liquid chromatography detection of the neochlorogenic acid may be gradient elution, and the volume ratio of the aqueous acetic acid and acetonitrile may be as shown in Table 1; the flow rate of the mobile phase may be 1.0 mL / min; the detection wavelength of the high performance liquid chromatography detection may be 330 nm, and the injection volume may be 10 μL.

[0064] Table 1 Gradient elution parameters for HPLC detection of neochlorogenic acid

[0065] time Acetic acid aqueous solution Acetonitrile 0~15min 92～83％ 8～17％ 15-30 minutes 83～82％ 17～18％ 30-40 minutes 82～81％ 18～19％ 40-42 minutes 81～76％ 19～24％ 42-50 minutes 76～60％ 24～40％ 50-65 minutes 60～59.8％ 40～40.2％

[0066] In the present invention, the parameters for HPLC detection of chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B and isochlorogenic acid C may be the same as those for neochlorogenic acid, and will not be repeated here.

[0067] After obtaining a stevia powder sample, the present invention performs near-infrared spectroscopy analysis on a target component in the stevia powder sample to obtain near-infrared spectral data of the target component. In the present invention, the near-infrared spectroscopy analysis equipment can be a near-infrared spectrometer; the near-infrared spectrometer can be a German Bruker MPA Fourier transform near-infrared spectrometer; the near-infrared spectroscopy analysis can use an integrating sphere diffuse reflectance method, and the sampling range can be 4000 to 10000 cm -1 , the resolution can be 8.0~64.0cm -1 , specifically 8.0cm -1 、16.0cm -1 、32.0cm -1 or 64.0cm -1 The number of scans may be 16 to 64 times, specifically 16, 32 or 64 times, and the number of acquisitions may be 3 to 6 times, specifically 3, 4, 5 or 6 times.

[0068] After obtaining the near-infrared spectrum data of the target component, the present invention preprocesses the near-infrared spectrum data of the target component and selects the characteristic wavelength corresponding to the target component.

[0069] In the present invention, the characteristic wavelength of stevioside is 6248.5~7143.4cm -1 , the dimension is 6, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of rebaudioside A is 6248.5~7143.4cm -1 and 7691.1~9091.2cm -1 , the dimension is 4, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of the new chlorogenic acid is 3999.8~5303.5cm -1 and 5936.1~6248.5cm -1 , the dimension is 5, and the second-order derivative is used to preprocess the near-infrared spectral data; the characteristic wavelength of chlorogenic acid is 3999.8~5303.5cm -1 , the dimension is 5, and the second-order derivative is used to preprocess the near-infrared spectral data; the characteristic wavelength of the cryptochlorogenic acid is 3999.8~5303.5cm -1 , the dimension is 7, and the near-infrared spectral data are preprocessed by eliminating the constant offset; the characteristic wavelength of the isochlorogenic acid B is 3999.8~5303.5cm -1 and 5623.7~6248.5cm -1 , the dimension is 4, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of the isochlorogenic acid A is 5623.7~6248.5cm -1, the dimension is 4, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of the isochlorogenic acid C is 3999.8~5303.5cm -1 and 5623.7~6248.5cm -1 , the dimension is 3, and the first-order derivative + vector normalization is used to preprocess the near-infrared spectral data.

[0070] The present invention establishes a near-infrared quantitative calibration model for the target component of stevia based on the Quant 2 module of OPUS 7.0 software. The characteristic wavelength of the target component determined in step (4) is used to fit the near-infrared spectral data with the content of the target component using the partial least squares method to obtain a calibration model for near-infrared quantitative analysis of stevia components. The present invention uses the partial least squares method, which has good multivariate modeling capabilities even when the sample size is limited.

[0071] The present invention adopts the cross-validation method to evaluate the performance of the calibration model for near-infrared quantitative analysis of stevia components, ensure the reliability and generalizability of the calibration model, and obtain the optimal near-infrared quantitative calibration model.

[0072] In the present invention, the steps of the cross-validation method are as follows: there are N sample data, one of which is used as a validation sample each time, and the remaining N-1 samples are used as training samples. N training and prediction cycles are performed, and finally the average value of the N results is used as the overall evaluation indicator of the model performance.

[0073] In the present invention, the cross-validation index may include the cross-validation determination coefficient (R 2 : Indicates the degree of fit of the model to the actual value of the sample. The closer the value is to 1, the stronger the explanatory power of the model), root mean square error of cross validation (RMSECV: reflects the average level of model prediction error. The smaller the value, the higher the prediction accuracy) and residual prediction deviation (RPD: greater than 2.5 means good prediction ability).

[0074] The calibration model for near-infrared quantitative analysis of stevia components obtained in the present invention has good prediction performance for each target component as shown in Table 2.

[0075] Table 2 Prediction performance of the correction model of the present invention for each target component

[0076] <![CDATA[R 2 ]]> RMSECV RPD Stevioside 0.9023 0.068100 3.20 Rebaudioside A 0.9728 0.075800 6.07 Neochlorogenic acid 0.9829 0.001820 7.65 Chlorogenic acid 0.9931 0.003830 12.0 Cryptochlorogenic acid 0.9653 0.005370 5.37 Isochlorogenic acid A 0.9773 0.015600 6.64 Isochlorogenic acid B 0.9575 0.000724 4.87 Isochlorogenic acid C 0.9262 0.012700 3.68

[0077] The present invention also provides a method for near-infrared quantitative analysis of stevia components, comprising the following steps:

[0078] (1) crushing and sieving the stevia sample in sequence to obtain a stevia powder sample;

[0079] (2) performing high performance liquid chromatography on one or more of stevioside, rebaudioside A, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B, and isochlorogenic acid C in the stevia powder sample to obtain the content of the target component;

[0080] (3) performing near-infrared spectroscopy analysis on the target component in the stevia powder sample to obtain near-infrared spectral data of the target component;

[0081] (4) Preprocessing the near-infrared spectrum data of the target component to select the characteristic wavelength corresponding to the target component; the characteristic wavelength of stevioside is 6248.5-7143.4 cm -1 , the dimension is 6, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of rebaudioside A is 6248.5~7143.4cm -1 and 7691.1~9091.2cm -1 , the dimension is 4, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of the new chlorogenic acid is 3999.8~5303.5cm -1 and 5936.1~6248.5cm -1 , the dimension is 5, and the second-order derivative is used to preprocess the near-infrared spectral data; the characteristic wavelength of chlorogenic acid is 3999.8~5303.5cm -1 , the dimension is 5, and the second-order derivative is used to preprocess the near-infrared spectral data; the characteristic wavelength of the cryptochlorogenic acid is 3999.8~5303.5cm -1 , the dimension is 7, and the near-infrared spectral data are preprocessed by eliminating the constant offset; the characteristic wavelength of the isochlorogenic acid B is 3999.8~5303.5cm -1 and 5623.7~6248.5cm -1 , the dimension is 4, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of the isochlorogenic acid A is 5623.7~6248.5cm -1 , the dimension is 4, and the near-infrared spectral data are preprocessed using linear baseline correction; the characteristic wavelength of the isochlorogenic acid C is 3999.8~5303.5cm -1 and 5623.7~6248.5cm -1 , the dimension is 3, and the first-order derivative + vector normalization is used to preprocess the near-infrared spectral data;

[0082] (5) Based on the Quant 2 module of OPUS 7.0 software, a near-infrared quantitative calibration model of the target component of stevia was established. The characteristic wavelength of the target component determined in step (4) was used, and the partial least squares method was used to fit the near-infrared spectral data with the content of the target component to obtain a calibration model for near-infrared quantitative analysis of stevia components;

[0083] (6) The stevia to be tested is successively crushed and sieved to obtain a stevia powder to be tested, and a target component in the stevia powder to be tested is analyzed by near-infrared spectroscopy to obtain near-infrared spectral data of the target component. The component content of the stevia is obtained according to a calibration model for near-infrared quantitative analysis of stevia components.

[0084] Steps (1) to (5) in the detection method provided by the present invention are consistent with the modeling method and will not be described in detail here.

[0085] In the present invention, the equipment and parameters for crushing, screening and near-infrared spectroscopy analysis in step (6) may be the same as those in steps (1) and (3), and will not be described in detail here.

[0086] In order to further illustrate the present invention, the scheme of the present invention is described in detail below with reference to the accompanying drawings and embodiments, but they should not be understood as limiting the scope of protection of the present invention.

[0087] Example 1

[0088] This example tests the contents of stevioside, rebaudioside A, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B, and isochlorogenic acid C in Stevia:

[0089] This example uses stevia samples collected from the Yili region of Xinjiang as the research subject. High-performance liquid chromatography (HPLC) was used to determine the contents of stevioside, rebaudioside A, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B, and isochlorogenic acid C in the stevia samples, providing accurate reference data for near-infrared (NIR) modeling. A total of 105 stevia samples were collected in this example to serve as a calibration set for NIR modeling. All stevia samples were pulverized, passed through a 40-mesh sieve, and sealed in ziplock bags for future use.

[0090] The instruments, reagents and modeling methods used in this example are as follows:

[0091] 1. Instrument:

[0092] High performance liquid chromatography (Waters, E2695), precision electronic balance (Mettler-Toledo Instrument Co., Ltd., ME204E / 02), rotary evaporator (Shanghai Aibo Instrument Co., Ltd., SB-2000), digital control ultrasonic cleaner (Kunshan Ultrasonic Instrument Co., Ltd., KQ5200DE), amino liquid chromatography column ( HPLC COLUMN NH2, 250 mm × 4.6 mm, 5 μm), C18 liquid chromatography column (Shim-pack GIST C18, 250 mm × 4.6 mm, 5 μm).

[0093] 2. Reagents:

[0094] Stevioside reference substance (Shanghai Jizhi Biochemical Technology Yangpu Co., Ltd.), rebaudioside A reference substance (Xinjiang Huijia Biotechnology Co., Ltd.), neochlorogenic acid reference substance (Shanghai Yuanfan Biotechnology Co., Ltd.), chlorogenic acid reference substance (Shanghai Maclean Biochemical Technology Co., Ltd.), cryptochlorogenic acid reference substance (Nantong Jingwei Biotechnology Co., Ltd.), isochlorogenic acid A reference substance (Shanghai Maclean Biochemical Technology Co., Ltd.), isochlorogenic acid B reference substance (Shanghai Maclean Biochemical Technology Co., Ltd.), isochlorogenic acid C reference substance (Nantong Jingwei Biotechnology Co., Ltd.), chromatographic methanol (Tianjin Biaoshiqi Technology Development Co., Ltd.), chromatographic acetonitrile (Anhui Tiandi High Purity Solvent Co., Ltd.), anhydrous ethanol (Tianjin Jindong Tianzheng Fine Chemical Reagent Factory), purified water (Wahaha Co., Ltd.), glacial acetic acid (Tianjin Jindong Tianzheng Fine Chemical Reagent Factory).

[0095] 3. Test method:

[0096] 3.1 Simultaneous extraction of stevioside, rebaudioside A, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B and isochlorogenic acid C:

[0097] 1g of stevia sample, crushed and passed through a 40-mesh sieve, was placed in a 250mL three-necked flask. 8mL of a 30% ethanol-water solution was added at a material-liquid ratio of 1:8 (m / v) to obtain a stevia solution. The stevia solution was heated to 60°C and extracted at this temperature for 1 hour. The solution was then cooled and filtered to collect the first filtrate. The remaining filter cake was transferred back to the three-necked flask. 6mL of a 30% ethanol-water solution was added at a material-liquid ratio of 1:6 (m / v). The solution was heated to 60°C again and extracted at this temperature for 0.5 hour. The solution was cooled again and filtered to collect the second filtrate. The first and second filtrates were combined and rotary evaporated to dryness to obtain a dry extract. The extract was stored in a brown vial in the dark and frozen at -20°C until use for HPLC analysis and modeling.

[0098] 3.2 Preparation of reference solution:

[0099] Stevioside, rebaudioside A, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B and isochlorogenic acid C were respectively taken, with the mass of each reference substance being 20 mg, 40 mg, 1 mg, 2 mg, 1 mg, 1 mg, 5 mg and 4 mg, respectively. After accurate weighing, they were placed in 5 mL volumetric flasks, and 30% volume fraction of chromatographic methanol aqueous solution was added to make the volume to 1 mL to obtain reference solution for content calibration and curve fitting of high performance liquid chromatography analysis.

[0100] 3.3 Preparation of test solution:

[0101] Weigh 10 mg of the extract prepared above, place it in a 5 mL volumetric flask, add 30% volume fraction of chromatographic methanol aqueous solution, and dilute to 1 mL to obtain the test solution.

[0102] 3.4 HPLC Characterization of Stevioside and Rebaudioside A:

[0103] High performance liquid chromatography was used to characterize stevioside and rebaudioside A in the extract. The specific chromatographic conditions were as follows: chromatographic column was used. HPLC column NH2 (250mm×4.6mm, 5μm), mobile phase is acetonitrile and purified water in a volume ratio of 80:20, flow rate is 1.2mL / min, detection wavelength is 210nm, column temperature is 30℃, injection volume is 10μL, and the obtained HPLC chromatogram is shown as follows Figure 2 shown.

[0104] according to Figure 2 It can be seen that stevioside and rebaudioside A can be clearly separated in the chromatogram, verifying the effectiveness of the extraction and detection methods.

[0105] 3.5 HPLC Characterization of Neochlorogenic Acid, Chlorogenic Acid, Cryptochlorogenic Acid, Isochroogenic Acid A, Isochroogenic Acid B and Isochroogenic Acid C:

[0106] High performance liquid chromatography was used to separate and quantitatively analyze neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B and isochlorogenic acid C in the extract. The specific chromatographic conditions were as follows: a Shim-pack-GISTC18 column (4.6×250 mm, 5 μm) was used, the column temperature was 30°C, the injection volume was 10 μL, and the detection wavelength was 330 nm; the mobile phase flow rate was 1.0 mL / min, the mobile phases were mobile phase A and mobile phase B, mobile phase A was a 1% volume fraction of acetic acid aqueous solution, and mobile phase B was acetonitrile, with gradient elution (0-15 min, 8-17% B; 15-30 min, 17-18% B; 30-40 min, 18-19% B; 40-42 min, 19-24% B; 42-50 min, 24-40% B; 50-65 min, 40-40.2% B). The obtained HPLC chromatogram is as follows Figure 3 shown.

[0107] according to Figure 3 It can be seen that under this condition, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B and isochlorogenic acid C can be effectively separated and stable chromatographic peaks can be obtained.

[0108] 3.6 The results of the test on the contents of stevioside, rebaudioside A, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B and isochlorogenic acid C in Stevia rebaudiana in this example are shown in Table 3.

[0109] Table 3 HPLC content detection results of Example 1

[0110] Serial number Component name Content percentage range Standard deviation Average content 1 Stevioside 2.56％-27.42％ 4.51％ 8.52％ 2 Rebaudioside A 17.64％-54.57％ 7.20％ 30.65％ 3 Neochlorogenic acid 0.06％-0.88％ 0.17％ 0.42％ 4 Chlorogenic acid 0.13％-2.56％ 0.53％ 1.19％ 5 Cryptochlorogenic acid 0.06％-0.43％ 0.07％ 0.21％ 6 Isochlorogenic acid B 0.17％-0.48％ 0.06％ 0.28％ 7 Isochlorogenic acid A 0.22％-9.94％ 1.52％ 2.79％ 8 Isochlorogenic acid C 0.55％-6.71％ 1.30％ 3.24％

[0111] Example 2

[0112] This example establishes a near-infrared quantitative analysis model for stevioside, rebaudioside A, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B, and isochlorogenic acid C in Stevia:

[0113] Based on the target component content detection results of 105 stevia samples obtained in Example 1, this example constructs a near-infrared quantitative analysis model for rapid, non-destructive quantitative analysis of the target component. The Quant 2 module in OPUS 7.0 spectral analysis software was used to establish a calibration model using the partial least squares method, and performance was verified by cross-validation. The following three core indicators were used to comprehensively evaluate the prediction performance and stability of the model: determination coefficient R 2 , cross-validation root mean square error RMSECV and residual prediction deviation RPD.

[0114] The specific steps of this embodiment are as follows:

[0115] 2.1 Spectral acquisition

[0116] The spectrum of stevia samples was collected using a German Bruker MPA Fourier transform near-infrared spectrometer with the following parameters: the sampling range was 4000-10000 cm -1 The measurement method is integrating sphere diffuse reflectance spectroscopy with a resolution of 8.0 cm -1 The number of scans was 64, and three spectra were collected for each sample. The sample cup was shaken before measurement to ensure sample uniformity. The near-infrared spectrum of the stevia sample collected is shown in the figure below. Figure 4 shown.

[0117] according to Figure 4 It can be seen that the spectral absorption peaks overlap severely, making it difficult to discern valid information from the original spectrum. Therefore, this embodiment uses chemometric methods to eliminate interference from background, noise, and other information and extract valid information from the spectrum, which is spectral preprocessing.

[0118] 2.2 Spectral preprocessing and characteristic wavelength

[0119] To improve the prediction accuracy and robustness of the near-infrared quantitative analysis calibration model, the optimal preprocessing methods and characteristic wavelengths of each target component are shown in Table 4.

[0120] Table 4 Preprocessing methods and characteristic wavelengths of each target component

[0121]

[0122] 2.3 Establishment of near-infrared quantitative analysis calibration model

[0123] Based on the optimal pretreatment method and characteristic wavelength, the Quant 2 module in OPUS 7.0 software was used to establish the near-infrared quantitative calibration model of stevioside, rebaudioside A, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B, and isochlorogenic acid C using the partial least squares method. During the model training process, based on the calibration set data, the following key indicators were output: qualitative coefficient R 2 The modeling results of each target component are shown in Table 5.

[0124] Table 5 Model calibration results for each target component

[0125] Component name Preprocessing methods <![CDATA[R 2 ]]> RMSEE RPD Stevioside Linear baseline correction 0.9932 0.019400 12.10 Rebaudioside A Linear baseline correction 0.9982 0.021000 23.60 Neochlorogenic acid Second-order derivative 0.9896 0.001470 9.83 Chlorogenic acid Second-order derivative 0.9958 0.003110 15.50 Cryptochlorogenic acid Eliminating constant offsets 0.9838 0.000764 7.86 Isochlorogenic acid B Linear baseline correction 0.9730 0.000604 6.09 Isochlorogenic acid A Linear baseline correction 0.9844 0.013300 6.64 Isochlorogenic acid C First-order derivative + vector normalization 0.9703 0.008440 5.80

[0126] According to the results in Table 5, it can be concluded that the near-infrared quantitative calibration model of stevioside, rebaudioside A, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B and isochlorogenic acid C was successfully established.

[0127] 2.4 Validation of near-infrared quantitative calibration model

[0128] After the near-infrared quantitative analysis calibration model of each target component was established, the cross-validation method was used to systematically verify the model in order to verify the stability and prediction performance of the model. The model was verified by cross-validation. According to the cross-validation determination coefficient R 2 , cross-validation root mean square error RMSECV and residual prediction deviation RPD are used to verify whether the correction model is the best model. At the same time, the cross-validation prediction diagram of each target component is as follows Figures 5 to 12 As shown, it further intuitively reflects the fitting between the model prediction value and the measured value, and verifies the accuracy and practicality of the constructed model in predicting different components.

[0129] Table 6 Cross-validation results of each component model

[0130] Component name <![CDATA[R 2 ]]> RMSECV RPD Bias Stevioside 0.9023 0.068100 3.20 -0.38200 Rebaudioside A 0.9728 0.075800 6.07 -1.40000 Neochlorogenic acid 0.9829 0.001820 7.65 -0.04090 Chlorogenic acid 0.9931 0.003830 12.00 -0.06080 Cryptochlorogenic acid 0.9653 0.005370 5.37 -0.00499 Isochlorogenic acid B 0.9575 0.000724 4.87 -0.05860 Isochlorogenic acid A 0.9773 0.015600 6.64 0.54200 Isochlorogenic acid C 0.9262 0.012700 3.68 0.27000

[0131] According to Table 6 and Figures 5 to 12 It can be seen that the near-infrared quantitative analysis calibration model established in the present invention has strong predictive ability, high accuracy and good practicality.

[0132] It can be seen from the above examples that the near-infrared quantitative analysis calibration model provided by the present invention has a stronger explanatory power and a higher prediction accuracy, and has good prediction ability.

[0133] Although the above embodiment provides a detailed description of the present invention, it is only a part of the embodiments of the present invention, not all of the embodiments. Other embodiments can be obtained based on this embodiment without creativity, and these embodiments all fall within the scope of protection of the present invention.

Claims

1. A modeling method for near-infrared quantitative analysis of stevia components, characterized in that: The following steps are involved: (1) crushing and sieving the stevia sample in sequence to obtain a stevia powder sample; (2) performing high performance liquid chromatography on one or more of stevioside, rebaudioside A, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B, and isochlorogenic acid C in the stevia powder sample to obtain the content of the target component; The preparation method of the test solution for high performance liquid chromatography detection comprises the following steps: Mixing the stevia powder sample with a first ethanol solution to perform a first extraction and then a first solid-liquid separation to obtain a first solid phase and a first liquid phase; Mixing the first solid phase with a second ethanol solution to perform a second extraction and then a second solid-liquid separation to obtain a second solid phase and a second liquid phase; mixing the first liquid phase and the second liquid phase and removing the solvent to obtain an extract; The extract and chromatographic methanol-water solution are mixed to obtain a test solution; (3) performing near-infrared spectroscopy analysis on the target component in the stevia powder sample to obtain near-infrared spectral data of the target component; (4) preprocessing the near-infrared spectrum data of the target component to select the characteristic wavelength corresponding to the target component; The characteristic wavelength of stevioside is 6248.5-7143.4 cm -1 , the dimension is 6, and linear baseline correction is used to preprocess the near-infrared spectral data; The characteristic wavelength of rebaudioside A is 6248.5-7143.4 cm -1 and 7691.1~9091.2cm -1 , the dimension is 4, and linear baseline correction is used to preprocess the near-infrared spectral data; The characteristic wavelength of the new chlorogenic acid is 3999.8 to 5303.5 cm -1 and 5936.1~6248.5cm -1 , the dimension is 5, and the second-order derivative is used to preprocess the near-infrared spectral data; The characteristic wavelength of chlorogenic acid is 3999.8-5303.5 cm -1 , the dimension is 5, and the second-order derivative is used to preprocess the near-infrared spectral data; The characteristic wavelength of the cryptochlorogenic acid is 3999.8 to 5303.5 cm -1 , the dimension is 7, and the near-infrared spectral data are preprocessed by eliminating the constant offset; The characteristic wavelength of isochlorogenic acid B is 3999.8 to 5303.5 cm -1 and 5623.7~6248.5cm -1 , the dimension is 4, and linear baseline correction is used to preprocess the near-infrared spectral data; The characteristic wavelength of isochlorogenic acid A is 5623.7-6248.5 cm -1 , the dimension is 4, and linear baseline correction is used to preprocess the near-infrared spectral data; The characteristic wavelength of isochlorogenic acid C is 3999.8 to 5303.5 cm -1 and 5623.7~6248.5cm -1 , the dimension is 3, and the first-order derivative + vector normalization is used to preprocess the near-infrared spectral data; (5) Based on the Quant 2 module of OPUS 7.0 software, a near-infrared quantitative calibration model of the target component of stevia was established. The characteristic wavelength of the target component determined in step (4) was used, and the partial least squares method was used to fit the near-infrared spectral data with the content of the target component to obtain a calibration model for near-infrared quantitative analysis of stevia components.

2. The modeling method according to claim 1, characterized in that The first ethanol solution is an ethanol aqueous solution, and the volume content of the first ethanol solution is 30%; The ratio of the mass of the stevia powder sample to the volume of the first ethanol solution is 1 g:8 mL.

3. The modeling method according to claim 1 or 2, characterized in that: The temperature of the first extraction is 60° C., and the insulation time is 1 hour.

4. The modeling method according to claim 1, characterized in that The second ethanol solution is an ethanol aqueous solution, and the volume content of the second ethanol solution is 30%; The ratio of the mass of the stevia powder sample to the volume of the second ethanol solution is 1 g:6 mL.

5. The modeling method according to claim 1 or 4, characterized in that: The temperature of the second extraction is 60° C., and the holding time is 0.5 h.

6. The modeling method according to claim 1, characterized in that The solvent removal is performed by rotary evaporation, the temperature of the rotary evaporation is 30 to 50° C., and the insulation time is 10 to 40 minutes.

7. The modeling method according to claim 1, characterized in that The volume fraction of methanol in the chromatographic methanol aqueous solution is 10 to 70%; The extract concentration in the test solution is 1-20 g / L.

8. The modeling method according to claim 1, characterized in that: The chromatographic column used for the high performance liquid chromatography detection of stevioside is HPLC COLUMN NH2; The mobile phase of the chromatographic column includes acetonitrile and water, the volume ratio of the acetonitrile to water is 80:20, and the flow rate of the mobile phase is 1.2 mL / min; The detection wavelength of the HPLC detection was 210 nm and the injection volume was 10 μL; The parameters for the HPLC detection of rebaudioside A are the same as those for stevioside; The chromatographic column used for the high performance liquid chromatography detection of the neochlorogenic acid is Shim-pack-GIST C18; The mobile phase of the chromatographic column includes acetic acid aqueous solution and acetonitrile, the volume content of acetic acid in the acetic acid aqueous solution is 1%, the high performance liquid chromatography detection of the neochlorogenic acid is gradient elution, and the volume ratio of the acetic acid aqueous solution and acetonitrile is: 92-83% acetic acid aqueous solution + 8-17% acetonitrile at 0-15 min, 83-82% acetic acid aqueous solution + 17-18% acetonitrile at 15-30 min, 82-81% acetic acid aqueous solution + 18-19% acetonitrile at 30-40 min, 81-76% acetic acid aqueous solution + 19-24% acetonitrile at 40-42 min, 76-60% acetic acid aqueous solution + 24-40% acetonitrile at 42-50 min, 60-59.8% acetic acid aqueous solution + 40-40.2% acetonitrile at 50-65 min, the flow rate of the mobile phase is 1.0 mL / min; the detection wavelength of the high performance liquid chromatography detection is 330 nm, and the injection volume is 10 μL; The parameters for high performance liquid chromatography detection of the chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B and isochlorogenic acid C are the same as those for neochlorogenic acid.

9. The modeling method according to claim 1 or 8, characterized in that: The near-infrared spectroscopy analysis adopts the integrating sphere diffuse reflectance method, and the sampling range is 4000~10000cm -1 , resolution of 8.0~64.0cm -1 , the number of scans is 16 to 64 times, and the number of acquisitions is 3 to 6 times.

10. A method for near-infrared quantitative analysis of stevia components, characterized in that: The following steps are involved: (1) crushing and sieving the stevia sample in sequence to obtain a stevia powder sample; (2) performing high performance liquid chromatography on one or more of stevioside, rebaudioside A, neochlorogenic acid, chlorogenic acid, cryptochlorogenic acid, isochlorogenic acid A, isochlorogenic acid B, and isochlorogenic acid C in the stevia powder sample to obtain the content of the target component; The preparation method of the test solution for high performance liquid chromatography detection comprises the following steps: Mixing the stevia powder sample with a first ethanol solution to perform a first extraction and then a first solid-liquid separation to obtain a first solid phase and a first liquid phase; Mixing the first solid phase with a second ethanol solution to perform a second extraction and then a second solid-liquid separation to obtain a second solid phase and a second liquid phase; mixing the first liquid phase and the second liquid phase and removing the solvent to obtain an extract; The extract and chromatographic methanol-water solution are mixed to obtain a test solution; (3) performing near-infrared spectroscopy analysis on the target component in the stevia powder sample to obtain near-infrared spectral data of the target component; (4) preprocessing the near-infrared spectrum data of the target component to select the characteristic wavelength corresponding to the target component; The characteristic wavelength of stevioside is 6248.5-7143.4 cm -1 , the dimension is 6, and linear baseline correction is used to preprocess the near-infrared spectral data; The characteristic wavelength of rebaudioside A is 6248.5-7143.4 cm -1 and 7691.1~9091.2cm -1 , the dimension is 4, and linear baseline correction is used to preprocess the near-infrared spectral data; The characteristic wavelength of the new chlorogenic acid is 3999.8 to 5303.5 cm -1 and 5936.1~6248.5cm -1 , the dimension is 5, and the second-order derivative is used to preprocess the near-infrared spectral data; The characteristic wavelength of chlorogenic acid is 3999.8-5303.5 cm -1 , the dimension is 5, and the second-order derivative is used to preprocess the near-infrared spectral data; The characteristic wavelength of the cryptochlorogenic acid is 3999.8 to 5303.5 cm -1 , the dimension is 7, and the near-infrared spectral data are preprocessed by eliminating the constant offset; The characteristic wavelength of isochlorogenic acid B is 3999.8 to 5303.5 cm -1 and 5623.7~6248.5cm -1 , the dimension is 4, and linear baseline correction is used to preprocess the near-infrared spectral data; The characteristic wavelength of isochlorogenic acid A is 5623.7-6248.5 cm -1 , the dimension is 4, and linear baseline correction is used to preprocess the near-infrared spectral data; The characteristic wavelength of isochlorogenic acid C is 3999.8 to 5303.5 cm -1 and 5623.7~6248.5cm -1 , the dimension is 3, and the first-order derivative + vector normalization is used to preprocess the near-infrared spectral data; (5) Based on the Quant 2 module of OPUS 7.0 software, a near-infrared quantitative calibration model of the target component of stevia was established. The characteristic wavelength of the target component determined in step (4) was used, and the partial least squares method was used to fit the near-infrared spectral data with the content of the target component to obtain a calibration model for near-infrared quantitative analysis of stevia components; (6) The stevia to be tested is successively crushed and sieved to obtain a stevia powder to be tested, and a target component in the stevia powder to be tested is analyzed by near-infrared spectroscopy to obtain near-infrared spectral data of the target component. The component content of the stevia is obtained according to a calibration model for near-infrared quantitative analysis of stevia components.

Images