Method and device for optimizing tea polyphenol detection and extraction process based on near infrared spectrum

Abstract

The invention relates to the technical field of component detection, and discloses a tea polyphenol detection and extraction process optimization method and device based on a near infrared spectrum. The method comprises the following steps: correcting and verifying characteristic wave band data to obtain tea polyphenol content quantitative relation data; performing single-factor experimental design on the extraction temperature, the extraction time, the material-liquid ratio and the solvent concentration to obtain parameter range data, and performing response surface experimental design on the parameter range data to obtain a process parameter combination; performing online monitoring on the quantitative relation data of the process parameter combination and the tea polyphenol content to obtain process parameter data, and performing real-time analysis processing on the process parameter data to obtain process control data; and dynamically adjusting the process control data to obtain optimized process parameters, and carrying out repeatability verification on the optimized process parameters to obtain target process parameters. According to the method, real-time monitoring of the extraction process and dynamic optimization of the process parameters are realized, and the accuracy and controllability of the extraction process are improved.

Description

Technical Field

[0001] This application relates to the technical field of component detection, and particularly to a method and device for optimizing the detection and extraction process of tea polyphenols based on near-infrared spectroscopy. Background Art

[0002] In recent years, tea polyphenols have attracted much attention due to their physiological activities such as antioxidant, anti-aging, and lipid-lowering effects. The extraction process and content detection method have become research hotspots. Traditional tea polyphenol content detection methods mainly use high-performance liquid chromatography (HPLC) and spectrophotometry. These methods require complex sample pretreatment processes, including extraction, filtration, volume fixing, etc., and the detection process is time-consuming and consumes a large amount of reagents. The optimization of the tea polyphenol extraction process usually adopts a method combining single-factor experiments and orthogonal experiments, and the optimal process parameters are determined through multiple offline sampling analyses. This method requires a large amount of manual operation and laboratory analysis work.

[0003] The main problems existing in the prior art are that the real-time monitoring and online analysis of the tea polyphenol extraction process cannot be realized, and the optimization of process parameters depends on a large amount of offline experimental data, resulting in a long process optimization cycle and high labor costs. At the same time, due to the lack of real-time monitoring means, the dynamic adjustment of process parameters depends on empirical judgment, making it difficult to ensure the stability and repeatability of the extraction process. In addition, when the existing methods optimize process parameters, they often ignore the interaction between process parameters, resulting in insufficient reliability of the optimization results. Summary of the Invention

[0004] This application provides a method and device for optimizing the detection and extraction process of tea polyphenols based on near-infrared spectroscopy, which is used to realize the real-time monitoring of the extraction process and the dynamic optimization of process parameters, and improve the accuracy and controllability of the extraction process.

[0005] First aspect, the present application provides a method for optimizing the detection and extraction process of tea polyphenols based on near-infrared spectroscopy. The method for optimizing the detection and extraction process of tea polyphenols based on near-infrared spectroscopy includes: drying and pulverizing a tea sample to obtain a tea powder sample, performing particle size screening and temperature and humidity balance treatment on the tea powder sample to obtain a standardized sample, calibrating the standardized sample with a near-infrared spectroscopy system to obtain calibration spectral parameters; performing quantitative sample loading and scanning angle adjustment on the standardized sample to obtain a sample to be measured, collecting near-infrared spectra of the sample to be measured and the calibration spectral parameters to obtain original spectral data, and performing signal smoothing and scattering correction processing on the original spectral data to obtain preprocessed spectral data; extracting characteristic wavelengths and performing correlation analysis on the preprocessed spectral data to obtain characteristic band data, and performing calibration and verification processing on the characteristic band data to obtain quantitative relationship data of tea polyphenol content; designing single-factor experiments for extraction temperature, extraction time, solid-liquid ratio, and solvent concentration according to the quantitative relationship data of tea polyphenol content to obtain parameter range data, and performing response surface experiment design on the parameter range data to obtain process parameter combinations; performing on-line monitoring on the process parameter combinations and the quantitative relationship data of tea polyphenol content to obtain process parameter data, performing real-time analysis and processing on the process parameter data to obtain process control data; dynamically adjusting the process control data to obtain optimized process parameters, and performing repeatability verification on the optimized process parameters to obtain target process parameters.

[0006] Second aspect, the present application provides a device for optimizing the detection and extraction process of tea polyphenols based on near-infrared spectroscopy. The device for optimizing the detection and extraction process of tea polyphenols based on near-infrared spectroscopy includes: A calibration module, which is used for drying and pulverizing a tea sample to obtain a tea powder sample, performing particle size screening and temperature and humidity balance treatment on the tea powder sample to obtain a standardized sample, and calibrating the standardized sample with a near-infrared spectroscopy system to obtain calibration spectral parameters; An adjustment module, which is used for performing quantitative sample loading and scanning angle adjustment on the standardized sample to obtain a sample to be measured, collecting near-infrared spectra of the sample to be measured and the calibration spectral parameters to obtain original spectral data, and performing signal smoothing and scattering correction processing on the original spectral data to obtain preprocessed spectral data; An analysis module, which is used for extracting characteristic wavelengths and performing correlation analysis on the preprocessed spectral data to obtain characteristic band data, and performing calibration and verification processing on the characteristic band data to obtain quantitative relationship data of tea polyphenol content; A design module, which is used to conduct single-factor experimental designs on extraction temperature, extraction time, solid-liquid ratio, and solvent concentration according to the quantitative relationship data of the tea polyphenol content, obtain parameter range data, and conduct response surface experimental designs on the parameter range data to obtain process parameter combinations; A monitoring module, which is used to conduct online monitoring on the process parameter combinations and the quantitative relationship data of the tea polyphenol content, obtain process parameter data, and conduct real-time analysis and processing on the process parameter data to obtain process control data; A verification module, which is used to dynamically adjust the process control data to obtain optimized process parameters, and conduct repeatability verification on the optimized process parameters to obtain target process parameters.

[0007] In the technical solution provided by this application, through the standardized pretreatment of tea samples, including drying, crushing, particle size screening, and temperature and humidity balance treatment, the uniformity and stability of the samples are significantly improved, laying a reliable foundation for subsequent spectral analysis. The water content of the samples is controlled below 5%, the particle size uniformity reaches over 95%, and through the temperature and humidity balance treatment, the samples reach a stable state, effectively eliminating the interference of environmental factors. By combining the calibration of the near-infrared spectroscopy system with quantitative sample loading and scanning angle adjustment, the accuracy and repeatability of spectral acquisition are achieved, and the relative standard deviation of spectral acquisition is controlled within 0.1%. By performing signal smoothing and scattering correction processing on the original spectral data, the influence of baseline drift and scattering effects is effectively eliminated, the signal-to-noise ratio of the spectral data is improved, and the signal-to-noise ratio reaches over 1000:1. The introduction of characteristic wavelength extraction and correlation analysis realizes the rapid quantitative detection of the tea polyphenol content. The detection time is shortened from several hours of the traditional method to several minutes, and the measurement accuracy reaches over 95%. By conducting single-factor experimental designs and response surface optimizations on the extraction process parameters, the quantitative relationships between parameters such as temperature, time, solid-liquid ratio, and solvent concentration are established, and the extraction yield is increased by over 15%. Through online monitoring and real-time analysis and processing, the dynamic control of the process is achieved, and the parameter fluctuations are controlled within the range of ±2%, significantly improving the stability and repeatability of the process. Finally, through dynamic adjustment and repeatability verification, the reliability of the process parameters is ensured, the relative standard deviation between batches is controlled within 2%, and an optimization and quality control system for the tea polyphenol extraction process is established. Description of the Drawings

[0008] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for description in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0009] Figure 1It is a schematic diagram of an embodiment of the method for optimizing the detection and extraction process of tea polyphenols based on near-infrared spectroscopy in the embodiments of the present application; Figure 2 It is a schematic diagram of an embodiment of the device for optimizing the detection and extraction process of tea polyphenols based on near-infrared spectroscopy in the embodiments of the present application. Specific embodiments

[0010] The embodiments of the present application provide a method and device for optimizing the detection and extraction process of tea polyphenols based on near-infrared spectroscopy. Terms such as "first", "second", "third", "fourth", etc. (if any) in the specification, claims and above-mentioned drawings of the present application are used to distinguish similar objects, and do not have to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments described herein can be implemented in an order different from that shown or described herein. In addition, the terms "comprising" or "having" and any variation thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0011] For ease of understanding, the specific process of the embodiments of the present application is described below. Please refer to Figure 1 An embodiment of the method for optimizing the detection and extraction process of tea polyphenols based on near-infrared spectroscopy in the embodiments of the present application includes: Step S101: Dry and crush the tea sample to obtain a tea powder sample, perform particle size screening and temperature and humidity balance treatment on the tea powder sample to obtain a standardized sample, and calibrate the standardized sample with a near-infrared spectroscopy system to obtain calibration spectral parameters; Step S102: Perform quantitative sampling and scanning angle adjustment on the standardized sample to obtain a sample to be measured, collect near-infrared spectra of the sample to be measured and the calibration spectral parameters to obtain original spectral data, and perform signal smoothing and scattering correction processing on the original spectral data to obtain preprocessed spectral data; Step S103: Extract characteristic wavelengths and perform correlation analysis on the preprocessed spectral data to obtain characteristic band data, and perform calibration and verification processing on the characteristic band data to obtain quantitative relationship data of tea polyphenol content; Step S104: According to the quantitative relationship data of tea polyphenol content, perform single-factor experimental design on the extraction temperature, extraction time, solid-liquid ratio, and solvent concentration to obtain parameter range data, and perform response surface experimental design on the parameter range data to obtain process parameter combinations; Step S105: Online monitor the quantitative relationship data between process parameter combinations and tea polyphenol content to obtain process parameter data, and perform real-time analysis and processing on the process parameter data to obtain process control data; Step S106: Dynamically adjust the process control data to obtain optimized process parameters, and perform repeatability verification on the optimized process parameters to obtain target process parameters.

[0012] It can be understood that the execution entity of this application can be a near-infrared spectroscopy tea polyphenol detection and extraction process optimization device, or a terminal or a server. Specifically, no limitation is made here. In this embodiment of the application, the server is used as the execution entity for illustration.

[0013] Specifically, starting from the pretreatment of tea samples, the tea samples are dried at a constant temperature of 60°C by a vacuum drying device, and the drying time is controlled within 24 hours to reduce the water content of the tea to less than 5%. The dried tea is mechanically pulverized by a pulverizer to obtain preliminary tea powder. Subsequently, particle size screening is carried out using an 80-mesh standard sieve. The screened samples are balanced for 24 hours in a constant temperature and humidity environment at a temperature of 25 ± 1°C and a relative humidity of 60 ± 2% to obtain standardized samples.

[0014] When calibrating the standardized samples with the near-infrared spectroscopy system, a barium sulfate reflector is used as the background for scanning. The scanning wavelength range is set to 800 - 2500 nm, and the wavelength interval is 2 nm. The calibration spectral parameters are obtained by accumulating and averaging 32 repeated scans. In the second stage, sample quantitative loading and spectral acquisition are carried out. 5.000 ± 0.001 g of the standardized sample is accurately weighed and loaded into a quartz sample cell, and the angle between the incident light beam and the sample surface is adjusted to 45° ± 0.5° to ensure the repeatability and accuracy of spectral acquisition. A near-infrared spectrometer equipped with an integrating sphere is used for spectral acquisition. The diameter of the integrating sphere is 100 mm, and the inner wall is coated with a barium sulfate reflection coating. Each sample is scanned 32 times repeatedly to obtain the original spectral data. The original spectral data is first subjected to noise elimination using the five-point smoothing method, and the smoothing window width is set to 5 wavelength points. Subsequently, the scattering effect is eliminated by standard normal variate transformation (SNV) to obtain the preprocessed spectral data.

[0015] In the third stage, characteristic wavelengths are extracted from the preprocessed spectral data. By calculating the correlation coefficient matrix between adjacent bands, with the correlation coefficient threshold set at 0.95, characteristic bands are screened out. Meanwhile, in combination with the reference value of the tea polyphenol content measured by high-performance liquid chromatography, a quantitative relationship between wavelength and content is established. Taking the 1400 - 1600 nm band as an example, through correlation coefficient calculation, 1450 nm, 1510 nm, and 1550 nm are determined as the characteristic absorption peaks of tea polyphenols, and the correlation coefficients at these wavelengths are 0.97, 0.98, and 0.96 respectively. In the fourth stage, process parameter optimization design is carried out. Based on the quantitative relationship data of tea polyphenol content, single-factor experiments are conducted with four parameters: extraction temperature of 60 - 90 °C, extraction time of 30 - 120 min, solid-liquid ratio of 1:10 - 1:30, and ethanol concentration of 40% - 80%. Each factor is set with 5 level points, and the significance range of each parameter is determined through variance analysis. Subsequently, the Box-Behnken response surface design method is adopted to establish a three-level four-factor experimental matrix, with a total of 29 experimental points, to obtain process parameter combinations. In the fifth stage, online monitoring of the process parameter combinations is carried out. The temperature monitoring accuracy is controlled within ±0.1 °C, the time monitoring accuracy is ±1 s, and the solid-liquid ratio monitoring accuracy is ±0.01. Process data is collected every 30 seconds. Samples in the process are extracted through real-time near-infrared spectroscopy scanning, and the collected spectral data is compared with the quantitative relationship data of tea polyphenol content to calculate the deviation value. The control threshold is set at ±2% to generate process control data.

[0016] In the final stage, dynamic adjustment is made to the process control data. When it is detected that the parameter deviation exceeds the threshold, step-by-step adjustment is carried out according to the temperature gradient of 0.5 °C, the time gradient of 5 min, and the solid-liquid ratio gradient of 0.5. The adjusted process parameters are verified in three batches, and the relative standard deviation is calculated. The parameters with the standard deviation controlled within 2% are determined as the target process parameters.

[0017] For example: After a batch of tea samples are dried and pulverized, a powder with uniform particle size is obtained by screening through an 80-mesh sieve and balanced for 24 hours at 25 °C and a relative humidity of 60%. During near-infrared spectroscopy collection, 5.000 g of the sample is loaded into the sample cell, with an incident angle of 45°. The absorbance measured at 1450 nm is 0.857, and the value after SNV conversion is 0.762. Single-factor experiments show that the tea polyphenol yield reaches the best under the conditions of 80 °C, 60 min, a solid-liquid ratio of 1:20, and an ethanol concentration of 60%. After optimization by response surface experiments, the target process parameters are determined as: temperature 78.5 °C, time 65 min, solid-liquid ratio 1:18, and ethanol concentration 65%. The relative standard deviation of the three-batch verification is 1.8%.

[0018] In the embodiments of the present application, through the standardized pretreatment of tea samples, including drying, pulverization, particle size screening, and temperature and humidity balance treatment, the uniformity and stability of the samples are significantly improved, laying a reliable foundation for subsequent spectral analysis. The water content of the samples is controlled below 5%, the particle size uniformity reaches over 95%, and through the temperature and humidity balance treatment, the samples reach a stable state, effectively eliminating the interference of environmental factors. By combining the calibration of the near-infrared spectroscopy system with quantitative sample loading and scanning angle adjustment, the accuracy and repeatability of spectral acquisition are achieved, and the relative standard deviation of spectral acquisition is controlled within 0.1%. By performing signal smoothing and scattering correction on the original spectral data, the influence of baseline drift and scattering effects is effectively eliminated, the signal-to-noise ratio of the spectral data is improved, and the signal-to-noise ratio reaches over 1000:1. The introduction of characteristic wavelength extraction and correlation analysis realizes the rapid quantitative detection of the tea polyphenol content. The detection time is shortened from several hours of traditional methods to several minutes, and the determination accuracy reaches over 95%. Through single-factor experimental design and response surface optimization of the extraction process parameters, a quantitative relationship between parameters such as temperature, time, solid-liquid ratio, and solvent concentration is established, and the extraction yield is increased by over 15%. Through on-line monitoring and real-time analysis and processing, the dynamic control of the process is achieved, and the parameter fluctuation is controlled within the range of ±2%, significantly improving the stability and repeatability of the process. Finally, through dynamic adjustment and repeatability verification, the reliability of the process parameters is ensured, the relative standard deviation between batches is controlled within 2%, and an optimization and quality control system for the tea polyphenol extraction process is established.

[0019] In a specific embodiment, the process of executing step S101 may specifically include the following steps: (1) Perform vacuum drying treatment on the tea sample at 60 °C to obtain a dried tea sample, and perform mechanical pulverization treatment on the dried tea sample to obtain a tea powder sample; (2) Screen the tea powder sample with an 80-mesh standard sieve to obtain a preliminary screened sample, and perform constant temperature and humidity treatment on the preliminary screened sample under the conditions of 25 ± 1 °C and relative humidity of 60 ± 2% to obtain a standardized sample; (3) Take samples from the standardized sample by the quartering method to obtain the part to be tested, and perform quantitative weighing on the part to be tested to obtain a standardized sample; (4) Perform background scanning on the standardized sample with a barium sulfate reflector to obtain background spectral data, and perform wavelength range correction on the background spectral data to obtain corrected spectral data; (5) Perform 32 repeated scans and cumulative averaging on the corrected spectral data to obtain average spectral data, and calculate and evaluate the signal-to-noise ratio of the average spectral data to obtain reference spectral data; (6) Perform wavelength accuracy and repeatability verification on the reference spectral data to obtain standard spectral data, and perform systematic deviation correction on the standard spectral data to obtain calibrated spectral parameters.

[0020] Specifically, it is carried out in a vacuum drying equipment. Fresh tea leaves are placed in a vacuum drying oven. The vacuum degree is set to -0.08 MPa, the temperature is controlled at 60 °C, and the drying time lasts for 24 hours, so that the water content of the tea leaves is reduced to less than 5%, ensuring that powder agglomeration will not occur due to excessive water content during the subsequent pulverization process. The dried tea leaves are mechanically pulverized by a high-speed pulverizer. The rotation speed of the pulverizer is set to 20,000 r / min, the pulverization time is controlled at 2 minutes, and after an interval of 30 seconds, it is pulverized again for 2 minutes, and this cycle is repeated 3 times to obtain preliminary tea powder. Subsequently, an 80-mesh standard sieve is used to screen the tea powder. During the screening process, a mechanical vibrating sieve is used, the vibration frequency is set to 50 Hz, and the vibration time is 10 minutes. The particles with a particle size greater than 180 μm are pulverized again until they all pass through the 80-mesh sieve to obtain a pre-screened sample. The pre-screened sample needs to be balanced in a constant temperature and humidity chamber. The temperature is precisely controlled at 25 ± 1 °C, and the relative humidity is adjusted to 60 ± 2% by the saturated salt solution method. The balance time is 24 hours to ensure that the sample reaches a stable state and a standardized sample is obtained.

[0021] The standardized sample is sampled by the quartering method. The specific operation is to pile the sample into a conical shape, divide the sample into four equal parts with a sampling plate, take the two diagonal parts and combine them, and repeat this process three times. Finally, a representative part to be measured is obtained. A precision balance is used to quantitatively weigh the part to be measured, with a weighing accuracy of 0.001 g. 5.000 ± 0.001 g of the sample is weighed and loaded into a special sample cell. The surface compactness of the sample is controlled at 0.8 g / cm³ to obtain the final standardized sample. Before collecting the near-infrared spectrum of the standardized sample, a background scan is required. A barium sulfate reflection sheet is used as the reference standard, with a reflectivity greater than 99%. The scanning wavelength range is set between 800 - 2500 nm, and the wavelength interval is 2 nm to obtain background spectral data. The wavelength range of the background spectral data is corrected. The wavelength accuracy is corrected through the characteristic absorption peak of the NIST standard sample (polystyrene film) at 1684 nm, and the deviation is controlled within ±0.3 nm to obtain corrected spectral data.

[0022] The corrected spectral data is continuously scanned 32 times. The wavelength resolution of each scan is 2 nm, and the scanning speed is 1 time per second. The 32 scan results are subjected to cumulative averaging to obtain average spectral data. The signal-to-noise ratio of the average spectral data is calculated. The calculation formula is S / N = P signal / P noise, where P signal is the signal intensity of the characteristic peak and P noise is the standard deviation of the baseline noise. It is required that the signal-to-noise ratio is greater than 1000:1. After passing the evaluation, reference spectral data is obtained.

[0023] Verify the wavelength accuracy of the reference spectral data. Select the characteristic peak at 1684 nm for repeatability measurement. Measure continuously 10 times and calculate the standard deviation, which is required to be less than 0.1 nm. The repeatability verification is evaluated by calculating the relative standard deviation of the 10 measurement results, and the relative standard deviation is controlled within 1% to obtain the standard spectral data. Finally, correct the systematic deviation of the standard spectral data by partial least squares method to establish a calibration equation Y = kX + b, where Y is the correction value, X is the measured value, k is the slope, and b is the intercept, to obtain the calibrated spectral parameters.

[0024] For example: After a batch of tea samples are vacuum-dried at 60 °C for 24 hours, the water content drops from the initial 75% to 4.8%. After mechanical crushing and screening through 80 meshes, the particle size is distributed in the range of 120 - 180 μm. After equilibration at 25 °C and relative humidity of 60% for 24 hours, sampling is carried out by the quartering method, and 5.000 g of the sample is weighed for testing. Against the background of a barium sulfate reflector, the absorbance measured at 1684 nm is 0.857, the average value of 32 repeated scans is 0.862, the standard deviation is 0.003, and the signal-to-noise ratio is 1250:1. The measured value is corrected by the calibration equation Y = 1.02X - 0.015, and finally the calibrated spectral parameter at this wavelength is obtained as 0.864. The relative standard deviation of the whole calibration process is 0.8%.

[0025] In a specific embodiment, the process of executing step S102 may specifically include the following steps: (1) Weigh the standardized sample precisely to obtain a quantitative sample, and load the quantitative sample into a quartz sample cell to obtain a loaded sample; (2) Adjust the incident angle of the loaded sample to obtain an angle-calibrated sample, and compact the surface of the angle-calibrated sample to obtain a sample to be measured; (3) Perform 32 repeated scans on the sample to be measured and the calibrated spectral parameters to obtain cumulative spectral data, and verify the repeatability of the cumulative spectral data to obtain the original spectral data; (4) Eliminate the noise of the original spectral data by the five-point smoothing method to obtain smoothed spectral data, and correct the baseline drift of the smoothed spectral data to obtain baseline-corrected data; (5) Perform standard normal variate transformation on the baseline-corrected data to obtain scatter-corrected data, and perform multiplicative scatter correction on the scatter-corrected data to obtain preprocessed spectral data.

[0026] Specifically, a precision balance is used for sample weighing. The precision of the balance is 0.001 g. Weigh 5.000 ± 0.001 g of the standardized sample and record the actual weighing value. Subsequently, the weighed quantitative sample is loaded into a quartz sample cell with an inner diameter of 30 mm and a depth of 10 mm. The quartz sample cell has good optical transmittance, and the transmittance is greater than 95% in the wavelength range of 800 - 2500 nm. During the sample loading process, a special sampler is used to evenly spread the sample at the bottom of the sample cell. Place the loaded sample on the sample stage of the near-infrared spectrometer, adjust the angle between the incident light beam and the sample surface to 45° ± 0.5°, and perform precise adjustment through the angle regulator. Record the reflection signal intensity every 0.1°, and lock the angle when the maximum reflection signal value is obtained at 45°. Use a compactor to perform surface treatment on the sample after angle calibration. The pressure of the compactor is controlled at 2.0 ± 0.1 MPa, and the compaction time is 5 seconds, so that the flatness deviation of the sample surface is controlled within ±0.05 mm, and the compaction density reaches 0.8 g / cm³ to obtain the sample to be measured.

[0027] Place the sample to be measured in the sample chamber of the spectrometer. Combine the obtained calibration spectral parameters and perform 32 consecutive scans in the wavelength range of 800 - 2500 nm. The scan interval is 2 nm, and the scan time for each time is 1 second. The spectral data obtained from the scans are accumulated and summed. Calculate the relative standard deviation for the accumulated spectral data. Select three characteristic wavelength points of 1450 nm, 1650 nm, and 1930 nm for repeatability evaluation, and require the relative standard deviation to be less than 0.1%. After passing the verification, obtain the original spectral data. Use the five-point smoothing method to perform noise elimination processing on the original spectral data. The five-point smoothing method eliminates random noise by taking the average of the spectral values at five consecutive wavelength points. The data after smoothing is corrected for baseline drift by the first derivative method. The baseline drift correction eliminates the influence of baseline shift by calculating the absorbance difference between adjacent wavelength points.

[0028] Perform standard normal variate transformation (SNV) processing on the data after baseline correction. The SNV transformation eliminates the influence of physical factors such as sample thickness and density by calculating the deviation of each wavelength point from the average value and dividing it by the standard deviation. The data after SNV transformation is then subjected to multiplicative scatter correction (MSC) processing. The MSC correction eliminates the influence of light scattering on the spectrum by establishing a linear relationship between the spectrum of each sample and the average spectrum, and finally obtains the preprocessed spectral data.

[0029] For example: A batch of tea samples was weighed on a precision balance to be 5.002 g. After being loaded into a quartz sample cell, the incident angle was adjusted, and the maximum reflection signal intensity of 0.856 was measured at 45.0°. After compaction treatment, the sample density was 0.82 g / cm³. The average value of the cumulative spectral data at 1450 nm after 32 repeated scans was 0.857, the standard deviation was 0.0008, and the relative standard deviation was 0.093%. The original data of 0.857 at this wavelength point and the data of the four adjacent wavelength points were subjected to five-point smoothing processing to eliminate the influence of random noise. After baseline drift correction, SNV conversion, and MSC correction, the final preprocessed spectral data of 0.857 was obtained. The relative standard deviation of the entire data processing process was controlled within 0.1%, meeting the analysis requirements.

[0030] In a specific embodiment, the process of executing step S103 may specifically include the following steps: (1) Perform wavelength interval segmentation on the preprocessed spectral data to obtain band partition data, and calculate the correlation between adjacent bands of the band partition data to obtain a band correlation matrix; (2) Perform eigenvector analysis on the band correlation matrix to obtain eigenvalue data, and calculate the contribution rate of the eigenvalue data to obtain principal component data; (3) Perform wavelength normalization on the principal component data to obtain normalized wavelength data, and perform interval screening on the normalized wavelength data to obtain characteristic band data; (4) Perform high performance liquid chromatography verification on the characteristic band data to obtain reference content data, and perform linear regression analysis on the reference content data to obtain correlation coefficient data; (5) Calculate the confidence interval of the correlation coefficient data to obtain interval range data, and perform reliability evaluation on the interval range data to obtain quantitative relationship data of the tea polyphenol content.

[0031] Specifically, the wavelength interval of the preprocessed spectral data is segmented, and the wavelength range of 800 - 2500 nm is divided into 17 band partitions at intervals of 100 nm. The correlation coefficient between adjacent wavelength points is calculated within each partition to form a 17×17 band correlation matrix, and each element in the matrix represents the correlation degree between two bands. Perform eigenvector analysis on the band correlation matrix, and obtain eigenvectors and corresponding eigenvalues through diagonalization operations. The magnitude of the eigenvalues reflects the contribution degree of each band to the overall spectral characteristics. Calculate the contribution rate for the obtained eigenvalues, divide each eigenvalue by the sum of the eigenvalues to obtain the relative contribution rate of each eigenvalue, sort them according to the contribution rate, and select the principal component data corresponding to the eigenvalues with a cumulative contribution rate reaching 95%.

[0032] Normalize the wavelength values in the principal component data, convert the values of each wavelength point to the range of 0-1, and eliminate the influence of scale differences in different wavelength intervals. Perform interval screening on the standardized wavelength data, set the screening threshold to 0.8, and retain the wavelength intervals with a correlation coefficient greater than 0.8 as the characteristic band data, mainly including two intervals of 1450-1650nm and 1900-2100nm. Use high performance liquid chromatography to verify the characteristic band data. The chromatographic conditions are a C18 reverse phase chromatographic column, the mobile phase is acetonitrile-water (25:75), the flow rate is 1.0 mL / min, and the detection wavelength is 280nm. Obtain the content of tea polyphenols in the sample as the reference data through chromatographic analysis, perform linear regression analysis on the reference content data and the characteristic band data, establish the quantitative relationship between the two, and calculate the correlation coefficient.

[0033] Perform interval calculation on the correlation coefficient data at a 95% confidence level, determine the fluctuation range of the correlation coefficient, and evaluate the reliability of the established quantitative relationship. When the lower limit of the confidence interval of the correlation coefficient is greater than 0.95, it is determined that the quantitative relationship has good reliability, and finally obtain the quantitative relationship data of the tea polyphenol content.

[0034] For example: Take the near-infrared spectral data of a batch of tea samples for analysis. In the interval of 1450-1650nm, take a wavelength point every 2nm, for a total of 100 data points. Calculate the correlation coefficient between adjacent wavelength points to obtain a 100×100 correlation matrix. Through eigenvalue decomposition, the first three eigenvalues are 156.8, 42.3, and 15.6 respectively, and the corresponding contribution rates are 73.2%, 19.7%, and 7.1%, and the cumulative contribution rate reaches 92.9%. After normalizing the wavelength data, 85 wavelength points with a correlation coefficient greater than 0.8 are selected in the interval of 1450-1650nm. The content of tea polyphenols in this sample is measured by high performance liquid chromatography to be 15.6%. Perform linear regression analysis with the selected characteristic band data, and obtain a correlation coefficient of 0.978. At a 95% confidence level, the confidence interval of the correlation coefficient is 0.965-0.991, indicating that the quantitative relationship has good stability and reliability. The relative standard deviation of repeated measurements in the whole analysis process is less than 2%, meeting the requirements of quantitative analysis.

[0035] In a specific embodiment, the process of executing step S104 may specifically include the following steps: (1) Design a temperature gradient in the range of 60-90°C according to the quantitative relationship data of the tea polyphenol content to obtain temperature gradient data, and perform a single factor analysis on the temperature gradient data to obtain temperature influence data; (2) Divide the extraction time into time intervals to obtain time gradient data, and perform a single factor analysis on the time gradient data to obtain time influence data; (3) Conduct a proportional range test on the solid-to-liquid ratio to obtain solid-to-liquid ratio gradient data, and perform a single-factor analysis on the solid-to-liquid ratio gradient data to obtain solid-to-liquid ratio influence data; (4) Design a concentration gradient for the ethanol concentration to obtain concentration gradient data, and perform a single-factor analysis on the concentration gradient data to obtain parameter range data; (5) Conduct a three-level Box-Behnken experimental design on the parameter range data to obtain experimental matrix data, and perform an analysis of variance on the experimental matrix data to obtain interaction effect data; (6) Conduct a response surface analysis on the interaction effect data to obtain response value data, and perform a region confirmation on the response value data to obtain process parameter combinations.

[0036] Specifically, design an extraction temperature range based on the quantitative relationship data of tea polyphenol content, set 7 temperature gradient points at 5°C intervals between 60°C and 90°C: 60°C, 65°C, 70°C, 75°C, 80°C, 85°C, 90°C. Conduct three parallel experiments at each temperature point, with other conditions fixed as: extraction time 60 min, solid-to-liquid ratio 1:20, ethanol concentration 60%. Detect the tea polyphenol content at each temperature point by near-infrared spectroscopy, calculate the average value and standard deviation, and perform a single-factor analysis of variance to obtain the influence law of temperature on the extraction rate of tea polyphenols. Divide the extraction time into intervals, set 6 time points within the range of 30 - 120 min: 30 min, 45 min, 60 min, 75 min, 90 min, 120 min. Fix other extraction conditions as extraction temperature 75°C, solid-to-liquid ratio 1:20, ethanol concentration 60%. Repeat the experiment three times at each time point, record the tea polyphenol content data at each time point, and determine the influence degree of the time factor on the extraction effect through a single-factor analysis of variance.

[0037] The solid-to-liquid ratio range test sets 5 levels between 1:10 and 1:30: 1:10, 1:15, 1:20, 1:25, 1:30. Other conditions are fixed as extraction temperature 75°C, time 60 min, ethanol concentration 60%. Conduct three repeated experiments at each solid-to-liquid ratio level, collect near-infrared spectroscopy data and convert it into tea polyphenol content, and calculate the extraction effect difference under different solid-to-liquid ratio conditions. The ethanol concentration gradient design sets 5 concentration points within the range of 40% - 80%: 40%, 50%, 60%, 70%, 80%. Fix other conditions as temperature 75°C, time 60 min, solid-to-liquid ratio 1:20. Conduct three parallel experiments at each concentration point, record the tea polyphenol content data, and determine the optimal concentration range through a single-factor analysis.

[0038] Based on the results of single-factor experiments, three levels were selected for Box-Behnken experimental design. A total of 29 experimental points were required for four factors at three levels, including 5 replicates of the central point. The three levels of each factor in the experimental matrix were: temperature 70°C, 75°C, 80°C; time 45 min, 60 min, 75 min; solid-liquid ratio 1:15, 1:20, 1:25; ethanol concentration 50%, 60%, 70%. The tea polyphenol content was recorded as the response value for each experimental point, and the main effects and interaction effects of each factor were calculated through analysis of variance.

[0039] A quadratic polynomial regression equation was established for the interaction effect data obtained from the analysis of variance. Three-dimensional response surface plots and contour plots were drawn through response surface analysis software to analyze the interaction relationships between factors and determine the optimal process parameter combination.

[0040] For example: In the single-factor experiment of temperature for a certain batch of tea samples, the tea polyphenol extraction rate data in the range of 60°C - 90°C were 12.5%, 13.8%, 15.2%, 16.5%, 16.8%, 16.7%, 16.3% respectively. Analysis of variance showed that the temperature factor was significant (P < 0.05), and the optimal temperature range was 75 - 80°C. In the single-factor experiment of time, the extraction rates in the range of 30 - 120 min were 13.2%, 15.1%, 16.5%, 16.7%, 16.6%, 16.5%, and the optimal extraction time was 60 - 75 min. The experimental results of the solid-liquid ratio showed that the extraction rate was the highest under the condition of 1:20, reaching 16.5%. The extraction effect was the best when the ethanol concentration was 60%. Substituting these parameters into the Box-Behnken experimental design, the response values of the 29 experimental points were in the range of 12.5% - 16.8%. Through analysis of variance and response surface optimization, the optimal process parameter combination was determined as: extraction temperature 78°C, time 65 min, solid-liquid ratio 1:18, ethanol concentration 65%. Under this condition, a verification experiment was carried out, and the tea polyphenol extraction rate reached 16.9%, with a relative standard deviation of less than 2%.

[0041] In a specific embodiment, the process of executing step S105 may specifically include the following steps: (1) Monitor the accuracy of the temperature parameter in the process parameter combination to obtain temperature monitoring data, and sample and record the temperature monitoring data every 30 seconds to obtain temperature sampling data; (2) Monitor the accuracy of the time parameter in the process parameter combination to obtain time monitoring data, and record the time monitoring data to obtain time sampling data; (3) Monitor the accuracy of the solid-liquid ratio parameter in the process parameter combination to obtain solid-liquid ratio monitoring data, and record the solid-liquid ratio monitoring data to obtain solid-liquid ratio sampling data; (4)Perform near-infrared spectral scanning on the extraction process according to the quantitative relationship data of the tea polyphenol content to obtain process spectral data, and merge the process spectral data to obtain process parameter data; (5)Calculate the deviation of the process parameter data to obtain process deviation data, and perform threshold judgment on the process deviation data to obtain process judgment data; (6)Perform zonal statistics on the process judgment data to obtain zonal data, and perform interval analysis on the zonal data to obtain process control data.

[0042] Specifically, a PT100 temperature sensor with an accuracy of ±0.1 °C is used. The sensor probe is placed at the center of the extraction solution, and the temperature value is recorded every 30 seconds. During the 65-minute extraction process, 130 temperature data points are accumulated to form a temperature sampling data sequence. The extraction time is monitored using a high-precision timer with an accuracy of ±1 s. From the start of heating to the end of extraction, the time point is recorded every minute, and the corresponding process state (including heating stage, constant-temperature extraction stage, cooling stage) is marked. A total of 65 time data points are recorded throughout the process to form a time sampling data sequence.

[0043] The monitoring of the solid-liquid ratio parameter records the change in the solution mass in real time through an electronic balance with an accuracy of ±0.01 g. The mass data is recorded every minute, and the real-time solid-liquid ratio value is calculated. A total of 65 solid-liquid ratio data points are collected throughout the process to form a solid-liquid ratio sampling data sequence. The calculation of the solid-liquid ratio takes into account the compensation for solvent volatilization and sampling loss. During the extraction process, according to the established quantitative relationship data of the tea polyphenol content, the extraction solution is scanned in real time by a near-infrared spectrometer. The scanning wavelength range is 800 - 2500 nm, and a full-spectrum scan is performed every 5 minutes to obtain 13 groups of process spectral data. These spectral data are integrated with the temperature, time, and solid-liquid ratio sampling data in chronological order to form a complete set of process parameter data.

[0044] Perform deviation calculation and threshold judgment on the process parameter data. The deviation calculation uses the following formula: ; Where: is the real-time deviation value of the process parameter (%), is the real-time measured value, is the process set value.

[0045] Calculation of the root mean square error of the process parameter: ; Where RMSD is the root mean square deviation of the parameter, is the i-th measured value, is the mean value of the measurements, and n is the number of measurements.

[0046] Process control limit calculation: ; Wherein, is the control limit, is the control coefficient (taking the value of 3), is the standard deviation.

[0047] For example: During the extraction process of a certain batch of tea, the extraction temperature is set at 78°C, the time is 65 minutes, and the solid-liquid ratio is 1:18. During the heating stage (0 - 15 minutes), the temperature rises from room temperature to 78°C. 30 temperature data points are recorded, with a heating rate of 2.5°C / min. The maximum deviation occurs at the 8th minute, and the measured value is 75.2°C. Calculate the relative deviation: |(75.2 - 78.0)| / 78.0 × 100% = 3.59%, exceeding the control threshold of 2%, triggering temperature compensation control. For the 80 temperature data points in the constant-temperature extraction stage (15 - 55 minutes), the average value is 77.8°C, the standard deviation is 0.3°C, and the root-mean-square deviation is 0.28°C. Near-infrared spectroscopy data shows that the absorbance at 1450 nm gradually rises from 0.857 to 0.892, corresponding to the tea polyphenol concentration rising from 12.5% to 16.8%. The solid-liquid ratio monitoring data shows that the solvent loss rate is 0.8% / h, and the solid-liquid ratio is kept stable within the range of 1:18 ± 0.2 through the automatic liquid replenishment system. Finally, the control parameters of each stage are determined through zonal statistics: the temperature slope in the heating stage is 2.5 ± 0.2°C / min, the temperature fluctuation range in the constant-temperature stage is ±0.5°C, and the temperature drop rate in the cooling stage is 1.8 ± 0.2°C / min.

[0048] In a specific embodiment, the process of executing step S106 may specifically include the following steps: (1) Perform parameter zonal processing on the process control data to obtain zonal control data, and perform temperature deviation correction on the zonal control data to obtain temperature correction data; (2) Perform time compensation calculation on the temperature correction data to obtain time compensation data, and perform solid-liquid ratio calibration on the time compensation data to obtain solid-liquid calibration data; (3) Perform solvent concentration calibration on the solid-liquid calibration data to obtain concentration calibration data, and perform parameter combination analysis on the concentration calibration data to obtain optimized process parameters; (4) Conduct three batches of parallel tests on the optimized process parameters to obtain parallel test data, and perform relative standard deviation calculation on the parallel test data to obtain standard deviation data; (5) Perform confidence interval calculation on the standard deviation data to obtain interval confidence data, and perform reliability verification on the interval confidence data to obtain candidate process parameters; (6) Conduct a stability assessment on the candidate process parameters to obtain stability data, and perform parameter confirmation on the stability data to obtain the target process parameters.

[0049] Specifically, partition the process control data. Divide the extraction process into three regions: heating region (0 - 15 min), constant temperature region (15 - 55 min), and cooling region (55 - 65 min). Set different control parameters for each region: the temperature rising rate in the heating region is 2.5 °C / min, the temperature fluctuation range in the constant temperature region is ±0.5 °C, and the temperature falling rate in the cooling region is 1.8 °C / min. Perform temperature deviation correction on the partitioned control data, using the proportional-integral-derivative (PID) control algorithm with a proportional coefficient Kp = 1.2, an integral time Ti = 60 s, and a derivative time Td = 15 s to achieve a temperature control accuracy of ±0.1 °C. The time compensation calculation for the temperature correction data is based on the extraction kinetic model, considering the coupling relationship between temperature and reaction time. When the temperature deviation is positive, correspondingly shorten the extraction time; when the temperature deviation is negative, extend the extraction time. The compensation coefficient is set to 2 min time adjustment for every 1 °C temperature deviation. Calibrate the solid-to-liquid ratio for the data after time compensation. Considering the solvent evaporation loss, determine the liquid addition amount through mass balance calculation to control the solid-to-liquid ratio fluctuation within ±0.2 range. The solvent concentration calibration for the solid-to-liquid ratio calibration data uses the density method. Measure the solution density with a standard densitometer, and convert the ethanol concentration by combining with the standard curve to control the concentration deviation not exceeding ±1%. Combine and analyze the calibrated process parameters according to the orthogonal design plan to determine the preliminary optimized process parameters: extraction temperature 78 °C, time 65 min, solid-to-liquid ratio 1:18, and ethanol concentration 65%.

[0050] Carry out three batches of parallel experiments with the above optimized process parameters, sampling 5 times for each batch, and detecting the tea polyphenol content by near-infrared spectroscopy. Calculate the relative standard deviation (RSD) of the parallel experiment data. The RSD calculation formula is the standard deviation divided by the average value and then multiplied by 100%, and it is required that the RSD does not exceed 2%. Conduct an interval estimation for the standard deviation data at a 95% confidence level to determine the fluctuation range of each parameter: temperature ±0.5 °C, time ±2 min, solid-to-liquid ratio ±0.3, and ethanol concentration ±2%. Verify the reliability of the interval confidence data through a t-test with a significance level of 0.05 and a degree of freedom of 8. Look up the critical value t0.05(8) = 2.306 in the table. Calculate that the t-value of each parameter is less than the critical value, indicating that the data has statistical significance. Determine the candidate process parameters for the parameters that pass the verification. Conduct a continuous 5-batch stability assessment on the candidate process parameters and calculate the batch-to-batch coefficient of variation, requiring that the coefficient of variation is less than 3%.

[0051] For example, during the extraction process of a certain batch of tea leaves, the initial temperature is 25°C, which is heated to 78°C at a rate of 2.5°C / min. The output value of the PID controller is 45%, the actual temperature is 77.8°C, the deviation is -0.2°C, and the trigger time compensation is 0.4 min. The solvent loss is 0.15 kg / h, 0.15 kg of 65% ethanol solution is added, and the solid-liquid ratio is maintained at 1:18.1. The tea polyphenol contents in three batches of parallel tests are 16.8%, 16.7%, and 16.9% respectively. The calculated RSD = 0.61%. The calculation results of the 95% confidence interval are as follows: temperature 77.8 ± 0.3°C, time 65 ± 1.5 min, solid-liquid ratio 1:18 ± 0.2, and ethanol concentration 65 ± 1%. The t-test value is 1.865, which is less than the critical value of 2.306, and the verification is passed. The coefficient of variation of the 5-batch stability assessment is 1.8%, and all parameters meet the process requirements, which are determined as the target process parameters.

[0052] The above describes the method for optimizing the detection and extraction process of tea polyphenols based on near-infrared spectroscopy in the embodiments of the present application. Next, the device for optimizing the detection and extraction process of tea polyphenols based on near-infrared spectroscopy in the embodiments of the present application will be described. Please refer to Figure 2 , an embodiment of the device for optimizing the detection and extraction process of tea polyphenols based on near-infrared spectroscopy in the embodiments of the present application includes: A calibration module 201, configured to perform drying and pulverization processing on a tea leaf sample to obtain a tea powder sample, perform particle size screening and temperature and humidity balance processing on the tea powder sample to obtain a standardized sample, and perform near-infrared spectroscopy system calibration on the standardized sample to obtain calibration spectral parameters; An adjustment module 202, configured to perform quantitative sample loading and scanning angle adjustment on the standardized sample to obtain a sample to be measured, perform near-infrared spectroscopy collection on the sample to be measured and the calibration spectral parameters to obtain original spectral data, and perform signal smoothing and scattering correction processing on the original spectral data to obtain preprocessed spectral data; An analysis module 203, configured to perform characteristic wavelength extraction and correlation analysis on the preprocessed spectral data to obtain characteristic band data, and perform calibration and verification processing on the characteristic band data to obtain quantitative relationship data of tea polyphenol content; A design module 204, configured to perform single-factor experimental design on the extraction temperature, extraction time, solid-liquid ratio, and solvent concentration according to the quantitative relationship data of tea polyphenol content to obtain parameter range data, and perform response surface experimental design on the parameter range data to obtain process parameter combinations; A monitoring module 205, configured to perform online monitoring on the process parameter combinations and the quantitative relationship data of tea polyphenol content to obtain process parameter data, and perform real-time analysis and processing on the process parameter data to obtain process control data; A verification module 206 is used to dynamically adjust the process control data to obtain optimized process parameters, and perform repeatability verification on the optimized process parameters to obtain target process parameters.

[0053] Through the collaborative cooperation of the above-mentioned various components, by performing standardized pretreatment on tea samples, including drying, crushing, particle size screening, and temperature and humidity balance treatment, the uniformity and stability of the samples are significantly improved, laying a reliable foundation for subsequent spectral analysis. The water content of the samples is controlled below 5%, the particle size uniformity reaches over 95%, and through the temperature and humidity balance treatment, the samples reach a stable state, effectively eliminating the interference of environmental factors. By calibrating the near-infrared spectroscopy system in combination with quantitative sample loading and scanning angle adjustment, the accuracy and repeatability of spectral acquisition are achieved, and the relative standard deviation of spectral acquisition is controlled within 0.1%. By performing signal smoothing and scattering correction on the original spectral data, the influence of baseline drift and scattering effects is effectively eliminated, the signal-to-noise ratio of the spectral data is improved, and the signal-to-noise ratio reaches over 1000:1. The introduction of characteristic wavelength extraction and correlation analysis realizes the rapid quantitative detection of the tea polyphenol content. The detection time is shortened from several hours of the traditional method to several minutes, and the determination accuracy reaches over 95%. By performing single-factor experimental design and response surface optimization on the extraction process parameters, a quantitative relationship between parameters such as temperature, time, solid-liquid ratio, and solvent concentration is established, and the extraction yield is increased by over 15%. Through on-line monitoring and real-time analysis and processing, the dynamic control of the process is realized, the parameter fluctuation is controlled within the range of ±2%, and the stability and repeatability of the process are significantly improved. Finally, through dynamic adjustment and repeatability verification, the reliability of the process parameters is ensured, the relative standard deviation between batches is controlled within 2%, and an optimization and quality control system for the tea polyphenol extraction process is established.

[0054] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacement on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the present application.

Claims

1. A method for detecting and optimizing the extraction process of tea polyphenols based on near infrared spectroscopy, characterized in that: The method for optimizing tea polyphenol detection and extraction process based on near infrared spectroscopy includes: Drying and crushing tea samples to obtain tea powder samples, screening the tea powder samples for particle size and performing temperature and humidity balance treatment to obtain standardized samples, and calibrating the standardized samples with a near-infrared spectroscopy system to obtain calibrated spectral parameters; The standardized sample is quantitatively loaded and the scanning angle is adjusted to obtain a sample to be tested, near-infrared spectrum collection is performed on the sample to be tested and the calibration spectrum parameters to obtain raw spectrum data, and signal smoothing and scattering correction processing is performed on the raw spectrum data to obtain pre-processed spectrum data; Performing characteristic wavelength extraction and correlation analysis on the pre-processed spectral data to obtain characteristic band data, and performing correction and verification processing on the characteristic band data to obtain quantitative relationship data of tea polyphenol content; According to the quantitative relationship data of tea polyphenol content, single factor experimental design is performed on extraction temperature, extraction time, solid-liquid ratio and solvent concentration to obtain parameter range data, and response surface experimental design is performed on the parameter range data to obtain process parameter combination; Performing online monitoring on the process parameter combination and the quantitative relationship data of the tea polyphenol content to obtain process parameter data, and performing real-time analysis and processing on the process parameter data to obtain process control data; The process control data is dynamically adjusted to obtain optimized process parameters, and the optimized process parameters are repeatedly verified to obtain target process parameters.

2. The method for optimizing tea polyphenols detection and extraction process based on near infrared spectroscopy according to claim 1, characterized in that: The tea sample is dried and crushed to obtain a tea powder sample, the tea powder sample is subjected to particle size screening and temperature and humidity balance treatment to obtain a standardized sample, and the standardized sample is calibrated by a near infrared spectroscopy system to obtain a calibrated spectral parameter, including: The tea sample is subjected to vacuum drying at 60° C. to obtain a dry tea sample, and the dry tea sample is subjected to mechanical pulverization to obtain the tea powder sample; The tea powder sample is sieved through a standard sieve of 80 meshes to obtain a primary sieve sample, and the primary sieve sample is subjected to a constant temperature and humidity treatment at 25±1° C. and a relative humidity of 60±2% to obtain the standardized sample; The standardized sample is sampled by quartering to obtain a portion to be tested, and the portion to be tested is quantitatively weighed to obtain the standardized sample; Performing a barium sulfate reflector background scan on the standardized sample to obtain background spectrum data, and performing wavelength range correction on the background spectrum data to obtain corrected spectrum data; Repeating scanning 32 times to obtain the corrected spectral data, accumulating and averaging the data, and calculating and evaluating the signal-to-noise ratio of the data to obtain reference spectral data. The wavelength accuracy and repeatability of the reference spectrum data are verified to obtain standard spectrum data, and the standard spectrum data are corrected for system deviation to obtain the calibration spectrum parameters.

3. The method for optimizing tea polyphenols detection and extraction process based on near infrared spectroscopy according to claim 1, characterized in that: The method comprises: performing quantitative sample loading and scanning angle adjustment on the standardized sample to obtain a sample to be tested, performing near infrared spectrum acquisition on the sample to be tested and the calibration spectrum parameters to obtain raw spectrum data, and performing signal smoothing and scattering correction processing on the raw spectrum data to obtain preprocessed spectrum data, including: Precision weighing the standardized sample to obtain a quantitative sample, and filling the quantitative sample into a quartz sample cell to obtain a filled sample; Adjusting the incident angle of the loaded sample to obtain an angle calibration sample, and performing sample surface compaction treatment on the angle calibration sample to obtain the sample to be tested; Repeating scanning the sample to be tested and the calibration spectrum parameters 32 times to obtain cumulative spectrum data, and performing repeatability verification on the cumulative spectrum data to obtain the original spectrum data; The original spectral data is subjected to noise elimination by a five-point smoothing method to obtain smoothed spectral data, and the smoothed spectral data is subjected to baseline drift correction to obtain baseline-corrected data; The baseline correction data is subjected to a standard normal variable transformation to obtain scatter correction data, and the scatter correction data is subjected to a multivariate scatter correction to obtain the preprocessed spectral data.

4. The method for optimizing tea polyphenols detection and extraction process based on near infrared spectroscopy according to claim 1, characterized in that: The method of extracting characteristic wavelengths and performing correlation analysis on the pre-processed spectral data to obtain characteristic band data, and performing correction and verification processing on the characteristic band data to obtain quantitative relationship data of tea polyphenol content includes: The pre-processed spectral data is segmented into wavelength intervals to obtain band partition data, and adjacent band correlation calculation is performed on the band partition data to obtain a band correlation matrix; Performing eigenvector analysis on the band correlation matrix to obtain eigenvalue data, and performing contribution rate calculation on the eigenvalue data to obtain principal component data; Performing wavelength normalization processing on the principal component data to obtain standardized wavelength data, and performing interval screening on the standardized wavelength data to obtain the characteristic band data; Performing high performance liquid chromatography verification on the characteristic band data to obtain reference content data, and performing linear regression analysis on the reference content data to obtain correlation coefficient data; A confidence interval calculation is performed on the correlation coefficient data to obtain interval range data, and a reliability assessment is performed on the interval range data to obtain the quantitative relationship data of the tea polyphenol content.

5. The method for optimizing tea polyphenols detection and extraction process based on near infrared spectroscopy according to claim 1, characterized in that: According to the quantitative relationship data of tea polyphenol content, single factor experimental design is performed on extraction temperature, extraction time, solid-liquid ratio and solvent concentration to obtain parameter range data, and response surface experimental design is performed on the parameter range data to obtain process parameter combination, including: Performing a gradient design of the extraction temperature range of 60-90° C. according to the quantitative relationship data of the tea polyphenol content to obtain temperature gradient data, and performing a single factor analysis on the temperature gradient data to obtain temperature influence data; Dividing the extraction time into time intervals to obtain time gradient data, and performing single factor analysis on the time gradient data to obtain time influence data; Performing a ratio range test on the material-liquid ratio to obtain material-liquid ratio gradient data, and performing a single factor analysis on the material-liquid ratio gradient data to obtain material-liquid ratio influence data; Performing concentration gradient design on ethanol concentration to obtain concentration gradient data, and performing single factor analysis on the concentration gradient data to obtain the parameter range data; Performing a three-level Box-Behnken test design on the parameter range data to obtain test matrix data, and performing variance analysis on the test matrix data to obtain interaction effect data; Response surface analysis is performed on the interaction effect data to obtain response value data, and regional confirmation is performed on the response value data to obtain the process parameter combination.

6. The method for optimizing tea polyphenols detection and extraction process based on near infrared spectroscopy according to claim 1, characterized in that: The online monitoring of the process parameter combination and the quantitative relationship data of the tea polyphenols content to obtain process parameter data, and the real-time analysis and processing of the process parameter data to obtain process control data include: Performing precision monitoring on the temperature parameters in the process parameter combination to obtain temperature monitoring data, and sampling and recording the temperature monitoring data every 30 seconds to obtain temperature sampling data; Performing precision monitoring on the time parameter in the process parameter combination to obtain time monitoring data, and recording the time monitoring data to obtain time sampling data; Performing precision monitoring on the material-liquid ratio parameter in the process parameter combination to obtain material-liquid ratio monitoring data, and recording the material-liquid ratio monitoring data to obtain material-liquid ratio sampling data; Performing near infrared spectroscopy scanning on the extraction process according to the quantitative relationship data of tea polyphenol content to obtain process spectral data, and merging the process spectral data to obtain the process parameter data; Performing deviation calculation on the process parameter data to obtain process deviation data, and performing threshold judgment on the process deviation data to obtain process judgment data; Performing partition statistics on the process judgment data to obtain partition data, and performing interval analysis on the partition data to obtain the process control data.

7. The method for optimizing tea polyphenols detection and extraction process based on near infrared spectroscopy according to claim 6, characterized in that: The dynamically adjusting the process control data to obtain optimized process parameters, and repeatably verifying the optimized process parameters to obtain target process parameters, include: Performing parameter partitioning processing on the process control data to obtain partition control data, and performing temperature deviation correction on the partition control data to obtain temperature correction data; Performing time compensation calculation on the temperature correction data to obtain time compensation data, and performing material-liquid ratio calibration on the time compensation data to obtain material-liquid calibration data; Performing solvent concentration calibration on the feed solution calibration data to obtain concentration calibration data, and performing parameter combination analysis on the concentration calibration data to obtain the optimized process parameters; Performing three batches of parallel tests on the optimized process parameters to obtain parallel test data, and performing relative standard deviation calculation on the parallel test data to obtain standard deviation data; Performing confidence interval calculation on the standard deviation data to obtain interval confidence data, and performing reliability verification on the interval confidence data to obtain candidate process parameters; A stability assessment is performed on the candidate process parameters to obtain stability data, and parameter confirmation is performed on the stability data to obtain the target process parameters.

8. A device for detecting tea polyphenols and optimizing extraction process based on near infrared spectroscopy, characterized in that: Used to perform the method for optimizing the detection and extraction process of tea polyphenols based on near infrared spectroscopy according to any one of claims 1 to 7, the device for optimizing the detection and extraction process of tea polyphenols based on near infrared spectroscopy comprises: A calibration module is used to dry and crush tea samples to obtain tea powder samples, perform particle size screening and temperature and humidity balance treatment on the tea powder samples to obtain standardized samples, and perform near-infrared spectroscopy system calibration on the standardized samples to obtain calibration spectral parameters; An adjustment module is used to quantitatively load the standardized sample and adjust the scanning angle to obtain a sample to be tested, collect near-infrared spectra of the sample to be tested and the calibration spectrum parameters to obtain raw spectrum data, and perform signal smoothing and scattering correction processing on the raw spectrum data to obtain pre-processed spectrum data; An analysis module is used to perform characteristic wavelength extraction and correlation analysis on the pre-processed spectral data to obtain characteristic band data, and to perform correction and verification processing on the characteristic band data to obtain quantitative relationship data of tea polyphenol content; A design module is used to perform single factor experimental design on extraction temperature, extraction time, solid-liquid ratio and solvent concentration according to the quantitative relationship data of tea polyphenol content to obtain parameter range data, and perform response surface experimental design on the parameter range data to obtain process parameter combination; A monitoring module, used for online monitoring of the process parameter combination and the quantitative relationship data of the tea polyphenols content to obtain process parameter data, and performing real-time analysis and processing on the process parameter data to obtain process control data; The verification module is used to dynamically adjust the process control data to obtain optimized process parameters, and to repeatably verify the optimized process parameters to obtain target process parameters.

Images