Optimal control method and system for decoction temperature of traditional Chinese medicine based on effective component analysis rate

Abstract

The present application provides a traditional Chinese medicine decoction temperature optimization control method and system based on effective component analysis rate, relates to the technical field of traditional Chinese medicine decooking equipment, and comprises the following steps: obtaining the medicinal material components and dosage of the traditional Chinese medicine to be decocted, combining a pre-configured medicinal material database, and determining the effective components of the traditional Chinese medicine to be decocted; obtaining the temperature data and near-infrared spectrum data of the traditional Chinese medicine to be decocted in the decoction process, combining a dynamic mathematical model constructed in advance through the ecological niche adaptation theory, taking the maximum effective component analysis rate as the objective function, and determining the temperature optimization strategy; based on the temperature optimization strategy, the decoction temperature of the traditional Chinese medicine to be decocted is monitored in real time and the heating power is dynamically adjusted through a closed-loop control system, so as to realize temperature optimization control of traditional Chinese medicine decoction. The present application can quickly respond to changes in decoction temperature, realize precise and stable temperature optimization control, ensure efficient and reliable quality in the traditional Chinese medicine decoction process, and fully exert the medicinal effect of medicinal materials.

Description

Technical Field

[0001] This invention relates to the field of traditional Chinese medicine decoction equipment technology, and more specifically, to a method and system for optimizing the control of decoction temperature based on the effective component extraction rate. Background Technology

[0002] Medicinal herbs, whether in powder form or in packets, require multiple processing steps to be prepared into a medicinal product. Taking the preparation of a decoction as an example, the first step is to precisely prepare the powdered or packetized medicinal herbs, which are then placed in a special container for decoction. During the decoction process, the temperature and time settings are crucial, as different combinations significantly affect the quality of the decoction and the extraction of medicinal components. After decoction, the resulting liquid must be carefully collected to form the final, drinkable decoction. The entire process is interconnected, with each step requiring strict control to ensure the decoction fully exerts its therapeutic effects, providing patients with a safe and effective traditional Chinese medicine preparation.

[0003] In the field of traditional Chinese medicine decoction equipment technology, medicinal slices or medicine packets need to undergo multiple steps to be processed into decoctions, among which temperature control during the decoction process is crucial. However, current related technologies have significant problems: on the one hand, after determining the effective components of traditional Chinese medicine, it is difficult to accurately formulate a decoction temperature control strategy, and it cannot fully consider the characteristics and interrelationships of each effective component in the decoction system, resulting in the inability to optimize with the goal of maximizing the extraction rate of effective components; on the other hand, after obtaining temperature data and near-infrared spectral data during the decoction process, the near-infrared spectral data is easily affected by temperature interference and lacks effective correction methods, making the prediction of the extraction rate of effective components based on spectral data inaccurate, thus affecting the formulation of temperature optimization strategies.

[0004] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention

[0005] In view of this, the present invention provides a method and system for optimizing the control of the decoction temperature of traditional Chinese medicine based on the effective component extraction rate, so as to solve the above-mentioned problems.

[0006] To solve the above problems, the specific technical solution adopted by the present invention is as follows:

[0007] According to one aspect of the present invention, a method for optimizing and controlling the decoction temperature of traditional Chinese medicine based on the extraction rate of effective components is provided, comprising the following steps:

[0008] S1. Obtain the medicinal ingredients and dosage of the Chinese medicine to be decocted, and determine the effective ingredients of the Chinese medicine to be decocted by combining it with a pre-configured medicinal material database.

[0009] S2. Obtain temperature data and near-infrared spectral data of the Chinese medicine to be decocted during the decoction process, and combine them with a dynamic mathematical model pre-constructed based on the ecological niche adaptation theory to determine the temperature optimization strategy with the objective function of maximizing the effective component extraction rate.

[0010] S3. Based on the temperature optimization strategy, the closed-loop control system monitors the decoction temperature of the Chinese medicine to be decocted in real time and dynamically adjusts the heating power to achieve optimized temperature control of the Chinese medicine decoction.

[0011] Preferably, the step of acquiring temperature data and near-infrared spectral data of the Chinese herbal medicine to be decocted during the decoction process, and combining them with a dynamic mathematical model pre-constructed based on niche adaptation theory, to determine the temperature optimization strategy with the objective function of maximizing the extraction rate of effective components, includes the following steps:

[0012] S21. During the decoction process of the Chinese medicine to be decocted, temperature data and near-infrared spectral data are collected simultaneously and standardized respectively to obtain standardized temperature data and standardized near-infrared spectral data.

[0013] S22. Based on the orthogonal signal correction method, and combined with standardized temperature data, the standardized near-infrared spectral data is compensated and corrected to obtain the corrected near-infrared spectral data.

[0014] S23. Based on calibrated near-infrared spectral data, a pre-trained quantitative inversion model is used to predict the current precipitation concentration of the active ingredients in real time, and the cumulative precipitation rate is calculated by integrating the decoction time.

[0015] S24. The real-time precipitation rate is used as a state feedback input to the dynamic mathematical model constructed based on the ecological niche adaptation theory. The objective function is to maximize the effective precipitation rate. The rolling time domain optimization method is used to solve for the optimal temperature setpoint for the future preset period, thus forming a temperature optimization strategy.

[0016] Preferably, the step of performing compensation and correction processing on the standardized near-infrared spectral data based on the orthogonal signal correction method and combined with standardized temperature data to obtain corrected near-infrared spectral data includes the following steps:

[0017] S221. Construct a near-infrared spectral matrix and a temperature compensation matrix based on standardized near-infrared spectral data and standardized temperature data, respectively.

[0018] S222. Based on the orthogonality between the temperature compensation matrix and the near-infrared spectral matrix, calculate the temperature-related orthogonal components and construct the orthogonal correction vector.

[0019] S223. The near-infrared spectral matrix is ​​iteratively updated using orthogonal correction vectors, and the corrected near-infrared spectral data is generated by combining derivative processing.

[0020] Preferably, the step of calculating the temperature-related orthogonal components and constructing the orthogonal correction vector based on the orthogonality of the temperature compensation matrix and the near-infrared spectral matrix includes the following steps:

[0021] S2221. Perform principal component analysis on the near-infrared spectral matrix and extract the principal component score vector.

[0022] S2222. Orthogonalize the principal component score vector with the temperature compensation matrix to remove temperature-related interference signals and obtain temperature-related orthogonal components.

[0023] S2223. Perform convergence judgment on temperature-related orthogonal components, and construct an orthogonal correction vector by combining the regression coefficients when the preset convergence conditions are met.

[0024] Preferably, the step of determining the convergence of the temperature-related orthogonal components and constructing an orthogonal correction vector by combining the regression coefficients when the preset convergence condition is met includes the following steps:

[0025] S22231. Calculate the change in variance contribution rate of the temperature-related orthogonal components and compare the change in variance contribution rate with a preset threshold. If the change in variance contribution rate is greater than the preset threshold, the preset convergence condition is not met, and return to step S2221. Otherwise, the preset convergence condition is met, and step S22232 is executed.

[0026] S22232. Using the temperature compensation matrix as the independent variable and the converged temperature-related orthogonal components as the dependent variable, the least squares method is used to calculate the regression coefficient vector.

[0027] S22233. Construct an orthogonal correction vector using the regression coefficient vector and the temperature compensation matrix.

[0028] Preferably, the step of iteratively updating the near-infrared spectral matrix using orthogonal correction vectors and combining derivative processing to generate corrected near-infrared spectral data includes the following steps:

[0029] S2231. Based on the orthogonal correction vector and temperature compensation matrix, calculate the temperature interference projection component to be removed, and subtract the projection component from the near-infrared spectral matrix to obtain the preliminary correction spectral matrix.

[0030] S2232. Determine whether the current iteration number has reached the preset maximum number of principal factors. If not, use the preliminary corrected spectral matrix as the new near-infrared spectral matrix and return to step S2221 for iterative update until the iteration number reaches the preset maximum number of principal factors. Otherwise, use the obtained preliminary corrected near-infrared spectral matrix as the final converged corrected near-infrared spectral matrix.

[0031] S2233. Perform first-order derivative transformation on the finally converged calibration spectral matrix to obtain the calibrated near-infrared spectral data.

[0032] Preferably, the step of using the real-time precipitation rate as state feedback input to a dynamic mathematical model constructed based on niche adaptation theory, with maximizing the effective component precipitation rate as the objective function, and employing a rolling time-domain optimization method to solve for the optimal temperature setpoint for a future preset period, thereby forming a temperature optimization strategy, includes the following steps:

[0033] S241. Obtain historical decoction experiment data and construct a dynamic mathematical model based on the ecological niche adaptation theory;

[0034] S242. Using the current real-time precipitation rate as the initial state of the dynamic mathematical model, predict the dynamic precipitation process of the effective components under different temperature trajectories within the preset prediction time domain.

[0035] S243. Taking the maximization of the cumulative effective component extraction rate at the end of the prediction period as the objective function, and combining the niche constraints of the effective components, the optimal temperature control sequence in the prediction time domain is solved to obtain the temperature optimization strategy.

[0036] Preferably, the step of acquiring historical decoction experiment data and constructing a dynamic mathematical model based on niche adaptation theory includes the following steps:

[0037] S2411. Obtain historical decoction experiment data, including historical temperature data, historical effective component data, decoction time data, and related process parameters;

[0038] S2412. Preprocess the historical decoction experiment data, and calculate the niche parameters of each effective component based on the preprocessed historical decoction experiment data using the niche adaptation theory.

[0039] S2413. The niche parameters and dissolution kinetics of each effective component are integrated and processed, and a dynamic mathematical model is constructed.

[0040] Preferably, the preprocessing of historical decoction experiment data, and the calculation of the niche parameters of each effective component based on the preprocessed historical decoction experiment data using niche adaptation theory, includes the following steps:

[0041] S24121. Perform preprocessing on the historical decoction experiment data, including data cleaning, data standardization and data alignment.

[0042] S24122. Based on the pre-processed historical decoction experimental data, the niche center was calculated by Gaussian fitting; the niche width was calculated by the Levins index; and the niche overlap was calculated by the Pianka index.

[0043] S24123. Integrate the niche center, niche width, and niche overlap to obtain the niche parameters of each effective component.

[0044] According to another aspect of the present invention, a traditional Chinese medicine decoction temperature optimization control system based on the effective component extraction rate is provided, the system comprising:

[0045] The data acquisition module is used to acquire the medicinal ingredients and dosage of the Chinese medicine to be decocted, and to determine the effective components of the Chinese medicine to be decocted by combining it with a pre-configured medicinal material database.

[0046] The temperature optimization module is used to acquire temperature data and near-infrared spectral data of the Chinese medicine to be decocted during the decoction process, and combine them with a dynamic mathematical model pre-constructed based on the ecological niche adaptation theory to determine the temperature optimization strategy with the objective function of maximizing the effective component extraction rate.

[0047] The temperature adjustment module is used to monitor the decoction temperature of the Chinese medicine to be decocted in real time and dynamically adjust the heating power based on the temperature optimization strategy through a closed-loop control system, so as to achieve optimized temperature control of the decoction of Chinese medicine.

[0048] The beneficial effects of this invention are as follows:

[0049] 1. This invention utilizes near-infrared spectral data and a dynamic model based on niche adaptation theory to formulate a temperature optimization strategy. It can scientifically consider the characteristics and interrelationships of components, aiming to maximize the extraction rate of components and improve the quality of traditional Chinese medicine. In terms of control effect, the closed-loop control system monitors and dynamically adjusts the heating power in real time, which can quickly respond to changes in decoction temperature, achieve precise and stable temperature optimization control, ensure that the decoction process of traditional Chinese medicine is efficient and reliable, and give full play to the efficacy of medicinal materials.

[0050] 2. This invention constructs a near-infrared spectral matrix and a temperature compensation matrix, and utilizes orthogonality to calculate temperature-related orthogonal components and construct orthogonal correction vectors. This enables precise removal of temperature interference signals and improves the quality of spectral data. By assessing the convergence of temperature-related orthogonal components and constructing orthogonal correction vectors appropriately, the accuracy and scientific validity of the correction are ensured. Iteratively updating the near-infrared spectral matrix using orthogonal correction vectors and combining derivative processing to generate corrected data effectively eliminates noise and interference, highlighting effective spectral features. This provides a more accurate and reliable data foundation for subsequent analysis of traditional Chinese medicine components based on spectral data, contributing to improved precision in the quality control and analysis of traditional Chinese medicine.

[0051] 3. This invention utilizes niche adaptation theory to calculate niche parameters, enabling in-depth analysis of the ecological characteristics of each effective component in the decoction system. By integrating this with dissolution kinetics to construct a dynamic mathematical model, it can scientifically simulate the dynamic precipitation process of components. With real-time precipitation rate feedback and using a rolling time-domain optimization method to solve for the optimal temperature setpoint, it forms an optimization strategy that can adapt to changes in decoction in real time. With the goal of maximizing the extraction rate of components, it fully considers niche constraints, accurately formulates temperature control schemes, and effectively improves the quality of traditional Chinese medicine decoction and the extraction efficiency of effective components. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly described below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0053] Figure 1 This is a flowchart of a method for optimizing the temperature control of traditional Chinese medicine decoction based on the effective component extraction rate according to an embodiment of the present invention.

[0054] Figure 2 This is a block diagram illustrating the principle of a traditional Chinese medicine decoction temperature optimization control system based on the effective component extraction rate according to an embodiment of the present invention.

[0055] In the picture:

[0056] 1. Data acquisition module; 2. Temperature optimization module; 3. Temperature adjustment module. Detailed Implementation

[0057] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this application.

[0058] According to embodiments of the present invention, a method and system for optimizing and controlling the decoction temperature of traditional Chinese medicine based on the effective component extraction rate are provided.

[0059] The present invention will now be further described in conjunction with the accompanying drawings and specific embodiments, such as... Figure 1 As shown, according to an embodiment of the present invention, a method for optimizing and controlling the decoction temperature of traditional Chinese medicine based on the effective component extraction rate is provided, comprising the following steps:

[0060] S1. Obtain the medicinal ingredients and dosage of the Chinese medicine to be decocted, and determine the effective ingredients of the Chinese medicine to be decocted by combining it with a pre-configured medicinal material database.

[0061] It should be noted that when obtaining the ingredients and dosage of medicinal materials, a complete list of medicinal materials to be decocted and their corresponding dosages can be obtained through the Chinese medicine prescription input system, such as 30g of Astragalus membranaceus, 15g of Angelica sinensis, and 10g of Atractylodes macrocephala.

[0062] The pre-configured medicinal herb database is a knowledge base linking single medicinal herbs, active ingredients, and efficacy characteristics. It is constructed in advance through literature review and experimental verification (such as high-performance liquid chromatography / HPLC detection) and is regularly updated. Its core contains three key types of information, providing a basis for screening active ingredients. The specific determination process is as follows: Based on the name and dosage of the medicinal herb, all chemical components of that herb are extracted from the database, and chemical substances labeled as "active ingredients" are screened. The interactions between components are analyzed in conjunction with the compatibility relationships of multiple medicinal herbs. For example, if the prescription contains both Scutellaria baicalensis and Coptis chinensis, the synergistic antibacterial effect of baicalin and berberine (the active ingredient in Coptis chinensis) is considered. The absolute content of each active ingredient is calculated based on the dosage of the medicinal herb; for example, the content of baicalin in 30g of Scutellaria baicalensis is approximately 30g × 0.5% = 0.15g.

[0063] S2. Obtain temperature data and near-infrared spectral data of the Chinese medicine to be decocted during the decoction process, and combine them with a dynamic mathematical model pre-constructed based on the ecological niche adaptation theory to determine the temperature optimization strategy with the objective function of maximizing the effective component extraction rate.

[0064] In a preferred embodiment, the step of acquiring temperature data and near-infrared spectral data of the Chinese herbal medicine to be decocted during the decoction process, and combining this with a dynamic mathematical model pre-constructed based on niche adaptation theory, to determine the temperature optimization strategy with the objective function of maximizing the extraction rate of effective components, includes the following steps:

[0065] S21. During the decoction process of the Chinese medicine to be decocted, temperature data and near-infrared spectral data are collected simultaneously and standardized respectively to obtain standardized temperature data and standardized near-infrared spectral data.

[0066] It should be noted that temperature data acquisition uses a high-precision temperature sensor (temperature control accuracy ±0.1℃), and the sensor probe needs to be directly inserted into the decoction of traditional Chinese medicine; near-infrared spectral data acquisition uses a portable ground object spectrometer (wavelength range 350~2500nm, spectral resolution ≤3nm), and the spectrometer probe needs to be kept perpendicular to the surface of the decoction during acquisition, with a fixed distance of 15~20cm.

[0067] Standardization processing includes noise processing and handling of inconsistencies in dimensions.

[0068] S22. Based on the orthogonal signal correction method, and combined with standardized temperature data, the standardized near-infrared spectral data is compensated and corrected to obtain the corrected near-infrared spectral data.

[0069] In a preferred embodiment, the step of performing compensation and correction processing on the standardized near-infrared spectral data based on the orthogonal signal correction method and combined with standardized temperature data to obtain corrected near-infrared spectral data includes the following steps:

[0070] S221. Construct a near-infrared spectral matrix and a temperature compensation matrix based on standardized near-infrared spectral data and standardized temperature data, respectively.

[0071] Specifically, when constructing the near-infrared spectral matrix, the standardized near-infrared spectral data can be aligned according to timestamps to construct an m×n dimensional matrix (where m is the number of cooking time points and n is the number of wavelength points). For example, if the cooking time is 2 hours and the spectrum is collected once per minute, the matrix dimension is 120×n, and each element corresponds to the absorbance value at a specific time point and at a specific wavelength.

[0072] When constructing the temperature compensation matrix, a p×q dimensional matrix can be built based on standardized temperature data (where p is the number of time points and q is the number of temperature-related variables). For example, it can include derived variables such as the main temperature value, temperature gradient (e.g., heating rate), and temperature fluctuation variance to quantify the combined effect of temperature on the spectrum.

[0073] S222. Based on the orthogonality between the temperature compensation matrix and the near-infrared spectral matrix, calculate the temperature-related orthogonal components and construct the orthogonal correction vector.

[0074] In a preferred embodiment, the step of calculating the temperature-related orthogonal components and constructing the orthogonal correction vector based on the orthogonality of the temperature compensation matrix and the near-infrared spectral matrix includes the following steps:

[0075] S2221. Perform principal component analysis on the near-infrared spectral matrix and extract the principal component score vector.

[0076] It should be noted that Principal Component Analysis (PCA) projects the original near-infrared spectral data into a new coordinate system (principal component space) through orthogonal transformation. The first principal component captures the largest variance, the second principal component captures the second largest variance, and so on. For example, if the spectral matrix has a dimension of 120×500 (120 time points × 500 wavelengths), PCA can compress it into a 120×k-dimensional score matrix (k being the number of principal components), retaining more than 95% of the original variance. Specifically, the mean is subtracted from each column (wavelength) of the near-infrared spectral matrix to obtain a centered matrix. The covariance matrix of the centered matrix is ​​calculated, and eigenvalues ​​and eigenvectors (i.e., principal component directions) are obtained through eigenvalue decomposition or singular value decomposition (SVD). The dot product of the original near-infrared spectral matrix and the principal component direction vectors generates the score vector.

[0077] S2222. Orthogonalize the principal component score vector with the temperature compensation matrix to remove temperature-related interference signals and obtain temperature-related orthogonal components.

[0078] It should be noted that orthogonal operations are performed on the row dimensions (time dimension) of the score vector and the temperature compensation matrix to ensure that the dot product of the processed vector and all column vectors of the temperature compensation matrix is ​​0; this can be achieved through the projection formula: Orthogonalization is achieved, where t2 is the temperature-independent orthogonalized principal component vector, t1 is the principal component score vector, and I is an m×m identity matrix. This is the transpose of the temperature compensation matrix; the formula essentially transforms... Project onto a subspace orthogonal to Y, and peel off The components linearly related to Y are obtained; the final vector is the temperature-related orthogonal component, each element of which only reflects the interference of temperature change on the spectrum and is unrelated to the effective component signal.

[0079] S2223. Perform convergence judgment on temperature-related orthogonal components, and construct an orthogonal correction vector by combining the regression coefficients when the preset convergence conditions are met.

[0080] In a preferred embodiment, the step of determining the convergence of the temperature-related orthogonal components and constructing an orthogonal correction vector by combining the regression coefficients when the preset convergence condition is met includes the following steps:

[0081] S22231. Calculate the change in variance contribution rate of the temperature-related orthogonal components and compare the change in variance contribution rate with a preset threshold. If the change in variance contribution rate is greater than the preset threshold, the preset convergence condition is not met, and return to step S2221. Otherwise, the preset convergence condition is met, and step S22232 is executed.

[0082] It should be noted that the variance contribution rate refers to the proportion of the variance of the temperature-related orthogonal component to the total variance of the original near-infrared spectral matrix. It reflects the explanatory power of this orthogonal component (temperature interference) on the variation in spectral data. The higher the variance contribution rate, the more complete the temperature interference information contained in that component. The change in the variance contribution rate (ΔR) 2 The variance contribution rate of the temperature-related orthogonal components obtained in this orthogonalization process is the difference between the result of the previous orthogonalization process.

[0083] Furthermore, in the first iteration (k=1), since only one orthogonalization process is performed and there is no historical comparison data, the variance contribution rate of this component can be calculated by default. ),like ≥85% (indicating sufficient interference signal extraction) can be directly determined to meet the convergence condition; if If the value is less than 85%, the condition is deemed unsatisfactory, and the process returns to S2221 to re-extract principal components and repeat orthogonalization. For multiple iterations (k≥2), ΔR is calculated. 2 If ΔR 2 The preset threshold indicates that each iteration can still significantly extract new temperature interference information, but the interference has not been fully removed. Therefore, it is necessary to return to S2221 and re-execute the PCA principal component extraction → orthogonalization stripping process until ΔR is reached. 2 ≤Preset threshold.

[0084] S22232. Using the temperature compensation matrix as the independent variable and the converged temperature-related orthogonal components as the dependent variable, the least squares method is used to calculate the regression coefficient vector.

[0085] It should be noted that, assuming that the temperature interference signal and the temperature characteristics satisfy a linear relationship, a regression equation can be constructed: t′2=Y·w+ε, where t′2 is the converged temperature-related orthogonal component vector, w is the regression coefficient vector (dimension k×1), Y is the temperature compensation matrix, and ε is the random error term (which must satisfy normal distribution and homogeneity of variance). The core of the least squares method is to find the optimal regression coefficient vector w that minimizes the sum of squares of the error term ε.

[0086] S22233. Construct an orthogonal correction vector using the regression coefficient vector and the temperature compensation matrix.

[0087] It should be noted that the core of the orthogonal correction vector is to map the temperature-interference correlation onto each band of the spectrum, and the formula is as follows: ;

[0088] In the formula, Y is the transpose of the near-infrared spectral matrix (dimension n×m, where n is the number of spectral bands); Y is the temperature compensation matrix; w is the regression coefficient vector (k×1); denominator The magnitude of the vector is used for normalization, ensuring uniformity of the correction vector's dimensions and preventing differences in magnitude from affecting the correction effect. The regression coefficient vector w provides the correlation weights between temperature and disturbance, and the temperature compensation matrix Y provides temperature characteristic data; combining the two yields the disturbance contribution value. This is achieved through the transpose of the spectral matrix. By mapping the interference contribution to the spectral bands, the resulting correction vector v can accurately match the temperature interference distribution in each band of the spectrum, ensuring the targeted nature of the correction.

[0089] S223. The near-infrared spectral matrix is ​​iteratively updated using orthogonal correction vectors, and the corrected near-infrared spectral data is generated by combining derivative processing.

[0090] In a preferred embodiment, the step of iteratively updating the near-infrared spectral matrix using orthogonal correction vectors and combining derivative processing to generate corrected near-infrared spectral data includes the following steps:

[0091] S2231. Based on the orthogonal correction vector and temperature compensation matrix, calculate the temperature interference projection component to be removed, and subtract the projection component from the near-infrared spectral matrix to obtain the preliminary correction spectral matrix.

[0092] It should be noted that the orthogonal correction vector and temperature compensation matrix quantify and remove temperature interference signals in the near-infrared spectrum through matrix operations. Specifically, the orthogonal correction vector is used to locate key temperature-affected bands in the spectrum, and the temperature compensation matrix is ​​used to establish a quantitative correlation between temperature change and spectral interference. The temperature interference projection component is calculated (this component is an m×n matrix with the same dimensions as the original near-infrared spectral matrix, and each element corresponds to the pure temperature interference intensity at a specific time point and in a specific band). By subtracting the temperature interference projection component from the original near-infrared spectral matrix, redundant temperature-related interference is eliminated, and the core signal dominated by the dissolution changes of effective components in the spectrum is retained, finally obtaining the preliminary corrected spectral matrix.

[0093] S2232. Determine whether the current iteration number has reached the preset maximum number of principal factors. If not, use the preliminary corrected spectral matrix as the new near-infrared spectral matrix and return to step S2221 for iterative update until the iteration number reaches the preset maximum number of principal factors. Otherwise, use the obtained preliminary corrected near-infrared spectral matrix as the final converged corrected near-infrared spectral matrix.

[0094] It should be noted that the preset maximum number of principal factors is an empirical value (usually 3 to 5) based on the dimensions of the near-infrared spectrum and the complexity of temperature interference. The number of principal factors corresponds to the temperature interference dimensions that need to be stripped. The more principal factors there are, the more comprehensive the interference dimensions can be stripped, but too many will lead to overfitting (removing effective component signals).

[0095] S2233. Perform first-order derivative transformation on the finally converged calibration spectral matrix to obtain the calibrated near-infrared spectral data.

[0096] Specifically, the first derivative transformation can achieve two key effects by calculating the difference between spectral values ​​of adjacent bands: (1) Baseline drift is mostly linear or slow nonlinear change. The derivative transformation can cancel this overall shift, making the spectral peak shape clearer; (2) The spectral response of effective components usually manifests as peak shape changes in specific bands. The derivative transformation can highlight the slope change of the peak, making weak peak signals more significant (such as characteristic peaks of alkaloids and flavonoids).

[0097] The first-order derivative transformation uses the Savitzky-Golay derivative algorithm (consistent with the data smoothing algorithm mentioned above, ensuring compatibility), with a window size of 11-15 points (balancing peak shape preservation and noise suppression), and a polynomial order of 2.

[0098] S23. Based on calibrated near-infrared spectral data, a pre-trained quantitative inversion model is used to predict the current precipitation concentration of the active ingredients in real time, and the cumulative precipitation rate is calculated by integrating the decoction time.

[0099] It should be noted that the pre-trained quantitative inversion model is constructed based on historical calibrated near-infrared spectral data and the corresponding measured concentrations of effective components (obtained by HPLC detection). The model type can be an algorithm adapted to the high-dimensional characteristics of near-infrared spectroscopy, such as Support Vector Machine (SVM) or Partial Least Squares Regression (PLS). During the decoction process, the calibrated near-infrared spectral data at the current moment is input into the model, and the instantaneous precipitation concentration values ​​of each effective component are quickly output. To comprehensively reflect the dissolution effect of the components, the concentration-time curve from the start of decoction to the current moment is integrated using the trapezoidal integral method with decoction time as the horizontal axis and instantaneous precipitation concentration as the vertical axis to obtain the cumulative precipitation amount of each effective component. This is then combined with the theoretical total content of each component determined through the medicinal material database (single herb dosage × component content percentage) to calculate the cumulative precipitation rate (cumulative precipitation rate = cumulative precipitation amount / theoretical total content × 100%).

[0100] S24. The real-time precipitation rate is used as a state feedback input to the dynamic mathematical model constructed based on the ecological niche adaptation theory. The objective function is to maximize the effective precipitation rate. The rolling time domain optimization method is used to solve for the optimal temperature setpoint for the future preset period, thus forming a temperature optimization strategy.

[0101] As a preferred embodiment, the step of using the real-time precipitation rate as state feedback input to a dynamic mathematical model constructed based on niche adaptation theory, with maximizing the effective component precipitation rate as the objective function, and employing a rolling time-domain optimization method to solve for the optimal temperature setpoint for a future preset period to form a temperature optimization strategy includes the following steps:

[0102] S241. Obtain historical decoction experiment data and construct a dynamic mathematical model based on the ecological niche adaptation theory;

[0103] As a preferred embodiment, the step of acquiring historical decoction experiment data and constructing a dynamic mathematical model based on niche adaptation theory includes the following steps:

[0104] S2411. Obtain historical decoction experiment data, including historical temperature data, historical effective component data, decoction time data, and related process parameters;

[0105] It should be noted that historical temperature data records temperature values ​​at different time points (e.g., per minute) during the decoction process, and must include complete temperature curves for the heating, isothermal, and cooling stages. Historical active ingredient data is determined by HPLC or UV-Vis spectroscopy to measure the concentration of target components (e.g., baicalin, chlorogenic acid) in the decoction, and must be synchronized with the temperature data. Decoction time data includes the total decoction time and the time allocation for each stage (e.g., 5 min preheating, 30 min isothermal decoction). Related process parameters include the amount of medicinal materials added, the amount of water added, the material of the decoction container, and the stirring frequency, which may interact to affect the temperature-component relationship.

[0106] S2412. Preprocess the historical decoction experiment data, and calculate the niche parameters of each effective component based on the preprocessed historical decoction experiment data using the niche adaptation theory.

[0107] As a preferred embodiment, the preprocessing of historical decoction experiment data and the calculation of niche parameters of each effective component based on the preprocessed historical decoction experiment data using niche adaptation theory include the following steps:

[0108] S24121. Perform preprocessing on the historical decoction experiment data, including data cleaning, data standardization and data alignment.

[0109] S24122. Based on the pre-processed historical decoction experimental data, the niche center was calculated by Gaussian fitting; the niche width was calculated by the Levins index; and the niche overlap was calculated by the Pianka index.

[0110] Specifically, the niche center is represented by the optimum temperature with the highest extraction rate, and the extraction rate R is... i Gaussian function fitting is performed on temperature T:

[0111] ;

[0112] In the formula, R i (t) represents the precipitation rate at time t, a i σ represents the proportion of maximum precipitation.i μ represents the standard deviation parameter. i Indicates the ecological niche center (optimal temperature).

[0113] The formula for calculating niche breadth using the Levins index is as follows:

[0114] ;

[0115] In the formula, B i The Levins niche breadth index represents the effective ingredient i, K represents the number of sampling temperature points, and T represents the number of sampling temperature points. k T represents the k-th sampling temperature point. m R represents the m-th sampling temperature point. i (T k ) indicates that the active ingredient i is at temperature T k The precipitation rate, p ik This indicates that the active ingredient i is at temperature T k The proportion of resources utilized.

[0116] ;

[0117] In the formula, O ij p represents the Piana niche overlap index between component i and component j. jk This indicates that the active ingredient j is at temperature T. k The proportion of resources utilized.

[0118] S24123. Integrate the niche center, niche width, and niche overlap to obtain the niche parameters of each effective component.

[0119] S2413. The niche parameters and dissolution kinetics of each effective component are integrated and processed, and a dynamic mathematical model is constructed.

[0120] It should be noted that during the decoction process of traditional Chinese medicine, the dissolution behavior of each active ingredient is not only affected by physical factors such as temperature and time, but also closely related to their niche relationships within the decoction system. Niche parameters (such as niche center, width, and overlap) reflect the ability of each ingredient to utilize environmental resources such as temperature and their competitive relationships, while dissolution kinetics describes the rate at which ingredients dissolve from the medicinal materials into the decoction. Integrating these two approaches allows for a comprehensive consideration of the ecological and dissolution characteristics of the ingredients, thereby more realistically simulating and predicting the dynamic changes of the ingredients during the decoction process, and providing a theoretical basis for optimizing the decoction process.

[0121] The constructed dynamic mathematical model, based on the law of mass action and niche theory, describes the rate of change of active ingredient concentration over time in the form of differential equations. The specific formula is:

[0122] ;

[0123] ;

[0124] In the formula, R i R(t) represents the precipitation rate of active ingredient i at time t. j (t) represents the precipitation rate of active ingredient j at time t, k i,sol (T) represents temperature. The dependent dissolution rate constant reflects the optimal dissolution conditions in the basic ecological niche, k i,deg (T) represents temperature. The degradation rate constant, which depends on the basic ecological niche, reflects the thermal stability (tolerance limit) in the α-niche. ij denoted as the niche interaction coefficient, i.e., the niche competition coefficient between effective component j and effective component i, where n represents the total number of effective components and c represents the regulatory factor.

[0125] S242. Using the current real-time precipitation rate as the initial state of the dynamic mathematical model, predict the dynamic precipitation process of the effective components under different temperature trajectories within the preset prediction time domain.

[0126] It should be noted that a reasonable prediction time domain is preset based on the process requirements of traditional Chinese medicine decoction and actual production needs. For example, if the decoction process is expected to continue for another 30 minutes, the prediction time domain can be set to different time periods within the next 30 minutes, such as every 5 minutes as a prediction interval. Specifically, the real-time cumulative extraction rate of each effective component is combined with the determined theoretical total content of the components to convert it into an initial concentration, which is used as the initial state input of the dynamic mathematical model. Based on the niche parameters of each effective component, multiple sets of candidate temperature trajectories covering suitable temperature ranges are generated. Each set of temperature trajectories is sequentially input into the dynamic mathematical model, and the instantaneous concentration changes and cumulative extraction rate trends of the effective components at each time point within the next 30 minutes are simulated by solving differential equations. The correspondence between temperature trajectory, time, and component dissolution data is output, providing a comprehensive prediction basis for selecting the optimal temperature strategy.

[0127] S243. Taking the maximization of the cumulative effective component extraction rate at the end of the prediction period as the objective function, and combining the niche constraints of the effective components, the optimal temperature control sequence in the prediction time domain is solved to obtain the temperature optimization strategy.

[0128] It should be noted that the objective function is constructed by maximizing the cumulative effective component extraction rate at the end of the prediction period. Simultaneously, constraints are set based on the niche characteristics of each effective component. An optimization algorithm is used to solve for the optimal temperature control sequence, forming a feasible temperature optimization strategy. Specifically, the objective function needs to incorporate efficacy weighting coefficients (based on the principle of monarch-minister-assistant-guide drug pairing, with the monarch drug having the highest weight for its core effective component, followed by the minister drugs, and then the assistant drugs), ensuring that the optimization results focus on the core efficacy.

[0129] Niche constraints limit the core suitable range of each component's niche center to ±5℃ within the predicted time domain, with a heating rate not exceeding 10℃ / min, while simultaneously ensuring that the cumulative precipitation rate of each component is not less than 1.2 times the current level (to ensure the effectiveness of optimization).

[0130] The solution process employs a rolling time-domain optimization algorithm, combined with gradient descent to traverse multiple candidate temperature trajectories generated earlier, and selects the temperature control sequence that satisfies all constraints and maximizes the objective function value. Finally, it outputs a time-segmented temperature setting scheme. This strategy can be directly used in subsequent closed-loop control systems to guide the dynamic adjustment of heating power and achieve the core objective of maximizing the effective component output rate.

[0131] S3. Based on the temperature optimization strategy, the closed-loop control system monitors the decoction temperature of the Chinese medicine to be decocted in real time and dynamically adjusts the heating power to achieve optimized temperature control of the Chinese medicine decoction.

[0132] It should be noted that after obtaining the temperature optimization strategy, the closed-loop control system can monitor the actual decoction temperature of the Chinese herbal medicine in real time and continuously. The measured temperature is compared instantly with the target temperature set in the optimization strategy, and the heating power is dynamically adjusted based on the deviation using a specific control algorithm. This real-time monitoring and dynamic adjustment form a highly efficient feedback regulation mechanism that can quickly respond to temperature fluctuations caused by environmental changes and differences in the characteristics of the medicinal materials during the decoction process. This ensures that the decoction temperature always closely matches the requirements of the optimization strategy, thereby achieving refined and intelligent temperature optimization control of the Chinese herbal medicine decoction process. This maximizes the extraction of effective components from the Chinese herbal medicine, improving the quality and efficacy of the decoction.

[0133] like Figure 2 As shown, according to another embodiment of the present invention, a traditional Chinese medicine decoction temperature optimization control system based on the effective component extraction rate is provided, the system comprising:

[0134] Data acquisition module 1 is used to acquire the medicinal ingredients and dosage of the Chinese medicine to be decocted, and to determine the effective ingredients of the Chinese medicine to be decocted by combining it with a pre-configured medicinal material database.

[0135] Temperature optimization module 2 is used to acquire temperature data and near-infrared spectral data of the Chinese medicine to be decocted during the decoction process, and combine them with a dynamic mathematical model pre-constructed based on the ecological niche adaptation theory to determine the temperature optimization strategy with the objective function of maximizing the effective component extraction rate.

[0136] Temperature adjustment module 3 is used to monitor the decoction temperature of the Chinese medicine to be decocted in real time and dynamically adjust the heating power based on the temperature optimization strategy through a closed-loop control system, so as to achieve optimized temperature control of the decoction of Chinese medicine.

[0137] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, etc.) containing computer-usable program code.

[0138] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for optimizing and controlling the decoction temperature of traditional Chinese medicine based on the effective component extraction rate, characterized in that, Includes the following steps: S1. Obtain the medicinal ingredients and dosage of the Chinese medicine to be decocted, and determine the effective ingredients of the Chinese medicine to be decocted by combining it with a pre-configured medicinal material database. S2. During the decoction process of the Chinese herbal medicine to be decocted, temperature data and near-infrared spectral data are collected simultaneously and standardized to obtain standardized temperature data and standardized near-infrared spectral data. Based on the orthogonal signal correction method, the standardized near-infrared spectral data is compensated and corrected in combination with the standardized temperature data to obtain calibrated near-infrared spectral data. Based on the calibrated near-infrared spectral data, a pre-trained quantitative inversion model is used to predict the current precipitation concentration of the effective components in real time, and the cumulative precipitation rate is calculated by integrating the decoction time. Historical decoction experimental data are obtained, and a dynamic mathematical model is constructed based on the niche adaptation theory. The current real-time precipitation rate is used as the initial state of the dynamic mathematical model, and the dynamic precipitation process of the effective components under different temperature trajectories is predicted within the preset prediction time domain. With the maximization of the cumulative effective component precipitation rate at the end of the prediction period as the objective function, the optimal temperature control sequence within the prediction time domain is solved in combination with the niche constraints of the effective components to obtain the temperature optimization strategy. S3. Based on the temperature optimization strategy, the closed-loop control system monitors the decoction temperature of the Chinese medicine to be decocted in real time and dynamically adjusts the heating power to achieve optimized temperature control of the Chinese medicine decoction.

2. The method for optimizing and controlling the decoction temperature of traditional Chinese medicine based on the effective component extraction rate according to claim 1, characterized in that, The method based on orthogonal signal correction, combined with standardized temperature data, performs compensation and correction processing on standardized near-infrared spectral data to obtain corrected near-infrared spectral data, including the following steps: S221. Construct a near-infrared spectral matrix and a temperature compensation matrix based on standardized near-infrared spectral data and standardized temperature data, respectively. S222. Based on the orthogonality between the temperature compensation matrix and the near-infrared spectral matrix, calculate the temperature-related orthogonal components and construct the orthogonal correction vector. S223. The near-infrared spectral matrix is ​​iteratively updated using orthogonal correction vectors, and the corrected near-infrared spectral data is generated by combining derivative processing.

3. The method for optimizing and controlling the decoction temperature of traditional Chinese medicine based on the effective component extraction rate according to claim 2, characterized in that, The calculation of temperature-related orthogonal components and the construction of orthogonal correction vectors based on the orthogonality of the temperature compensation matrix and the near-infrared spectral matrix include the following steps: S2221. Perform principal component analysis on the near-infrared spectral matrix and extract the principal component score vector. S2222. Orthogonalize the principal component score vector with the temperature compensation matrix to remove temperature-related interference signals and obtain temperature-related orthogonal components. S2223. Perform convergence judgment on temperature-related orthogonal components, and construct an orthogonal correction vector by combining the regression coefficients when the preset convergence conditions are met.

4. The method for optimizing and controlling the decoction temperature of traditional Chinese medicine based on the effective component extraction rate according to claim 3, characterized in that, The process of determining the convergence of temperature-related orthogonal components and constructing an orthogonal correction vector by combining regression coefficients when a preset convergence condition is met includes the following steps: S22231. Calculate the change in variance contribution rate of the temperature-related orthogonal components and compare the change in variance contribution rate with a preset threshold. If the change in variance contribution rate is greater than the preset threshold, the preset convergence condition is not met, and return to step S2221. Otherwise, the preset convergence condition is met, and step S22232 is executed. S22232. Using the temperature compensation matrix as the independent variable and the converged temperature-related orthogonal components as the dependent variable, the least squares method is used to calculate the regression coefficient vector. S22233. Construct an orthogonal correction vector using the regression coefficient vector and the temperature compensation matrix.

5. The method for optimizing and controlling the decoction temperature of traditional Chinese medicine based on the effective component extraction rate according to claim 4, characterized in that, The process of iteratively updating the near-infrared spectral matrix using orthogonal correction vectors and combining this with derivative processing to generate corrected near-infrared spectral data includes the following steps: S2231. Based on the orthogonal correction vector and temperature compensation matrix, calculate the temperature interference projection component to be removed, and subtract the projection component from the near-infrared spectral matrix to obtain the preliminary correction spectral matrix. S2232. Determine whether the current iteration number has reached the preset maximum number of principal factors. If not, use the preliminary corrected spectral matrix as the new near-infrared spectral matrix and return to step S2221 for iterative update until the iteration number reaches the preset maximum number of principal factors. Otherwise, use the obtained preliminary corrected near-infrared spectral matrix as the final converged corrected near-infrared spectral matrix. S2233. Perform first-order derivative transformation on the finally converged calibration spectral matrix to obtain the calibrated near-infrared spectral data.

6. The method for optimizing and controlling the decoction temperature of traditional Chinese medicine based on the effective component extraction rate according to claim 5, characterized in that, The process of acquiring historical decoction experiment data and constructing a dynamic mathematical model based on niche adaptation theory includes the following steps: S2411. Obtain historical decoction experiment data, including historical temperature data, historical effective component data, decoction time data, and related process parameters; S2412. Preprocess the historical decoction experiment data, and calculate the niche parameters of each effective component based on the preprocessed historical decoction experiment data using the niche adaptation theory. S2413. The niche parameters and dissolution kinetics of each effective component are integrated and processed, and a dynamic mathematical model is constructed.

7. The method for optimizing and controlling the decoction temperature of traditional Chinese medicine based on the effective component extraction rate according to claim 6, characterized in that, The process of preprocessing historical decoction experiment data and calculating the niche parameters of each effective component based on the preprocessed historical decoction experiment data using niche adaptation theory includes the following steps: S24121. Perform preprocessing on the historical decoction experiment data, including data cleaning, data standardization and data alignment. S24122. Based on the pre-processed historical decoction experimental data, the niche center was calculated by Gaussian fitting; the niche width was calculated by the Levins index; and the niche overlap was calculated by the Pianka index. S24123. Integrate the niche center, niche width, and niche overlap to obtain the niche parameters of each effective component.

8. A traditional Chinese medicine decoction temperature optimization control system based on the effective component extraction rate, used to implement the traditional Chinese medicine decoction temperature optimization control method based on the effective component extraction rate as described in any one of claims 1-7, characterized in that, The system includes: The data acquisition module is used to acquire the medicinal ingredients and dosage of the Chinese medicine to be decocted, and to determine the effective components of the Chinese medicine to be decocted by combining it with a pre-configured medicinal material database. The temperature optimization module is used to acquire temperature data and near-infrared spectral data of the Chinese medicine to be decocted during the decoction process, and combine them with a dynamic mathematical model pre-constructed based on the ecological niche adaptation theory to determine the temperature optimization strategy with the objective function of maximizing the effective component extraction rate. The temperature adjustment module is used to monitor the decoction temperature of the Chinese medicine to be decocted in real time and dynamically adjust the heating power based on the temperature optimization strategy through a closed-loop control system, so as to achieve optimized temperature control of the decoction of Chinese medicine.

Images