Dandelion effective component extraction process optimization method based on I-WO-SVR model

Abstract

The invention relates to a dandelion effective component extraction process optimization method based on an I-WO-SVR model, and the method comprises the steps: obtaining to-be-predicted factors which comprise ethanol concentration, ultrasonic time, ultrasonic temperature and ultrasonic power; inputting a to-be-predicted factor into the I-WO-SVR model to obtain the extraction rate of the dandelion flower stem chlorogenic acid; the I-WO-SVR model is obtained by training a support vector regression SVR prediction model by using a training set and optimizing penalty parameters and kernel parameters of the SVR prediction model by using an improved sea image optimizer I-WO; wherein the training set comprises original factors and chlorogenic acid extraction rates corresponding to the original factors. The traditional Chinese medicine effective component extraction optimization method is designed by combining the modern artificial intelligence technology and the traditional Chinese medicine field, the application prospect is wide, and a new thought can be provided for traditional Chinese medicine extraction and intelligent industrial production thereof.

Description

Technical Field

[0001] The invention relates to the technical field of modern extraction of traditional Chinese medicine, and in particular to an optimization method for extracting effective components of dandelions based on an I-WO-SVR model. Background Art

[0002] Dandelion, commonly known as dandelion and yellow flowered dandelion, is a perennial herbaceous plant of the genus Taraxacum in the Asteraceae family. Its dried whole herb can be used as medicine, and it is known as the "Queen of Herbs" and "Natural Penicillin". Chlorogenic acid, as the main component of dandelion leaves, has been internationally recognized as "plant gold" and has a wide range of functions such as lowering blood lipids, inhibiting bacteria, anti-oxidation, anti-tumor, protecting the liver and promoting choleresis. Therefore, the extraction of chlorogenic acid in dandelions has become a hot spot in the research and development of functional foods.

[0003] Traditional extraction process optimization methods usually use partial experiments instead of comprehensive experiments, so they cannot analyze the experimental effects and their interactions one by one like all experiments. Machine learning methods can effectively solve nonlinear problems and perform global optimization, which can reduce experimental costs and time. When applied to the optimization of extraction processes, good results can be achieved.

[0004] Support vector regression (SVR) is a classic algorithm in machine learning, which can provide technical support for the optimization of effective ingredient extraction from traditional Chinese medicine. SVR can map data to high-dimensional space by introducing kernel functions, perform regression analysis in high-dimensional space, and can handle nonlinear problems well. However, the performance of SVR is highly dependent on the selection of its hyperparameters, and the walrus optimization algorithm can achieve SVR model parameter optimization.

[0005] Therefore, there is an urgent need for an optimization method for the extraction process of dandelion active ingredients based on the I-WO-SVR model. By constructing a prediction model that integrates the improved walrus optimizer and SVR, the optimization method for the extraction process of dandelion active ingredients can be determined, which can improve the efficiency of Chinese medicine ingredient extraction. Summary of the invention

[0006] In order to solve the problems existing in the above-mentioned prior art, the purpose of the present invention is to provide a method for optimizing the extraction process of dandelion active ingredients based on the I-WO-SVR model. The method can combine artificial intelligence algorithms with traditional Chinese medicine extraction processes, and obtain the optimal parameters of the extraction process by constructing a prediction model for process parameters and optimizing the model.

[0007] To achieve the above object, the present invention provides the following solutions:

[0008] A method for optimizing the extraction process of effective components of dandelion based on an I-WO-SVR model, comprising:

[0009] Obtaining factors to be predicted, wherein the factors to be predicted include: ethanol concentration, ultrasound time, ultrasound temperature and ultrasound power;

[0010] The factors to be predicted are input into the I-WO-SVR model to obtain the chlorogenic acid extraction rate of dandelion stems; the I-WO-SVR model uses the training set to train the SVR prediction model, and uses the improved walrus optimizer I-WO to optimize the penalty parameters and kernel parameters of the SVR prediction model; wherein the training set includes: extraction influencing factors and corresponding chlorogenic acid extraction rates.

[0011] Optionally, the penalty parameters and kernel parameters of the SVR prediction model optimized by the improved walrus optimizer I-WO include:

[0012] Determine the SVR objective function, select the radial basis function as the kernel function of the SVR prediction model, and determine the penalty factor C and kernel function parameter g of the SVR prediction model through the improved walrus optimizer I-WO:

[0013] The SVR objective function expression is:

[0014]

[0015]

[0016] Among them, ω and b are SVR model parameters, x h ,y h They represent the input variables and output variables of the model, n is the number of support vectors, and ε represents the tolerance deviation;

[0017] The kernel function The expression is:

[0018]

[0019] in, is the support vector, x is the original sample set, and g is the kernel function parameter.

[0020] Optionally, optimizing the SVR prediction model using the improved walrus optimizer I-WO includes:

[0021] Step 1: Initialize the walrus population using infinite folding iterative chaotic mapping;

[0022] Step 2: When the danger signal of the walrus environment is too high, the walrus population is migrated, and adaptive weights are introduced for the first time to improve the walrus population migration process;

[0023] Step 3: When the walrus is attacked by a natural predator while foraging underwater, it sends out the danger signal and flees from the current activity area. The adaptive weight is introduced for the second time to improve the walrus escape process.

[0024] Step 4: Repeat steps 2 to 3 until the maximum number of iterations is reached, and output the optimal penalty parameters and kernel parameters.

[0025] Optionally, the formula for initializing the walrus population using infinite folding iterative chaotic mapping is:

[0026]

[0027] Among them, δ is the control parameter, X i and X i+1 They represent the values ​​of the i-th walrus and the (i+1)-th walrus in the sequence generated by the infinite folding iterative chaotic map.

[0028] Optionally, the formula for first introducing adaptive weights to improve the walrus population migration process is:

[0029]

[0030] in, represents the new position of the i-th walrus in the j-th dimension at the d+1-th iteration of the migration process, represents the current position of the i-th walrus on the j-th dimension at the d-th iteration during the migration process. Two patrol walruses are set during the migration process, and their positions are and α1 represents the adaptive weight introduced for the first time.

[0031] Optionally, the formula for obtaining the adaptive weight is:

[0032]

[0033] Among them, D is the maximum number of iterations, d is the current number of iterations, and e is a natural constant.

[0034] Optionally, the formula for introducing adaptive weights for the second time to improve the walrus escape process is:

[0035]

[0036] in, represents the distance between the current walrus and the best walrus, Λ is the risk factor, and α2 is the adaptive weight introduced for the second time.

[0037] Optionally, obtaining the training set includes:

[0038] Drawing a standard curve, using the standard curve, calculating the concentration of chlorogenic acid in dandelion flower stems, and then calculating the chlorogenic acid extraction rate;

[0039] Based on the single factor experiment, a response surface experiment with 4 factors and 3 levels was conducted with chlorogenic acid extraction rate as the evaluation index to obtain experimental sample data;

[0040] According to the experimental results, a sample set is constructed, the sample set is screened and data enhanced, the expanded sample set is standardized, and the sample data is set as a training set.

[0041] The beneficial effects of the present invention are:

[0042] The present invention can determine the optimal process parameters for extracting effective ingredients from dandelions, thereby improving production efficiency while ensuring product quality, and bringing a new method to traditional Chinese medicine extraction and its intelligent industrial production. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0044] Figure 1 The figure is a schematic diagram of a method for optimizing the extraction process of effective components of dandelion based on the I-WO-SVR model according to an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0047] like Figure 1As shown, this embodiment discloses a method for optimizing the extraction process of effective components of dandelion based on the I-WO-SVR model, comprising: obtaining factors to be predicted, the factors to be predicted include: ethanol concentration, ultrasonic time, ultrasonic temperature and ultrasonic power; inputting the factors to be predicted into the I-WO-SVR model to obtain the chlorogenic acid extraction rate of dandelion; the I-WO-SVR model is obtained by training the SVR prediction model with a training set, and optimizing the penalty parameters and kernel parameters of the SVR prediction model with the improved walrus optimizer I-WO; wherein the training set includes: original factors and the chlorogenic acid extraction rates corresponding to the original factors.

[0048] Furthermore, obtaining a training set includes:

[0049] Draw a standard curve, use the standard curve to calculate the concentration of chlorogenic acid in dandelion flower stems, and then calculate the chlorogenic acid extraction rate;

[0050] Based on the single factor experiment, a response surface experiment with 4 factors and 3 levels was conducted with chlorogenic acid extraction rate as the evaluation index to obtain experimental sample data;

[0051] According to the experimental results, a sample set is constructed, the sample set is screened and data enhanced, the expanded sample set is standardized, and the sample data is divided into a training set.

[0052] Specifically:

[0053] Step (1) Construct a data sample set:

[0054] Step (1.1) Preparation of chlorogenic acid standard curve:

[0055] Accurately weigh 5 mg of chlorogenic acid standard in a small beaker, add an appropriate amount of anhydrous ethanol, transfer to a 100 mL volumetric flask with ultrasonic assistance until completely dissolved, and dilute to the mark with anhydrous ethanol as the chlorogenic acid standard solution. Transfer 1, 3, 5, 7, and 9 mL of the chlorogenic acid standard solution to a 10 mL volumetric flask, dilute with anhydrous ethanol, and let stand for 15 minutes. Use anhydrous ethanol as a blank control group to measure its absorbance at 329 nm. Each group is measured in parallel 3 times to obtain the average value. Draw a standard curve with the absorbance of the chlorogenic acid standard solution as the ordinate and the concentration of the chlorogenic acid standard solution as the abscissa.

[0056] Step (1.2) Extraction of chlorogenic acid from dandelion flower stems:

[0057] Collect fresh dandelion flower stems, wash them, put them in a 60°C oven to dry to constant weight, crush them with a grinder, pass them through a 100-mesh sieve to obtain dandelion flower stem powder, and put them in a brown ground-mouth bottle for later use. Accurately weigh 1.00 g of dandelion flower stem powder in a conical flask, add a 30% ethanol aqueous solution with a solid-liquid ratio of 1:20 (g / mL), shake well, put them in a 300W ultrasonic cleaner at 50°C for ultrasonic-assisted extraction for 1 hour, cool to room temperature, centrifuge at 4000r / min for 10 minutes, absorb the supernatant volume, dilute, and measure the absorbance of the solution at 329nm. Calculate the chlorogenic acid concentration in the extract according to the standard curve, and calculate the chlorogenic acid extraction rate using the following formula:

[0058]

[0059] Wherein: η is the extraction rate of chlorogenic acid; ρ is the concentration of chlorogenic acid calculated from the standard curve (μg / mL); V is the volume of the supernatant (mL); F is the dilution factor of the solution; m is the mass of dandelion dry powder (g).

[0060] Step (1.3) Single factor experiment:

[0061] Four single factors, namely, ethanol concentration, ultrasonic temperature, ultrasonic time and ultrasonic power, were selected as the investigation factors, and the chlorogenic acid extraction rate was used as the investigation index for single factor experiment. Ethanol concentration was 20%, 30%, 40%, 50%, 60%; ultrasonic time was 30min, 45min, 60min, 75min, 90min; ultrasonic temperature was 30℃, 40℃, 50℃, 60℃, 70℃; ultrasonic power was 240W, 300W, 360W, 420W, 480W, and the extraction was carried out twice in parallel to determine the content of chlorogenic acid in the extract. Except for the changed single factors, the appropriate conditions of each factor in the pre-experiment were adopted, namely, ethanol concentration was 30%, ultrasonic time was 60min, ultrasonic temperature was 50℃, and ultrasonic power was 300W.

[0062] Step (1.4) Response surface experimental design with 4 factors and 3 levels:

[0063] On the basis of the single factor experiment, the chlorogenic acid extraction rate was used as the evaluation index to carry out a 4-factor 3-level response surface experimental design to obtain the experimental sample data, as shown in Table 1. Table 1 shows the experimental factors, levels and codes.

[0064] Table 1

[0065]

[0066] Step (1.5) Data preprocessing:

[0067] Since there are attribute differences among various input parameters in the sample data, the prediction accuracy of the extraction rate may be inaccurate. Standardization processing can handle the differences among various input parameters and ensure the accuracy of SVR model prediction.

[0068] Step (1.5.1) Construct a sample set: Screen the data obtained from the experiment, delete the sample data with outlier characteristics, and obtain the experimental sample data. Use data enhancement technology to expand the data sample set.

[0069] Step (1.5.2) Standardization: Standardize the expanded sample data set using the following formula to map the sample data to [0,1]:

[0070]

[0071] In the above formula, x o is the sample set after standardization, x is the original sample set, and x min is the minimum sample in the sample set, x max is the largest sample in the sample set.

[0072] Step (1.5.3) uses the K-fold cross-validation method to divide the sample data into a training set and a validation set:

[0073] The original sample set N is divided into K parts, one of which is used as the model validation set, and the remaining K-1 parts are used as the model training set. After that, the validation set division needs to be repeated K times, and the sample data of each subset will be selected in sequence. After the validation set data is selected, the remaining data is used as the training set. In each round, the training set data is imported into the SVR model constructed in step (2) to train the model, and the results are applied to the validation set.

[0074] Furthermore, the SVR prediction model is optimized using the improved walrus optimizer including:

[0075] Step 1: Initialize the walrus population using infinite folding iterative chaotic mapping;

[0076] Step 2: When the danger signal of the walrus' environment is too high, the walrus population is migrated, and adaptive weights are introduced to improve the walrus population migration process;

[0077] Step 3: When the walrus is attacked by natural predators while foraging underwater, it sends out a danger signal and flees from the current activity area. Adaptive weights are introduced again to improve the walrus escape process.

[0078] Step 4: Repeat steps 2 to 3 until the maximum number of iterations is reached, and output the optimal penalty parameters and kernel parameters.

[0079] Specifically:

[0080] Step (2) constructing the I-WO-SVR model:

[0081] The SVR model is a branch of the support vector machine (SVM) and can be used to solve nonlinear regression problems. The SVR model has high prediction accuracy, can effectively avoid overfitting problems, improve the generalization ability of the model, and can also obtain good statistical laws for small sample data. The SVR model has a certain fault tolerance compared to the traditional regression model. Even if there is a certain deviation in the predicted value, the predicted value can be considered reasonable. The present invention solves the nonlinear relationship between different ethanol concentrations, ultrasonic time, ultrasonic temperature, ultrasonic power and dandelion chlorogenic acid extraction rate by constructing an SVR model.

[0082] Step (2.1) constructs the SVR prediction model:

[0083] Step (2.1.1) Establish the regression equation and objective function: If the given training sample A = {(x1, y1), (x2, y2), ..., (x h ,y h ),...,(x N ,y N )}, x and y represent the input variables (ethanol concentration, ultrasonic time, ultrasonic temperature, ultrasonic power) and output variables (chlorogenic acid extraction rate of dandelion) of the model respectively. For the training samples, the regression equation F(x) can be established as follows:

[0084]

[0085] Among them, ω and b are SVR model parameters, ω T is the transpose operation of ω, is the eigenvector after mapping of x.

[0086] The SVR objective function can be expressed as:

[0087]

[0088]

[0089] Among them, C is the penalty factor, which indicates the tolerance to error. n is the number of support vectors, and ε indicates the tolerance deviation.

[0090] Step (2.1.2) Determine the kernel function of the SVR model: When establishing a support vector machine model, the most important thing is to select the kernel function of the model.

[0091] The kernel function has an important influence on the performance of the support vector machine model. It can map the input samples to a high-dimensional space, thereby making the problem easier to handle. Common kernel functions include linear kernel function, polynomial kernel function, radial basis kernel function (RBF) and sigmoid kernel function, and RBF kernel function is the first choice because of its simplicity and powerful nonlinear modeling ability. Therefore, the present invention selects radial basis function (RBF) as the kernel function of the SVR prediction model, and its specific expression is as follows:

[0092]

[0093] in, is the support vector, and g is the parameter of the RBF kernel function.

[0094] Step (2.2) uses infinite folding iterative chaotic mapping to initialize the population, and adaptive distribution improves the migration and escape processes of the walrus population. The improvements to WO include:

[0095] WO is a swarm intelligence optimization method designed by simulating the walrus's choice of migration, reproduction, habitat, foraging, collection and escape behaviors after receiving key signals (danger signals and safety signals). However, WO still has room for improvement in terms of convergence accuracy and solution speed. To this end, the present invention improves WO by the following steps to solve the optimal C and g parameters in the SVR model:

[0096] Step (2.2.1) Infinite folding iterative chaotic mapping to initialize the population: Using chaotic variables for optimization search can evenly distribute the initial population in the search space, which is more advantageous than using random variables for random search. The ergodicity and randomness of chaos can prevent the search process from falling into the local optimum. Through chaotic mapping, the diversity of the initial state of the walrus population is increased, which effectively prevents falling into the local optimum and premature convergence, and improves global convergence and accuracy. The expression of infinite folding iterative chaotic mapping is as follows:

[0097]

[0098] Among them, δ is a control parameter, which takes values ​​between (0,1). i and X i+1 They represent the values ​​of the i-th walrus and the (i+1)-th walrus in the sequence generated by the infinite folding iterative chaotic map.

[0099] Step (2.2.2) Adaptive migration: When the danger signal of the walrus environment is too high, the walrus population will migrate to an area that is more suitable for survival. At this stage, adaptive weights are introduced to improve the walrus population migration process, and the new walrus position is updated as follows:

[0100]

[0101] in, represents the new position of the i-th walrus in the j-th dimension at the d+1-th iteration of the migration process, represents the current position of the i-th walrus on the j-th dimension at the d-th iteration during the migration process. Two patrol walruses are set during the migration process, and their positions are and α1 represents the adaptive weight introduced for the first time, which is determined by the following formula:

[0102]

[0103] Among them, D is the maximum number of iterations, and d is the current number of iterations.

[0104] Step (2.2.3) Adaptive escape: Walruses are also attacked by natural predators when foraging underwater. They will escape from the current activity area according to the danger signals sent by their companions. This behavior occurs in the late iteration of WO. A certain degree of disturbance to the population helps the walrus to conduct global exploration. The present invention chooses to use adaptive weights again to adjust the escape process, effectively coordinating the relationship between global search and local search. The escape process after replacement is:

[0105]

[0106] in represents the distance between the current walrus and the best walrus, Λ is the risk factor, and α2 is the adaptive weight introduced for the second time.

[0107] Step (2.3) Establish the I-WO-SVR model: Use the I-WO algorithm to optimize the penalty parameters and kernel parameters of SVR.

[0108] The penalty factor C and the kernel function parameter g in the SVR model are two key indicators that affect the model prediction results. The larger the C, the less tolerant the model is to errors, and it is easy to produce overfitting; the smaller the C, the worse the generalization ability, which will cause underfitting. The size of the kernel function parameter g will affect the number of support vectors. The larger the g, the fewer support vectors. On the contrary, the more support vectors, the larger or smaller the C and g are, the faster and more accurate the estimation will be. At the same time, different combinations of parameters will also affect the performance of the model. Therefore, it is particularly important to determine the optimal parameters C and g for the prediction effect of the model. In order to find the optimal parameters of the model more conveniently and quickly, the present invention uses an improved walrus optimizer to find the optimal C and g parameters of the model, further establishes an I-WO-SVR model, and improves the accuracy of model prediction.

[0109] Step (3), determine the optimal extraction scheme for dandelion active ingredients: input the training set into the I-WO-SVR model for training, and apply the trained model to the validation set. 2 The difference between the predicted value (chlorogenic acid extraction rate predicted by I-WO-SVR model) and the true value (chlorogenic acid extraction rate in the training set) was calculated using three indicators: root mean square error (RMSE) and mean absolute error (MAE) to measure the accuracy of the SVR model. The calculation formulas of the three indicators are as follows:

[0110]

[0111]

[0112]

[0113] Where y is the predicted value, E[y] is the mean of the predicted values, z is the true value, and E[z] is the mean of the true values.

[0114] In general, R 2 The larger the value, the better, and the smaller the RMSE and MAE values, the better. 2 The value infinitely approaches 1 and the error infinitely approaches 0, which means that the closer the model prediction value is to the true value, the better the model fit is.

[0115] When R 2 When the value is close to 1 and the RMSE and MAE are both small, it means that the I-WO-SVR model has a high prediction accuracy and can effectively predict the extraction rate of chlorogenic acid from dandelion stems. After the I-WO-SVR model is trained, the corresponding extraction rate can be output by adjusting the input of different factors to be predicted (ethanol concentration, ultrasonic time, ultrasonic temperature and ultrasonic power). When the extraction rate is the highest, the obtained process parameter combination is the optimal extraction solution.

[0116] This embodiment can couple the relationship between the extraction process parameters and the extraction rate through the I-WO-SVR model, so as to determine the optimal extraction process parameters, for example:

[0117] The extraction rate is the highest when the ethanol concentration is 30%, the ultrasonic time is 60 minutes, the ultrasonic temperature is 50°C, and the ultrasonic power is 300w, indicating that the best extraction process is 30% ethanol concentration, 60 minutes ultrasonic time, 50°C ultrasonic temperature, and 300w ultrasonic power.

[0118] The present invention can improve production efficiency while ensuring product quality, and provides a new method for traditional Chinese medicine extraction and intelligent industrial production.

[0119] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for optimizing the extraction process of effective components from dandelion based on the I-WO-SVR model, characterized in that: include: Obtaining factors to be predicted, wherein the factors to be predicted include: ethanol concentration, ultrasound time, ultrasound temperature and ultrasound power; The factors to be predicted are input into the I-WO-SVR model to obtain the chlorogenic acid extraction rate of dandelion stems; the I-WO-SVR model uses the training set to train the SVR prediction model, and uses the improved walrus optimizer I-WO to optimize the penalty parameters and kernel parameters of the SVR prediction model; wherein the training set includes: extraction influencing factors and corresponding chlorogenic acid extraction rates.

2. The method for optimizing the extraction process of effective components from dandelion based on the I-WO-SVR model according to claim 1, characterized in that: The penalty parameters and kernel parameters of the SVR prediction model optimized by the improved walrus optimizer I-WO include: Determine the SVR objective function, select the radial basis function as the kernel function of the SVR prediction model, and determine the penalty factor C and kernel function parameter g of the SVR prediction model through the improved walrus optimizer I-WO: The SVR objective function expression is: Among them, ω and b are SVR model parameters, x h ,y h They represent the input variables and output variables of the model, n is the number of support vectors, and ε represents the tolerance deviation; The kernel function The expression is: in, is the support vector, x is the original sample set, and g is the kernel function parameter.

3. The method for optimizing the extraction process of effective components from dandelion based on the I-WO-SVR model according to claim 1, characterized in that: The SVR prediction model is optimized using the improved walrus optimizer I-WO including: Step 1: Initialize the walrus population using infinite folding iterative chaotic mapping; Step 2: When the danger signal of the walrus environment is too high, the walrus population is migrated, and adaptive weights are introduced for the first time to improve the walrus population migration process; Step 3: When the walrus is attacked by a natural predator while foraging underwater, it sends out the danger signal and flees from the current activity area. The adaptive weight is introduced for the second time to improve the walrus escape process. Step 4: Repeat steps 2 to 3 until the maximum number of iterations is reached, and output the optimal penalty parameters and kernel parameters.

4. The method for optimizing the extraction process of effective components from dandelion based on the I-WO-SVR model according to claim 3, characterized in that: The formula for initializing the walrus population using infinite folding iterative chaotic mapping is: Among them, δ is the control parameter, X i and X i+1 They represent the values ​​of the i-th walrus and the (i+1)-th walrus in the sequence generated by the infinite folding iterative chaotic map.

5. The method for optimizing the extraction process of effective components from dandelion based on the I-WO-SVR model according to claim 3, characterized in that: The formula for improving the migration process of walrus populations by introducing adaptive weights for the first time is: in, represents the new position of the i-th walrus in the j-th dimension at the d+1-th iteration of the migration process, represents the current position of the i-th walrus on the j-th dimension at the d-th iteration during the migration process. Two patrol walruses are set during the migration process, and their positions are and α1 represents the adaptive weight introduced for the first time.

6. The method for optimizing the extraction process of effective components from dandelion based on the I-WO-SVR model according to claim 5, characterized in that: The formula for obtaining the adaptive weight is: Among them, D is the maximum number of iterations, d is the current number of iterations, and e is a natural constant.

7. The method for optimizing the extraction process of effective components from dandelion based on the I-WO-SVR model according to claim 3, characterized in that: The formula for the second introduction of adaptive weights to improve the walrus escape process is: in, represents the distance between the current walrus and the best walrus, Λ is the risk factor, and α2 is the adaptive weight introduced for the second time.

8. The method for optimizing the extraction process of effective components from dandelion based on the I-WO-SVR model according to claim 1, characterized in that: Acquiring the training set includes: Drawing a standard curve, using the standard curve, calculating the concentration of chlorogenic acid in dandelion flower stems, and then calculating the chlorogenic acid extraction rate; Based on the single factor experiment, a response surface experiment with 4 factors and 3 levels was conducted with chlorogenic acid extraction rate as the evaluation index to obtain experimental sample data; According to the experimental results, a sample set is constructed, the sample set is screened and data enhanced, the expanded sample set is standardized, and the sample data is set as a training set.

Images