CN112915579A - Effective safflower component ultrasonic extraction process based on LS-SVM model - Google Patents
Effective safflower component ultrasonic extraction process based on LS-SVM model Download PDFInfo
- Publication number
- CN112915579A CN112915579A CN201911365093.7A CN201911365093A CN112915579A CN 112915579 A CN112915579 A CN 112915579A CN 201911365093 A CN201911365093 A CN 201911365093A CN 112915579 A CN112915579 A CN 112915579A
- Authority
- CN
- China
- Prior art keywords
- evaluation value
- extraction
- safflower
- svm
- comprehensive evaluation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
- 235000003255 Carthamus tinctorius Nutrition 0.000 title claims abstract description 38
- 244000020518 Carthamus tinctorius Species 0.000 title claims abstract description 38
- 238000000034 method Methods 0.000 title claims abstract description 36
- 230000008569 process Effects 0.000 title claims abstract description 24
- 238000002137 ultrasound extraction Methods 0.000 title claims abstract description 21
- 238000011156 evaluation Methods 0.000 claims abstract description 46
- IAVUBSCVWHLRGE-UXEKTNMQSA-N (6e)-2,5-dihydroxy-6-[(e)-1-hydroxy-3-(4-hydroxyphenyl)prop-2-enylidene]-2,4-bis[(2s,3r,4r,5s,6r)-3,4,5-trihydroxy-6-(hydroxymethyl)oxan-2-yl]cyclohex-4-ene-1,3-dione Chemical compound O[C@@H]1[C@@H](O)[C@H](O)[C@@H](CO)O[C@H]1C(C(C(O)([C@H]1[C@@H]([C@@H](O)[C@H](O)[C@@H](CO)O1)O)C1=O)=O)=C(O)\C1=C(/O)\C=C\C1=CC=C(O)C=C1 IAVUBSCVWHLRGE-UXEKTNMQSA-N 0.000 claims abstract description 31
- ZZMASNSDVDSYKO-UHFFFAOYSA-N hydroxysafflor yellow A Natural products OCC1OC(C(O)C(O)C1O)C2=C(O)C(O)(C3OC(CO)C(O)C(O)C3O)C(=O)C(=C2O)C(=O)C=Cc4ccc(O)cc4 ZZMASNSDVDSYKO-UHFFFAOYSA-N 0.000 claims abstract description 31
- 238000012843 least square support vector machine Methods 0.000 claims abstract description 29
- 238000004458 analytical method Methods 0.000 claims abstract description 11
- 238000004088 simulation Methods 0.000 claims abstract description 3
- 238000000605 extraction Methods 0.000 claims description 58
- LFQSCWFLJHTTHZ-UHFFFAOYSA-N Ethanol Chemical compound CCO LFQSCWFLJHTTHZ-UHFFFAOYSA-N 0.000 claims description 24
- 239000011344 liquid material Substances 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 13
- 238000002790 cross-validation Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 5
- 238000013459 approach Methods 0.000 claims 1
- 239000006185 dispersion Substances 0.000 claims 1
- 238000012360 testing method Methods 0.000 description 15
- 238000005303 weighing Methods 0.000 description 11
- OKKJLVBELUTLKV-UHFFFAOYSA-N Methanol Chemical compound OC OKKJLVBELUTLKV-UHFFFAOYSA-N 0.000 description 9
- 238000002474 experimental method Methods 0.000 description 8
- 239000000243 solution Substances 0.000 description 8
- 238000005457 optimization Methods 0.000 description 7
- 239000013558 reference substance Substances 0.000 description 7
- 238000013461 design Methods 0.000 description 6
- 230000004044 response Effects 0.000 description 6
- 238000011160 research Methods 0.000 description 5
- 239000000523 sample Substances 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 4
- 238000001514 detection method Methods 0.000 description 4
- 239000003814 drug Substances 0.000 description 4
- 239000012085 test solution Substances 0.000 description 4
- WEVYAHXRMPXWCK-UHFFFAOYSA-N Acetonitrile Chemical compound CC#N WEVYAHXRMPXWCK-UHFFFAOYSA-N 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 239000000843 powder Substances 0.000 description 3
- 238000011084 recovery Methods 0.000 description 3
- 239000012488 sample solution Substances 0.000 description 3
- 239000011550 stock solution Substances 0.000 description 3
- 238000012706 support-vector machine Methods 0.000 description 3
- 238000002604 ultrasonography Methods 0.000 description 3
- 239000004480 active ingredient Substances 0.000 description 2
- 230000017531 blood circulation Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000013401 experimental design Methods 0.000 description 2
- 239000000706 filtrate Substances 0.000 description 2
- 238000004128 high performance liquid chromatography Methods 0.000 description 2
- 238000002347 injection Methods 0.000 description 2
- 239000007924 injection Substances 0.000 description 2
- -1 ketone compounds Chemical class 0.000 description 2
- BDAGIHXWWSANSR-UHFFFAOYSA-N methanoic acid Natural products OC=O BDAGIHXWWSANSR-UHFFFAOYSA-N 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003012 network analysis Methods 0.000 description 2
- 238000002360 preparation method Methods 0.000 description 2
- 239000012088 reference solution Substances 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- OSWFIVFLDKOXQC-UHFFFAOYSA-N 4-(3-methoxyphenyl)aniline Chemical compound COC1=CC=CC(C=2C=CC(N)=CC=2)=C1 OSWFIVFLDKOXQC-UHFFFAOYSA-N 0.000 description 1
- 206010002660 Anoxia Diseases 0.000 description 1
- 241000976983 Anoxia Species 0.000 description 1
- 201000006474 Brain Ischemia Diseases 0.000 description 1
- 206010008120 Cerebral ischaemia Diseases 0.000 description 1
- 206010021143 Hypoxia Diseases 0.000 description 1
- 208000001431 Psychomotor Agitation Diseases 0.000 description 1
- 206010063837 Reperfusion injury Diseases 0.000 description 1
- 206010038743 Restlessness Diseases 0.000 description 1
- 238000002835 absorbance Methods 0.000 description 1
- 230000007953 anoxia Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000740 bleeding effect Effects 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 210000000748 cardiovascular system Anatomy 0.000 description 1
- 206010008118 cerebral infarction Diseases 0.000 description 1
- 229940126678 chinese medicines Drugs 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000010828 elution Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 235000019253 formic acid Nutrition 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003907 kidney function Effects 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000002175 menstrual effect Effects 0.000 description 1
- 239000012982 microporous membrane Substances 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003647 oxidation Effects 0.000 description 1
- 238000007254 oxidation reaction Methods 0.000 description 1
- 230000000144 pharmacologic effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- QEVHRUUCFGRFIF-MDEJGZGSSA-N reserpine Chemical compound O([C@H]1[C@@H]([C@H]([C@H]2C[C@@H]3C4=C(C5=CC=C(OC)C=C5N4)CCN3C[C@H]2C1)C(=O)OC)OC)C(=O)C1=CC(OC)=C(OC)C(OC)=C1 QEVHRUUCFGRFIF-MDEJGZGSSA-N 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 238000013112 stability test Methods 0.000 description 1
- 230000004936 stimulating effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000000967 suction filtration Methods 0.000 description 1
- 239000006228 supernatant Substances 0.000 description 1
- 238000010998 test method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000010200 validation analysis Methods 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B01—PHYSICAL OR CHEMICAL PROCESSES OR APPARATUS IN GENERAL
- B01D—SEPARATION
- B01D11/00—Solvent extraction
- B01D11/02—Solvent extraction of solids
- B01D11/0261—Solvent extraction of solids comprising vibrating mechanisms, e.g. mechanical, acoustical
- B01D11/0265—Applying ultrasound
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16C—COMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
- G16C10/00—Computational theoretical chemistry, i.e. ICT specially adapted for theoretical aspects of quantum chemistry, molecular mechanics, molecular dynamics or the like
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computing Systems (AREA)
- Acoustics & Sound (AREA)
- General Health & Medical Sciences (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Mechanical Engineering (AREA)
- Chemical & Material Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Medicines Containing Plant Substances (AREA)
Abstract
The invention discloses an LS-SVM model-based ultrasonic extraction process for effective components of safflower, which comprises the following steps: s1, ultrasonically extracting the effective components HSYA and AHSYB of the safflower according to the Box-BenhnkenDesign principle; s2, establishing a least square support vector machine model LS-SVM; s3, optimizing a model parameter, namely a kernel parameter g and a penalty factor C, wherein a kernel function adopts a radial basis kernel function RBF and is edited by Matlab software; s4, inputting the optimal parameters g and C in S3 into an LS-SVM model to obtain a comprehensive predictive evaluation value of the Box-BenhnkenDesign analysis scheme in S1; s5, evaluating the simulation performance of the LS-SVM model by using mean square error MSE, wherein the MSE is equal to m, m is the data set number, Yi is the comprehensive evaluation value of the LS-SVM for prediction, and Yi is the real comprehensive evaluation value. The invention solves the problem of unstable high-latitude nonlinear problem in the prior art.
Description
Technical Field
The invention relates to the technical field of traditional Chinese medicine extraction, in particular to an ultrasonic extraction process of effective components of safflower based on an LS-SVM model.
Background
Safflower is a traditional Chinese medicine and has the efficacies of removing blood stasis, stopping bleeding, promoting blood circulation, stimulating the menstrual flow, clearing away the heart-fire, relieving restlessness and the like. The safflower has the functions of protecting cardiovascular system, improving blood circulation, resisting oxidation, protecting cerebral ischemia, protecting reperfusion injury, enhancing anoxia resistance, improving renal function and the like in the pharmacological aspect. The active ingredients of safflower mainly comprise two types, namely fat-soluble safflower ketone compounds and water-soluble phenolic acid compounds, wherein the active ingredients playing a positive role are dozens of types.
A Support Vector Machine (SVM) is a machine learning model for solving a regression problem, and solves a complex nonlinear problem by introducing a kernel function, thereby avoiding a dimension disaster problem of high-latitude space calculation. The least square support vector machine (LS-SVM) is a further improvement of the LS-SVM, and improves the operation speed and reduces the calculation complexity by converting an objective function and optimizing equation conditions. At present, many researchers often analyze Chinese medicine extraction data by operating a neural network model in a Matlab environment, but researches on optimizing the process conditions for extracting effective substances of Chinese medicines by using a least square support vector machine are rarely reported.
Disclosure of Invention
In order to solve the problem of unstable high-latitude nonlinear problem in the prior art, the invention aims to provide an ultrasonic extraction process of effective safflower components based on an LS-SVM model.
In order to achieve the purpose, the invention adopts the following technical scheme: an ultrasonic extraction process of effective components of safflower based on an LS-SVM model comprises the following steps:
s1, carrying out ultrasonic extraction on the safflower effective components HSYA and AHSYB according to the Box-BenhnkenDesign principle, carrying out experimental determination on the effective component extraction rate and the real comprehensive evaluation value of each process, and obtaining an m-group analysis scheme experimental data set D { (X)1，Y1)，(X2，Y2)，…，(Xm，Ym)}，X1，X2…XmIs a combination of m extraction factor processes, Y1，Y2…YmA real comprehensive evaluation value of m extraction rates corresponding to the extraction factor process combination;
s2, establishing a least square support vector machine LS-SVM which is based on Matlab language environment,establishing an extraction factor xiAnd comprehensively evaluating the value y to obtain an extraction factor x1，x2…xiA quantitative relation with the predictive comprehensive evaluation value y of the extraction rate;
s3, optimizing a model parameter kernel parameter g and a penalty factor C, wherein the kernel function adopts a radial basis kernel function RBF, editing by Matlab software, performing cross validation on g and C simultaneously by a cross validation method, and checking each pair of parameter effects one by one in a parameter matrix consisting of g and C to obtain optimal parameters g and C;
s4, inputting the optimal parameters g and C in S3 into an LS-SVM model to obtain a comprehensive predictive evaluation value of the Box-BenhnkenDesign analysis scheme in S1;
s5, evaluating the simulation performance of the LS-SVM model by using mean square error MSE, wherein
m is the number of data sets, yiPredictive comprehensive evaluation value, Y, for LS-SVMiThe evaluation value is a real comprehensive evaluation value.
Preferably, in S1, the real comprehensive evaluation value is a sum of values obtained by multiplying the process extraction rate by a weight, and the weight is calculated by a standard deviation method.
Preferably, in the S1, the optimum ultrasonic extraction conditions of the effective components HSYA and AHSYB of safflower are extraction temperature of 85 ℃, ultrasonic time of 52min, liquid-material ratio of 12:1 and ethanol concentration of 90%.
Preferably, in S5, the MSE is a range in which the predicted overall evaluation value is close to the true overall evaluation value and the model reliability data.
Compared with the prior art, the invention has the following beneficial effects: the LS-SVM model establishes a nonlinear model by using a batch of mutually corresponding input and output data of an ultrasonic extraction process optimization experiment of the effective components of the safflower obtained by the Box-BenhnkenDesign principle, reveals the quantitative relation between the data, has small model error and high reliability, and provides a new thought and reference for the optimization research of the extraction process of the effective components of the traditional Chinese medicine; according to the method, a regression model of the comprehensive evaluation value about the extraction factors is established in a high-dimensional feature space by using a least square support vector machine, the regression model is converted into an optimization model related to the kernel function, and finally, the predictive comprehensive evaluation value with the minimum error with the real comprehensive evaluation value under the optimal parameters is continuously learned through a data set so as to find out the extracted optimal process factors.
Drawings
The invention is described in further detail below with reference to the following figures and detailed description:
FIG. 1 is an HPLC chart of the mixed control solution (a) and the safflower test solution (b) of the present invention;
FIG. 2 is a graph of optimal parameter results for the present invention;
FIG. 3 is a graph of the results of error analysis of experimental data and model predictive data in accordance with the present invention;
FIG. 4 is a comparison of experimental data and model predictive data analysis in accordance with the present invention.
Detailed Description
The following description of the embodiments of the present invention is provided for illustrative purposes, and other advantages and effects of the present invention will become apparent to those skilled in the art from the present disclosure.
Please refer to fig. 1 to 4. It should be understood that the structures, ratios, sizes, and the like shown in the drawings and described in the specification are only used for matching with the disclosure of the specification, so as to be understood and read by those skilled in the art, and are not used to limit the conditions under which the present invention can be implemented, so that the present invention has no technical significance, and any structural modification, ratio relationship change, or size adjustment should still fall within the scope of the present invention without affecting the efficacy and the achievable purpose of the present invention. In addition, the terms "upper", "lower", "left", "right", "middle" and "one" used in the present specification are for clarity of description, and are not intended to limit the scope of the present invention, and the relative relationship between the terms and the terms is not to be construed as a scope of the present invention.
Example 1:
HPLC determination of SalB and Tan IIA:
chromatographic conditions are as follows: the column was Hypersil BDSC18(4.6 mm. times.300 mm, 5 μm); the mobile phase was acetonitrile (a), 0.1% formic acid (B), with gradient elution: 0-12min, 10% -22% A; 12-30min, 22% -30% A; 30-38min, 30% -45% A; 38-50min, 45% -90% A; 50-55min, 90% -90% A; 55-56min, 90-95% A; 56-60min, 95% -10% A; the flow rate is 1 mL/min; the column temperature is 25 ℃; the detection wavelength is 280 nm; the amount of the sample was 10. mu.L.
Preparation of control solutions: precisely weighing proper amounts of HSYA and AHSYB reference substances, respectively placing in 10mL volumetric flasks, adding 75% methanol to scale marks, and fully shaking to obtain single-component reference substance stock solutions containing HSYA1.00mg/mL and AHSYB 1.00mg/mL. Precisely sucking the 2 stock solutions 4.00mL and 0.50mL respectively, placing in the same 10mL measuring flask, adding 75% methanol to dilute to scale, and shaking to obtain mixed reference solution (HSYA0.40mg/mL, AHSYB0.05mg/mL).
Preparation of a test solution: precisely weighing 5g of safflower powder sample, placing in a conical flask, performing ultrasonic extraction for 50min under the conditions of liquid-material ratio of 12:1, temperature of 75 ℃, ethanol concentration of 80% and ultrasonic frequency of 40kHz, and performing suction filtration to obtain filtrate. Centrifuging the filtrate at 5000r/min for 10min, and filtering the supernatant with 0.22 μm microporous membrane to obtain test solution.
Drawing a standard curve: accurately sucking appropriate amounts of 1.00mg/mLHSYA and AHSYB reference substance stock solutions respectively, placing in volumetric flasks of different specifications, adding 75% methanol to scale marks, shaking up, and preparing No. 1-5 mixed reference substance solutions with different concentrations, wherein the concentrations of HSYA in the No. 1-5 solutions are respectively as follows: 0.30, 0.40, 0.70, 0.90, 1.00 mg/mL; the concentrations of AHSYB were: 0.03, 0.04, 0.05, 0.06, 0.08 mg/mL. According to sample injection detection under the chromatographic condition, a standard curve is drawn by taking the solution concentration as an abscissa (X) and the absorbance as an ordinate (Y), and the obtained HSYA standard curve is as follows: y is 9.30 multiplied by 106X +1.60 multiplied by 105(R2 is 0.9990), and has good linear relation in the concentration range of 0.30-1.00 mg/mL; the AHSYB standard curve is: y is 2.10X 107X-6.24X 103(R2 is 0.9995), and has good linear relation in the concentration range of 0.03-0.08 mg/mL. When the signal-to-noise ratio is 3:1, the obtained HSYA and AHSYB minimum detection limits are 0.05 and 0.01mg/mL respectively.
Example 2:
specificity test: based on the example 1, the mixed reference solution and the safflower sample solution with proper amount are precisely absorbed and respectively measured according to the chromatographic conditions, the HSYA and AHSYB peaks are completely separated, the peak shape is better, the separation degree is more than 1.50, and the method is shown in figure 1, which shows that the specificity of the method is good.
And (3) precision test: accurately sucking a proper amount of HSYA of 0.50mg/mL and AHSYB reference substance solution of 0.05mg/mL, respectively, and repeatedly and parallelly determining for 6 times according to chromatographic conditions to obtain peak areas RSD of 0.69% and 0.76%, which indicates that the precision of the instrument is good.
And (3) stability test: and (3) sampling and measuring the sample solution for 0, 2, 4, 6 and 8 hours under chromatographic conditions and room temperature conditions, recording the chromatographic peak areas of HSYA and AHSYB, and calculating RSD values to be 1.24% and 1.68% respectively, which indicates that the sample solution is stable within 8 hours.
And (3) repeatability test: precisely weighing 5g of safflower powder in the same batch, preparing 6 parts of supply solution, and measuring under a chromatographic condition to obtain the RSD value of the HSYA peak area of 0.35 percent and the RSD value of the AHSYB peak area of 0.28 percent, which indicates that the method has good repeatability.
Sample recovery rate test: precisely weighing 5g of safflower powder of the same batch, and 6 parts in total, adding 0.50mg of HSYA reference substance and 0.10mg of AHSYB reference substance into each part to prepare a test solution, and measuring according to chromatographic conditions, wherein the average recovery rate of the HSYA is 98.39%, and the RSD is 0.52%; the average recovery of AHSYB was 97.8%, and the RSD was 0.57%.
Example 3:
one-factor experiment (based on example 1):
influence of ethanol concentration on extraction yield: precisely weighing 5 parts of 5g of safflower respectively, and performing ultrasonic extraction for 40min under the conditions of liquid-material ratio of 12:1, temperature of 55 ℃ and ethanol concentration of 50%, 60%, 70%, 80% and 90% and ultrasonic frequency of 40kHz respectively, wherein experimental results show that the extraction rates of HSYA and AHSYB are highest when the ethanol concentration is 80%, so 70%, 80% and 90% are selected as three levels of the ethanol concentration in response surface design.
Influence of ultrasound time on extraction yield: precisely weighing 5 parts of 5g of safflower respectively, and performing ultrasonic extraction for 20min, 30min, 40min, 50min and 60min under the conditions of liquid-material ratio of 12:1, temperature of 55 ℃ and ethanol concentration of 80% and ultrasonic frequency of 40kHz respectively, wherein experimental results show that the extraction rates of HSYA and AHSYB are highest when the ultrasonic time is 50min, so that 40min, 50min and 60min are selected as three levels of ultrasonic time in response surface design.
Influence of ultrasound temperature on extraction yield: precisely weighing 5 parts of 5g safflower respectively, and respectively measuring the safflower at 45, 55, 65, 75, 55, 75 and 75 ℃ under the conditions of liquid-material ratio of 12:1, ethanol concentration of 80% and ultrasonic frequency of 40kHz,
Ultrasonic extraction is carried out for 50min at the temperature of 85 ℃, and experimental results show that when the ultrasonic temperature is 75 ℃, the extraction rates of HSYA and AHSYB are the highest, so that 65 ℃, 75 ℃ and 85 ℃ are selected as three levels of the ultrasonic temperature in response surface design.
Influence of liquid-material ratio on extraction rate: precisely weighing 5g of safflower respectively, and performing ultrasonic extraction for 50min at ethanol concentration of 80%, temperature of 75 deg.C and ultrasonic frequency of 40kHz according to liquid-material ratio of 8:1, 10:1, 12:1, 14:1 and 16: 1. The experimental result shows that when the liquid-material ratio is 12: at 1, the extraction rates of HSYA and AHSYB are highest, so 10:1, 12:1 and 14:1 are selected as three levels of liquid-to-material ratio in response surface design.
Influence of extraction times on extraction yield: respectively and precisely weighing 5 parts of 5g of safflower, and respectively extracting for 1, 2, 3 and 4 times under the conditions of 80% ethanol concentration, 50min time, 75 ℃ temperature and 12:1 liquid-material ratio, wherein experimental results show that when the extraction times are 3, the extraction rates of HSYA and AHSYB are relatively high, and the extraction rates tend to be gentle after 3 times, which indicates that the extraction of HSYA and AHSYB is finished at the moment.
Example 4:
response surface method optimization process:
factor level design: on the basis of the single-factor test in the embodiment 3, because the extraction frequency has little influence on the extraction rate, the ultrasonic time (A), the liquid-material ratio (B), the ultrasonic temperature (C) and the ethanol concentration (D) are selected as four factors, and finally, the response surface test design of 4-factor 3 level is carried out, so that the optimal extraction process parameters are optimized, and the test factors and specific level values are shown in the table 1.
TABLE 1 BBD test factors and codings
The test method comprises the following steps: precisely weighing 30 parts of 5g of safflower, selecting factor levels of each experimental group number according to a BBD principle, carrying out sample injection detection under a chromatographic condition to obtain peak area values of HSYA and AHSYB, respectively calculating the concentrations of the HSYA and the AHSYB according to a regression equation, and then combining an extraction rate formula c v/m 1000, wherein c is the concentration (mg/mL) of the HSYA and the AHSYB obtained by calculation through the calibrated regression equation, v is the total volume (mL) of an extracting solution, m is the mass (g) of the safflower to calculate the respective extraction rates of the HSYA and the AHSYB, and the experimental group number and results are shown in a table 2.
And (3) calculating a comprehensive evaluation value: the weighting values were calculated by a network analysis method, and the weighting coefficients of HSYA and AHSYB were 0.48 and 0.52, respectively, and the calculation results of the comprehensive evaluation values, i.e., HSYA extraction ratio × 0.48+ AHSYB extraction ratio × 0.52, are shown in table 2.
TABLE 2BBD analysis protocol and Experimental results
Example 5:
modeling and analyzing by a least square support vector machine:
establishing a least square support vector machine model:
the idea of the LS-SVM model is to input X for the ith set of data of the experimental plani＝(x1,x2…xn) Representing the corresponding n values of the extraction factor and the ith data output YiRepresenting the overall evaluation value for the extraction yield, given the known m sets of BBD analysis protocol experimental data sets D { (X)1,Y1),(X2,Y2),…,(Xm,Ym) And exploring the quantitative relation between the extraction factors and the comprehensive evaluation value. Specifically for the present experiment, least squares support is utilizedThe vector machine establishes a regression model of the comprehensive evaluation value about extraction factors, introduces a Lagrange multiplier according to a structure risk minimization principle, converts the regression model into an optimization model related to a kernel function under an optimization theory Kuhn-Tucker condition (KKT), and finally continuously learns through an experimental actual data set and finds out a predictive comprehensive evaluation value with the minimum error with a real comprehensive evaluation value under an optimal parameter so as to simulate the optimal technological factor of safflower extraction. In the research, four extraction process data are set as x based on Matlab2014b language environmentiRespectively, liquid to material ratio (x)1) Ultrasonic temperature (x)2) Ultrasonic power (x)3) Time of ultrasound (x)4) And a comprehensive evaluation value Y, wherein Y is a comprehensive evaluation value calculated by linear weighting of extraction rates of the HSYA and AHSYB in the safflower through a network analysis method.
Optimizing model parameters:
for the small data volume of the experiment, the gaussian kernel function (RBF) has high performance and application range, so when selecting the gaussian kernel function, a proper kernel parameter g and a penalty factor C need to be selected to optimize the LS-SVM model. Firstly, 30 groups of experimental result data of the BBD analysis scheme are preprocessed, and an experimental data matrix with 30 rows and 5 columns is divided into an input variable matrix with 30 rows and 4 columns and an output variable matrix with 30 rows and 1 columns, which are imported into a platform of Matlab2014b software, wherein the input variables are replaced by specific factor level values (the specific values are shown in Table 1). In order to avoid too large dimension difference of input and output variables, the research simultaneously performs normalization processing on experimental data, namely dividing each extracted index factor level value by the mean value of the extracted index factor level values, programming by using Matlab2014b software, modeling by using an LS-SVM (least squares support vector machine), using a Gaussian kernel function as the kernel function, finally performing cross validation on a kernel parameter g and a penalty factor C by using a cross validation method, checking each pair of parameter effects in a parameter matrix formed by g and C one by one, traversing to obtain an optimal kernel parameter g which is 3.4822 and an optimal penalty factor C which is 0.3789, and the traversing process is shown as the following figure 2.
Prediction results and analysis
The optimal kernel parameter g-3.4822 and the penalty factor C-0.3789 can be obtained by analyzing the optimal parameter result diagram of fig. 2 and combining the program results. The comprehensive evaluation prediction values of 30 BBD analysis schemes obtained by returning to the LS-SVM model after obtaining the appropriate nuclear parameter g and penalty factor C are shown in the following table 3.
Table 3 comprehensive evaluation prediction value of LS-SVM model
In this study, the predicted comprehensive evaluation value is compared with the real comprehensive evaluation value of the experiment, the correlation coefficient R between the predicted value and the experimental value is 0.97, the maximum error is 3.8165%, the minimum error is 0.1224%, and the error result is shown in fig. 3 below.
And then using Mean-square error (MSE) to evaluate the performance of the LS-SVM model, wherein a specific formula is as follows:
wherein n is 30 data sets; y isiFor the predicted value of the LS-SVM,
actual values obtained by experiments.
And finally, substituting 30 groups of original experimental values and predicted values into a formula to calculate MSE and drawing a comparison graph of a real comprehensive evaluation value and a predicted comprehensive evaluation value of the experiment shown in the following figure 4.
Fig. 4 shows that the comprehensive evaluation values of the 30 sets of actual data and predicted data are very close, which indicates that the trained data is fitted to the actual test data, and then the MSE is calculated to be 0.003014, indicating that the model training and prediction effects are better.
Conditional optimization and outcome prediction of the model:
based on BBD experimental design, combined with actual reachable conditions of experiments, 34 groups of data sets to be predicted are added in a gradient mode through Matlab2014b software according to 4 factors (the gradient of three factors including ultrasonic time, ultrasonic temperature and ultrasonic power is 0.1, and the gradient of liquid-material ratio is 0.25), and then an LS-SVM model is used for predicting the optimal combination. The optimal process conditions for the final extraction are as follows: the temperature is 68 ℃, the ultrasonic time is 47min, the liquid-material ratio is 12:1, the ultrasonic power is 200W, and the predictive comprehensive evaluation value of the optimal condition is 1.1034.
Example 6:
and (3) verification test:
and weighing 4 parts of safflower medicinal material according to the optimal extraction process conditions, wherein each part is 25g, and carrying out ultrasonic extraction. The extraction rates of HSYA and AHSYB were measured and the overall evaluation value was calculated, and the results are shown in table 4. As can be seen from table 4, the average comprehensive evaluation value of the ultrasonic testing method is 1.1025, the predicted value is 1.1034, and the relative error between the actual value of the test and the predicted value of the model is 0.08157%. The verification tests show that the error is within the allowable range of the tests, so that the LS-SVM model has certain value in optimizing the safflower extraction process, namely the best extraction process condition of the research is the result given by the LS-SVM model: the temperature is 68 ℃, the ultrasonic time is 47min, the liquid-material ratio is 12:1, and the ultrasonic power is 200W.
Table 4 validation test under optimum process
In conclusion, the BBD experimental design and the LS-SVM model are utilized to optimize the extraction process of two effective components, namely HSYA and AHSYB, in the safflower, and the optimal extraction conditions obtained after adjustment according to actual conditions are as follows: the temperature is 68 ℃, the ultrasonic time is 47min, the liquid-material ratio is 12:1, the ultrasonic power is 200W, the comprehensive predictive evaluation value is 1.1034, the average verified true value under the condition is 1.1025, the relative error is 0.08157%, and the error is small, so that the model can reflect the extraction conditions of HSYA and AHSYB to a certain extent.
The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.
Claims (4)
1. An ultrasonic extraction process of effective components of safflower based on an LS-SVM model is characterized by comprising the following steps:
s1, carrying out ultrasonic extraction on the safflower effective components HSYA and AHSYB according to the Box-BenhnkenDesign principle, carrying out experimental determination on the effective component extraction rate and the real comprehensive evaluation value of each process, and obtaining an m-group analysis scheme experimental data set D { (X)1，Y1)，(X2，Y2)，…，(Xm，Ym)}，X1，X2…XmIs a combination of m extraction factor processes, Y1，Y2…YmA real comprehensive evaluation value of m extraction rates corresponding to the extraction factor process combination;
s2, establishing a least square support vector machine LS-SVM, and establishing an extraction factor x based on Matlab language environmentiAnd comprehensively evaluating the value y to obtain an extraction factor x1，x2…xiA quantitative relation with the predictive comprehensive evaluation value y of the extraction rate;
s3, optimizing a model parameter kernel parameter g and a penalty factor C, wherein the kernel function adopts a radial basis kernel function RBF, editing by Matlab software, performing cross validation on g and C simultaneously by a cross validation method, and checking each pair of parameter effects one by one in a parameter matrix consisting of g and C to obtain optimal parameters g and C;
s4, inputting the optimal parameters g and C in S3 into an LS-SVM model to obtain a comprehensive predictive evaluation value of the Box-BenhnkenDesign analysis scheme in S1;
s5, evaluating the simulation performance of the LS-SVM model by using mean square error MSE, wherein
m is the number of data sets, yiPredictive comprehensive evaluation value, Y, for LS-SVMiThe evaluation value is a real comprehensive evaluation value.
2. The ultrasonic extraction process of effective components of safflower based on LS-SVM model as claimed in claim 1, characterized in that: in S1, the real comprehensive evaluation value is a sum of values obtained by multiplying the process extraction rate by a weight, and the weight is calculated by a standard dispersion method.
3. The ultrasonic extraction process of effective components of safflower based on LS-SVM model as claimed in claim 1, characterized in that: in the S1, the optimum ultrasonic extraction conditions of the effective components HSYA and AHSYB of the safflower are that the extraction temperature is 85 ℃, the ultrasonic time is 52min, the liquid-material ratio is 12:1, and the ethanol concentration is 90%.
4. The ultrasonic extraction process of effective components of safflower based on LS-SVM model as claimed in claim 1, characterized in that: in S5, the MSE is a range where the predicted comprehensive evaluation value approaches the true comprehensive evaluation value and the model reliability data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911365093.7A CN112915579A (en) | 2019-12-26 | 2019-12-26 | Effective safflower component ultrasonic extraction process based on LS-SVM model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911365093.7A CN112915579A (en) | 2019-12-26 | 2019-12-26 | Effective safflower component ultrasonic extraction process based on LS-SVM model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112915579A true CN112915579A (en) | 2021-06-08 |
Family
ID=76162132
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911365093.7A Withdrawn CN112915579A (en) | 2019-12-26 | 2019-12-26 | Effective safflower component ultrasonic extraction process based on LS-SVM model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112915579A (en) |
-
2019
- 2019-12-26 CN CN201911365093.7A patent/CN112915579A/en not_active Withdrawn
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103913433A (en) | Online detection method for double-effect concentration process of danhong injection | |
| CN101231274B (en) | Method for rapid measuring allantoin content in yam using near infrared spectrum | |
| CN102621092A (en) | Method for detecting Danhong injection ethanol precipitation process on line | |
| CN110988153B (en) | Ultrasonic extraction process optimization method for effective components of salvia miltiorrhiza based on LS-SVM model | |
| CN102058682A (en) | NIR online detection method for paeoniflorin content of white paeony root extract | |
| CN102252992A (en) | Method for performing near-infrared on-line detection in process of extracting Chinese medicines | |
| CN110376312B (en) | Method for detecting quality grade of traditional Chinese medicine | |
| CN110501435A (en) | The finger print quality detecting method of leaf of Moringa lotus leaf ginseng solid beverage | |
| CN106053384A (en) | Rapid quantitative detection method for sweet wormwood and honeysuckle alcohol precipitation concentration process | |
| CN105203496A (en) | Near-infrared online detecting method for lame impediment tablet concentration process | |
| CN105675750A (en) | Method for constructing HPLC characteristic chromatogram of Chinese patent medicine 'Yishen Bugu liquid' | |
| CN110006886A (en) | A kind of nanosizing color sensitive sensor and its method for differentiating wheat moulding ability | |
| CN112915579A (en) | Effective safflower component ultrasonic extraction process based on LS-SVM model | |
| CN110579548B (en) | Wild chrysanthemum flower medicinal material quality evaluation method and application | |
| Zhang et al. | Assessment of quality consistency in traditional Chinese medicine using multi‐wavelength fusion profiling by integrated quantitative fingerprint method: Niuhuang Jiedu pill as an example | |
| CN108051396A (en) | A kind of rapid detection method of Xin Ke Shu ' tablet for treating coronary heart disease active constituent content | |
| CN107029038B (en) | Response surface method optimized extraction process of longan leaf total flavonoids | |
| CN105092744B (en) | The characteristic spectrum and its method for building up of Desmodium styracifolium extractive of general flavone and application | |
| CN107655849A (en) | A kind of herbal tea near infrared online detection method | |
| CN109223881A (en) | A method of optimization Taohe Chengqi decoction decoction preparation process | |
| CN103335960A (en) | Rapid detection method of key indicators in cinobufagin extraction and concentration processes | |
| CN102323236A (en) | The method of the multiple component content of near infrared ray kuh-seng leaching process | |
| CN108152384B (en) | Analysis method for distinguishing pericarpium citri reticulatae and pericarpium citri reticulatae by combining fingerprint spectrum with chemometrics | |
| CN112881561A (en) | System and method for determining sanshool in rattan pepper | |
| CN113244647A (en) | Rectifying tower VOFFLC temperature control method and system based on matrix decoupling |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210608 |