A method and system for preparing a sweet orange blossom essential oil formula

Through the AMPD algorithm and density peak clustering algorithm, combined with response surface analysis and BP neural network, the sweet orange blossom essential oil formula is optimized, which solves the problems of complex configuration process and low optimization accuracy of the sweet orange blossom essential oil formula, and achieves high-quality sweet orange Flower essential oil production.

CN118657062BActive Publication Date: 2024-11-01WELFIN (BEIJING) TECH DEV CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411125333.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-16
Publication Date
2024-11-01
Estimated Expiration
2044-08-16

AI Technical Summary

Technical Problem

The configuration process of the sweet orange flower essential oil formula is complex and the optimization accuracy is low, resulting in poor product quality. It is difficult for traditional methods to identify the relevant dependencies between process parameters and formula, resulting in errors and waste of resources.

Method used

AMPD algorithm and density peak clustering algorithm are used to obtain independent significant impact index of essential oil factors. Combined with response surface analysis and BP neural network, the sweet orange flower essential oil formula is optimized, and the interaction relationship between multiple factors is considered, and the best formula is adaptively obtained and Optimal process parameters.

Benefits of technology

It improves the accuracy of the formulation configuration of the sweet orange blossom essential oil, ensures product quality, avoids the problem of poor adaptability of the formula and process flow, and enhances resource utilization.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118657062B_ABST
    Figure CN118657062B_ABST
Patent Text Reader

Abstract

The present application relates to the technical field of sweet orange flower essential oil production, and specifically to a method and system for configuring a sweet orange flower essential oil formula, the method comprising: collecting essential oil influencing factors, dividing test groups and level grades; obtaining the comprehensive quality coefficient of the essential oil; then obtaining the test quality stability factor of each essential oil influencing factor; calculating the factor significant assurance coefficient of each essential oil influencing factor; then obtaining the independent significant influence index; obtaining the essential oil factor interaction intensity between any two essential oil influencing factors; then obtaining the essential oil feedback priority index, and configuring the sweet orange flower essential oil formula. The present application solves the problem that the sweet orange flower essential oil formula configuration process is complicated and the optimization accuracy is low.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the technical field of sweet orange blossom essential oil production, and in particular to a method and system for configuring a sweet orange blossom essential oil formula. Background Art

[0002] Sweet orange (Citrus sinensis) is a plant of the genus Citrus in the Rutaceae family. Sweet orange has a wide planting area and high yield, but every year a large number of sweet orange flowers are male-degenerated sterile and deformed flowers. If they are thrown away directly, it will lead to a waste of resources and even affect the local ecosystem. Moroccan bitter orange flower essential oil is a natural active substance extracted from bitter orange flowers. It is widely used in food, daily chemical products and French aromatherapy, but there are problems such as inconvenient transportation, long delivery time and high cost. Sweet orange and bitter orange belong to the same genus of plants. Extracting sweet orange flower essential oil from sweet orange flowers has extremely high economic value, and can improve the overall resource utilization rate of the sweet orange industry and protect the environment.

[0003] When preparing sweet orange flower essential oil using a steam distillation device, factors such as the crushing power of the sweet orange flower, the distillation time, and the solid-liquid ratio will affect the quality of the sweet orange flower essential oil obtained by distillation. The traditional formula configuration uses the control variable method, which only configures the formula. There are problems such as the influence of process parameters on the formula and formula configuration errors. At the same time, both process parameters and formulas are influencing factors, and there are extremely high correlations and dependencies between different influencing factors. The control variable method has the problem of difficulty in identifying these correlations and dependencies, resulting in a complex optimization process for the sweet orange flower essential oil formulation method and low optimization accuracy, and the resulting sweet orange flower essential oil product is of poor quality. Summary of the invention

[0004] In order to solve the above technical problems, the purpose of this application is to provide a method and system for preparing a sweet orange flower essential oil formula, and the technical solutions adopted are as follows:

[0005] In a first aspect, the present application provides a method for preparing a sweet orange flower essential oil formula, the method comprising the following steps:

[0006] Collecting various essential oil influencing factors of sweet orange flower essential oil, dividing each test group and each level grade; each level grade is determined by evenly dividing the value range of the corresponding essential oil influencing factor under a preset number;

[0007] Obtain the comprehensive quality coefficient of essential oils of all levels of each test group in all essential oil influencing factors; obtain the quality sample points of each test group of each essential oil influencing factor based on the comprehensive quality coefficient of essential oils;

[0008] Based on the distance between the quality sample points and the comprehensive quality coefficient of the essential oil, the experimental quality stability factor of each essential oil influencing factor is obtained; based on the experimental quality stability factor and the comprehensive quality coefficient of the essential oil, the factor significance assurance coefficient of each essential oil influencing factor is obtained; based on the factor significance assurance coefficient, the independent significant influence index of each essential oil influencing factor is obtained;

[0009] Based on the comprehensive quality coefficient of essential oils, the interaction intensity of essential oil factors between any two essential oil influencing factors is obtained; based on the independent significant impact index and the interaction intensity of essential oil factors, the essential oil feedback priority index of each essential oil influencing factor is obtained;

[0010] The sweet orange blossom essential oil formula was configured based on the essential oil feedback priority index.

[0011] Further, the method for obtaining the comprehensive quality coefficient of the essential oil is:

[0012] For each level of each test group in each essential oil influencing factor, each quality index in the sweet orange flower essential oil of the level is obtained, and a vector composed of all quality indexes in the sweet orange flower essential oil of the level is used as the essential oil quality vector of each level of each test group in each essential oil influencing factor;

[0013] The quality indicators are the yield, number of common compounds, and total monoterpene content difference of sweet orange flower essential oil;

[0014] The essential oil quality vector is used as the input of the comprehensive index method to obtain the essential oil comprehensive quality coefficient of each level of each test group in each essential oil influencing factor.

[0015] Furthermore, the method for obtaining the quality sample points is:

[0016] For each test group of each essential oil influencing factor, the comprehensive quality coefficient of all essential oils is used as the ordinate, and the horizontal level of the comprehensive quality coefficient of all essential oils is used as the abscissa to construct the factor influencing quality polyline;

[0017] For the factor influence quality broken line of each test group of each essential oil influencing factor, a peak detection algorithm is used to obtain the maximum peak value and peak width in the factor influence quality broken line, and the vector composed of the maximum peak value and the peak width is used as the essential oil quality peak vector of each factor influence quality broken line;

[0018] For each essential oil quality peak vector, the maximum peak value is used as the horizontal coordinate and the peak width is used as the vertical coordinate to construct the quality sample points of each essential oil quality peak vector.

[0019] Furthermore, the method for obtaining the test quality stability factor is:

[0020] The area with each quality sample point as the center and a preset distance as the radius is taken as the neighborhood quality similarity range of each quality sample point;

[0021] In the adjacent quality similarity range of the quality sample points of each test group of each essential oil influencing factor, the proportion of all quality sample points of each test group of each essential oil influencing factor in the adjacent quality similarity range is used as the relative proximity of the test of each test group of each essential oil influencing factor;

[0022] For each essential oil influencing factor, the sum of the relative proximity of all test groups is calculated as the first test sum, and the sum of the distance between the factor influence quality polylines of any two test groups among the essential oil influencing factors and the first preset parameter adjustment factor is calculated as the first distance sum;

[0023] The test quality stability factor of each essential oil influencing factor is obtained according to the first test sum value and the first distance sum value, and the test quality stability factor is positively correlated with the first test sum value and negatively correlated with the first distance sum value.

[0024] Furthermore, the method of significantly ensuring the coefficient of the factor is:

[0025] For each essential oil influencing factor, the sum of the test quality stability factors of each essential oil influencing factor and any other essential oil influencing factor is calculated as the quality stability weight of each essential oil influencing factor; the mean of the comprehensive quality coefficient of essential oils of all test groups in the essential oil influencing factor is calculated as the essential oil quality peak value of each essential oil influencing factor; the mean of the peak width of all test groups of the essential oil influencing factor is calculated as the quality peak span of each essential oil influencing factor;

[0026] Calculate the difference between the essential oil quality peak value of each essential oil influencing factor and the essential oil quality peak value of any other essential oil influencing factor as the first peak difference of each essential oil influencing factor; calculate the difference between the quality peak span of each essential oil influencing factor and the quality peak span of any other essential oil influencing factor as the first span difference of each essential oil influencing factor;

[0027] The factor significant assurance coefficient is obtained according to the quality stability weight, the first peak difference and the first span difference. The factor significant assurance coefficient is positively correlated with the quality stability weight, positively correlated with the first peak difference, and negatively correlated with the first span difference.

[0028] Furthermore, the method for obtaining the independent significant impact index is:

[0029] The set of all quality sample points is taken as the quality sample set, and the clustering algorithm is used to process the quality sample set to obtain each cluster, which is used as the quality similarity class of each essential oil;

[0030] For each essential oil influencing factor, the essential oil quality similarity class of the test group with the largest number of quality sample points is taken as the representative quality similarity class of each essential oil influencing factor;

[0031] For each essential oil influencing factor, the number of quality sample points of the test group of any other essential oil influencing factor in the representative quality similarity class of the essential oil influencing factor is counted as the essential oil quality similarity between each quality essential oil factor and any other essential oil influencing factor;

[0032] The independent significant influence index of each essential oil influencing factor is obtained according to the significant ensuring coefficient of the factor and the essential oil quality similarity. The independent significant influence index is positively correlated with the significant ensuring coefficient of the factor and negatively correlated with the essential oil quality similarity.

[0033] Further, the method for obtaining the interaction intensity of the essential oil factors is:

[0034] The response surface analysis algorithm is used to obtain the response surface analysis stereogram between any two essential oil influencing factors;

[0035] For any two essential oil influencing factors, the maximum values ​​in the factor influence quality polylines of all test groups of the any two essential oil influencing factors are obtained, which are respectively used as the first quality maximum value and the second quality maximum value, and the maximum value between the first quality maximum value and the second quality maximum value is used as the single factor optimal essential oil quality between the any two essential oil influencing factors;

[0036] For the corresponding surface analysis stereogram of any two essential oil influencing factors, the response surface point where the essential oil comprehensive quality coefficient is greater than or equal to the single factor best essential oil quality is taken as the interactive quality enhancement point, and the ratio of the number of all interactive quality enhancement points to the number of response surface points at all levels of the any two essential oil influencing factors is counted, and recorded as the interactive significant interval proportion of each essential oil influencing factor in the any two essential oil influencing factors;

[0037] For any two essential oil influencing factors, the sum of the interactive significant interval proportions of the two essential oil influencing factors is calculated as the first significant sum value, the sum of the independent significant impact indexes of the two essential oil influencing factors is calculated as the first independent sum value, and the difference between the essential oil comprehensive quality coefficient of each interactive quality enhancement point in the response surface analysis stereogram of the two essential oil influencing factors and the single factor optimal essential oil quality of the two essential oil influencing factors is calculated as the first quality difference;

[0038] The interaction strength of the essential oil factor is obtained according to the first significant sum value, the first independent sum value and the first quality difference value, wherein the interaction strength of the essential oil factor is positively correlated with the first significant sum value, negatively correlated with the first independent sum value, and positively correlated with the first quality difference value.

[0039] Further, the method for obtaining the essential oil feedback priority index is:

[0040] For each essential oil influencing factor, the sum of the essential oil factor interaction intensities of the essential oil influencing factor and all other essential oil influencing factors is calculated as the interaction intensity cumulative coefficient of each essential oil influencing factor, and the difference between the interaction intensity cumulative coefficient of each essential oil influencing factor and the interaction intensity cumulative coefficient of other essential oil influencing factors is calculated as each first intensity difference of each essential oil influencing factor;

[0041] The essential oil feedback priority index is obtained according to the first intensity difference and the independent significant impact index. The essential oil feedback priority index is positively correlated with the first intensity difference and is positively correlated with the independent significant impact index.

[0042] Further, the configuration of the sweet orange flower essential oil formula based on the essential oil feedback priority index includes:

[0043] The essential oil feedback priority indexes of all essential oil influencing factors are sorted in descending order to obtain the optimized sequence, and the optimized sequence and all response surface analysis stereograms are used as inputs of the essential oil formula optimization model. The essential oil formula optimization model is obtained by BP neural network training, and the optimal formula and optimal process parameters of sweet orange flower essential oil are output.

[0044] In a second aspect, an embodiment of the present application further provides a sweet orange blossom essential oil formula configuration system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein when the processor executes the computer program, the steps of any one of the above-mentioned sweet orange blossom essential oil formula configuration methods are implemented.

[0045] This application has at least the following beneficial effects:

[0046] The present application relates to a method and system for configuring a sweet orange flower essential oil formula. According to the essential oil quality of each essential oil influencing factor at different levels in the production process of the sweet orange flower essential oil, an AMPD algorithm and a density peak clustering algorithm are used to obtain the independent significant influence index of all essential oil influencing factors, reflecting the influence of independent essential oil influencing factors on the quality of the sweet orange flower essential oil. The beneficial effect is that multiple test groups are comprehensively considered to avoid errors in single test analysis and improve the stability of single factor tests. Based on the interactive correlation between the essential oil influencing factors, the essential oil feedback priority index of all essential oil influencing factors can be obtained to reflect the priority of configuration selection of each essential oil influencing factor. A neural network is used to adaptively obtain the best formula and optimal process parameters of the sweet orange flower essential oil according to the comprehensive influence of the essential oil influencing factors on the quality of the sweet orange flower essential oil. The beneficial effect is that the problem of poor adaptability between the formula and the process flow is avoided by directly configuring the formula and ignoring the process parameters. At the same time, considering the extremely high correlation dependency between different influencing factors, the problem of complex configuration process and low optimization accuracy of the sweet orange flower essential oil formula configuration is solved, the accuracy of the formula configuration is improved, and the quality of the sweet orange flower essential oil product is ensured. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0048] Figure 1 A flowchart of a method for preparing a sweet orange flower essential oil formula provided in one embodiment of the present application;

[0049] Figure 2 This is a schematic diagram of the sweet orange flower essential oil test framework;

[0050] Figure 3 Obtain a flow chart for the Essential Oil Feedback Priority Index. DETAILED DESCRIPTION

[0051] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following is a detailed description of a method and system for configuring a sweet orange flower essential oil formula proposed in the present application, its specific implementation, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0052] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0053] The specific scheme of a sweet orange blossom essential oil formula configuration method and system provided by the present application is described in detail below with reference to the accompanying drawings.

[0054] See also Figure 1 , which shows a step flow chart of a method for preparing a sweet orange flower essential oil formula provided by an embodiment of the present application, the method comprising the following steps:

[0055] Step S1, obtaining the factor influence quality polyline of each test group among all essential oil influencing factors through single factor test and GCMS test.

[0056] The main steps of producing the sweet orange flower essential oil are as follows: weighing the sweet orange flowers and pouring them from the top of a feeding cylinder, using a tissue crusher, setting the crushing power, and crushing the sweet orange flowers, the purpose of which is to break the cell structure of the sweet orange flowers and release the essential oil components therein, so as to prepare for subsequent distillation and extraction, wherein the sweet orange flowers are sweet orange flowers from Gannan, Jiangxi Province, China; weighing the crushed sweet orange flowers, adding an amount of sweet orange flowers, placing them in a distillation bottle, then adding double distilled water of an added amount of double distilled water, setting the distillation temperature and the distillation time, separating and collecting, and obtaining the sweet orange flower essential oil.

[0057] First, this embodiment selects crushing power, amount of sweet orange flower added, amount of double distilled water added, distillation temperature and distillation time as essential oil influencing factors in turn. This embodiment obtains the value range of each essential oil influencing factor, and divides the value range of each essential oil influencing factor into a preset number of m sub-ranges, and takes the boundary values ​​of all sub-ranges as horizontal levels in turn, where the preset number m is 6 in this embodiment, and the total number of horizontal levels is m+1. For example, for the distillation time of sweet orange flower essential oil production, the minimum distillation time of 5 hours and the maximum distillation time of 8 hours are obtained as the value range of the distillation time [5,8], and the horizontal levels are 5, 5.5, 6, 6.5, 7, 7.5, and 8 in turn.

[0058] Secondly, in order to avoid errors in single test analysis, this embodiment sets N test groups for each test, and N in this embodiment is 5. Single factor tests are performed on all test groups of each essential oil influencing factor using the control variable method, wherein the levels of each factor in the single factor test are the levels of each essential oil influencing factor, and the sweet orange flower essential oil corresponding to each level of each test group in each essential oil influencing factor is obtained, and the yield of sweet orange flower essential oil is counted, wherein the single factor test is a well-known technology, and the specific process is not repeated here.

[0059] Furthermore, 0.5 μL of sweet orange flower essential oil was taken for GCMS testing, and the total ion flow diagram of the sweet orange flower essential oil corresponding to each level of each test group in each essential oil influencing factor was obtained. Through spectral library retrieval and manual analysis, the number of common compounds between sweet orange flower essential oil and Moroccan bitter orange flower essential oil was counted, and the difference in the total monoterpene content between sweet orange flower essential oil and Moroccan bitter orange flower essential oil was calculated. For specific steps, please refer to the literature: Comparative study on the components of sweet orange flower essential oil and Moroccan bitter orange flower essential oil in Gannan, Jiangxi, China.

[0060] Furthermore, the yield, number of common compounds and total monoterpene content difference of sweet orange flower essential oil were taken as quality indicators of sweet orange flower essential oil to form the essential oil quality vector. Secondly, the essential oil quality vectors of all levels of each experimental group in all essential oil influencing factors were taken as input, and the CIM (Comprehensive Index Method) comprehensive index method was used to obtain the essential oil comprehensive quality coefficients of each level of each experimental group in all essential oil influencing factors.

[0061] Furthermore, the standard index method is used to eliminate the dimensional influence of each quality index, and the standard deviation weight method is used to obtain the weight of each quality index in the CIM comprehensive index method. The CIM comprehensive index method, the standard deviation weight method and the standard index method are all well-known technologies, and the specific acquisition process will not be repeated. The test framework of sweet orange flower essential oil is as follows: Figure 2 shown.

[0062] At this point, according to the above steps, the essential oil comprehensive quality coefficient corresponding to all level grades of each test group in all essential oil influencing factors can be obtained. For each essential oil influencing factor in the same test group, the obtained essential oil comprehensive quality coefficient is recorded in the form of a broken line in ascending order of level grades, which is recorded as the factor influencing quality broken line, wherein each test group of each essential oil influencing factor has a factor influencing quality broken line.

[0063] Step S2, according to the production process of sweet orange flower essential oil, obtain the independent significant influence index of all essential oil influencing factors, reflecting the influence of independent essential oil influencing factors on the quality of sweet orange flower essential oil; based on the interactive correlation between the essential oil influencing factors, the essential oil feedback priority index of all essential oil influencing factors can be obtained.

[0064] In the process of obtaining sweet orange flower essential oil by distillation, each essential oil influencing factor will have a significant impact on the quality of sweet orange flower essential oil. For example, the crushing power is set to crush the sweet orange flower. The level of the crushing power will affect the degree of crushing of the sweet orange flower. The higher the crushing power, the higher the degree of crushing of the sweet orange flower. The more serious the cell structure of the sweet orange flower is destroyed, the more essential oil components contained in the sweet orange flower are released, which directly affects the quality of the sweet orange flower essential oil. In the process of formula configuration, it is often necessary to first determine the configuration order of each essential oil influencing factor. Therefore, the purpose of the present invention is to obtain the configuration priority weight of each essential oil influencing factor according to the degree of influence of each essential oil influencing factor on the quality of the sweet orange flower essential oil.

[0065] The effect of each essential oil influencing factor on the quality of sweet orange flower essential oil may not be a linear relationship. For example, when the crushing power is too low, the sweet orange flower cannot fully release the essential oil components, resulting in a low extraction rate of sweet orange flower essential oil, low quality of sweet orange flower essential oil, and a small comprehensive quality coefficient of essential oil. When the crushing power is too high, the sweet orange flower cells are overly damaged, and some of the compounds in them, such as monoterpenes, may be fully exposed to the air and undergo oxidation reactions, resulting in low quality of sweet orange flower essential oil and a small comprehensive quality coefficient of essential oil. The greater the impact of the essential oil influencing factors on the quality of sweet orange flower essential oil, the more critical the essential oil influencing factors are to the configuration of the formulation method, and the more they should be configured first.

[0066] The factor-affecting quality curve of each test group in each essential oil influencing factor is used as the input of the AMPD (Adaptive Multi-scale Peak Detection) algorithm, and the output is the maximum peak value and peak width in the factor-affecting quality curve. The peak width refers to the duration of the peak value, which constitutes the essential oil quality peak vector. The AMPD algorithm is a well-known technology and will not be described in detail here.

[0067] The essential oil quality peak vectors of the factor-affecting quality polyline of each test group in each essential oil influencing factor are used as quality sample points to construct a quality sample set, and the quality sample set is used as input, and the density peak DPC clustering algorithm is used. The cross-validation method is used to obtain the cutoff distance of the DPC clustering algorithm, and K clusters are output. The factor-affecting quality polylines corresponding to each quality sample point in the same cluster have similar essential oil quality characteristics, and each cluster is recorded as an essential oil quality similarity class. The density peak DPC clustering algorithm is a well-known technology and will not be described in detail here.

[0068] Furthermore, each quality sample point is taken as the center, and the area within the radius with the cutoff distance is taken as the neighboring quality similarity range of each quality sample point.

[0069] For each essential oil influencing factor, the essential oil quality similarity class of the test group containing the largest number of quality sample points is taken as the representative quality similarity class of each essential oil influencing factor.

[0070] Based on the above analysis, the experimental quality stability factor of each essential oil influencing factor can be obtained:

[0071] In the adjacent quality similarity range of the quality sample points of each test group of each essential oil influencing factor, the proportion of all quality sample points of each test group of each essential oil influencing factor in the adjacent quality similarity range is taken as the relative proximity of the test of each test group of each essential oil influencing factor.

[0072] For each essential oil influencing factor, the sum of the relative proximity of all test groups is calculated as the first test sum, and the sum of the distance between the factor influence quality polylines of any two test groups in the essential oil influencing factors and the first preset parameter adjustment factor is calculated as the first distance sum. In this embodiment, the sum of the first preset parameter adjustment factor is 1.

[0073] The test quality stability factor of each essential oil influencing factor is obtained according to the first test sum value and the first distance sum value, and the test quality stability factor is positively correlated with the first test sum value and negatively correlated with the first distance sum value.

[0074] Specifically, the present application proposes an embodiment in which the ratio of the first test sum value to the first distance sum value is used as a test quality stabilization factor.

[0075] It should be noted that the positive correlation relationship can be a multiplication, addition, or idempotent relationship of an exponential function, which can be determined based on actual conditions and will not be elaborated in this application; the negative correlation relationship can be a subtraction relationship, a division relationship, etc., which can be determined based on actual conditions and will not be elaborated in this application.

[0076] Furthermore, the significant factor assurance coefficient of each essential oil influencing factor is obtained according to the test quality stability factor and the essential oil comprehensive quality coefficient, and the acquisition method is:

[0077] For each essential oil influencing factor, the sum of the test quality stability factors of each essential oil influencing factor and any other essential oil influencing factor is calculated as the quality stability weight of each essential oil influencing factor; the mean of the comprehensive quality coefficient of essential oils of all test groups in the essential oil influencing factor is calculated as the essential oil quality peak of each essential oil influencing factor; the mean of the peak width of all test groups of the essential oil influencing factor is calculated as the quality peak span of each essential oil influencing factor.

[0078] The difference between the essential oil quality peak value of each essential oil influencing factor and the essential oil quality peak value of any other essential oil influencing factor is calculated as the first peak difference of each essential oil influencing factor, and the difference between the quality peak span of each essential oil influencing factor and the quality peak span of any other essential oil influencing factor is calculated as the first span difference of each essential oil influencing factor.

[0079] The factor significant assurance coefficient is obtained according to the quality stability weight, the first peak difference and the first span difference. The factor significant assurance coefficient is positively correlated with the quality stability weight, positively correlated with the first peak difference, and negatively correlated with the first span difference.

[0080] Specifically, the present application proposes an embodiment, in which the ratio of the first peak difference to the first span difference is calculated as the first kurtosis ratio, and the product of the first kurtosis ratio and the quality stability weight is calculated as the first quality product. For each essential oil influencing factor, the sum of all the first quality products of the essential oil influencing factors is used as the factor significance assurance coefficient of each essential oil influencing factor.

[0081] Furthermore, the independent significant impact index of each essential oil influencing factor is obtained according to the significant factor ensuring coefficient, and the acquisition method is:

[0082] For each essential oil influencing factor, the number of quality sample points of the test group of any other essential oil influencing factor in the representative quality similarity class of the essential oil influencing factor is counted as the essential oil quality similarity between each quality essential oil factor and any other essential oil influencing factor.

[0083] The independent significant influence index of each essential oil influencing factor is obtained according to the significant ensuring coefficient of the factor and the essential oil quality similarity. The independent significant influence index is positively correlated with the significant ensuring coefficient of the factor and negatively correlated with the essential oil quality similarity.

[0084] Specifically, the present application proposes an embodiment, calculating the sum of the similarity of each essential oil influencing factor to the essential oil quality of any essential oil influencing factor and the second preset parameter adjustment factor as the first similarity sum value, calculating the sum of all the first similarity sum values ​​of each essential oil influencing factor as the first summation result, calculating the ratio of the factor significant guarantee coefficient of each essential oil influencing factor to the first summation result as the factor significant guarantee coefficient of each essential oil influencing factor. In this embodiment, the value of the second preset parameter adjustment factor is 1, and the implementer can select other values ​​according to actual conditions.

[0085] The more similar the factor influence quality broken lines of all test groups in the essential oil influencing factors are, and the higher the relative proximity of the tests of all test groups in the essential oil influencing factors is, the smaller the distance between the factor influence quality broken lines is, and the larger the first test and value is, it means that the analysis results of each test group of sweet orange flower under the fixed essential oil influencing factors are more similar as a whole, the more accidental interference factors of the test can be eliminated, and the accuracy of the formula configuration can be improved. The more stable the quality of sweet orange flower essential oil reflected in all test groups is, the more research value there is for subsequent analysis, the larger the test quality stability factor is, and the greater the quality stability weight is.

[0086] The larger the essential oil quality peak of the essential oil influencing factor and the smaller the quality peak span, that is, the larger the first kurtosis ratio, the more significant the impact of the essential oil influencing factor on the quality of sweet orange flower essential oil, and the shorter the duration of the significant impact. When configuring the sweet orange flower essential oil formula, the more priority should be given to ensuring the level of the essential oil influencing factor, and the larger the factor significance ensuring coefficient.

[0087] The lower the similarity between the essential oil influencing factors and other essential oil influencing factors in essential oil quality, the more unique and independent the local peak characteristics of all test groups of the essential oil influencing factors are, that is, the more independent and significant the impact of the essential oil influencing factors on the quality of sweet orange flower essential oil is, and the greater the independent significant impact index is.

[0088] At this point, the independent significant impact index of each essential oil influencing factor is obtained.

[0089] In the process of formula configuration of sweet orange flower essential oil, there are two or more essential oil influencing factors that influence each other and have a strong correlation. For example, when the distillation temperature is increased, the essential oil components in the crushed sweet orange flower evaporate faster, and a longer distillation time is not required. Too fast distillation time may cause the decomposition of heat-sensitive or highly volatile compounds in the sweet orange flower, thereby reducing the concentration in the sweet orange flower essential oil, which has a greater impact on the quality of the sweet orange flower essential oil, that is, it also has a significant impact on the formula configuration of the sweet orange flower essential oil, but the independent significant influence index is small at this time, and there is a large error in the optimization of the configuration order of the sweet orange flower essential oil formula configuration.

[0090] In order to further study the interaction between the factors affecting essential oils, the configuration order of each factor affecting essential oils was determined. According to all the levels of all the factors affecting essential oils, two unequal crop variables of the factors affecting essential oils were selected each time, and the comprehensive quality coefficient of essential oils was used as the response value. A response surface analysis experiment was performed, and a response surface analysis RSM algorithm was used to obtain a response surface analysis stereogram of any two unequal factors affecting essential oils, and the response surface analysis stereogram was a three-dimensional image. Among them, the response surface analysis RSM algorithm is a well-known technology and will not be described in detail here.

[0091] Furthermore, for any two essential oil influencing factors, the maximum values ​​in the factor influence quality polylines of all test groups of the any two essential oil influencing factors are obtained as the first quality maximum value and the second quality maximum value, respectively, and the maximum value between the first quality maximum value and the second quality maximum value is taken as the single factor optimal essential oil quality between the any two essential oil influencing factors.

[0092] Furthermore, for the corresponding surface analysis stereogram of any two essential oil influencing factors, the response surface point where the essential oil comprehensive quality coefficient is greater than or equal to the single factor best essential oil quality is taken as the interactive quality enhancement point, where it should be noted that the horizontal coordinate of the response surface point is the essential oil influencing factor, and the vertical coordinate is the horizontal level of the essential oil influencing factor.

[0093] Specifically, for the response surface analysis stereogram of any two essential oil influencing factors, the ratios of the number of all interactive quality enhancement points to the number of response surface points at all horizontal levels of the any two essential oil influencing factors are counted and recorded as the proportion of the interactive significant interval of each essential oil influencing factor in the any two essential oil influencing factors.

[0094] Based on the above analysis, the essential oil factor interaction strength between any two essential oil influencing factors is obtained by:

[0095] For any two essential oil influencing factors, the sum of the proportions of the interactive significant intervals of the two essential oil influencing factors is calculated as the first significant sum, the sum of the independent significant influence indexes of the two essential oil influencing factors is calculated as the first independent sum, and the difference between the essential oil comprehensive quality coefficient of each interactive quality enhancement point in the response surface analysis stereogram of the two essential oil influencing factors and the single factor optimal essential oil quality of the two essential oil influencing factors is calculated as the first quality difference. The interaction intensity of the essential oil factors is positively correlated with the first significant sum, negatively correlated with the first independent sum, and positively correlated with the first quality difference.

[0096] Specifically, the present application proposes an embodiment, calculating the ratio of the first significant sum value to the first independent sum value as the first ratio, calculating the sum of all the first quality differences between two essential oil influencing factors as the first quality sum value, and taking the product of the first ratio and the first quality sum value as the essential oil factor interaction intensity of the two essential oil influencing factors.

[0097] Furthermore, the essential oil feedback priority index of each essential oil influencing factor is obtained according to the independent significant impact index and the interaction intensity of the essential oil factors. The acquisition method is:

[0098] For each essential oil influencing factor, the sum of the essential oil factor interaction intensities of the essential oil influencing factor and all other essential oil influencing factors is calculated as the interaction intensity cumulative coefficient of each essential oil influencing factor, and the difference between the interaction intensity cumulative coefficient of each essential oil influencing factor and the interaction intensity cumulative coefficient of other essential oil influencing factors is calculated as each first intensity difference of each essential oil influencing factor. The essential oil feedback priority index is positively correlated with the first intensity difference and is positively correlated with the independent significant influence index. The essential oil feedback priority index acquisition flow chart is as follows: Figure 3 shown.

[0099] Specifically, the present application proposes an embodiment, for each essential oil influencing factor, the sum of all the first intensity differences of the essential oil influencing factor is calculated as the first intensity sum value, and the normalized value of the product of the first intensity sum value and the independent significant influence index is calculated as the essential oil feedback priority index of each essential oil influencing factor.

[0100] When the independent significant influence index of the two essential oil influencing factors is larger, the first independent sum value is larger, which means that the independence and irrelevance of the two essential oil influencing factors are lower, and the interactive influence characteristics between the two essential oil influencing factors should be considered more. When the proportion of the interactive significant interval of the two essential oil influencing factors is larger, that is, the first significant sum value is larger, it means that the interval of the significant correlation effect of the two essential oil influencing factors is wider, and the interactive correlation influence of the two essential oil influencing factors at all levels is closer, and the influence intensity of the two essential oil influencing factors should be enhanced.

[0101] When the comprehensive quality coefficient of the essential oil at the interaction quality enhancement point is greater than the single factor best essential oil quality, that is, the first quality difference is greater, it means that the interaction between the two essential oil influencing factors is more significant than the single factor best essential oil quality of the single factor experiment. At this time, the higher the interaction intensity between the two essential oil influencing factors, that is, the greater the interaction intensity of the essential oil factors.

[0102] The more independent and significant the impact of the essential oil influencing factors on the quality of sweet orange flower essential oil, the higher the interaction intensity of the essential oil influencing factors on other essential oil influencing factors. The larger the independent significant influence index and the larger the cumulative coefficient of interaction intensity, the greater the impact of the essential oil influencing factors on the quality of sweet orange flower essential oil, and the more likely it is that it is the dominant factor in the production process of sweet orange flower essential oil, and the more it should be given priority in the configuration of sweet orange flower essential oil formula, that is, the larger the essential oil feedback priority index.

[0103] Step S3, obtaining the essential oil feedback priority index of each essential oil influencing factor, combining the response surface analysis stereogram of any two essential oil influencing factors, and obtaining the optimal formula and optimal process parameters of sweet orange flower essential oil.

[0104] For the essential oil feedback priority index of all essential oil influencing factors, the optimization sequence is obtained by sorting them in descending order, and the historically obtained optimization sequence and all response surface analysis stereograms are used as training sets. The training set is trained using a BP neural network to obtain an essential oil formula optimization model. The currently obtained optimization sequence and all response surface analysis stereograms are used as inputs to the essential oil formula optimization model. The optimizer uses the Adam algorithm, and the loss function uses the MSE function to output the optimal formula and optimal process parameters of sweet orange flower essential oil, among which the optimal process parameters include crushing power, distillation temperature and distillation time.

[0105] Based on the same inventive concept as the above method, an embodiment of the present application also provides a sweet orange flower essential oil formula configuration system, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of a sweet orange flower essential oil formula configuration method described in any one of the above-mentioned sweet orange flower essential oil formula configuration methods are implemented.

[0106] It should be noted that the above sequence of the embodiments of the present application is for description only and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0107] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0108] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for preparing a sweet orange flower essential oil formula, characterized in that: The method comprises the following steps: Collecting various essential oil influencing factors of sweet orange flower essential oil, dividing each test group and each level grade; each level grade is determined by evenly dividing the value range of the corresponding essential oil influencing factor under a preset number; Obtain the comprehensive quality coefficient of essential oils of all levels of each test group in all essential oil influencing factors; obtain the quality sample points of each test group of each essential oil influencing factor based on the comprehensive quality coefficient of essential oils; Based on the distance between the quality sample points and the comprehensive quality coefficient of the essential oil, the experimental quality stability factor of each essential oil influencing factor is obtained; based on the experimental quality stability factor and the comprehensive quality coefficient of the essential oil, the factor significance assurance coefficient of each essential oil influencing factor is obtained; based on the factor significance assurance coefficient, the independent significant influence index of each essential oil influencing factor is obtained; Based on the comprehensive quality coefficient of essential oils, the interaction intensity of essential oil factors between any two essential oil influencing factors is obtained; based on the independent significant impact index and the interaction intensity of essential oil factors, the essential oil feedback priority index of each essential oil influencing factor is obtained; The sweet orange blossom essential oil formula was configured based on the essential oil feedback priority index; The method for obtaining the comprehensive quality coefficient of the essential oil is: For each level of each test group in each essential oil influencing factor, each quality index in the sweet orange flower essential oil of the level is obtained, and a vector composed of all quality indexes in the sweet orange flower essential oil of the level is used as the essential oil quality vector of each level of each test group in each essential oil influencing factor; The quality indicators are the yield, number of common compounds, and total monoterpene content difference of sweet orange flower essential oil; Using the essential oil quality vector as the input of the comprehensive index method to obtain the essential oil comprehensive quality coefficient of each level of each test group in each essential oil influencing factor; The method for obtaining the quality sample points is: For each test group of each essential oil influencing factor, the comprehensive quality coefficient of all essential oils is used as the ordinate, and the horizontal level of the comprehensive quality coefficient of all essential oils is used as the abscissa to construct the factor influencing quality polyline; For the factor influence quality broken line of each test group of each essential oil influencing factor, a peak detection algorithm is used to obtain the maximum peak value and peak width in the factor influence quality broken line, and the vector composed of the maximum peak value and the peak width is used as the essential oil quality peak vector of each factor influence quality broken line; For each essential oil quality peak vector, the maximum peak value is used as the horizontal coordinate and the peak width is used as the vertical coordinate to construct the quality sample points of each essential oil quality peak vector; The method for obtaining the test quality stability factor is: The area with each quality sample point as the center and a preset distance as the radius is taken as the neighborhood quality similarity range of each quality sample point; In the adjacent quality similarity range of the quality sample points of each test group of each essential oil influencing factor, the proportion of all quality sample points of each test group of each essential oil influencing factor in the adjacent quality similarity range is used as the relative proximity of the test of each test group of each essential oil influencing factor; For each essential oil influencing factor, the sum of the relative proximity of all test groups is calculated as the first test sum, and the sum of the distance between the factor influence quality polylines of any two test groups among the essential oil influencing factors and the first preset parameter adjustment factor is calculated as the first distance sum; Obtaining a test quality stability factor of each essential oil influencing factor according to the first test sum value and the first distance sum value, wherein the test quality stability factor has a positive correlation with the first test sum value and a negative correlation with the first distance sum value; The method for obtaining the interaction intensity of the essential oil factors is: The response surface analysis algorithm is used to obtain the response surface analysis stereogram between any two essential oil influencing factors; For any two essential oil influencing factors, the maximum values ​​in the factor influence quality polylines of all test groups of the any two essential oil influencing factors are obtained, which are respectively used as the first quality maximum value and the second quality maximum value, and the maximum value between the first quality maximum value and the second quality maximum value is used as the single factor optimal essential oil quality between the any two essential oil influencing factors; For the corresponding surface analysis stereogram of any two essential oil influencing factors, the response surface point where the essential oil comprehensive quality coefficient is greater than or equal to the single factor best essential oil quality is taken as the interactive quality enhancement point, and the ratio of the number of all interactive quality enhancement points to the number of response surface points at all levels of the any two essential oil influencing factors is counted, and recorded as the interactive significant interval proportion of each essential oil influencing factor in the any two essential oil influencing factors; For any two essential oil influencing factors, the sum of the interactive significant interval proportions of the two essential oil influencing factors is calculated as the first significant sum value, the sum of the independent significant impact indexes of the two essential oil influencing factors is calculated as the first independent sum value, and the difference between the essential oil comprehensive quality coefficient of each interactive quality enhancement point in the response surface analysis stereogram of the two essential oil influencing factors and the single factor optimal essential oil quality of the two essential oil influencing factors is calculated as the first quality difference; The interaction strength of the essential oil factor is obtained according to the first significant sum value, the first independent sum value and the first quality difference value, wherein the interaction strength of the essential oil factor is positively correlated with the first significant sum value, negatively correlated with the first independent sum value, and positively correlated with the first quality difference value.

2. A method for preparing a sweet orange flower essential oil formula as claimed in claim 1, characterized in that: The factors significantly ensure the coefficients by: For each essential oil influencing factor, the sum of the test quality stability factors of each essential oil influencing factor and any other essential oil influencing factor is calculated as the quality stability weight of each essential oil influencing factor; the mean of the comprehensive quality coefficient of essential oils of all test groups in the essential oil influencing factor is calculated as the essential oil quality peak value of each essential oil influencing factor; the mean of the peak width of all test groups of the essential oil influencing factor is calculated as the quality peak span of each essential oil influencing factor; Calculate the difference between the essential oil quality peak value of each essential oil influencing factor and the essential oil quality peak value of any other essential oil influencing factor as the first peak difference of each essential oil influencing factor; calculate the difference between the quality peak span of each essential oil influencing factor and the quality peak span of any other essential oil influencing factor as the first span difference of each essential oil influencing factor; The factor significant assurance coefficient is obtained according to the quality stability weight, the first peak difference and the first span difference. The factor significant assurance coefficient is positively correlated with the quality stability weight, positively correlated with the first peak difference, and negatively correlated with the first span difference.

3. A method for preparing a sweet orange flower essential oil formula as claimed in claim 1, characterized in that: The method for obtaining the independent significant impact index is: The set of all quality sample points is taken as the quality sample set, and the clustering algorithm is used to process the quality sample set to obtain each cluster, which is used as the quality similarity class of each essential oil; For each essential oil influencing factor, the essential oil quality similarity class of the test group with the largest number of quality sample points is taken as the representative quality similarity class of each essential oil influencing factor; For each essential oil influencing factor, the number of quality sample points of the test group of any other essential oil influencing factor in the representative quality similarity class of the essential oil influencing factor is counted as the essential oil quality similarity between each quality essential oil factor and any other essential oil influencing factor; The independent significant influence index of each essential oil influencing factor is obtained according to the significant ensuring coefficient of the factor and the essential oil quality similarity. The independent significant influence index is positively correlated with the significant ensuring coefficient of the factor and negatively correlated with the essential oil quality similarity.

4. A method for preparing a sweet orange flower essential oil formula as claimed in claim 1, characterized in that: The method for obtaining the essential oil feedback priority index is: For each essential oil influencing factor, the sum of the essential oil factor interaction intensities of the essential oil influencing factor and all other essential oil influencing factors is calculated as the interaction intensity cumulative coefficient of each essential oil influencing factor, and the difference between the interaction intensity cumulative coefficient of each essential oil influencing factor and the interaction intensity cumulative coefficient of other essential oil influencing factors is calculated as each first intensity difference of each essential oil influencing factor; The essential oil feedback priority index is obtained according to the first intensity difference and the independent significant impact index. The essential oil feedback priority index is positively correlated with the first intensity difference and is positively correlated with the independent significant impact index.

5. A method for preparing a sweet orange flower essential oil formula as claimed in claim 1, characterized in that: The method of configuring the sweet orange flower essential oil formula based on the essential oil feedback priority index includes: The essential oil feedback priority indexes of all essential oil influencing factors are sorted in descending order to obtain the optimized sequence, and the optimized sequence and all response surface analysis stereograms are used as inputs of the essential oil formula optimization model. The essential oil formula optimization model is obtained by BP neural network training, and the optimal formula and optimal process parameters of sweet orange flower essential oil are output.

6. A sweet orange flower essential oil formula configuration system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method for configuring a sweet orange blossom essential oil formula as described in any one of claims 1 to 5 are implemented.