Method, device, medium and electronic equipment for detecting the age of liquor base wine

Abstract

The present application discloses a method, device, medium and electronic device for detecting the year of liquor base wine, belonging to the field of data processing technology. The method includes: in response to the acquired target hydroxyl displacement intensity data of the liquor base wine to be detected, the target hydroxyl displacement intensity data is input into the year detection model to obtain the predicted year of the liquor base wine to be detected. The establishment method of the year detection model includes: using the liquor base wine year data and the corresponding hydroxyl displacement intensity data detected in history as samples, dividing into a training set and a test set, the year data is an independent variable, and the hydroxyl displacement intensity data is a dependent variable; training an orthogonal partial least squares discriminant model with the training set, and cross-validating the trained model to obtain an initial year detection model, and validating the initial year detection model with the test set to obtain an optimal year detection model. The present application can improve the accuracy of the target predicted year.

Description

Technical Field

[0001] The present application belongs to the field of data processing technology, and in particular relates to a method, device, medium and electronic equipment for detecting the year of a base liquor of a liquor. Background Art

[0002] Liquor base wine mainly refers to the original liquor obtained after fermentation, distillation and other steps in the production process of liquor. At the same time, it is also the liquor without post-processing such as blending and seasoning, and its degree is high and the flavor is strong. Therefore, the year and quality of liquor base wine determine the year and quality of finished liquor. In the process of liquor brewing, aging is a key step, and as the storage time of aging is longer, physical and chemical reactions are continuously carried out in the liquor until the chemical molecules in the liquor are in a state of dynamic equilibrium. The total content of water and ethanol in liquor accounts for about 98%, and trace components account for about 2%, and there are more free ethanol molecules in the new liquor, and in the process of storage of the liquor, water and ethanol have a strong association ability, and a new molecular association group is formed through association, which can make more free ethanol molecules bound, thereby reducing the stimulation of wine to the taste organs. However, in addition to ethanol-water molecules, there are 2% trace components in the liquor. During the aging process, with the physical and chemical changes, the trace components also show an increase in acid and a decrease in ester. The formation rules of various alcohols, aldehydes, acids, ketones, and esters can be expressed by the hydroxyl displacement intensity between molecules in the liquor system.

[0003] Under natural conditions, organic acids and alcohols in the liquor base undergo reversible conversion reactions with esters, and within a certain period of time, this reversible conversion reaction is conducive to the stabilization of product quality. In the dynamic equilibrium process of the Fischer esterification reaction, various molecules in the liquor base will undergo clustering and a series of reactions. The formation of these cluster molecules and macromolecular groups will increasingly suppress the intensity of hydroxyl displacement in the ethanol-water molecules in the liquor, and a minimum point of hydroxyl displacement can be formed within a period of time. Then, as the acid gradually increases, the electronegativity of the liquor system becomes stronger and stronger, and the electron cloud density around the hydrogen nucleus gradually weakens, forming a deshielding effect, partially breaking the dynamic reaction equilibrium, and the hydrogen nucleus moves from high field to low field. The hydroxyl displacement intensity will gradually increase with the aging time. At this time, the liquor tends to be stable and easier to store.

[0004] The quality identification of liquor base wine is mainly the process of judging and distinguishing the quality grade of liquor base wine to ensure the quality and style of liquor products. In the related art, the year of liquor base wine is identified by professional wine tasting. However, when the year of liquor base wine is identified by professional wine tasting, the judgment is highly subjective and easily affected by the external environment, which leads to inaccurate year of the identified liquor base wine. Summary of the invention

[0005] The purpose of this application is to overcome the defects of the prior art and provide a method, device, medium and electronic equipment for detecting the year of liquor base wine, aiming to solve the problem that the year identification results of liquor base wine are not objective and accurate.

[0006] The purpose of this application is achieved through the following technical solutions:

[0007] A method for detecting the year of a liquor base wine, the method comprising:

[0008] In response to the acquired target hydroxyl displacement intensity data of the white wine base wine to be tested, the target hydroxyl displacement intensity data is input into a year detection model to obtain a predicted year of the white wine base wine to be tested;

[0009] The method for establishing the year detection model includes:

[0010] The year data of the liquor base wine detected in history and the corresponding hydroxyl displacement intensity data are used as samples and divided into a training set and a test set, wherein the year data is an independent variable and the hydroxyl displacement intensity data is a dependent variable;

[0011] The orthogonal partial least squares discriminant model is trained with the training set, and the trained model is cross-validated to obtain an initial year detection model, and the initial year detection model is verified with the test set to obtain an optimal year detection model.

[0012] Furthermore, the target hydroxyl shift intensity data is obtained by measuring the white wine base wine to be tested by nuclear magnetic resonance spectroscopy.

[0013] Furthermore, the method further comprises:

[0014] The standard deviation of the optimal year detection model is calculated, and then the error range is obtained from the standard deviation. When the error range exceeds the first error threshold, the year detection model is re-established with new historical detected white wine base wine year data and corresponding hydroxyl displacement intensity data.

[0015] Furthermore, the step of validating the initial year detection model with the test set to obtain the optimal year detection model specifically includes:

[0016] Inputting the hydroxyl displacement intensity data in the test set into the initial year detection model to obtain a first predicted year;

[0017] When the deviation between the first predicted year and the base liquor year data in the test set is less than a first year threshold, the year detection model is determined as the optimal year detection model.

[0018] Furthermore, the step of validating the initial year detection model with the test set to obtain the optimal year detection model specifically includes:

[0019] Obtaining aldehyde substance information and ketone substance information corresponding to the liquor base liquor in the test set;

[0020] Obtaining a first score corresponding to the aldehyde substance information and a second score corresponding to the ketone substance information by a principal component analysis method, and constructing a first scatter plot with the first score as the abscissa and the second score as the ordinate;

[0021] Inputting the historical hydroxyl displacement intensity data in the test set into the year detection model to obtain a second predicted year, and constructing a second scatter plot using the second predicted year as the abscissa and the hydroxyl displacement intensity data in the test set as the ordinate;

[0022] When the scatter point trend in the first scatter point diagram is the same as the scatter point trend in the second scatter point diagram, the year detection model is determined as the optimal year detection model.

[0023] Furthermore, the step of validating the initial year detection model with the test set to obtain the optimal year detection model specifically includes:

[0024] Inputting the historical hydroxyl displacement intensity data in the test set into the year detection model to obtain a third predicted year;

[0025] When the third predicted year is within a preset confidence interval, the year detection model is determined as the optimal year detection model.

[0026] Furthermore, the step of validating the initial year detection model with the test set to obtain the optimal year detection model specifically includes:

[0027] Inputting the historical hydroxyl displacement intensity data in the test set into the year detection model according to a first preset order to obtain a fourth predicted year;

[0028] Inputting the historical hydroxyl displacement intensity data in the test set into the year detection model according to a second preset order to obtain a fifth predicted year;

[0029] When the fourth predicted year is the same as the fifth predicted year, the year detection model is determined as the optimal year detection model.

[0030] On the other hand, the present application also provides a device for detecting the year of a liquor base wine, the device comprising:

[0031] A prediction module, in response to the acquired target hydroxyl displacement intensity data of the white wine base wine to be tested, inputs the target hydroxyl displacement intensity data into a year detection model to obtain a predicted year of the white wine base wine to be tested;

[0032] A modeling module, wherein the modeling module uses the historically detected liquor base wine year data and the corresponding hydroxyl displacement intensity data as samples, and is divided into a training set and a test set, wherein the year data is an independent variable, and the hydroxyl displacement intensity data is a dependent variable;

[0033] The orthogonal partial least squares discriminant model is trained with the training set, and the trained model is cross-validated to obtain an initial year detection model, and the initial year detection model is verified with the test set to obtain an optimal year detection model.

[0034] On the other hand, the present application also provides a computer device, which includes a processor and a memory, in which a computer program is stored. The computer program is loaded and executed by the processor to implement any of the above-mentioned methods for detecting the year of a base liquor.

[0035] On the other hand, the present application also provides a computer-readable storage medium, in which a computer program is stored. The computer program is loaded and executed by a processor to implement any of the above-mentioned methods for detecting the year of a liquor base wine.

[0036] The beneficial effects of this application are:

[0037] The present application obtains target hydroxyl shift intensity data of the white wine base wine to be tested, and the target hydroxyl shift intensity data is obtained by measuring the white wine base wine to be tested by nuclear magnetic resonance spectroscopy; the target hydroxyl shift intensity data is input into the optimal year detection model to obtain the target predicted year of the white wine base wine to be tested, wherein the optimal year detection model is used to predict the year of the white wine base wine. By processing the target hydroxyl shift intensity data with the optimal year detection model to obtain the target predicted year of the white wine base wine to be tested, the subjectivity of tasting the white wine base wine can be reduced, and thus the accuracy of the target predicted year can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a flowchart of a method for detecting the year of a liquor base liquor in an embodiment of the present application;

[0039] Figure 2 It is a schematic diagram of the spectrum of sample data of white wine base wine by nuclear magnetic resonance spectroscopy in an embodiment of the present application;

[0040] Figure 3 It is a line graph of hydroxyl displacement intensity data of sample data of liquor base wine by nuclear magnetic resonance spectroscopy in an embodiment of the present application;

[0041] Figure 4 This is a structural block diagram of a device for detecting the year of a liquor base wine according to an embodiment of the present application;

[0042] Figure 5 It is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0043] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0044] Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative work shall fall within the scope of protection of this application.

[0045] When the year of the base liquor is identified by professional wine tasting, the judgment is highly subjective and easily affected by the external environment, which may lead to inaccurate year of the identified base liquor.

[0046] In order to solve the above technical problems, the following embodiments of the method, device, medium and electronic device for detecting the year of the base liquor of the present application are proposed.

[0047] Reference Figure 1 ,like Figure 1 The figure is a flowchart of a method for detecting the year of a liquor base liquor in an embodiment of the present application. Specifically, the method comprises the following steps:

[0048] S101: Establish a year detection model.

[0049] S102: In response to the acquired target hydroxyl shift intensity data of the white wine base wine to be tested, the target hydroxyl shift intensity data is input into a year detection model to obtain a predicted year of the white wine base wine to be tested.

[0050] Through the above technical scheme, the target hydroxyl shift intensity data of the white wine base wine to be tested is obtained, and the target hydroxyl shift intensity data is obtained by measuring the white wine base wine to be tested by nuclear magnetic resonance spectroscopy; the target hydroxyl shift intensity data is input into the optimal year detection model to obtain the target predicted year of the white wine base wine to be tested, wherein the optimal year detection model is used to predict the year of the white wine base wine. By processing the target hydroxyl shift intensity data through the optimal year detection model to obtain the target predicted year of the white wine base wine to be tested, the subjectivity of tasting the white wine base wine can be reduced, and the accuracy of the target predicted year can be improved.

[0051] The target hydroxyl displacement intensity data can be the interaction between intramolecular or intermolecular hydrogen bonds, resulting in a change in the chemical displacement of the hydrogen atoms on the hydroxyl group. And the target hydroxyl displacement intensity data can be used to evaluate the specificity and flavor of the base liquor of liquor. Among them, the target hydroxyl displacement intensity data can be measured by nuclear magnetic resonance spectroscopy (NMR, Nuclear Magnetic Resonance), and nuclear magnetic resonance spectroscopy can be used to measure the magnetic resonance phenomenon of atomic nuclei in molecules, thereby obtaining molecular structure and dynamic information. This embodiment of the present disclosure does not make specific limitations on this. Reference Figure 2 ,like Figure 2 The diagram is a schematic diagram of the spectrum of sample data of the liquor base liquor by nuclear magnetic resonance spectroscopy in the embodiment of the present application. Figure 2 In the figure, we can see the spectra of the sample data LZ-1, LZ-3, LZ-5, LZ-8, and LZ-15 of the liquor base wine detected by nuclear magnetic resonance spectroscopy, as well as the corresponding hydroxyl displacement intensity data. Among them, LZ-1 indicates that the base wine is 1 year old, LZ-3 indicates that the base wine is 3 years old, LZ-5 indicates that the base wine is 5 years old, LZ-8 indicates that the base wine is 8 years old, and LZ-15 indicates that the base wine is 15 years old.

[0052] Reference Figure 3 ,like Figure 3 The figure shows a line graph of the hydroxyl displacement intensity data of the sample data of the liquor base liquor by nuclear magnetic resonance spectroscopy in the embodiment of the present application. Figure 3It can be obtained that in the early stage of aging of the liquor base wine, the hydrogen bond association ability between the ethanol and water molecules in the liquor base wine is weak, and the substances are not yet rich, so the hydroxyl displacement intensity data is at a high point. As the liquor aging reaction continues, substances such as acids, esters, and alcohols gradually increase, small molecules gradually become large molecules, and the intermolecular clustering effect is enhanced. The flavor of the liquor base wine is more and more coordinated, and the taste is richer and richer. When the reaction gradually matures, the molecules in the liquor base wine are more and more closely associated, achieving the effect of reducing the degree and removing impurities. At the same time, the dynamic changes of flavor substances continue, and the flavor substances are gradually enriched, which inhibits the hydroxyl displacement intensity. At this time, the base wine reaches the best tasting period. As the reaction further develops, the organic acid content of the liquor base wine becomes higher and higher, and some balances are destroyed. The deshielding effect of the electron cloud near the hydrogen nucleus in the system is weakened, and the hydrogen nucleus gradually moves from the high energy field to the low energy field. At this time, the liquor base wine is the most stable and more and more suitable for storage.

[0053] In this embodiment, the target hydroxyl shift intensity data in the white wine base wine to be tested can be obtained by nuclear magnetic resonance spectroscopy, and the target hydroxyl shift intensity data can be used in the subsequent target prediction year of the white wine base wine to be tested.

[0054] For example, the optimal year detection model can be used to detect the year of the base liquor of liquor. When the target hydroxyl displacement intensity data is input into the optimal year detection model, the target predicted year corresponding to the base liquor of liquor can be obtained. The year detection model is expressed by the following calculation formula:

[0055] A = β*t+α;

[0056] Among them, A is the target hydroxyl displacement intensity data, β is the slope, t is the target year, and α is the intercept.

[0057] Among them, β=(n*Σab-Σa*Σb) / (n*Σa 2- Σa 2 ), α=(Σb-β*Σa) / n, a can be the year, b can be the hydroxyl displacement intensity data, and n can be the total amount of data.

[0058] After obtaining the year detection model, you can evaluate the accuracy of the year detection model by calculating the residual and standard deviation. You can express it as follows:

[0059] c = A-(β*t+α);

[0060] SSE = Σ(c^2);

[0061] MSE = SSE / (n - 2);

[0062] SD = √MSE;

[0063] Among them, c is the residual, SSE is the residual sum of squares, MSE is the mean square error, and SD is the standard deviation.

[0064] Specifically, when there are five historical years, and the historical years are 1, 3, 5, 8, and 15, the corresponding historical hydroxyl displacement intensity data are 4.0058, 3.9336, 4.0258, 3.8876, and 4.0289, respectively. The above calculation formula can be used to calculate SSE=0.001, MSE=0.000000333, SD=0.000577, then the error range is 1.96*SD, and the calculation result is 0.00113. According to the fitting results, the estimated standard deviation is 0.000577, and the error range is ±0.00113. If the error range exceeds a certain threshold, the year detection model can be reconstructed with a new data sample to improve the accuracy of the final year detection model.

[0065] By setting the optimal year detection model to detect the year of the liquor base wine, the year detection of the liquor base wine can be realized intelligently. Compared with the liquor year detected by relevant personnel through wine tasting in related technologies, it can reduce subjective judgment and thus improve the accuracy of the target predicted year.

[0066] Among the possible approaches, the optimal year detection model is obtained by:

[0067] Obtain the historical years of the historically tested liquor base wine and its corresponding historical hydroxyl displacement intensity data;

[0068] Divide the historical years and their corresponding historical hydroxyl displacement intensity data into a training data set and a test data set;

[0069] The historical year is used as the independent variable, the historical hydroxyl displacement intensity data is used as the dependent variable, and the orthogonal partial least squares discriminant model is trained through the training data set, and the trained model is cross-validated to obtain the year detection model;

[0070] The year detection model is verified through the test data set to obtain the optimal year detection model.

[0071] It should be understood that after the liquor base wine has been stored for a certain number of years, the hydroxyl displacement intensity data contained in the liquor base wine will be detected, and the hydroxyl displacement intensity data corresponding to the liquor base wine of different years are different. Therefore, the year detection model can be determined based on the historical year of the liquor base wine obtained by historical detection and the corresponding historical hydroxyl displacement intensity data. Among them, when determining the year detection model, the historical years and the corresponding historical hydroxyl displacement intensity data can be divided into a training data set and a test data set, wherein the historical years can include multiple years, and the corresponding historical hydroxyl displacement intensity data can also include multiple historical years. The training data set can include multiple historical years and corresponding historical hydroxyl displacement intensity data, and the test data set can also include multiple historical years and corresponding historical hydroxyl displacement intensity data. The embodiment of the present disclosure does not make specific limitations on this.

[0072] In the training data set, the historical year can be used as the independent variable and the historical hydroxyl displacement intensity data as the dependent variable to train the orthogonal partial least squares discriminant model (OPLS-DA, Orthogonal Partial Least Squares Discriminant Analysis). Among them, the orthogonal partial least squares discriminant model can be a multivariate statistical analysis method of partial least squares regression and orthogonal signal correction. It is mainly used for data classification and feature selection. After that, the trained model can be cross-validated to obtain the year detection model. By cross-validating the model obtained after training, the robustness of the year detection model and the ability to predict new and unseen data can be improved.

[0073] In a possible way, the year detection model is verified by a test data set to obtain the optimal year detection model, including:

[0074] The hydroxyl displacement intensity data in the test set is input into the year detection model to obtain the first predicted year;

[0075] When the deviation between the first predicted year and the historical year in the test set is less than the first preset threshold, the year detection model is determined as the optimal year detection model. If the deviation exceeds the first preset threshold, the year detection model can be rebuilt with new data samples.

[0076] It should be understood that after the orthogonal partial least squares discriminant model is trained through the training data set to obtain the year detection model, the year detection model can be tested through the test data set. The historical hydroxyl displacement intensity data in the test data set can be input into the year detection model to obtain the first predicted year, and the first predicted year obtained is compared with the historical year in the test data set, and the comparison result is used to determine whether the year detection model is the optimal year detection model.

[0077] For example, when the test data set includes multiple historical years and their corresponding historical hydroxyl displacement intensity data, each historical hydroxyl displacement intensity data is input into the year detection model to obtain a first predicted year, and the first predicted year is compared with the historical year corresponding to the historical hydroxyl displacement intensity data. When the difference between the first predicted year and the historical year is less than a first preset threshold, the year detection model can be determined as the optimal year detection model, wherein the first preset threshold can be the maximum value of the difference between the first predicted year and the historical year when the year detection model is the optimal year detection model.

[0078] By verifying the year detection model based on the difference between the first predicted year and the historical year, the accuracy of the optimal year detection model can be improved, and thus the accuracy of the target year can be improved.

[0079] In a possible manner, the detection method also includes:

[0080] Acquire the aldehyde substance information and ketone substance information of the liquor base wine corresponding to the test data set, and process the aldehyde substance information and ketone substance information by the principal component analysis method to obtain the first score corresponding to the aldehyde substance information and the second score corresponding to the ketone substance information, and use the first score as the horizontal axis and the second score as the vertical axis to construct a first scatter plot;

[0081] The year detection model is verified through the test data set, and the optimal year detection model is obtained, including:

[0082] Input the historical hydroxyl displacement intensity data in the test data set into the year detection model to obtain the second predicted year, and use the second predicted year as the horizontal axis and the hydroxyl displacement intensity data in the test data set as the vertical axis to construct a second scatter plot;

[0083] When the scatter point trend in the first scatter plot is the same as the scatter point trend in the second scatter plot, the year detection model is determined as the optimal year detection model. The determination method of the trend can be:

[0084] The scatter plot is fitted to obtain a straight line, and the judgment is made by whether the difference in slope is less than the threshold.

[0085] It should be understood that when the vintage detection model is verified through a test data set to obtain the optimal vintage detection model, the scatter plot between different components in the liquor base wine detected by the principal component analysis method can be compared with the scatter plot constructed in the test data set for detection.

[0086] In the liquor base wine, the aldehyde substance information can be used to increase the aroma and flavor of the liquor base wine, and the aldehyde substance information can include acetaldehyde and aldehyde, which is not specifically limited in this disclosure. The ketone substance information mainly affects the aroma of the liquor base wine, wherein the ketone substance information can include 2-pentanone and 2-heptanone, which is not specifically limited in this embodiment of the disclosure.

[0087] After obtaining the test data set, the aldehyde substance information and ketone substance information corresponding to each liquor base wine in the test data set can be obtained, and the aldehyde substance information and ketone substance information can be processed by principal component analysis to obtain a first score corresponding to the aldehyde substance information and a second score corresponding to the ketone substance information. Among them, processing the aldehyde substance information and ketone substance information by principal component analysis can be processing the aldehyde substance information and ketone substance information by principal component analysis method in related technologies, which will not be described in detail in the embodiments of the present disclosure.

[0088] The first scatter plot is drawn with the first score of the aldehyde substance information as the horizontal coordinate and the second score of the ketone substance information as the vertical coordinate. The first scatter plot can intuitively show the contribution of the distribution of the test data set in the principal component space, and reflects the ability of the principal component to explain the sample differences. The cumulative difference score in the principal component analysis method can be used to represent the degree of explanation of the variation of the liquor base wine. Afterwards, the first score can be used as the horizontal coordinate and the second score as the vertical coordinate to construct the first scatter plot.

[0089] The historical hydroxyl displacement intensity data in the test data set is input into the year detection model to obtain the second predicted year. In the year detection model, the second predicted year corresponding to the historical hydroxyl displacement intensity data in the test data set can be predicted, and then the second predicted year can be used as the horizontal coordinate, and the historical hydroxyl displacement intensity data can be used as the vertical coordinate to construct a second scatter plot. The data corresponding to one of the years in each test data set corresponds to a point on the graph. By observing the second scatter plot, the historical year and historical hydroxyl displacement intensity data of the liquor base wine can be obtained, and the historical hydroxyl displacement intensity data show an increasing or decreasing trend. And in the second scatter plot, points that appear in the second scatter plot and are significantly deviated from other points may indicate that the sample has abnormalities or special circumstances and needs to be processed accordingly.

[0090] Then, it can be determined whether the year detection model is the optimal year detection model based on the trend between the first scatter plot and the second scatter plot. When the trend between the first scatter plot and the second scatter plot is the same, the year detection model can be determined as the optimal year detection model. By using the principal component analysis method in model verification, the target year of the liquor base wine detected can be made more accurate.

[0091] In a possible way, the year detection model is verified by a test data set to obtain the optimal year detection model, including:

[0092] Inputting the historical hydroxyl displacement intensity data in the test data set into the year detection model to obtain the third predicted year;

[0093] When the third predicted year is within a preset confidence interval, the year detection model is determined as the optimal year detection model, wherein the preset confidence interval is used to characterize the range of years corresponding to the hydroxyl displacement intensity data of the liquor base wine.

[0094] It should be understood that when verifying the year detection model, the historical hydroxyl displacement intensity data in the test data set can be input into the year detection model to obtain a third predicted year, and then the relationship between the third predicted year and the preset confidence interval can be used to determine whether the year detection model is the optimal year detection model. When the third predicted year is within the preset confidence interval, the year detection model can be determined as the optimal year detection model. The preset confidence interval can be a 95% confidence interval.

[0095] A 95% confidence interval is calculated based on the third predicted year in the test data set. This interval can be regarded as an estimate of whether the historical year corresponding to the base liquor and its historical hydroxyl shift intensity data are within the normal range. Each third predicted year and its historical hydroxyl shift intensity data are then compared with this 95% confidence interval. When there is a third predicted year and its corresponding historical hydroxyl shift intensity data within the 95% confidence interval, the third predicted year and its corresponding historical hydroxyl shift intensity data are normal.

[0096] By comparing the third predicted year with the preset confidence interval and then determining whether the detection model for this year is the optimal detection model, the accuracy of the optimal year detection model can be improved, thereby improving the accuracy of the target year estimation.

[0097] In a possible way, the year detection model is verified by a test data set to obtain the optimal year detection model, including:

[0098] Inputting the historical hydroxyl displacement intensity data in the test data set into the year detection model according to a first preset order to obtain a fourth predicted year;

[0099] Inputting the historical hydroxyl displacement intensity data in the test data set into the year detection model according to a second preset order to obtain a fifth predicted year;

[0100] When the fourth predicted year is the same as the fifth predicted year, the year detection model is determined as the optimal year detection model.

[0101] It should be understood that, in the historical years in the test data set, the historical years can be randomly arranged, and the year detection model can be inferred by the method of random arrangement whether the overfitting phenomenon occurs. Thus, the historical years in the test data set can be arranged according to the first preset order and the second preset order, respectively, and the first preset order is different from the second preset order. Afterwards, the historical hydroxyl displacement intensity data corresponding to the first preset order is input into the year detection model to obtain the fourth predicted year, and the historical hydroxyl displacement intensity data corresponding to the second preset order is input into the year detection model to obtain the fifth predicted year. When the fourth predicted year is the same as the fifth predicted year, the year detection model is determined as the optimal year detection model.

[0102] Reference Figure 4 ,like Figure 4 The figure is a structural block diagram of a detection device for the year of a liquor base wine according to an embodiment of the present application. The detection device 400 for the year of a liquor base wine comprises the following structures:

[0103] Prediction module 401, in response to the acquired target hydroxyl displacement intensity data of the base liquor to be tested, the prediction module 401 inputs the target hydroxyl displacement intensity data into the year detection model to obtain the predicted year of the base liquor to be tested;

[0104] Modeling module 402, modeling module 402 uses the historical detected liquor base wine year data and the corresponding hydroxyl displacement intensity data as samples, and divides them into a training set and a test set, with the year data as an independent variable and the hydroxyl displacement intensity data as a dependent variable;

[0105] The orthogonal partial least squares discriminant model is trained with the training set, and the trained model is cross-validated to obtain the initial year detection model. The initial year detection model is verified with the test set to obtain the optimal year detection model.

[0106] Reference Figure 5 ,like Figure 5 The electronic device 500 may include: a processor 501 and a memory 502. The electronic device 500 may also include one or more of a multimedia component 503, an input / output (I / O) interface 504, and a communication component 505.

[0107] The processor 501 is used to control the overall operation of the electronic device 500 to complete all or part of the steps in the above-mentioned method for detecting the age of the base liquor of the liquor. The memory 502 is used to store various types of data to support the operation of the electronic device 500, and these data may include, for example, instructions for any application or method used to operate on the electronic device 500, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 502 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, referred to as EPROM), programmable read-only memory (Programmable Read-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic memory, flash memory, magnetic disk or optical disk. The multimedia component 503 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone, which is used to receive external audio signals. The received audio signal may be further stored in the memory 502 or sent through the communication component 505. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 504 provides an interface between the processor 501 and other interface modules, and the above-mentioned other interface modules may be a keyboard, a mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 505 is used for wired or wireless communication between the electronic device 500 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 505 may include: Wi-Fi module, Bluetooth module, NFC module.

[0108] In an exemplary embodiment, the electronic device 500 can be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to execute the above-mentioned method for detecting the year of the base liquor.

[0109] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned method for detecting the year of the base liquor of the liquor are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 502 including program instructions, and the above-mentioned program instructions can be executed by the processor 501 of the electronic device 500 to complete the above-mentioned method for detecting the year of the base liquor of the liquor.

[0110] In another exemplary embodiment, a computer program product is also provided, which includes a computer program that can be executed by a processor, and when the computer program is executed by the processor, the steps of the above-mentioned method for detecting the year of the white wine base wine are implemented.

[0111] 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 and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A method for detecting the year of a liquor base wine, characterized in that: The method comprises: In response to the acquired target hydroxyl displacement intensity data of the white wine base wine to be tested, the target hydroxyl displacement intensity data is input into the optimal year detection model to obtain the predicted year of the white wine base wine to be tested, wherein the target hydroxyl displacement intensity data is the change in the chemical displacement of the hydrogen atoms on the hydroxyl group caused by the interaction between intramolecular or intermolecular hydrogen bonds; The method for establishing the optimal year detection model includes: The year data of the liquor base wine detected in history and the corresponding hydroxyl displacement intensity data are used as samples and divided into a training set and a test set, wherein the year data is an independent variable and the hydroxyl displacement intensity data is a dependent variable; The orthogonal partial least squares discriminant model is trained with the training set, and the trained model is cross-validated to obtain an initial year detection model, and the initial year detection model is verified with the test set to obtain an optimal year detection model.

2. The method for detecting the year of the liquor base wine according to claim 1, characterized in that: The target hydroxyl displacement intensity data is obtained by measuring the white wine base wine to be tested by nuclear magnetic resonance spectroscopy.

3. The method for detecting the year of the liquor base wine according to claim 1, characterized in that: The method further comprises: The standard deviation of the optimal year detection model is calculated, and then the error range is obtained from the standard deviation. When the error range exceeds the first error threshold, the year detection model is re-established with new historical detected white wine base wine year data and corresponding hydroxyl displacement intensity data.

4. The method for detecting the year of the liquor base wine according to claim 1, characterized in that: The step of validating the initial year detection model with the test set to obtain the optimal year detection model specifically includes: Inputting the hydroxyl displacement intensity data in the test set into the initial year detection model to obtain a first predicted year; When the deviation between the first predicted year and the base liquor year data in the test set is less than a first year threshold, the year detection model is determined as the optimal year detection model.

5. The method for detecting the year of the liquor base wine according to claim 1, characterized in that: The step of validating the initial year detection model with the test set to obtain the optimal year detection model specifically includes: Obtaining aldehyde substance information and ketone substance information corresponding to the liquor base liquor in the test set; Obtaining a first score corresponding to the aldehyde substance information and a second score corresponding to the ketone substance information by a principal component analysis method, and constructing a first scatter plot with the first score as the abscissa and the second score as the ordinate; Inputting the historical hydroxyl displacement intensity data in the test set into the year detection model to obtain a second predicted year, and constructing a second scatter plot using the second predicted year as the abscissa and the hydroxyl displacement intensity data in the test set as the ordinate; When the scatter point trend in the first scatter point diagram is the same as the scatter point trend in the second scatter point diagram, the year detection model is determined as the optimal year detection model.

6. The method for detecting the year of the liquor base wine according to claim 1, characterized in that: The step of validating the initial year detection model with the test set to obtain the optimal year detection model specifically includes: Inputting the historical hydroxyl displacement intensity data in the test set into the year detection model to obtain a third predicted year; When the third predicted year is within a preset confidence interval, the year detection model is determined as the optimal year detection model.

7. The method for detecting the year of the liquor base wine according to claim 1, characterized in that: The step of validating the initial year detection model with the test set to obtain the optimal year detection model specifically includes: Inputting the historical hydroxyl displacement intensity data in the test set into the year detection model according to a first preset order to obtain a fourth predicted year; Inputting the historical hydroxyl displacement intensity data in the test set into the year detection model according to a second preset order to obtain a fifth predicted year; When the fourth predicted year is the same as the fifth predicted year, the year detection model is determined as the optimal year detection model.

8. A device for detecting the year of a base liquor, characterized in that: The device comprises: A prediction module, in response to the acquired target hydroxyl displacement intensity data of the base liquor to be tested, the prediction module inputs the target hydroxyl displacement intensity data into the optimal year detection model to obtain the predicted year of the base liquor to be tested, wherein the target hydroxyl displacement intensity data is the change in the chemical displacement of the hydrogen atoms on the hydroxyl group caused by the interaction between intramolecular or intermolecular hydrogen bonds; A modeling module, wherein the modeling module uses the historically detected liquor base wine year data and the corresponding hydroxyl displacement intensity data as samples, and is divided into a training set and a test set, wherein the year data is an independent variable, and the hydroxyl displacement intensity data is a dependent variable; The orthogonal partial least squares discriminant model is trained with the training set, and the trained model is cross-validated to obtain an initial year detection model, and the initial year detection model is verified with the test set to obtain an optimal year detection model.

9. A computer device, characterized in that: The computer device includes a processor and a memory, wherein a computer program is stored in the memory, and the computer program is loaded and executed by the processor to implement the method for detecting the year of the liquor base wine as claimed in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, which is loaded and executed by a processor to implement a method for detecting the vintage of a liquor base wine as claimed in any one of claims 1 to 7.

Images