Method for decoupling influence of production area climate and brewing process on Baijiu fermentation effect based on model and application

Abstract

The invention belongs to the technical field of wine brewing, and relates to a method for decoupling influence of production area climate and a brewing process on a white wine fermentation effect based on a model, which comprises the following steps: collecting a sample; screening modeling indexes; measuring the screened modeling indexes to obtain variable values; performing data preprocessing; the modeling indexes are imported into an SEM model, and a model for decoupling the influence of the production area climate and the brewing technology on the Baijiu fermentation effect is obtained; and based on the model for decoupling the influence of the production area climate and the brewing process on the Baijiu fermentation effect, decoupling the influence of the production area climate and the brewing process on the Baijiu fermentation effect. Modeling indexes are screened, variable values are measured, the modeling indexes are preprocessed, a model is constructed to decouple the coupling relation among climate elements, brewing process parameters and the Baijiu fermentation effect, the influence chain of the coupling effect between the Baijiu brewing environment and the brewing process parameters is combed, the degree of the coupling effect is quantified, and the Baijiu fermentation effect is obtained. And the influence mechanism of production area climate elements, brewing process technological parameters and fermentation effect parameters is perfected.

Description

Technical Field

[0001] The invention belongs to the technical field of wine making, and relates to a method and application of decoupling the influence of climate of a production area and brewing technology on the fermentation effect of liquor based on a model. Background Art

[0002] Structural equation modeling (SEM) can handle the interrelationships between multiple variables in complex systems. Compared with traditional multivariate statistical methods such as multiple regression, principal component analysis, and cluster analysis, SEM can simultaneously indicate the strength of each relationship, and does not require the observed data to follow a normal distribution, and has no assumptions about the population or measurement scale. In recent years, it has been increasingly used in scientific fields such as econometrics, social sciences, ecology, and hydrology.

[0003] At present, there are many methods for applying SEM models to analyze impact mechanisms. For example, the document "A method for decoupling the impact of meteorological elements and vegetation dynamics on hydrological elements" is mainly in the field of hydrological analysis technology, involving a method for decoupling the impact of meteorological elements and vegetation dynamics on hydrological elements; the document "A method and system for obtaining ecological impact mechanisms" mainly involves the field of ecological protection technology. A method for obtaining ecological impact mechanisms; "A method for analyzing the impact of arbuscular mycorrhizal fungi on plant water use efficiency" involves a method for analyzing the impact of arbuscular mycorrhizal fungi on plant water use efficiency in the field of botany.

[0004] As a typical representative of traditional open natural fermentation, the fermentation effect of liquor is jointly affected by the brewing environment and process control parameters. Due to the production characteristics of open fermentation, the natural climate elements of the liquor production area directly affect the brewing environment. Among the natural climate elements of the production area, meteorological elements such as temperature and humidity can exchange with the brewing environment and affect the brewing environment, and regional precipitation can provide moisture for the brewing environment. On the one hand, climate elements have a direct impact on the brewing effect by affecting the temperature and humidity of the environment in which the brewing microorganisms are located. On the other hand, based on the temperature and humidity of the brewing environment, technicians control process parameters such as sugar and water through the brewing process, thereby indirectly affecting the brewing effect. Therefore, in the process of liquor brewing, there is a complex cause-and-effect relationship between climate elements and process parameter control, and they are coupled and jointly affect the final liquor brewing effect.

[0005] The application of SEM model in the field of liquor brewing, such as the patent "Method for predicting and controlling microbial flora in the process of koji making and its application", discloses a method of judging the fermentation effect of Daqu by observing variables, but does not disclose the influence of latent variables that cannot be directly and accurately observed or can be observed but still need to be integrated by other methods on the fermentation effect. Specifically, only the influence of indoor environmental indicators such as temperature and humidity of the fermentation chamber on the main brewing microbial species is disclosed, but it is impossible to determine the joint influence of the macro-environmental indicators of the brewing area and the regulation of the brewing process on the fermentation brewing effect. Summary of the invention

[0006] The purpose of the present invention is to provide a method and application for decoupling the influence of climate and brewing process in production areas on the fermentation effect of liquor based on a model. The SEM model is applied to remove the coupling relationship between climate factors, brewing process parameters and liquor fermentation effect, and the influence of the complex relationship between climate factors and brewing effect in liquor production areas on the fermentation effect is evaluated. The influence chain of the coupling effect between liquor brewing environment and brewing process parameters is sorted out, the degree of coupling effect is quantified, and finally the mutual influence mechanism of climate factors in production areas, brewing process parameters and fermentation effect parameters is improved to form a corresponding decoupling model. Through this model, the connection between elements such as meteorological data, brewing process data, brewing result data, etc. can be characterized at the level of latent variables, and the fermentation effects under different production environments can be judged, thereby guiding production and improving efficiency.

[0007] In a first aspect, the present invention provides a method for decoupling the influence of climate in a production area and brewing process on the fermentation effect of liquor based on a model, comprising the following steps:

[0008] S1. Sample collection: Collect liquor brewing samples at similar fermentation stages and different fermentation points;

[0009] S2, screening modeling indicators;

[0010] S3, indicator determination: Determine the modeling indicators screened in step S2 to obtain variable values ​​of the screened indicators;

[0011] S4. Data preprocessing: Use the min-max standardization method to preprocess data;

[0012] S5. Model establishment: importing the modeling index into the SEM model to obtain a model for decoupling the influence of the climate of the production area and the brewing process on the fermentation effect of liquor;

[0013] S6. Decoupling the influence of the climate of the production area and the brewing technology on the fermentation effect of liquor based on the model for decoupling the influence of the climate of the production area and the brewing technology on the fermentation effect of liquor.

[0014] In some embodiments, the modeling indicators include meteorological parameters, brewing process parameters and fermentation effect parameters.

[0015] In some embodiments, the meteorological parameter includes at least one of average daily temperature, average daily humidity, and average daily precipitation;

[0016] Preferably, the brewing process parameters include at least one of the acidity, sugar content, water content and starch content of the brewing sample; preferably, the fermentation effect parameters include at least one of the ethanol content, acetic acid content and lactic acid content of the brewing sample.

[0017] In some embodiments, in the data preprocessing step, the fermentation effect parameters correspond to the meteorological parameters of the production area and the process parameters of the brewing process 1-5 days ago;

[0018] Preferably, in the data preprocessing step, the fermentation effect parameters correspond to the meteorological parameters of the production area and the technological parameters of the brewing process 2-4 days ago.

[0019] In some embodiments, the model building step further comprises constructing a set of equations, wherein the set of equations is as follows:

[0020]

[0021] B=d3C+e5; wherein, A represents the meteorological parameters of the production area; A1 represents the average daily temperature; A2 represents the average daily humidity; A3 represents the average daily precipitation; B represents the process parameters of the brewing process; B1 represents the acidity of the brewing sample; B2 represents the sugar content; B3 represents the water content; B4 represents the starch content; C represents the fermentation effect parameter; C1 represents the ethanol content; C2 represents the acetic acid content; C3 represents the lactic acid content; a1, a2, a3 represent the load coefficients of the average daily temperature A1, the average daily humidity A2 and the average daily precipitation A3 on the meteorological parameter A of the production area, that is, the load coefficients of the indicator on the latent variable contribution; the b1, b2, b3, b4 represent the load coefficients of acidity B1, sugar content B2, water content B3 and starch content B4 on brewing process parameter B; the c1, c2, c3 represent the load coefficients of ethanol content C1, acetic acid content C2 and lactic acid content C3 on fermentation effect parameter C; the d1, d2 and d3 represent the path coefficients of production area meteorological parameter A on fermentation effect parameter C, brewing process parameter B on fermentation effect parameter C and production area meteorological parameter A on brewing process parameter B, that is, the size of the influence between latent variables; the e1, e2, e3, e4 and e5 are the correction coefficients of the above single equations respectively.

[0022] In some embodiments, the model building step also includes model testing, specifically: based on the SPSSPRO software, input data and the equation group, calculate and output GFI, CFI, RMR, NFI, and RMSEA evaluation indicators, and judge the effectiveness of the model based on the GFI, CFI, RMR, NFI, and RMSEA evaluation indicators.

[0023] In some embodiments, the data preprocessing distributes the uneven modeling indicators uniformly between [0, 1].

[0024] In some embodiments, the data preprocessing method is:

[0025] 1) Find the minimum and maximum values ​​of each modeling indicator;

[0026] 2) For each data point in the data set, the following formula is used for standardization: X'=(X-min) / (max-min), where X' is the standardized data and X is the original data.

[0027] In some embodiments, in the step of S1, collecting samples, the specific sampling method is as follows: sampling during the open spontaneous fermentation process, collecting 0.5-1.5 parts by weight of samples from the outer layer 0.1m-0.2m deep and the center >0.4m deep of the fermentation pile fermented for 1-5 days, and mixing them.

[0028] In a second aspect, the present invention provides an application of any one of the above methods in decoupling the effects of climate in a production area and brewing technology on the fermentation effect of liquor.

[0029] In summary, the present application includes at least one of the following beneficial technical effects:

[0030] By screening modeling indicators and measuring variable values, the modeling indicators are preprocessed with min-max standardization, and a model is constructed to decouple the coupling relationship between climate factors, brewing process parameters and liquor fermentation effect. The influence of the complex relationship between climate factors and brewing effect in liquor producing areas on fermentation effect is evaluated, the influence chain of the coupling effect between liquor brewing environment and brewing process parameters is sorted out, the degree of coupling effect is quantified, and finally the mutual influence mechanism of climate factors in the producing area, brewing process parameters and fermentation effect parameters is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a framework diagram of the influence of climate factors in the production area and brewing process parameters on the fermentation effect of liquor in the liquor brewing system in Example 1;

[0032] Figure 2This is a model diagram of the influence of climate factors in the production area and brewing process parameters on the fermentation effect of liquor in the liquor brewing system in Example 1; wherein, the solid line is a significant path with p≤0.05, and the dotted line is a non-significant path with p>0.05. DETAILED DESCRIPTION

[0033] The technical solution of the present invention is further described below by specific embodiments, which do not limit the protection scope of the present invention. Some non-essential modifications and adjustments made by others based on the concept of the present invention still fall within the protection scope of the present invention.

[0034] As mentioned above, among the climate elements of the natural environment of the production area, meteorological elements such as temperature and humidity can exchange with the brewing environment and affect the brewing environment, and regional precipitation can provide moisture for the brewing environment. Climate elements have a direct impact on the brewing effect by affecting the temperature and humidity of the environment in which the brewing microorganisms are located. Based on the temperature and humidity of the brewing environment, the brewing process indirectly affects the brewing effect by regulating process parameters such as sugar and water. Therefore, in the process of liquor brewing, there is a complex cause-and-effect relationship between climate elements and process parameter regulation, and they are coupled and jointly affect the final liquor brewing effect.

[0035] Based on this, the present invention attempts to use the SEM model to decouple the coupling relationship between climate factors, brewing process parameters and liquor fermentation effect, and evaluate the influence of the complex relationship between climate factors and brewing effect in liquor producing areas on the fermentation effect of fermented mash. Specifically, in view of the problem that the current regional "climate factors", liquor brewing "process parameters" and "fermentation effect" indicators cannot be directly observed or need to be synthesized by other methods, the present invention uses the average daily temperature, average daily humidity and average daily precipitation as the explicit variables of the latent variable "climate factors", the acidity, sugar, moisture and starch content of the brewing samples as the explicit variables of the latent variable "process parameters", and the ethanol, acetic acid and lactic acid content in the brewing samples as the explicit variables of the latent variable "fermentation effect".

[0036] Then, by fitting the climate factors, brewing process parameters and fermentation effects, the influence chain of the coupling effect between the liquor brewing environment and the brewing process parameters is sorted out, the degree of coupling effect is quantified, and finally the mutual influence mechanism of the climate factors, brewing process parameters and fermentation effect parameters in the production area is improved. The connection between elements such as meteorological data, brewing process data, brewing result data can be characterized in depth at the level of latent variables (which cannot be directly measured in actual work and are called "latent variables" in SEM), and the influence mechanism between the production area environment and brewing elements is carefully explored. Through the analysis of the mediating effect, a clear causal chain between the latent variables is obtained, thereby improving the influence mechanism.

[0037] In the process of trying to use the SEM model to decouple the coupling relationship between climate factors, brewing process parameters and liquor fermentation effect, the meteorological parameters A that characterize the production area were first screened, including: average daily temperature A1, average daily humidity A2, average daily precipitation A3, maximum daily temperature A4, minimum daily temperature A5, and average daily wind speed A6 collected by meteorological sensors; the brewing process parameters B included: acidity B1, sugar content B2, moisture content B3, and starch content B4 of the brewing samples measured by near-infrared technology; the fermentation effect parameters C included: ethanol content C1, acetic acid content C2, and lactic acid content C3 of the brewing samples measured by near-infrared technology. The evaluation indicators failed to meet the requirements and the overall model was invalid.

[0038] So we further tested the influence of various different indicators on the results, and finally found that: the meteorological parameters A characterizing the production area include: the average daily temperature A1, the average daily humidity A2, and the average daily precipitation A3 collected by the meteorological sensor; the process parameters B of the brewing process include: the acidity B1, sugar content B2, water content B3, and starch content B4 of the brewing sample measured by near infrared technology; the fermentation effect parameter C includes: the ethanol content C1, acetic acid content C2, and lactic acid content C3 of the brewing sample measured by near infrared technology. The evaluation index of the structural equation model diagram of the influence of the climate elements and brewing process parameters of the production area on the fermentation effect of the liquor in the liquor brewing system can meet the requirements, and the overall model is effective. On this basis, when trying to reduce or increase at least one indicator, the evaluation indicators of the equation model obtained are not as good as the results of the model obtained by the combination of indicators of this scheme, that is, not as good as the technical scheme of Example 1 of the present invention.

[0039] At the same time, in the data preprocessing process, the present invention initially corresponds the fermentation effect parameter C with the production area meteorological parameter A and the brewing process process parameter B of the day, and then uses the min-max standardization method to preprocess the data for the obtained index data set, resulting in poor evaluation indicators of the model. After the indicators in various aspects of debugging have failed to obtain a good improvement, considering that there is a certain lag in the impact of meteorology and technology on the fermentation effect, it is attempted to correspond the fermentation effect parameter C to the production area meteorological parameter A and the brewing process process parameter B 1-5 days ago, and according to the index data set obtained after the lag correspondence, the min-max standardization method is used for data preprocessing. The results show that the model index evaluation has a significant improvement effect, and with the increase of the lead time, the overall index evaluation effect shows a trend of first rising and then falling, and the fermentation effect parameter C corresponding to the production area meteorological parameter A and the brewing process process parameter B 3 days ago has the best effect.

[0040] In addition, the present invention also attempts to further optimize the solution through other data preprocessing methods, such as using a centralization method for data preprocessing, but the results of the model evaluation index are not as good as the min-max standardization method for data preprocessing. Therefore, the present invention finally adopts the min-max standardization method for data preprocessing.

[0041] Example 1

[0042] A method for decoupling the influence of climate in a production area and brewing technology on the fermentation effect of liquor based on a model, comprising the following steps:

[0043] S1. Sample collection: Collect liquor brewing samples at similar fermentation stages and different fermentation points;

[0044] S2. Screening modeling indicators: Select indicators that characterize the meteorological parameters of the production area, the process parameters of the brewing process, and the parameters that characterize the fermentation effect;

[0045] S3, indicator determination: Determine the modeling indicators screened in step S2 to obtain variable values ​​of the screened indicators;

[0046] S4. Data preprocessing: Use the min-max standardization method to preprocess data;

[0047] S5. Model establishment: The modeling characteristics characterizing the meteorological parameters of the production area and the process parameters characterizing the brewing process are imported into the model as independent variables, and the parameters characterizing the fermentation effect are imported into the model as dependent variables, so as to obtain a model that decouples the influence of the climate of the production area and the brewing process on the fermentation effect of liquor;

[0048] S6. Decouple the influence of climate of production area and brewing technology on the fermentation effect of liquor based on the established model.

[0049] In this embodiment, the sample collection step in S1 is specifically as follows: 100 samples of liquor fermented grains at similar fermentation stages and different fermentation points are collected daily as samples, and at the same time, meteorological parameters of the production area on the same day, daily, and daily average, brewing process parameters, fermentation effect parameters and other indicators are collected;

[0050] The specific sampling method is as follows: Brewing samples should be sampled during open spontaneous fermentation processes such as pile fermentation. Collect 1kg of samples from the outer layer (0.1m to 0.2m deep) and the center (>0.4m deep) of the fermentation pile that has been fermenting for 1 to 5 days. After quickly mixing, take out 0.5kg and place it in a sterile bag, and record the sampling time, team and cellar number.

[0051] In some embodiments, the fermentation sample is at least one of distiller's grains, daqu, and fermented grains, such as distiller's grains or daqu. In this embodiment, the fermentation sample is fermented grains.

[0052] In some embodiments, the number of fermentation samples is 50-500, such as 50, 100, 150, 200, 300, 350, 400, 500, etc. In this embodiment, the number of fermentation samples is 100.

[0053] In this embodiment, the specific method of S2 screening modeling indicators is:

[0054] The specific screening method is to select indicators around the three aspects of production area meteorology, brewing process technology and fermentation effect. Meteorological data is composed of daily average temperature, daily average humidity, daily average precipitation, daily maximum temperature, daily minimum temperature, and daily average wind speed data, striving to fully and comprehensively represent the multiple dimensions of meteorological data. However, as in Example 1, when the above 6 meteorological indicators are used to represent the meteorological data of the production area to construct the SEM model, the evaluation indicators of the structural equation model diagram of the influence of climate factors and brewing process parameters in the production area on the fermentation effect of liquor in the liquor brewing system do not meet the parameter requirements. Considering that the information required by the model may need to be further summarized, the above 6 meteorological indicators are removed one by one to simplify the model, and finally the daily average temperature, daily average humidity and daily average precipitation are selected as model indicators to summarize the meteorological information of the production area. The brewing process parameters select important control indicators of the brewing process (acidity, sugar content, water content, starch content), and comprehensively summarize the intensity of brewing process regulation on the brewing samples. The fermentation effect parameters include ethanol content, acetic acid content and lactic acid content, which include ethanol and main organic acids (acetic acid and lactic acid), which are important main products in the liquor brewing process.

[0055] Through the above method to screen the modeling indicators, the following indicators are finally determined as the features required for building the model:

[0056] Meteorological parameters A that characterize the production area include: average daily temperature A1, average daily humidity A2, and average daily precipitation A3;

[0057] Characterize the brewing process parameters B, including: brewing sample acidity B1, sugar content B2, water content B3 and starch content B4;

[0058] The parameters C that characterize the fermentation effect include: the ethanol content C1, the acetic acid content C2 and the lactic acid content C3 of the brewing sample.

[0059] In this embodiment, the specific method of measuring the index in S3 is:

[0060] The indicators screened in S2 are measured to obtain indicator measurement data, specifically:

[0061] The average daily temperature A1, average daily humidity A2, and average daily precipitation A3 collected by meteorological sensors;

[0062] The acidity B1, sugar content B2, moisture content B3 and starch content B4 of brewing samples measured by near-infrared technology are as follows: stir the brewing sample to be tested evenly, remove obvious impurities, and load it into the sample dish of the near-infrared spectrometer in a non-agglomerated state, and ensure that the irradiated surface is flat and seamless, and the sample does not precipitate liquid. Collect near-infrared spectra of the same sample three times, and synthesize to form the average near-infrared spectrum of the sample; then use the calibration model of the acidity, sugar, moisture and starch content of the fermented mash to obtain the acidity, sugar content, moisture content and starch content indicators of the fermented mash to be tested.

[0063] The ethanol content C1, acetic acid content C2 and lactic acid content C3 of the brewing sample measured by near-infrared technology are as follows: stir the brewing sample to be tested evenly, remove obvious impurities, and load it into the sample dish of the near-infrared spectrometer in a non-caking state, and ensure that the irradiated surface is flat and seamless, and the sample does not precipitate liquid. Collect the near-infrared spectrum of the same sample three times and synthesize it to form the average near-infrared spectrum of the sample. Then use the calibration model of the ethanol, acetic acid and lactic acid content of the fermented grains to obtain the ethanol content, acetic acid content and lactic acid content indicators of the fermented grains to be tested.

[0064] In this embodiment, the specific method of S4 data preprocessing is:

[0065] The brewing process parameter B and fermentation effect parameter C of the brewing sample set in the target analysis time period are screened out, and the daily average value of each index is taken as the average index of the brewing process parameter B and fermentation effect parameter C of the brewing sample set of that day. Considering the hysteresis effect of meteorology and process on the fermentation effect, the fermentation effect parameter C is corresponded to the meteorological parameter A and brewing process parameter B of the production area three days ago, and the min-max normalization method is used for data preprocessing based on the index data set obtained after the hysteresis correspondence. The specific method is as follows:

[0066] 1) Find the minimum (min) and maximum (max) values ​​of each observed variable;

[0067] 2) For each data point in the data set, the following formula is used for standardization: X'=(X-min) / (max-min) (where X' is the standardized data and X is the original data).

[0068] The standardized data X' will fall between [0,1].

[0069] The purpose of preprocessing is to evenly distribute the uneven variables between [0,1]. For example, the unprocessed temperature is distributed between [20,30]℃, the unprocessed acidity is distributed between [0,5], and the unprocessed precipitation is distributed between [0,1000]mm. In this way, due to the inconsistency of units and dimensions, it will cause the wrong result that the precipitation effect > temperature effect > acidity effect. Therefore, it is necessary to carry out standardization work to standardize temperature, acidity and precipitation to [0,1], so as to better explore the important influencing factors.

[0070] In this embodiment, the specific method of establishing the S5 model is as follows:

[0071] Based on SPSSPRO software, a structural equation model (SEM) was constructed, and the selected indicators were imported into data, path construction, and model testing in turn, as follows:

[0072] The selected modeling indicators are used as features to construct the model, that is, the modeling features characterize the meteorological parameters A of the production area, including: average daily temperature A1, average daily humidity A2, average daily precipitation A3; characterize the process parameters B of the brewing process, including: acidity B1, sugar content B2, moisture content B3 and starch content B4 of the brewing sample, which are imported into the model and used as independent variables; the parameters C characterizing the fermentation effect, including: ethanol content C1, acetic acid content C2 and lactic acid content C3 of the brewing sample, are imported into the model and used as dependent variables. According to the actual situation, the fermentation effect is jointly regulated by the macro-environmental parameters and the process parameters, and the process parameters will be adjusted individually due to changes in the macro-environmental parameters, so a path pointing to a specific latent variable is established ( Figure 1 ), construct the system of equations:

[0073]

[0074] Among them, a1, a2, a3 respectively represent the load coefficients of daily average temperature A1, daily average humidity A2 and daily average precipitation A3 on the production area meteorological parameter A (the contribution of the indicator to the latent variable); b1, b2, b3, b4 respectively represent the load coefficients of acidity B1, sugar content B2, water content B3 and starch content B4 on brewing process parameter B; c1, c2, c3 respectively represent the load coefficients of ethanol content C1, acetic acid content C2 and lactic acid content C3 on fermentation effect parameter C; d1, d2 and d3 respectively represent the path coefficients (the size of the influence between latent variables) of production area meteorological parameter A on fermentation effect parameter C, brewing process parameter B on fermentation effect parameter C and production area meteorological parameter A on brewing process parameter B; e1, e2, e3, e4 and e5 are the correction coefficients of the above single equations. Model analysis mainly focuses on path coefficients.

[0075] Based on SPSSPRO software, based on input data and set structural equation groups, the output evaluation indicators such as GFI, CFI, RMR, NFI, RMSEA, etc. are calculated and output, and those that meet the reference indicator range are qualified. The evaluation indicators of the structural equation model diagram of the influence of climate factors and brewing process parameters on the fermentation effect of liquor in the liquor brewing system obtained are shown in Table 1:

[0076] Table 1 Evaluation indicators of model graphs

[0077] Evaluation indicators GFI CFI RMR NFI RMSEA Reference indicators ≥0.9 ≥0.9 ≤0.1 ≥0.9 ≤0.08 Actual indicators 0.931 0.937 0.077 0.900 0.069

[0078] The goodness of fit index (GFI) was 0.931, the comparative fit index (CFI) was 0.937, the root mean square of the residual (RMR) was 0.077, the normed fit index (NFI) was 0.900, and the root mean square of the error of approximation (RMSEA) was 0.069. The GFI ≥ 0.90, CFI ≥ 0.90, RMR ≤ 0.10, NFI ≥ 0.90, and RMSEA ≤ 0.08 were met. The model was valid as a whole and was successfully constructed.

[0079] In this embodiment, S6 decouples the influence of the climate of the production area and the brewing process on the fermentation effect of liquor based on the established model for decoupling the influence of the climate of the production area and the brewing process on the fermentation effect of liquor. The specific method is:

[0080] Based on the calculation results of the model, the influence mechanism of climate factors and brewing process parameters in liquor brewing areas on liquor fermentation effect is explained, and the paths with significant coefficients are further selected to discuss the causal relationship. Figure 2 As shown in the figure, the paths between the variables of "macro-environment", "process parameters" and "fermentation effect" are all significant paths, so there is a partial mediating effect in the influence of climate factors and brewing process parameters in liquor brewing areas on liquor fermentation effect. That is, the macro-environment will directly affect the fermentation effect, and the process parameters will respond to changes in the macro-environment to affect the fermentation effect.

[0081] Comparative Example 1

[0082] A method for decoupling the influence of production area climate and brewing process on liquor fermentation effect based on a model, which differs from Example 1 in that:

[0083] In the step of screening modeling indicators, meteorological parameters A characterizing the production area include: meteorological sensors collect average daily temperature A1, average daily humidity A2, average daily precipitation A3, daily maximum temperature A4, daily minimum temperature A5, and daily average wind speed A6; process parameters B characterizing the brewing process include: acidity B1, sugar content B2, moisture content B3, and starch content B4 of brewing samples are imported into the model as independent variables; parameters C characterizing the fermentation effect include: ethanol content C1, acetic acid content C2, and lactic acid content C3 of brewing samples are imported into the model as dependent variables.

[0084] In the model building step, construct the equation system:

[0085]

[0086] Among them, a1, a2, a3, a4, a5, a6 respectively represent the load coefficients of daily average temperature A1, daily average humidity A2, daily average precipitation A3, daily maximum temperature A4, daily minimum temperature A5, and daily average wind speed A6 on the production area meteorological parameter A (the contribution of the indicator to the latent variable); b1, b2, b3, and b4 respectively represent the load coefficients of acidity B1, sugar content B2, water content B3, and starch content B4 on the brewing process parameter B; c1, c2, and c3 respectively represent the load coefficients of ethanol content C1, acetic acid content C2, and lactic acid content C3 on the fermentation effect parameter C; d1, d2, and d3 respectively represent the path coefficients (the size of the influence between latent variables) of the production area meteorological parameter A on the fermentation effect parameter C, the brewing process parameter B on the fermentation effect parameter C, and the production area meteorological parameter A on the brewing process parameter B; e1, e2, e3, e4, and e5 are the correction coefficients of the above single equations. The model analysis mainly focuses on the path coefficients.

[0087] The evaluation index of the structural equation model diagram of the influence of climate factors and brewing process parameters on the fermentation effect of liquor in the liquor brewing system obtained in this way is shown in Table 2:

[0088] Table 2 Evaluation indicators of model graphs

[0089] Evaluation indicators GFI CFI RMR NFI RMSEA Reference indicators ≥0.9 ≥0.9 ≤0.1 ≥0.9 ≤0.08 Actual indicators 0.513 0.541 0.315 0.641 0.431

[0090] The goodness of fit index (GFI) of the model was 0.513, the comparative fit index (CFI) was 0.541, the root mean square of the residual (RMR) was 0.315, the normed fit index (NFI) was 0.641, and the root mean square of the error of approximation (RMSEA) was 0.431. It did not meet GFI≥0.90, CFI≥0.90, RMR≤0.10, NFI≥0.90, and RMSEA≤0.08, and the model was invalid as a whole.

[0091] Comparative Example 2

[0092] A method for decoupling the influence of production area climate and brewing process on liquor fermentation effect based on a model, which differs from Example 1 in that:

[0093] In the data preprocessing step, the indicators of brewing process parameter B and fermentation effect parameter C of the brewing sample set in the target analysis time period are screened, and the daily average value of each indicator is taken as the average indicator of brewing process parameter B and fermentation effect parameter C of the brewing sample set of that day. The fermentation effect parameter C is matched with the production area meteorological parameter A and brewing process parameter B on that day, and the min-max standardization method is used to perform data preprocessing on the obtained indicator data set.

[0094] The evaluation index of the structural equation model diagram of the influence of climate factors and brewing process parameters on the fermentation effect of liquor in the liquor brewing system obtained in this way is shown in Table 3:

[0095] Table 3 Evaluation indicators of model graphs

[0096] Evaluation indicators GFI CFI RMR NFI RMSEA Reference indicators ≥0.9 ≥0.9 ≤0.1 ≥0.9 ≤0.08 Actual indicators 0.632 0.612 0.529 0.437 0.315

[0097] The goodness of fit index (GFI) of the model was 0.632, the comparative fit index (CFI) was 0.612, the root mean square of the residual (RMR) was 0.529, the normed fit index (NFI) was 0.437, and the root mean square of the error of approximation (RMSEA) was 0.315. It did not meet GFI≥0.90, CFI≥0.90, RMR≤0.10, NFI≥0.90, and RMSEA≤0.08, and the model was invalid as a whole.

[0098] Comparative Example 3

[0099] A method for decoupling the influence of production area climate and brewing process on liquor fermentation effect based on a model, which differs from Example 1 in that:

[0100] In the data preprocessing step, the indicators of brewing process parameter B and fermentation effect parameter C of the brewing sample set in the target analysis time period are screened, and the daily average value of each indicator is taken as the average indicator of brewing process parameter B and fermentation effect parameter C of the brewing sample set of that day. The fermentation effect parameter C is corresponded to the production area meteorological parameter A and brewing process parameter B three days ago, and the centralization method is used to perform data preprocessing based on the indicator data set obtained after lag correspondence.

[0101] The evaluation index of the structural equation model diagram of the influence of climate factors and brewing process parameters on the fermentation effect of liquor in the liquor brewing system obtained in this way is shown in Table 4:

[0102] Table 4 Evaluation indicators of model graphs

[0103] Evaluation indicators GFI CFI RMR NFI RMSEA Reference indicators ≥0.9 ≥0.9 ≤0.1 ≥0.9 ≤0.08 Actual indicators 0.792 0.815 0.223 0.750 0.148

[0104] The goodness of fit index (GFI) of the model was 0.792, the comparative fit index (CFI) was 0.815, the root mean square of the residual (RMR) was 0.223, the normed fit index (NFI) was 0.750, and the root mean square of the error of approximation (RMSEA) was 0.148. It did not meet GFI≥0.90, CFI≥0.90, RMR≤0.10, NFI≥0.90, and RMSEA≤0.08, and the model was invalid as a whole.

[0105] It is to be understood that the present invention is described by some embodiments, and it is known to those skilled in the art that various changes or equivalent substitutions may be made to these features and embodiments without departing from the spirit and scope of the present invention. In addition, under the teachings of the present invention, these features and embodiments may be modified to adapt to specific circumstances and materials without departing from the spirit and scope of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed herein, and all embodiments falling within the scope of the claims of this application are within the scope of protection of the present invention.

Claims

1. A method for decoupling the influence of climate in production area and brewing process on liquor fermentation effect based on a model, characterized in that: The following steps are involved: S1. Sample collection: Collect liquor brewing samples at similar fermentation stages and different fermentation points; S2, screening modeling indicators; S3, indicator determination: Determine the modeling indicators screened in step S2 to obtain variable values ​​of the screened indicators; S4. Data preprocessing: Use the min-max standardization method to preprocess data; S5. Model establishment: importing the modeling index into the SEM model to obtain a model for decoupling the influence of the climate of the production area and the brewing process on the fermentation effect of liquor; S6. Decoupling the influence of the climate of the production area and the brewing technology on the fermentation effect of liquor based on the model for decoupling the influence of the climate of the production area and the brewing technology on the fermentation effect of liquor.

2. The method for decoupling the influence of climate in the production area and brewing process on the fermentation effect of liquor based on the model as claimed in claim 1, characterized in that: The modeling indicators include meteorological parameters, brewing process parameters and fermentation effect parameters.

3. The method for decoupling the influence of climate in the production area and brewing process on the fermentation effect of liquor based on the model as claimed in claim 2, characterized in that: The meteorological parameters include at least one of average daily temperature, average daily humidity, and average daily precipitation; Preferably, the brewing process parameters include at least one of the acidity, sugar content, water content and starch content of the brewing sample; Preferably, the fermentation effect parameter includes at least one of the ethanol content, acetic acid content and lactic acid content of the brewing sample.

4. The method for decoupling the influence of climate in the production area and brewing process on the fermentation effect of liquor based on the model as claimed in claim 2, characterized in that: In the data preprocessing step, the fermentation effect parameters correspond to the meteorological parameters of the production area and the process parameters of the brewing process 1-5 days ago; Preferably, in the data preprocessing step, the fermentation effect parameters correspond to the meteorological parameters of the production area and the technological parameters of the brewing process 2-4 days ago.

5. The method for decoupling the influence of climate in the production area and brewing process on the fermentation effect of liquor based on the model as claimed in claim 1, characterized in that: The step of model building also includes constructing a set of equations, which are as follows: Wherein, A represents the meteorological parameters of the production area; A1 represents the average daily temperature; A2 represents the average daily humidity; A3 represents the average daily precipitation; B represents the process parameters of the brewing process; B1 represents the acidity of the brewing sample; B2 represents the sugar content; B3 represents the water content; B4 represents the starch content; C represents the fermentation effect parameter; C1 represents the ethanol content; C2 represents the acetic acid content; C3 represents the lactic acid content; a1, a2, and a3 represent the load coefficients of the average daily temperature A1, the average daily humidity A2, and the average daily precipitation A3 on the meteorological parameter A of the production area, that is, the contribution of the indicator to the latent variable; The b1, b2, b3, and b4 respectively represent the load coefficients of acidity B1, sugar content B2, moisture content B3, and starch content B4 on brewing process parameter B; the c1, c2, and c3 respectively represent the load coefficients of ethanol content C1, acetic acid content C2, and lactic acid content C3 on fermentation effect parameter C; the d1, d2, and d3 respectively represent the path coefficients of production area meteorological parameter A on fermentation effect parameter C, brewing process parameter B on fermentation effect parameter C, and production area meteorological parameter A on brewing process parameter B, that is, the size of the influence between latent variables; the e1, e2, e3, e4, and e5 are the correction coefficients of the above-mentioned single equations.

6. The method for decoupling the influence of climate in the production area and brewing process on the fermentation effect of liquor based on the model as claimed in claim 5, characterized in that: The model building step also includes model testing, specifically: based on SPSSPRO software, input data and the equation group, calculate and output GFI, CFI, RMR, NFI, and RMSEA evaluation indicators, and judge the effectiveness of the model according to the GFI, CFI, RMR, NFI, and RMSEA evaluation indicators.

7. The method for decoupling the influence of climate in the production area and brewing process on the fermentation effect of liquor based on the model as claimed in claim 1, characterized in that: The data preprocessing evenly distributes the uneven modeling indicators between [0, 1].

8. The method for decoupling the influence of climate in the production area and brewing process on the fermentation effect of liquor based on the model as claimed in claim 1, characterized in that: The data preprocessing method is: 1) Find the minimum and maximum values ​​of each modeling indicator; 2) For each data point in the data set, the following formula is used for standardization: X'=(X-min) / (max-min), where X' is the standardized data and X is the original data.

9. The method for decoupling the influence of the climate of the production area and the brewing process on the fermentation effect of liquor based on the model as claimed in claim 1, characterized in that: In the step S1, collecting samples, the specific sampling method is as follows: sampling during the open spontaneous fermentation process, collecting 0.5-1.5 parts by weight of samples from the outer layer 0.1m-0.2m deep and the center >0.4m deep of the fermentation pile fermented for 1-5 days, respectively, and mixing.

10. Application of the method according to any one of claims 1 to 9 in decoupling the effects of climate in production areas and brewing processes on liquor fermentation effects.

Images