Method and system for improving key flavor components of small-grain coffee

Abstract

The invention belongs to the technical field of coffee baking, and discloses a method and system for improving key flavor components of small coffee, and the method comprises the steps: based on the content of main flavor precursor components of small coffee raw beans, setting the baking temperature and time range, and based on a machine learning model, predicting the content of key beneficial volatile flavor components; acquiring temperature data and time data when the content of the key beneficial volatile flavor component is maximum; on the basis of the obtained temperature data and time data, the small coffee raw beans are baked, and the flavor of the small coffee is improved to the maximum extent. According to the method, the small green coffee beans are accurately collected and analyzed, and the key beneficial volatile flavor component content is predicted in combination with the machine learning model, so that the prediction accuracy and efficiency are improved, and the baking temperature and time are optimally set, so that the key beneficial volatile flavor component content is maximized, and the product quality is improved. And better coffee experience is brought to consumers.

Description

Technical Field

[0001] The invention belongs to the technical field of coffee roasting, and relates to a method and a system for improving key aroma components of small-grain coffee. Background Art

[0002] Due to the influence of factors such as the growing environment, field management and variety, the coefficient of variation of the main components of coffee beans is large. The main components of green beans and the degree of roasting are the key to determining the final flavor of coffee. Although the current coffee roasting process is relatively fixed and unified, there is no scientific guidance for the roasting of different green beans. Coffee roasters rely on personal experience and sensory evaluation to determine the roasting temperature and time in order to obtain the ideal coffee flavor. However, this method has many uncertainties, because different batches of small-grain coffee beans may have different contents of flavor precursor components, and the transformation and interaction of these components during the roasting process are extremely complex and difficult to accurately predict based on experience alone. In actual roasting, past experience often leads to unstable quality of roasted beans, and the best flavor is not obvious, which restricts the brand effect. The development of boutique coffee and the demand for personalized nutritional flavors urgently need to adopt scientific technology to guide the roasting of coffee beans to improve their flavor quality. Summary of the invention

[0003] The purpose of the present invention is to solve the problem that it is difficult to stably obtain the quality of roasted beans and cannot reflect the best flavor of roasted beans in the prior art, and to provide a method and system for improving the key aroma components of small-grain coffee.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] A method for improving key aroma components of small-grain coffee, comprising:

[0006] Collecting green coffee beans and analyzing the collected green coffee beans to obtain main flavor precursor components;

[0007] Based on the changes in the content of the main flavor precursor components of small-grain coffee beans at different times and temperatures, a flavor component database was constructed, and the content of key beneficial volatile flavor components was predicted based on a machine learning model.

[0008] Based on the predicted content of the key beneficial volatile flavor component, obtaining temperature data and time data when the content of the key beneficial volatile flavor component is maximum;

[0009] Based on the acquired temperature data and time data, the small-grain coffee beans are roasted to enhance the flavor of the roasted coffee beans.

[0010] A further improvement of the present invention is:

[0011] Furthermore, the main flavor precursor components include: chlorogenic acid and amino acids; the content ranges are 6%-16% and 6.5%-14.5% respectively.

[0012] Furthermore, based on the changes in the content of main flavor precursors of small-grain coffee beans at different times and temperatures, a flavor component database was constructed, and the content of key beneficial volatile flavor components was predicted based on a machine learning model. Specifically, the volatile flavor components of lightly roasted to darkly roasted coffee were analyzed, and ROVA value analysis of volatiles was performed to obtain key beneficial volatile flavor components. A large database was then constructed in combination with the main flavor precursors of green beans.

[0013] Furthermore, based on the machine learning model, the content of key beneficial volatile flavor components is predicted, specifically: based on the BaggingRegressor model trained on the big database, prediction is performed within the specified temperature range and time range; the temperature is from 160°C to 250°C, and prediction is made every 10°C; the time is from 10 minutes to 20 minutes, and prediction is made every 1 minute; for each combination of temperature and time, prediction is performed based on fixed chlorogenic acid and amino acid values ​​to obtain the corresponding content of key beneficial volatile flavor components.

[0014] Furthermore, the temperature data and time data when the content of key beneficial volatile flavor components is maximum are obtained, specifically: MATLAB is called to draw a prediction curve of the chemical content corresponding to different baking times at each temperature, and the temperature and time when the content of each key beneficial volatile flavor component reaches the maximum value are compared, and the baking temperature and baking time corresponding to the maximum content of each chemical substance are obtained.

[0015] Furthermore, light roasting to deep roasting includes: light roasting for 8-20 minutes, 170-190℃, before the first crack; medium roasting time is 10-25 minutes, 180-199℃, before the first crack and before the second crack; deep roasting time is 11-26 minutes, 185-215℃, at the second crack and after the second crack.

[0016] Furthermore, the key beneficial volatile flavor components are hydroxyacetone, 2-methylpyrazine, and 3-ethyl-2,5-dimethylpyrazine.

[0017] Furthermore, after roasting the green coffee beans, the method further includes: culturing the roasted coffee beans, specifically, in a sterile environment, at a temperature of 20-40°C, for a duration of more than 30 minutes, with an irradiation intensity of more than 70uW / cm2, and a culturing time of 24-48 hours.

[0018] A system for improving key aroma components of arabica coffee, comprising:

[0019] An analysis module, wherein the analysis module collects green coffee beans and analyzes the collected green coffee beans to obtain main flavor precursor components;

[0020] A prediction module, which constructs a flavor component database based on the changes in the content of main flavor precursor components of small-grain coffee beans at different times and temperatures, and predicts the content of key beneficial volatile flavor components based on a machine learning model;

[0021] An acquisition module, wherein the acquisition module acquires temperature data and time data when the content of the key beneficial volatile flavor component is maximum based on the predicted content of the key beneficial volatile flavor component;

[0022] The roasting module roasts the small-grain coffee beans based on the acquired temperature data and time data to enhance the flavor of the roasted coffee beans.

[0023] Compared with the prior art, the present invention has the following beneficial effects:

[0024] The present invention not only improves the accuracy and efficiency of prediction, but also achieves the optimization setting of roasting temperature and time by accurately collecting and analyzing raw coffee beans and combining machine learning models to predict the content of key beneficial volatile flavor components. The present invention can ensure that coffee beans are roasted under optimal conditions, thereby maximizing the content of key beneficial volatile flavor components. Targeted cultivation of roasted beans based on a specific environment effectively improves the overall flavor of the coffee beans, bringing consumers a better coffee experience. The present invention can provide strong support for flavor optimization in the coffee industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 A schematic flow chart of a method for improving key aroma components of small-grain coffee according to the present invention;

[0027] Figure 2 A schematic diagram of the structure of a system for improving key aroma components of small-grain coffee according to the present invention;

[0028] Figure 3 Schematic diagram of the relationship between the main aroma components and flavor precursors of roasted small-grain coffee and the roasting time and temperature. DETAILED DESCRIPTION

[0029] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0030] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0032] In the description of the embodiments of the present invention, it should be noted that if the terms "upper", "lower", "horizontal", "inner", etc. indicate an orientation or positional relationship based on the orientation or positional relationship shown in the drawings, or the orientation or positional relationship in which the product of the invention is usually placed when in use, it is only for the convenience of describing the present invention and simplifying the description, and does not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. In addition, the terms "first", "second", etc. are only used to distinguish the description, and cannot be understood as indicating or implying relative importance.

[0033] In addition, if the term "horizontal" appears, it does not mean that the component must be absolutely horizontal, but can be slightly tilted. For example, "horizontal" only means that its direction is more horizontal than "vertical", which does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0034] In the description of the embodiments of the present invention, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "set", "install", "connect", and "connect" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal connection of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0035] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0036] See also Figure 1 The present invention discloses a method for improving key aroma components of small-grain coffee, comprising:

[0037] S101: collecting green coffee beans and analyzing the collected green coffee beans to obtain main flavor precursor components;

[0038] The main flavor precursor components include: chlorogenic acid and amino acids; the content ranges are 6%-16% and 6.5%-14.5% respectively.

[0039] S102: Based on the changes in the content of the main flavor precursor components of small-grain coffee beans at different times and temperatures, a flavor component database was constructed, and the content of key beneficial volatile flavor components was predicted based on a machine learning model;

[0040] The method constructs a flavor component database based on the relationship between the content of the main flavor precursor components of small-grain coffee beans at different times and temperatures, specifically: analyzing the volatile flavor components of lightly roasted to darkly roasted coffee, performing volatile ROVA value analysis, obtaining key beneficial volatile flavor components, and constructing a large database in combination with the main flavor precursor components of green beans.

[0041] A large database is constructed to organize and analyze the volatile flavor components and ROVA values ​​of coffee at different roasting degrees. The large database includes a list of volatile flavor components and their ROVA values ​​of coffee at different roasting degrees, key beneficial volatile flavor components; a list of the main flavor precursor components of green coffee beans and their changing patterns at different roasting degrees, as well as the obtained volatile flavor components. By comparing the volatile flavor components and ROVA values ​​of coffee at different roasting degrees, the law of how coffee flavor changes with roasting degree is revealed.

[0042] Based on the machine learning model, the content of key beneficial volatile flavor components is predicted, specifically: based on the BaggingRegressor model trained on the big database, prediction is performed within the specified temperature range and time range; the temperature is from 160°C to 250°C, and prediction is made every 10°C; the time is from 10 minutes to 20 minutes, and prediction is made every 1 minute; for each combination of temperature and time, prediction is performed based on fixed chlorogenic acid and amino acid values ​​to obtain the corresponding content of key beneficial volatile flavor components.

[0043] S103: based on the predicted content of the key beneficial volatile flavor component, obtaining temperature data and time data when the content of the key beneficial volatile flavor component is maximum;

[0044] Call MATLAB to draw the prediction curve of the chemical content corresponding to different baking times at each temperature, compare the temperature and time when the content of each key beneficial volatile flavor component reaches the maximum value, and obtain the baking temperature and baking time corresponding to the maximum content of each chemical substance.

[0045] Light roasting to deep roasting includes: light roasting for 8-20 minutes, 170-190℃, before the first crack; medium roasting time is 10-25 minutes, 180-199℃, before the first crack and before the second crack; deep roasting time is 11-26 minutes, 185-215℃, at the second crack and after the second crack.

[0046] The contents of key beneficial volatile flavor components are hydroxyacetone, 2-methylpyrazine and 3-ethyl-2,5-dimethylpyrazine.

[0047] S104: roasting the small-grain coffee beans based on the acquired temperature data and time data to enhance the flavor of the roasted coffee beans.

[0048] After roasting the green coffee beans, the process further includes: culturing the roasted coffee beans, specifically, in a sterile environment, at a temperature of 20-40°C, for a duration of more than 30 minutes, with an irradiation intensity of more than 70uW / cm2, and a culturing time of 24-48 hours.

[0049] See also Figure 2 The present invention discloses a system for improving key aroma components of small-grain coffee, comprising:

[0050] An analysis module, wherein the analysis module collects green coffee beans and analyzes the collected green coffee beans to obtain main flavor precursor components;

[0051] A prediction module, which constructs a flavor component database based on the changes in the content of main flavor precursor components of small-grain coffee beans at different times and temperatures, and predicts the content of key beneficial volatile flavor components based on a machine learning model;

[0052] An acquisition module, wherein the acquisition module acquires temperature data and time data when the content of the key beneficial volatile flavor component is maximum based on the predicted content of the key beneficial volatile flavor component;

[0053] The roasting module roasts the small-grain coffee beans based on the acquired temperature data and time data to enhance the flavor of the roasted coffee beans.

[0054] Example:

[0055] The invention discloses a method for improving key aroma components of small-grain coffee, comprising: predictive model construction, raw material sources, determining flavor precursor content, selecting a roasting method and static bean cultivation.

[0056] Preferably, the roasting prediction model for small-grain coffee beans constructs a schematic diagram of the relationship between the main aroma components and flavor precursors of small-grain coffee roasting and the roasting time and temperature according to the correlation between chlorogenic acid, amino acids, roasting time, roasting temperature and key beneficial volatile flavor components, such as Figure 3 As shown, Figure 3 *P indicates significant correlation at the 0.05 level, and **P indicates significant correlation at the 0.01 level.

[0057] Preferably, the roasting prediction model of small-grain coffee beans is built using a Bagging Regressor regression algorithm. The steps are:

[0058] (1) Import data:

[0059] #Read data

[0060] import pandas as pd

[0061] import numpy as np

[0062] withpd.ExcelFile('coffee data.xlsx')as xls:

[0063] kf_df = pd.read_excel(xls,"data")

[0064] (2) Data processing:

[0065] x = kf_df[['baking temperature', 'baking time', 'chlorogenic acid', 'amino acid']].to_numpy()

[0066] y = kf_df[['Hydroxyacetone (1-hydroxy-2-propanone)','2-methylpyrazine','3-ethyl-2,5-dimethylpyrazine']].to_numpy()

[0067] (3) Training model:

[0068] fromsklearn.ensemble importBaggingRegressor

[0069] model_BaggingRegressor=BaggingRegressor()

[0070] model_BaggingRegressor.fit(x,y)

[0071] The data is learned through the BaggingRegressor regression algorithm to finally form a data model. According to the analysis of "baking temperature", "baking time", "chlorogenic acid" and "amino acid" in the test data, the maximum and minimum values ​​of each label are obtained. The data in this interval is brought into the model at fixed size intervals through enumeration to estimate the eigenvalue matrix T.

[0072] In actual use, the "chlorogenic acid" and "amino acid" data that meet the maximum and minimum values ​​of the labels are brought into the matrix for search, and the corresponding temperature, time and eigenvalue can be found. By finding the maximum value of the eigenvalue, the corresponding temperature and time can be known. The calculation formula is as follows:

[0073] Based on the BaggingRegressor prediction model, and enumeration to form the eigenvalue matrix space T

[0074]

[0075] Based on the determination of "chlorogenic acid", "amino acid" filtering to obtain the search eigenvalue matrix space T search

[0076]

[0077] (4) Given (chlorogenic acid amino acid), predict all data within this temperature range and time range based on the temperature range and time range in the sample data, and find the temperature data and time data when the indicator value is the largest based on the generated predicted data

[0078] #Temperature range is between 160℃ and 250℃, prediction every 10℃

[0079] temp = np.arange(160,260,10)

[0080] #Choose a baking time interval between 10 and 20, and predict once for each 1

[0081] time = np.arange(10,21,1)

[0082] labels = ['Hydroxyacetone (1-hydroxy-2-propanone)', '2-methylpyrazine', '3-ethyl-2,5-dimethylpyrazine']

[0083] #Assume that chlorogenic acid is determined

[0084] cga=12.5

[0085] #Assume that the amino acid is determined

[0086] amino=10.36

[0087] (5) Draw the data samples of each temperature range and calculate the temperature and time when the data of ('hydroxyacetone (1-hydroxy-2-propanone)', '2-methylpyrazine', '3-ethyl-2,5-dimethylpyrazine') is the maximum

[0088] importmatplotlib.pyplot as plt

[0089] importmatplotlib

[0090] frommatplotlib.font_manager importFontProperties

[0091] #Chinese font settings

[0092] font=FontProperties(fname='E:\workspace\python\jupyter\simsun.ttc')

[0093] res = {

[0094] 'Hydroxyacetone (1-hydroxy-2-propanone)':{

[0095] 'value':0,

[0096] 'temp':0,

[0097] 'time':0,

[0098] },

[0099] '2-Methylpyrazine':{

[0100] 'value':0,

[0101] 'temp':0,

[0102] 'time':0,

[0103] },

[0104] '3-Ethyl-2,5-dimethylpyrazine':{

[0105] 'value':0,

[0106] 'temp':0,

[0107] 'time':0,

[0108] },

[0109] }

[0110] fortemprature_itemin temp:

[0111] index=0

[0112] result=np.zeros((time.shape[0],len(labels)),dtype=float)

[0113] fortime_itemin time:

[0114] y_pred

[0115] =model_BaggingRegressor.predict([[temprature_item,time_item,cga,amino]])

[0116] if(res['Hydroxyacetone (1-hydroxy-2-propanone)']['value'] <y_pred[0][0]):

[0117] res['Hydroxyacetone (1-hydroxy-2-propanone)']['value'] = y_pred[0][0]

[0118] res['Hydroxyacetone (1-hydroxy-2-propanone)']['temp'] = temprature_item

[0119] res['Hydroxyacetone (1-hydroxy-2-propanone)']['time'] = time_item

[0120] if(res['2-methylpyrazine']['value'] <y_pred[0][1]):

[0121] res['2-methylpyrazine']['value'] = y_pred[0][1]

[0122] res['2-methylpyrazine']['temp'] = temprature_item

[0123] res['2-methylpyrazine']['time'] = time_item

[0124] if(res['3-ethyl-2,5-dimethylpyrazine']['value'] <y_pred[0][2]):

[0125] res['3-ethyl-2,5-dimethylpyrazine']['value'] = y_pred[0][2]

[0126] res['3-ethyl-2,5-dimethylpyrazine']['temp'] = temprature_item

[0127] res['3-ethyl-2,5-dimethylpyrazine']['time'] = time_item

[0128] result[index]=y_pred[0]

[0129] index+=1

[0130] for col in range(result.shape[1]):

[0131] plt.plot(time,result[:,col],label=labels[col])

[0132] plt.title('Chlorogenic acid:'+str(cga)+'; Amino acid:'+str(amino)+'; Baking temperature:'+str(temprature_item)+'Prediction curve of different baking time under different conditions', fontproperties=font)

[0133] plt.xlabel('baking time',fontproperties=font)

[0134] plt.ylabel("Material content",fontproperties=font)

[0135] plt.legend(prop=font)

[0136] plt.show()

[0137] print(result)

[0138] (6) Results:

[0139] {'Hydroxyacetone (1-hydroxy-2-propanone)':{'value':4.195586913040706,'temp':230,'time':13},

[0140] '2-Methylpyrazine':{'value':11.690991661198542,'temp':230,'time':13},

[0141] '3-ethyl-2,5-dimethylpyrazine':{'value':1.3461164098031522,'temp':230,'time':13}}

[0142] Find the maximum baking temperature and baking time values ​​for "Hydroxyacetone", "3-ethyl-2,5-dimethylpyrazine", and "2-methylpyrazine"

[0143] {baking temperature, baking time} = T search {max(hydroxyacetone, 3-ethyl-2,5-dimethylpyrazine, "dimethylpyrazine")}

[0144] Preferably, the raw material of the small-grain coffee beans is grown in Yunnan at an altitude of 1000 to 2000 m, and the beans are light green in appearance, fresh in smell, and free of odor after being fully matured and pre-treated with coffee cherries.

[0145] Preferably, the main flavor precursor components include chlorogenic acid content of 6% to 16% and amino acid content of 6.5% to 14.5%.

[0146] Preferably, the roasting degree is based on the prediction of the BaggingRegressor prediction model, using an automatic roaster, specifically light roasting for 8-20 minutes at 160-190°C, medium roasting for 10-25 minutes at 180-199°C, and deep roasting for 11-26 minutes at 185-250°C.

[0147] Preferably, the baking time is selected by calculating the curves of the contents of volatile aroma components hydroxyacetone, 2-methylpyrazine, and 3-ethyl-2,5-dimethylpyrazine and different baking temperatures.

[0148] Preferably, the aseptic environment for growing beans adopts an environment temperature of 20-40°C and an ultraviolet light irradiation intensity greater than 70uW / cm 2 , after the duration is greater than 30 minutes, the beans are placed for 24 to 48 hours.

[0149] The following describes in detail the specific embodiments of the present invention.

[0150] The present invention discloses a method for improving key aroma components of small-grain coffee, which specifically comprises:

[0151] Collecting raw materials of small-grain coffee beans: Small-grain coffee cherries grown in a province are picked from November to February of the following year. They must be fully mature and have uniform color. After pre-treatment, they must meet the grade requirements of commercial beans and above, and the moisture content must be controlled within 12%.

[0152] Determination of the main flavor precursor components of raw beans: Analytical instruments were used to analyze the chlorogenic acid and amino acids in raw beans to determine their content. The analysis results are shown in Table 1:

[0153] Table 1: Chlorogenic acid and amino acid content of raw beans

[0154]

[0155]

[0156] Green bean roasting: The green beans are roasted using a coffee intelligent roasting machine. The automatic roasting machine is preheated to 120℃~200℃, and the amount of green beans is 15kg. According to the content of green bean flavor precursors, the model predicts the content of key flavor components at different temperatures (160℃-250℃).

[0157] Determination of roasting time: Determine the roasting time based on the curve of the change of key volatile aroma components hydroxyacetone, 2-methylpyrazine, and 3-ethyl-2,5-dimethylpyrazine and roasting time according to the prediction model.

[0158] The roasting degree is determined based on the BaggingRegressor prediction model, as shown in Table 2 below:

[0159] Example Chlorogenic acid content of raw beans (%) Raw bean amino acid content (%) Optimal roasting degree 1 16.5 12.60 155℃,18min 2 13.0 8.28 235℃,13.7min 3 12.5 10.36 230℃,13.0min

[0160] After roasting, cool and grow the beans. When the selected roasting time is up, the automatic roaster gradually cools down. When the roaster stops rotating, the roasting is finished. Pour out the coffee beans and cool them to room temperature in the cooling plate. The roasted and cooled coffee beans are sterilized with ultraviolet light, and the irradiation intensity is greater than 70uW / cm 2 After sterilization, the raw beans are placed in this environment for curing, with the ambient temperature at 20-40°C and the curing time at 24-48 hours.

[0161] The contents of key volatile aroma components hydroxyacetone, 2-methylpyrazine, and 3-ethyl-2,5-dimethylpyrazine in roasted coffee beans of Examples 1, 2, and 3 of the present invention are shown in Table 3:

[0162] Table 3 Content of key flavor components in the products obtained in the examples of the present invention (μg / g)

[0163]

[0164] As shown in Table 3, the contents of hydroxyacetone, 2-methylpyrazine, and 3-ethyl-2,5-dimethylpyrazine, the main aroma components of small-grain coffee predicted by the present invention, are close to the actual analysis values. The prediction model of the present invention can not only better stimulate the main aroma components of small-grain coffee and improve the quality of roasted coffee products, but also provide a basis for qualitative roasting of specific flavor coffee.

[0165] Changes in roasted coffee bean flavor components predicted by chlorogenic acid and amino acids; roasting time and temperature determined based on changes in key flavor components predicted by the model; cooling and curing beans after roasting.

[0166] Changes in roasted coffee bean flavor components predicted by chlorogenic acid and amino acids. The chlorogenic acid content is between 6% and 16%, and the amino acid content is between 6.5% and 14.5%. Chlorogenic acid and amino acids are the main flavor precursors in coffee roasting. Chlorogenic acid decreases with the increase in the degree of roasting of coffee beans during the roasting process, and can react with some Maillard intermediates such as aldehydes to reduce the content of furans and certain volatiles (octanol, 2-methylbutanal, butane-2,3-dione, pentane-2,3-dione, etc.). Chlorogenic acid lactones formed by the dehydration and cyclization of chlorogenic acid during the roasting process of coffee have been identified as the key factor in the bitterness of coffee. Amino acids are the key precursors involved in the Maillard reaction. The reaction between sugars and amino acids produces furans, pyrazines, pyridines, etc., which are the source of the aroma of roasted coffee beans. The present invention obtains a large amount of data based on the preliminary analysis of flavor precursors and flavor components of different roasting degrees, and screens out three key flavor components of small-grain coffee, namely hydroxyacetone, 2-methylpyrazine, and 3-ethyl-2,5-dimethylpyrazine. On this basis, the BaggingRegressor regression algorithm is used to model and determine the roasting temperature and time.

[0167] The roasting time and temperature are determined according to the changes in key flavor components predicted by the model. The BaggingRegressor regression algorithm is used to model the key flavor components of small-grain coffee, such as hydroxyacetone, 2-methylpyrazine, and 3-ethyl-2,5-dimethylpyrazine, to determine the roasting temperature and time.

[0168] After baking, the beans are cooled and cured. After baking at the time and temperature determined by the prediction model, the beans are cured and placed in a sterile environment, the ambient temperature is 20-40°C, the sterilization time is greater than 30 minutes, and the irradiation intensity is greater than 70uW / cm 2 The purpose is to replace the carbon dioxide continuously produced in the raw beans and some irritating flavor components with the external air, so that the flavor reaches a stable state.

[0169] The terminal device provided in an embodiment of the present invention. The terminal device of this embodiment includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned method embodiments are implemented. Alternatively, when the processor executes the computer program, the functions of the modules / units in the above-mentioned device embodiments are implemented.

[0170] The computer program may be divided into one or more modules / units, and the one or more modules / units are stored in the memory and executed by the processor to accomplish the present invention.

[0171] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0172] The processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.

[0173] The memory may be used to store the computer programs and / or modules, and the processor implements various functions of the terminal device by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory.

[0174] If the module / unit integrated in the terminal device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0175] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for improving the key aroma components of small-grain coffee, characterized in that: include: Collecting green coffee beans and analyzing the collected green coffee beans to obtain main flavor precursor components; Based on the changes in the content of the main flavor precursor components of small-grain coffee beans at different times and temperatures, a flavor component database was constructed, and the content of key beneficial volatile flavor components was predicted based on a machine learning model. Based on the predicted content of the key beneficial volatile flavor component, obtaining temperature data and time data when the content of the key beneficial volatile flavor component is maximum; Based on the acquired temperature data and time data, the small-grain coffee beans are roasted to enhance the flavor of the roasted coffee beans.

2. The method for improving the key aroma components of small-grain coffee according to claim 1, characterized in that: The main flavor precursor components include: chlorogenic acid and amino acids; the content ranges are 6%-16% and 6.5%-14.5% respectively.

3. The method for improving the key aroma components of small-grain coffee according to claim 2, characterized in that: The method constructs a flavor component database based on the changes in the content of the main flavor precursor components of small-grain coffee beans at different times and temperatures, specifically: analyzing the volatile flavor components of lightly roasted to darkly roasted coffee, performing volatile ROVA value analysis, obtaining key beneficial volatile flavor components, and constructing a large database in combination with the main flavor precursor components of green beans.

4. The method for improving the key aroma components of small-grain coffee according to claim 3, characterized in that: The method for predicting the content of key beneficial volatile flavor components based on the machine learning model is as follows: based on the BaggingRegressor model trained on the big database, prediction is performed within the specified temperature range and time range; the temperature is from 160°C to 250°C, and prediction is made every 10°C; the time is from 10 minutes to 20 minutes, and prediction is made every 1 minute; for each combination of temperature and time, prediction is performed based on fixed chlorogenic acid and amino acid values ​​to obtain the corresponding content of key beneficial volatile flavor components.

5. The method for improving the key aroma components of small-grain coffee according to claim 4, characterized in that: The method of obtaining the temperature data and time data when the content of the key beneficial volatile flavor components is maximum is specifically as follows: calling MATLAB to draw a prediction curve of the chemical content corresponding to different baking times at each temperature, comparing the temperature and time when the content of each key beneficial volatile flavor component reaches the maximum value, and obtaining the baking temperature and baking time corresponding to the maximum content of each chemical substance.

6. The method for improving the key aroma components of small-grain coffee according to claim 5, characterized in that: The light roasting to deep roasting includes: light roasting for 8-20 minutes, 170-190°C, before the first crack; medium roasting for 10-25 minutes, 180-199°C, before the first crack and before the second crack; deep roasting for 11-26 minutes, 185-215°C, at the second crack and after the second crack.

7. The method for improving the key aroma components of small-grain coffee according to claim 6, characterized in that: The key beneficial volatile flavor components are hydroxyacetone, 2-methylpyrazine and 3-ethyl-2,5-dimethylpyrazine.

8. The method for improving the key aroma components of small-grain coffee according to claim 1, characterized in that: After roasting the green coffee beans, the process further includes: culturing the roasted coffee beans, specifically, in a sterile environment, at a temperature of 20-40°C, for a duration of more than 30 minutes, with an irradiation intensity of more than 70uW / cm2, and a culturing time of 24-48 hours.

9. A system for improving the key aroma components of small-grain coffee, characterized in that: include: An analysis module, wherein the analysis module collects green coffee beans and analyzes the collected green coffee beans to obtain main flavor precursor components; A prediction module, which constructs a flavor component database based on the changes in the content of main flavor precursor components of small-grain coffee beans at different times and temperatures, and predicts the content of key beneficial volatile flavor components based on a machine learning model; An acquisition module, wherein the acquisition module acquires temperature data and time data when the content of the key beneficial volatile flavor component is maximum based on the predicted content of the key beneficial volatile flavor component; The roasting module roasts the small-grain coffee beans based on the acquired temperature data and time data to enhance the flavor of the roasted coffee beans.

Images