Method for analyzing flavor of sweet potato brandy

Abstract

The invention belongs to the technical field of brandy production processes, and particularly relates to a sweet potato brandy flavor analysis method which comprises the following steps: S1, preparing sweet potato brandy; s2, measuring physical and chemical parameters of the sweet potato brandy in the step S1; s3, performing sensory evaluation test on the sweet potato brandy in the step S1; s4, performing volatile compound GC-IMS analysis on the sweet potato brandy in the step S1; step S5, performing contrast spectrum analysis on the result of the step S4; and S6, performing hierarchical clustering analysis on a result of the step S4. According to the invention, by combining gas chromatography-ion mobility spectrometry and multivariate data analysis, the research reveals that sweet potatoes processed differently have significant change on the spectrum of volatile organic compounds in the fermented and brewed brandy.

Description

Technical Field

[0001] The invention relates to a sweet potato brandy flavor analysis method and belongs to the technical field of brandy. Background Art

[0002] Brandy originally referred to strong liquor distilled from wine, and later gradually expanded to distilled liquor made from various fresh fruits or juices through fermentation, distillation, storage and blending. Brandy is famous for its golden and transparent color, harmonious fruity aroma, aged oak aroma and mellow wine aroma, elegant, rich, mellow, sweet, moist, delicate, full and continuous unique body style. Flavor is an important factor in determining the sensory perception and consumer acceptance of alcoholic products. The unique flavor characteristics in alcoholic beverages come from the complex interaction between non-volatile compounds (such as sugars, acids and phenols) and volatile compounds (such as esters, higher alcohols, fatty acids, aldehydes, ketones, terpenes and volatile phenols). Given the important role of volatile components in determining sensory attributes and overall flavor, their analysis is a key link in alcoholic beverage research. With the improvement of living standards, people's pursuit of the flavor of wine is getting higher and higher, and brandy lovers are also beginning to pursue brandy varieties with richer flavors and more flavor levels.

[0003] Sweet potato (Ipomoea batatas L.) is the sixth most important food crop in the world, with an annual output of more than 90 million tons, playing an important role in global food security. Yanshu No. 25, also known as honey potato, is a high-quality sweet potato variety bred by Yantai Academy of Agricultural Sciences in Shandong Province through sexual hybridization (the female parent is Lushu No. 3 and the male parent is Red Meat Red). Compared with conventional sweet potatoes, Yanshu No. 25 has more prominent nutritional value, with high mucin and soluble sugar levels far higher than conventional sweet potato varieties, and has a unique sweet and sticky flavor, golden potato flesh and medium β-carotene content.

[0004] However, the current research on using sweet potatoes to brew and ferment brandy is relatively limited. Moreover, the sweet potatoes must be pretreated before being used in the preparation of brandy. Different pretreatment methods have a direct relationship with the stimulation of nutrients in the sweet potatoes, and have a great influence on the quantity and level of flavor substances in the final brandy. However, how to more effectively and intuitively analyze the flavor substances in brandy obtained by different sweet potato pretreatment methods, so as to obtain what sweet potato pretreatment method should be adopted in the preparation of sweet potato brandy, so as to make full use of the nutrients in sweet potatoes such as Yanshu No. 25, and enrich and improve the flavor and flavor level of brandy, is a technical problem that technical personnel in this field urgently need to solve. Summary of the invention

[0005] The invention aims at solving the defects of the prior art and provides a method for analyzing the flavor of sweet potato brandy.

[0006] The technical solution of the present invention to solve the above technical problems is as follows: A sweet potato brandy flavor analysis method comprises the following steps: Step S1, preparation of sweet potato brandy; Step S2, measuring the physical and chemical parameters of the sweet potato brandy in step S1; Step S3, performing a sensory evaluation test on the sweet potato brandy of step S1; Step S4, performing GC-IMS analysis on volatile compounds of the sweet potato brandy of step S1; Step S5, performing comparative spectral analysis on the result of step S4; Step S6, performing hierarchical cluster analysis on the result of step S4.

[0007] On the basis of the above technical solution, the present invention can also make the following improvements: Further, in step S1, the preparation of the sweet potato brandy specifically comprises the following steps: (1) Sweet potato pretreatment; (2) peeling the pretreated sweet potatoes, stirring and homogenizing them into mashed potatoes, and transferring them to a fermentation container for fermentation; (3) After fermentation is complete, the fermented mash is added to the distiller and distilled to produce brandy.

[0008] Furthermore, in step (1), the sweet potato pretreatment includes washing, drying and weighing, or also includes any one of steaming, boiling, frying or baking.

[0009] Furthermore, in step (2), after the sweet potato is made into mashed potato, pectinase is added, and sugar and water are added to adjust the soluble solid content of the paste to 25%, the koji is evenly distributed on the surface of the paste, and the fermentation container is sealed. The fermentation is carried out in a well-ventilated environment with a temperature controlled at 20° C. to 25° C. and a pH value of 4.5 to 5.0. The fermentation cycle is 20 days, and the final alcohol content (volume) of the fermented paste is 11% to 13%.

[0010] Furthermore, in step (2), the amount of pectinase added is 0.04 g / L.

[0011] Furthermore, in step (2), the amount of sugar added is 23 g / 100 mL.

[0012] Furthermore, in step (3), the sweet potato brandy is prepared by double distillation, the temperature of the first distillation is controlled at 80°C-85°C, and the temperature of the second distillation is controlled at 60°C-65°C.

[0013] Furthermore, the sweet potato is Yanshu No. 25.

[0014] Furthermore, in step (1), the specific operation of steaming is: putting the sweet potato into a steamer, using a multi-stage temperature control method, first steaming at 90°C for 10 minutes, then steaming at 100°C for 20 minutes, and finally steaming at 110°C for 10 minutes, with a total steaming time of 40 minutes, and adding a small amount of salt during the steaming process to promote the Maillard reaction.

[0015] Furthermore, in step (1), the specific operation of cooking is: putting the sweet potatoes into boiling water, keeping it in a rolling state, cooking for 30 minutes, and adding citric acid to the water to enhance the sweetness and maintain color stability.

[0016] Furthermore, in step (1), the specific operation of frying is: immersing the sweet potatoes in edible oil at 170° C. and frying for 5 minutes, and adding cinnamon during the process to enhance the aroma and promote the Maillard reaction.

[0017] Furthermore, in step (1), the specific operation of baking is: putting the sweet potato into an oven and baking it at 200° C. for 90 minutes. After baking for 45 minutes, evenly applying a mixture of light soy sauce and honey on the surface of the sweet potato to form a caramelized crust to enhance the taste and flavor.

[0018] Furthermore, in step S2, the physicochemical parameters include pH value, reducing sugar content, alcohol content, soluble solid content, total ester content and total acidity.

[0019] Further, the specific steps of step S4 are: The brandy prepared by different sweet potato pretreatments in step S1 is sampled, the sample is 1.0 mL-1.5 mL, put into a headspace bottle, sealed with a magnetic screw cap, and incubated at 60° C. and 500 rpm for 10 minutes to 15 minutes to balance the volatile organic compounds in the headspace; after incubation, the headspace sample is drawn using a syringe heated to 85° C. and injected into the GC-IMS system in a splitless mode; the chromatographic column and the migration tube are maintained at 80° C. and 45° C., respectively; During the analysis, the carrier gas was high-purity nitrogen, and the flow rate program was: 2 mL / min for the first 2 minutes, then increased to 10 mL / min within 8 minutes, then increased to 100 mL / min within 10 minutes, and finally maintained at 100 mL / min until the end of the analysis. The analysis time was 30-35 minutes.

[0020] Furthermore, the specific steps of step S5 are: using the spectrum of one of the samples in step S4 as a reference, and generating a difference spectrum by comparing it with the spectrum of other samples.

[0021] Further, step S6 includes dynamic principal component analysis and fingerprint similarity analysis of the sample.

[0022] The fingerprint similarity analysis evaluates the similarity of VOC spectra between samples by calculating the Euclidean distance.

[0023] The beneficial effects of the present invention are: The present invention combines gas chromatography-ion mobility spectrometry (GC-IMS) and multivariate data analysis to obtain the results of the influence of sweet potatoes processed differently on the spectrum of volatile organic compounds (VOCs) in fermented brandy.

[0024] The present invention obtains the effects of different thermal processing methods on volatile organic compounds in sweet potato brandy by using gas chromatography-ion mobility spectrometry technology. The results show that different processing methods significantly change the composition of volatile organic compounds in sweet potato brandy and give it unique flavor characteristics.

[0025] The present invention further distinguishes the volatile organic compound spectra produced by different processing methods through principal component analysis and fingerprint similarity analysis based on Euclidean distance, highlighting the key role of processing technology in flavor development.

[0026] The invention provides a basis for optimizing the sweet potato processing method to produce sweet potato brandy with specific aromatic characteristics, and is of great significance for the food industry to optimize production technology to produce sweet potato brandy products that meet consumers' preferences for specific aroma and flavor. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is the sensory evaluation radar chart of the sweet potato brandy of test groups 1 to 5; Figure 2 The three-dimensional spectra of the volatile components of sweet potato brandy in test groups 1 to 5; Figure 3 It is a two-dimensional top view of the spectra of the volatile components of sweet potato brandy of test groups 1 to 5; Figure 4 The difference spectra of the volatile components of sweet potato brandy in experimental groups 1 to 5 are compared; Figure 5 The results of qualitative analysis of the volatile components of sweet potato brandy in experimental group 1 using library search; Figure 6 The fingerprint spectra of volatile compounds in sweet potato brandy of test groups 1 to 5 are shown; Figure 7 This is the PCA analysis of the volatile component data of sweet potato brandy from experimental groups 1 to 5; Figure 8 It is the Euclidean distance diagram between the sweet potato brandies of experimental groups 1 to 5. DETAILED DESCRIPTION

[0028] The principles and features of the present invention are described below in conjunction with the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0029] Chemical reagents and materials used in the present invention Yanshu No. 25 sweet potatoes were purchased from Yanda Market (Yantai, China); Golden Dragon Fish brand peanut oil (900 ml), purchased from Zhenhua Supermarket (Yantai, China); The koji used in the fermentation process was Wuliangye koji (Chunzhongchun, Sichuan, China); All chemical reagents used in the analysis were of commercial analytical grade.

[0030] A sweet potato brandy flavor analysis method of the present invention comprises the following steps: Step S1, preparation of sweet potato brandy: (1) Sweet potato pretreatment To ensure consistency, Yanshu No. 25 sweet potatoes of similar size and weight were selected. Fifteen medium-sized sweet potatoes (100 g to 150 g) were selected, cleaned, dried and weighed, and then randomly divided into five groups of three each, as experimental groups 1 to 5. Experimental group 1 was raw (denoted as DB1), that is, it was only washed, dried and weighed. Experimental groups 2 to 5 were steamed (denoted as DB2), boiled (denoted as DB3), fried (denoted as DB4) and baked (denoted as DB5), and the skin was not peeled before cooking.

[0031] In the steaming treatment of Experimental Group 2, the sweet potatoes were placed on a container and steamed in a Midea C30-IH3002 steamer using a multi-stage temperature control method for 40 minutes, including steaming at 90°C for 10 minutes, steaming at 100°C for 20 minutes, and steaming at 110°C for 10 minutes, and a small amount of salt was added to the water to promote the Maillard reaction.

[0032] During the cooking process of Experimental Group 3, the sweet potatoes were placed in 4 liters of boiling water, kept boiling for 30 minutes, and a small amount of citric acid was added to enhance the sweetness and maintain the color stability. The sweet potatoes were cooked using a Midea C30-IH3002 steamer.

[0033] The experimental group 4 was fried in edible oil at 170°C for 5 minutes. A small amount of cinnamon was added during the frying process to enrich the aroma and promote the Maillard reaction. The frying was performed using a Chigo ZG-BK-ZL-81 fryer equipped with a metal basket.

[0034] During baking, the sweet potatoes of Experimental Group 5 were placed in a Conca 13T1 oven and baked at 200°C for 90 minutes. After baking for 45 minutes, a mixture of light soy sauce and honey was evenly applied on the surface of the sweet potatoes to form a caramelized crust to enhance the taste and flavor.

[0035] (2) Mashed potato fermentation After different cooking treatments in experimental groups 1 to 5, the sweet potatoes were cooled to room temperature and then peeled. The peeled sweet potato flesh was blended into a puree using a blender and transferred to a fermentation container. Pectinase was added (pectinase addition amount: 0.04 g / L, enzyme activity parameter: 100,000 U / g, SAS SOFRALAB, France), ensuring thorough mixing to promote cold maceration, sugar was added, and the sugar addition amount was 23 g / 100 mL. Water was carefully added until the soluble solids content reached 25%, and 1 gram of Wuliangye liquor was evenly distributed on the surface of the paste. The fermentation process was designed to achieve a final alcohol content of 12% (v / v). The fermentation container was sealed and placed in a well-ventilated environment with a temperature control of 23°C for 20 days of natural fermentation. During this process, stirring was performed twice a day to ensure uniform fermentation.

[0036] (3) Distillation After fermentation, the fermented mash was added to the distiller, where the mash was distilled to produce brandy. The distillation process was carried out using a batch distillation unit with a capacity of 1 L, equipped with a 24x600 mm column and filled with 7x15 mm Raschig rings. The temperature in the flask was monitored by a T-type thermocouple with mineral insulation, which was threaded on the side of the flask and monitored in real time in Arduino 1.0.6 software during the distillation process. In addition, the temperature was monitored using a thermometer at the top of the column (the temperature was controlled at 80°C for the first distillation and 65°C for the second distillation). The distillate was cooled by circulating tap water at room temperature through a straight condenser and collected at the end of the system. The distilled brandy was bottled and stored in a cool, dark environment for stabilization before further analytical evaluation.

[0037] Step S2, measuring the physical and chemical parameters of the sweet potato brandy of step S1: specifically, measuring the pH value, reducing sugar content, alcohol content, soluble solids content, total ester content and total acidity of the sweet potato brandy.

[0038] Step S3, performing sensory evaluation on the sweet potato brandy of step S1.

[0039] Step S4, performing GC-IMS analysis on the volatile compounds of the sweet potato brandy of step S1.

[0040] Step S5, performing comparative spectral analysis on the result of step S4.

[0041] Step S6, performing hierarchical clustering analysis on the results of step S4, specifically, including (1) dynamic principal component analysis (PCA) of the samples; and (2) fingerprint similarity analysis based on Euclidean distance.

[0042] In the above step S2: During the pH determination process, a PHS-3E pH meter (Shanghai Jingke Scientific Instrument Co., Ltd., Shanghai, China) was used for measurement. The specific operation steps are as follows: First, the instrument was powered on and preheated for 10 minutes to ensure stable operation. Subsequently, the pH meter was calibrated using standard buffer solutions (pH 4.00, pH 6.86, and pH 9.18). The electrodes were immersed in the above standard solutions respectively. After the readings stabilized, the instrument was adjusted to make the displayed value consistent with the pH value of the standard buffer solution to complete the three-point calibration. After the calibration was completed, 20 mL of the sweet potato brandy sample to be tested was placed in a clean glass beaker, and the cleaned and dried pH electrode was immersed in the sample and gently stirred to ensure uniform contact. After the pH value displayed by the instrument stabilized (wait for 30 seconds), the reading was recorded. After each measurement, the electrode was thoroughly rinsed with distilled water and dried with filter paper to avoid cross contamination between samples. Each group of samples was measured three times, and the average value was taken as the final pH result.

[0043] Reducing sugars were determined by the 3,5-dinitrosalicylic acid (DNS) colorimetric method with glucose as the standard, and after distillation, the ethanol content was determined using a DA-130 N portable densitometer (Kyoto Electronics Manufacturing Co., Ltd., Tokyo, Japan).

[0044] Alcohol content is the volume fraction of ethanol measured by an alcohol meter.

[0045] The test methods for soluble solids content, total ester content and total acidity are: The total acidity was determined by acid-base titration, with tartaric acid as the reference standard, and the soluble solids content was measured using a handheld refractometer; the total ester content was calculated based on the stoichiometric relationship of the NaOH consumed in the reaction by saponifying the esters in the sample with a standard NaOH solution and then titrating the remaining NaOH with a standard H2SO4 solution.

[0046] Table 1 shows the changes in the main components of brandy fermented from sweet potatoes with different treatments.

[0047] Table 1 Main components of sweet potato wine processed by different methods

[0048] Significant differences between samples are indicated by letters (ac). DB1: Experimental group 1, brandy made from raw sweet potatoes. DB2: Experimental group 2, brandy made from steamed sweet potatoes; DB3: Experimental group 3, brandy made from boiled sweet potatoes; DB4: Experimental group 4, brandy made from fried sweet potatoes; DB5: Experimental group 5, brandy made from baked sweet potatoes.

[0049] Alcohol content is a key parameter in fermented fruit wine and plays a vital role in the traditional alcohol fermentation process. Soluble solids content is another key quality indicator that is closely related to consumer taste preferences. The alcohol content varied among the treatments, ranging from 22.00% (DB5) to 24.50% (DB1), with DB1 showing the highest alcohol concentration. This suggests that thermal treatments (such as steaming, boiling, frying, and baking) may reduce the fermentable sugars available for yeast metabolism, resulting in lower alcohol yields in these samples. A similar trend was seen in soluble solids content, with DB1 having the highest soluble solids content (10.5 ± 0.10%) and DB5 having the lowest (9.3 ± 0.09%). The reduction in soluble solids in DB5 and DB4 samples may be due to the Maillard reaction and caramelization of sugars, which reduces the solubility of fermentable sugars.

[0050] Esters are the main contributors to wine aroma and flavor, and the ester content in different samples varied significantly. DB1 had the highest ester content (480 ± 5.00 mg / L), while DB5 had the lowest ester content (430 ± 3.50 mg / L). The higher ester content in DB1 indicated a more active esterification process, probably due to the retention of volatile precursors in raw sweet potatoes. In contrast, the lower ester content in the heat-treated samples, especially DB5, might be due to the thermal degradation of ester precursors or their conversion to non-volatile compounds during high-temperature processing.

[0051] Reducing sugars, a key indicator of residual sweetness, also showed significant differences in the content of reducing sugars among the different treated samples. The highest reducing sugar content was found in DB5 (4.3 ± 0.06 g / L), probably due to the caramelization of sugars during baking. In contrast, the lowest reducing sugar content was found in the cooked sample (DB2) (2.4 ± 0.06 g / L), probably due to the leaching of sugars in the cooking medium. The increase in reducing sugar content in the baked samples suggests that thermal treatments, especially those involving dry heat, may enhance the residual sweetness of the final product.

[0052] Acidity is essential for imparting a refreshing feel to wine, while pH is another key factor affecting the quality of brandy. The total acidity of the different treatments was relatively stable, ranging from 3.9 to 4.5 g / L, indicating that the retention or generation of acid during processing changed little. However, the pH value varied slightly, with DB1 having the lowest pH value (3.6 ± 0.05) and DB5 having the highest (3.9 ± 0.05). The lower pH value in DB1 may be due to the higher retention of organic acids in raw sweet potatoes, while the higher pH value in DB5 may be due to the thermal degradation of organic acids during baking, resulting in a slight alkalinization of the matrix. Lower acidity helps retain fruity aroma, balances sweetness and acidity, and brings a pleasant taste. However, too low acidity may have an adverse effect on the taste of fruit wine, resulting in a sour or bitter feeling. The acidity levels of the wines in test groups 1 to 5 of the present invention are within a range that helps balance the taste.

[0053] In summary, different heat treatments play an important role in the chemical composition and sensory attributes of sweet potato brandy. Through the analytical method of the present invention, it was found that the wine made from raw sweet potatoes retained more fermentable sugars and volatile precursors, resulting in higher alcohol content and richer ester characteristics. In contrast, roasted and fried sweet potato wine showed higher residual sugars, but lower ester and alcohol contents, giving a sweeter flavor profile. These research results provide valuable insights for targeted regulation of sweet potato wine properties by controlling processing conditions, which will help develop high-quality, flavor-diverse sweet potato wine beverages.

[0054] In the above step S3: The aroma analysis sensory evaluation method was evaluated in accordance with ISO 8589–2007. The evaluation panel consisted of 10 trained members (5 males and 5 females) aged between 20 and 40 years old. The highest score was 100 points, and the average score of the evaluation panel was taken as the final sensory evaluation result. The sensory scoring criteria for sweet potato wine are detailed in Table 2. To ensure the reliability and accuracy of the descriptive scoring, the descriptive analysis test was conducted three times within a week, and each test lasted 60 minutes. The judges took a break between tasting four different samples and cleaned their mouths with water to reset their senses. Each sample was evaluated three times.

[0055] Table 2 Sensory evaluation criteria for sweet potato wine

[0056] The sensory evaluation results are shown in Figure 1. The sensory evaluation of the sweet potato brandies fermented from sweet potatoes with different processing methods in test groups 1 to 5 showed significant differences in multiple sensory dimensions such as appearance, aroma, taste and typicality, which ultimately affected the overall score. In terms of appearance, DB1 received the highest score (19 points), indicating that the unprocessed sweet potatoes retained the best color and transparency than the heat-treated samples. In contrast, DB5 received the lowest appearance score (16.5 points), which may be due to the darkening of color and caramelization effect caused by long-term exposure to high temperatures during the baking process.

[0057] The aroma scores showed a clear trend, with DB1 having the highest score (28.5) and DB5 having the lowest score (23.8). The high aroma score of DB1 can be attributed to the retention of volatile compounds such as esters and aldehydes, which are better preserved in raw sweet potatoes (see Figure 6 ). Thermal processing, especially baking and frying, tends to reduce the concentration of these volatile compounds, thereby weakening the aroma characteristics.

[0058] In terms of taste, DB3 outperformed the other samples with a score of 36.5. As a moist heat treatment method, steaming can enhance the sweetness of sweet potatoes and preserve the flavor while minimizing the degradation of important flavor compounds. In contrast, DB5 and DB2 had significantly lower taste scores, which was due to the breakdown or loss of delicate flavor compounds during the more intense heat treatment.

[0059] In terms of typicality, DB1 again received the highest score (9.5 points), indicating that the raw sweet potato brandy retains the most representative sensory attributes of the raw material. DB5 scored the lowest (7.2 points), due to the significant changes brought about by roasting that masked the original sweet potato flavor.

[0060] Overall, DB1 (94 points) and DB3 (93 points) received the highest total scores, highlighting the importance of retaining sensory quality in sweet potato brandy production, especially when minimal or moderate handling is used during processing. In contrast, DB5 received the lowest total score (85 points), indicating that although roasting may provide some flavor complexity, it may not meet consumer preferences for the sensory attributes of traditional sweet potato brandy.

[0061] and Figure 6The results presented are consistent, among which DB1 retains the highest levels of esters, alcohols and aldehydes, contributing to its excellent sensory properties. The significant differences found between DB1 and heat-treated samples (especially DB5) using the analytical method of the present invention highlight the key role of processing methods in shaping the flavor and aroma characteristics of fermented sweet potato products. This shows that although heat treatment such as baking may bring complex flavors, it may lead to a decrease in overall sensory quality, especially in beverages such as brandy, where the value of retaining the inherent properties of the raw materials is very high, which is of guiding significance for the preparation of brandy.

[0062] In the above step S4: The analytical instrument uses Flavorspec ® The system (GAS Instruments, Germany) was equipped with a MXT-WAX capillary column (30 m × 0.53 mm × 1 μm, Restek, USA). Before the experiment, the instrument was powered on and preheated to a stable state. The temperature control systems of the chromatographic column and migration tube were checked to operate normally, and were set and maintained at 80 °C and 45 °C, respectively. Each brandy sample (1.0 mL) was placed in a 20 mL headspace bottle, sealed with a magnetic screw cap, and incubated at 60 °C at 500 rpm for 15 minutes to equilibrate the volatile organic compounds (VOCs) in the headspace. During the incubation process, ensure that the sample is stirred evenly to avoid local overheating of the sample or loss of volatiles.

[0063] After incubation, 100 μL of headspace sample was drawn using a syringe heated to 85°C and injected into the GC-IMS system in splitless mode. The column and migration tube were maintained at 80°C and 45°C, respectively. During the analysis, the carrier gas was high-purity nitrogen (99.999%), and the flow rate program was: 2 mL / min for the first 2 minutes, then increased to 10 mL / min in the next 8 minutes, then increased to 100 mL / min in the next 10 minutes, and finally maintained at 100 mL / min until the end of the entire 30-minute analysis.

[0064] In this step, each sample was tested in triplicate, and the results were expressed as mean ± standard deviation (SD). Data were analyzed using IBM SPSS Statistics 26 software for one-way analysis of variance (ANOVA), and Duncan's multiple range test was used to determine significant differences between treatment groups (P < 0.05). Other analyses included orthogonal partial least squares discriminant analysis (OPLS-DA) using SIMCA-P 14.1 software, and graphics were generated using Origin 2021 software.

[0065] The volatile compounds in sweet potato fermented brandy processed by different methods were analyzed using gas chromatography-ion mobility spectrometry (GC-IMS). The GC-IMS used in the present invention provides a sensitive and direct way to detect volatile compounds, which can accurately capture the aromatic characteristics of brandy without the need for sample enrichment. The analysis principle of GC-IMS is based on differential ion migration time, which effectively separates volatile organic compounds (VOCs) in sweet potato brandy under different processing technologies.

[0066] The GC-IMS analysis results are presented in the form of a three-dimensional spectrum, such as Figure 2 As shown in the figure, the x-axis represents the drift time, the y-axis represents the retention time, and the z-axis represents the peak intensity. The three-dimensional graph clearly shows the differences in volatile organic compounds in different brandy samples. Figure 2 Visual inspection of the HPLC-MS / MS spectra revealed that, although the peak signal distributions of all samples were generally similar, there were small but significant differences in the signal intensities at the black circle markers, indicating that the volatile components of brandy were slightly different under different processing methods.

[0067] Figure 2 The differences marked by red circles provide more insights into possible sources of these differences. These differences may be caused by multiple signals generated by a single volatile compound at different concentrations and chemical properties, which can behave as monomers or dimers. In addition, the binding of the analyte ion to neutral molecules (such as trimers) during the analysis may also generate multiple signals during ion migration.

[0068] Through Figure 2 Analysis of the red circled area revealed that the brandy made from raw sweet potatoes (DB1) was significantly different from the other samples. This difference may be due to the fact that more original flavor components, such as aldehydes, were retained in the unheated sweet potatoes. Figure 2 The dense signal peaks in confirm the high content of these indigenous volatile compounds in raw sweet potato brandy, highlighting its aromatic complexity.

[0069] Figure 3 The 2D top-down view in further reveals the changes in volatile compounds. The background of the top-down view is mainly blue, with a prominent red vertical line at the x-coordinate of 1.0, representing the reactive ion peak (RIP). The RIP is a normalized feature that originates from the interaction of water in the headspace with high-energy electrons from the IMS tritium (³H) ionization source to form protonated water clusters. When volatile compounds enter the IMS ionization region, the intensity of the RIP may weaken or disappear, indicating the presence of these compounds.

[0070] Normalization is essential to standardize ion migration times to compensate for changes caused by temperature and pressure fluctuations during the detection process. The different masses and charges of volatile compounds, combined with their interaction with neutral gas molecules and the electric field in the ion drift region, enable the separation of ions. This separation is essential for qualitative and quantitative analysis because it relates ion migration times to the intensity of the ion response peak, allowing accurate assessment of volatile compounds.

[0071] exist Figure 3 Each dot next to the RIP corresponds to a unique volatile organic compound (VOC). The color intensity of the dots reflects the concentration of the compound, with white indicating lower levels and red indicating higher concentrations. The stronger the color, the higher the concentration. Figure 3 The y-axis represents the retention time determined by gas chromatography, and the x-axis represents the normalized ion drift time.

[0072] Figure 3 It shows that most of the signals are concentrated between retention time 200-720 seconds and drift time 1.0-1.80 milliseconds. The DB1 sample shows the highest VOCs content in the red box area, which is significantly different from the other samples. Figure 2 The findings are consistent with .

[0073] In the above step S5, Figure 3 The DB1 sample in the figure is used as a reference, and a difference spectrum is generated by comparing the spectra of the other samples. In the difference spectrum, a uniform white background indicates that the VOCs levels between the samples are equal. The red area indicates that the VOCs concentration is higher than that of the reference sample (DB1), and the intensity of the red reflects the magnitude of the increase. In contrast, dark blue indicates that the concentration of some compounds is lower than that of DB1.

[0074] In the difference comparison mode of step S5 of the present invention, Figure 4 The results showed that there were significant differences between DB1 and DB2 samples, which may be due to the effect of cooking. Cooking is known to inactivate enzymes such as lipase, thereby enhancing the complexity of flavor. In particular, the VOCs concentration of DB2 exceeded that of DB1 between retention time 200 and 305 seconds and drift time 1.0 and 1.2 milliseconds. This suggests that cooking may cause sweet potatoes to release a wider range of volatiles, including monoterpenes, benzene derivatives, and furan compounds, enhancing the aroma characteristics of brandy.

[0075] The difference spectrum of DB2 also shows a large dark blue area between retention time 210-330 seconds and drift time 1.2-1.65 milliseconds, indicating a significant difference in VOCs concentration compared to DB1. This change may be caused by mechanical stress during cooking, which may induce evaporation or decomposition of volatile organic compounds. At drift times 1.35 ms and 1.75 ms, DB2 shows two prominent red spots, indicating a higher concentration of aroma compounds such as sesquiterpenes and possible ketones or maltol, which is a characteristic aroma compound of cooked sweet potatoes.

[0076] Analysis of DB3, DB4, and DB5 samples—representing steamed, fried, and baked sweet potato brandy, respectively—showed similar changes in VOCs. These differences are generally associated with thermal processing and the Maillard reaction, a non-enzymatic browning process in which carbon-based compounds react with amino acids to produce unique flavor compounds. When comparing DB1 and DB5, the difference spectrum, particularly between drift times 1.0 and 1.7 milliseconds, showed a range of red and blue dots, indicating the range of concentration differences. The characteristic flavor of baked sweet potato is associated with pyrolysis processes, including the release of terpenoid glycosides, degradation of carotenoids, caramelization, Maillard reactions, and Stryker degradation. These reactions may enhance volatility or promote the accumulation of certain compounds, leading to changes in VOC concentrations in DB5.

[0077] The significant differences between DB2 and DB5 in retention time of 305-550 s and drift time of 1.3-1.75 ms highlight the aroma characteristics of roasted sweet potatoes relative to boiled sweet potatoes, indicating that the flavor outcomes under different processing methods are quite different.

[0078] The GC-IMS system is equipped with VOCal and three dedicated plug-ins, providing a powerful sample analysis platform. As the main tool, VOCal is responsible for qualitative and quantitative spectral interpretation, using NIST and IMS databases for substance identification, while allowing the database to be expanded through external standards. Its graphical interface associates each data point with a specific volatile organic compound (VOCs) and ensures accurate quantification of volatile components in the sample through appropriate calibration.

[0079] The volatile compounds were identified by comparing the IMS drift time and retention index with those of standard substances and combining the NIST and IMS databases in the GC-IMS system. The volatile components in the raw sweet potato fermented brandy sample were qualitatively analyzed and the detailed spectra of the identified compounds were obtained. Figure 5 As shown, the x-axis represents the difference time, the y-axis represents the resolution time, and the red numbers mark the corresponding compounds and are listed in Table 3.

[0080] Despite the wide coverage of the database, three compounds were not identified due to some limitations. Table 3 lists Figure 5 Among the compounds shown in the figure, esters are the most abundant, which significantly contribute to the fruity and floral aromas of brandy and enrich the flavor characteristics of sweet potatoes. The volatile organic compounds in the "Yanshu No. 25" sweet potato fermented brandy are mainly esters. The analysis detected 40 peaks and identified 37 different volatile compounds, including 22 esters, 7 alcohols, 4 ketones, 2 aldehydes, 1 terpenoid and 1 alkylbenzene, showing the unique volatile component characteristics of this variety.

[0081] In the drift region, multiple signals for a single compound were observed, likely due to the formation of adducts between the analyte ion and neutral molecules, including dimers. Compounds with high proton affinity may form dimers or trimers in the ion drift tube, a process affected by their concentration and half-life in the drift tube. The presence of monomers and dimers depends on the concentration and chemical stability of the volatile species, highlighting the complexity of ion behavior in the drift region. Eight volatile compounds were identified in this analysis as monomers and dimers: ethyl octanoate-D, isopentyl acetate-D, ethyl 2-methylbutyrate-M, ethyl butyrate-D, ethyl butyrate-M, ethyl 2-methylbutyrate-D, isopentyl acetate-M, and ethyl octanoate-M.

[0082] Table 3 Compound list

[0083] The present invention solves the limitation of traditional morphological diagrams and ion mobility spectroscopy that only provide an overall overview of compound composition and concentration through fingerprint spectroscopy analysis. These traditional methods cannot fully evaluate single volatile flavor compounds in sweet potato brandy. The present invention can clearly show the differences in volatile components in sweet potato fermented brandy samples treated by different processing methods through fingerprint spectroscopy, such as Figure 6 shown. Figure 6 In the graph, each row represents a signal peak in a single sample, and each column shows the signal intensity of a specific volatile compound in different samples. The color intensity of each point corresponds to the concentration of the compound, and brighter colors indicate higher concentrations. Figure 6 The numbers "1, 2, 3" without chemical names correspond to substances that have not yet been characterized in the migration spectrum library.

[0084] Figure 6The results showed that the diversity and concentration of volatile organic compounds in sample DB1 were the highest. Although samples DB2 to DB5 contained fewer volatile substances, they each presented unique flavor characteristics, which were related to their respective processing methods. Raw sweet potato brandy (DB1) was mainly composed of esters, alcohols and ketones, supplemented by aldehydes, terpenes and alkylbenzenes. Cooked sweet potato brandy (DB2) also had esters, alcohols and ketones, while steamed and fried brandies were dominated by esters and aldehydes. In contrast, baked sweet potato brandy (DB5) had higher concentrations of esters and ketones and contained significant methanol and acetone, which helped enhance the flavor after processing.

[0085] Esters were prevalent in all processed samples and played a key role in imparting fruity sensory characteristics to brandy. In DB1, esters such as isoamyl acetate and ethyl acetate imparted a winey aroma and sweet fruity flavor, respectively ( Figure 6 Region A in the sample). Ethyl butyrate brings apple and pineapple aromas to DB1. Meanwhile, ethyl formate and ethyl propionate contribute a variety of sweet, fruity, winey, floral, balsamic, and cocoa aromas. These esters indicate that raw sweet potato brandy is primarily fruity, with ethyl acetate being the signature marker of fruity aromas in spirits. In DB2, isobutyl isovalerate ( Figure 6 The unique banana aroma is introduced in region B, while the ethyl isovalerate in DB3 and DB4 ( Figure 6 Region C) and ethyl 2-methyl-3-methylbutyrate ( Figure 6 Area D in DB5 contributes to the aroma of fruit, apple, banana, and grass. The high concentration of hexyl acetate in DB5 gives the aroma of pineapple and grass. The octyl acetate in DB2 ( Figure 6 Region D in the middle adds an apricot-like aroma, while ethyl hexanoate enhances the fruity and herbal aromas, making the aroma of DB2 more delicate and smooth. The ester composition of DB2 to DB5 is different from that of DB1 because the heat treatment leads to the inactivation of enzymes, which in turn inhibits the formation of esters.

[0086] Alcohols also play an important role in enhancing the complexity of strong wine aromas. In DB1, 2-butanol and isobutanol impart wine-like and solvent-like aromas to the wine, while DB2 is characterized by methanol and 2-butoxyethanol, which bring fruity and herbal aromas. In DB5, ethyl acetate methanol imparts spicy, sweet, caramel and ethereal flavors. No unique alcohols were detected in DB3 or DB4. A common point in all samples is the presence of 3-methyl-1-butanol, which imparts the flavors of sweet potato brandy whiskey, malt and caramel to "Yanshu No. 25".

[0087] Aldehydes and ketones are key to enhancing wine complexity and regulating the release of aromatic compounds. In DB1, butyraldehyde contributes spicy and grassy notes, 3-hydroxy-2-butanone brings buttery and creamy aromas, and 2-octanone introduces soapy and gasoline-like flavors. In DB2, 3-octanone brings unique vanilla, buttery and resinous aromas. Trans-2-butenal in DB3 and DB4 provides floral aromas. DB5 lacks significant aldehyde and ketone compounds, but acetone, absent in DB1 and present in DB2, DB3, DB4 and DB5, imparts a refreshing fruity aroma. The differences in aldehydes and ketones among the samples may arise from differences in free amino acids, which serve as precursors for aldehyde formation through Strecker degradation.

[0088] In summary, the compounds listed in Table 3 and Figure 6 Comparison of the fingerprint spectra in the samples revealed significant differences in the volatile components, especially between DB1 and the other samples. The compounds showing the highest concentrations in DB1 included 3-hydroxy-2-butanone, ethyl acetate, and various esters such as ethyl 2-methylpropionate, ethyl 2-methylbutyrate, ethyl 3-methylbutyrate, ethyl butyrate, ethyl hexanoate, ethyl valerate, and ethyl formate. DB5 was characterized by higher levels of hydroxyacetone and ethyl octanoate, while ethylene glycol monobutyl ether, 3-octanone, and isoamyl acetate were mainly detected in DB2. The smallest difference was found between DB3 and DB4, indicating that the two samples were similarly affected by processing in terms of volatile components.

[0089] Table 4 Composition and odor characteristics of volatile compounds identified by GC-IMS

[0090] #Odor characteristics* mainly comes from the Flavornet database.

[0091] In the above step S6: (1) Dynamic principal component analysis (PCA) of samples Principal component analysis (PCA) is used to process complex multi-quantitative variable data sets and is widely used in cluster analysis to visualize and interpret high-dimensional data by reducing dimensionality, thereby highlighting patterns and differences between samples. Typically, a PCA model is considered valid when the combined explanatory power of the first two principal components (PC1 and PC2) exceeds 60%.

[0092] The present invention uses PCA to analyze the volatile compound data in the sweet potato brandy samples analyzed by GC-IMS, and the data are characterized by peak position and intensity. The data is processed by the dynamic PCA plug-in, which clearly reveals the differences between the sweet potato fermented brandies processed by different processing methods. Figure 7 As shown, the total contribution of the first two principal components PC1 and PC2 is 88%, of which PC1 accounts for 79% and PC2 accounts for 9% of the total variance. This far exceeds the 60% benchmark, indicating that the PCA model has a high explanatory power. The high variance explanation ensures that key information is retained in the reduced dimensionality dataset, accurately reflects the inherent differences of the original variables, and effectively distinguishes the brandy samples.

[0093] Figure 7 The PCA plot in Figure 1 distributes the DB1, DB3, DB4, and DB5 samples in different quadrants of the Cartesian coordinate system, with DB1 in the first quadrant, DB5 in the second quadrant, DB3 and DB4 both in the third quadrant, and DB2 on the negative x-axis. The clear separation of the samples in the plot emphasizes the significant differences in their characteristics. However, the close distance between DB3 and DB4 indicates that the variance between these two samples is small, indicating that the differences in the data set are more subtle.

[0094] also, Figure 8 These findings are further verified by the Euclidean distance in , which shows that DB3 and DB4 are close to each other, while DB1 and DB5 are the farthest apart. This distribution highlights the variation and similarity of the volatile organic compound (VOC) spectra in the brandy samples. It can be seen that the analytical method of the present invention can effectively distinguish the characteristics of volatile compounds in sweet potato brandy processed by different processing technologies.

[0095] (2) Fingerprint similarity analysis based on Euclidean distance Fingerprint similarity analysis evaluates the similarity of volatile organic compound (VOC) spectra between samples by calculating the Euclidean distance. This method relies on the distance coefficient, which is a key metric in cluster analysis. The larger the distance coefficient, the greater the difference between the samples, indicating that the size of the coefficient is proportional to the degree of dissimilarity of the samples. Conversely, a smaller coefficient indicates less difference and higher similarity between the samples.

[0096] like Figure 8 As shown, fingerprint similarity analysis based on Euclidean distance verifies Figure 6 and Figure 7 The brandy samples are clearly distinguished in the figure, with DB1 being significantly separated from DB2, DB3, DB4, and DB5. It is worth noting that the significant distance between DB1 and DB5 indicates the most obvious difference, which may be due to the chemical transformation of the sweet potato during the roasting process, giving it a unique flavor profile after fermentation.

[0097] In addition, DB3 and DB4 show the closest distance in Euclidean distance, reflecting the highest similarity between the sample pairs. This closeness indicates that the VOC spectra of DB3 and DB4 are less affected by their respective processing methods, emphasizing the effectiveness of the Euclidean distance measure in detecting small differences in sample characteristics.

[0098] The analytical method of the present invention enables detailed spectral comparative analysis through 2D and 3D spectra and difference formats, as well as analysis of VOCs fingerprints, providing qualitative and quantitative perspectives on sample changes. Dynamic principal component analysis clusters samples through dynamic PCA, which helps identify unknown compounds. In addition, the "nearest neighbor" analysis function uses the Euclidean distance matrix to improve the accuracy of comparative analysis by distinguishing closely related samples from relatively distant samples.

[0099] Sweet potatoes are rich in nutrients, providing 7.91 to 12.85 mg of beta-carotene per 100 g, as well as abundant essential minerals such as potassium (260 mg / 100 g), phosphorus (51 mg / 100 g) and calcium (29 mg / 100 g). It is worth noting that the bioavailability of sweet potato carotenoids is as high as 85%, which can be converted into vitamin A, and has important health significance for improving vitamin A deficiency in children and pregnant women. Traditional sweet potato cooking methods include steaming, microwaving, baking and frying, but in recent years, the innovative method of using sweet potatoes to brew fermented brandy has attracted increasing attention. This emerging method uses the natural sugars and aromatic compounds in sweet potatoes to enhance the complexity of flavor during the fermentation process, which is in line with the current food trend oriented towards healthy and diversified eating experiences. Compared with regular sweet potatoes, Yanshu No. 25 has more outstanding nutritional value, containing 30.2% high-viscosity protein and 3.48 times the soluble sugar of ordinary varieties per 100 g. Its unique sweet and sticky flavor, golden flesh and medium β-carotene content not only make it superior to other varieties in terms of edible quality, but also bring significant advantages to the processing of fermented alcoholic beverages. However, the current research on volatile aromatic substances in fermented sweet potato brandy, especially the effects of different cooking methods on them, is relatively limited. To fill this gap, the present invention adopts gas chromatography-ion mobility spectrometry (GC-IMS) technology, which is a highly sensitive volatile detection method that does not require sample pretreatment, is suitable for various matrices, and can accurately characterize the flavor characteristics of samples. In the present invention, GC-IMS is used to explore the potential flavor transformation pathways in fermented sweet potato brandy, providing new insights into the complex flavor matrix of brandy.

[0100] This paper systematically evaluates the volatile organic compounds (VOCs) in sweet potato brandy under different processing conditions, and uses GC-IMS technology to establish a comprehensive VOC spectrum that captures retention time and migration time for comparative analysis between different samples. This approach will reveal the impact of different processing methods on the volatile composition and sensory characteristics of sweet potato brandy.

[0101] The present invention combines gas chromatography-ion mobility spectrometry (GC-IMS) and multivariate data analysis to reveal the significant changes in volatile organic compounds (VOCs) caused by no heat treatment, steaming, boiling, frying and baking. The results show that different processing methods significantly change the concentration and composition of key volatile compounds, which in turn affect the aroma and taste characteristics of brandy. Esters, in particular, are particularly sensitive to processing methods. Test group 1 without heat treatment showed the richest and most diverse ester content. The processed samples generated new aromatic compounds due to thermal reactions, showing a unique volatile component spectrum. In addition, the processing methods also significantly changed alcohols, aldehydes and ketones, which have important contributions to the sensory properties of brandy.

[0102] Principal component analysis (PCA) and Euclidean distance analysis provided valuable clustering insights, highlighting similarities and differences between samples. These analytical approaches confirmed that each processing method imparted a unique VOC profile to brandy, highlighting the key role of processing in shaping the sensory profile of the final product.

[0103] The present invention uses gas chromatography-ion mobility spectrometry (GC-IMS) to analyze the effects of different processing methods on volatile organic compounds in brandy, so that the required processing method can be selected according to the type and concentration of volatile organic compounds. For example, raw sweet potatoes can better retain the fruity aroma brought by ester substances, while roasting can bring more caramel flavor. According to this information and the required flavor, the processing method can be selected or combined to increase the content of specific aroma substances in brandy to achieve the purpose of optimizing the flavor. The present invention uses fingerprints to know that brandy made from raw sweet potatoes retains the most esters and aldehyde compounds and has fruit and floral characteristics; steaming and boiling treatments make the taste more mellow and retain more original sweet potato aroma; while baking and frying add caramel and roasted aroma, but make part of the original aroma covered. Therefore, a suitable combination of processing methods can be selected according to the desired aroma characteristics of the product, such as steaming and boiling can be preferred to retain the natural aroma of sweet potatoes. The present invention uses dynamic principal component analysis (PCA) and Euclidean distance analysis to cluster samples produced by different processing methods and identify samples that are closest to the target flavor characteristics. Cluster analysis can help find the optimal combination of processing methods to ensure that the product maintains consistent flavor characteristics in different batches. The present invention uses trained tasters to conduct sensory scoring to evaluate the effects of different processing methods on the aroma, taste, and typicality of brandy. The sensory evaluation provides direct sensory feedback for flavor optimization, and the processing method can be further adjusted based on the sensory score. For example, the sweet potato pretreatment method of high-scoring samples can be selected as a way to optimize the flavor.

[0104] In summary, the present invention provides a method for analyzing the flavor of sweet potato brandy and comprehensively explores the effects of different processing methods on the volatile component spectrum and sensory attributes of sweet potato fermented brandy. The research results are of great significance for the food industry to optimize production technology to produce sweet potato brandy products that meet consumers' preferences for specific aroma and flavor.

[0105] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for analyzing the flavor of sweet potato brandy, characterized in that: The following steps are involved: Step S1, preparation of sweet potato brandy; Step S2, measuring the physical and chemical parameters of the sweet potato brandy in step S1; Step S3, performing a sensory evaluation test on the sweet potato brandy of step S1; Step S4, performing GC-IMS analysis on volatile compounds of the sweet potato brandy of step S1; Step S5, performing comparative spectral analysis on the result of step S4; Step S6, performing hierarchical cluster analysis on the result of step S4.

2. The sweet potato brandy flavor analysis method according to claim 1, characterized in that: In step S1, the preparation of the sweet potato brandy specifically comprises the following steps: (1) Sweet potato pretreatment; (2) peeling the pretreated sweet potatoes, stirring and homogenizing them into mashed potatoes, and transferring them to a fermentation container for fermentation; (3) After fermentation is completed, the fermented mash is distilled to produce brandy.

3. The sweet potato brandy flavor analysis method according to claim 2, characterized in that: In step (1), the sweet potato pretreatment includes washing, drying and weighing, or also includes any one of steaming, boiling, frying or baking.

4. The sweet potato brandy flavor analysis method according to claim 3, characterized in that: In step (2), after the sweet potato is made into mashed potato, pectinase is added, and sugar and water are added to adjust the soluble solid content of the paste to 25%, distiller's yeast is evenly distributed on the surface of the paste, and the fermentation container is sealed. The fermentation is carried out in a well-ventilated environment with a temperature controlled at 20° C. to 25° C. and a pH value of 4.5 to 5.

0. The fermentation cycle is 20 days, and the final alcohol content of the fermented paste is 11% to 13%.

5. The method for analyzing the flavor of sweet potato brandy according to claim 4, wherein: In step (3), the sweet potato brandy is prepared by double distillation method, the temperature is controlled at 80°C-85°C during the first distillation, and the temperature is controlled at 60°C-65°C during the second distillation.

6. The method for analyzing the flavor of sweet potato brandy according to claim 5, characterized in that: In step S2, the physicochemical parameters include pH value, reducing sugar content, alcohol content, soluble solid content, total ester content and total acidity.

7. The method for analyzing the flavor of sweet potato brandy according to claim 6, characterized in that: The specific steps of step S4 are: The brandy prepared by different sweet potato pretreatments in step S1 is sampled, the sample is 1.0 mL-1.5 mL, put into a headspace bottle, sealed with a magnetic screw cap, and incubated at 60° C. and 500 rpm for 10 minutes to 15 minutes to balance the volatile organic compounds in the headspace; after incubation, the headspace sample is drawn using a syringe heated to 85° C. and injected into the GC-IMS system in a splitless mode; the chromatographic column and the migration tube are maintained at 80° C. and 45° C., respectively; During the analysis, the carrier gas was high-purity nitrogen, and the flow rate program was: 2 mL / min for the first 2 minutes, then increased to 10 mL / min within 8 minutes, then increased to 100 mL / min within 10 minutes, and finally maintained at 100 mL / min until the end of the analysis. The analysis time was 30-35 minutes.

8. The method for analyzing the flavor of sweet potato brandy according to claim 7, characterized in that: The specific steps of step S5 are: using the spectrum of one of the samples in step S4 as a reference, and generating a difference spectrum by comparing it with the spectrum of other samples.

9. The method for analyzing the flavor of sweet potato brandy according to claim 8, characterized in that: Step S6 includes dynamic principal component analysis and fingerprint similarity analysis of the sample.

10. The method for analyzing the flavor of sweet potato brandy according to claim 9, characterized in that: The fingerprint similarity analysis evaluates the similarity of VOC spectra between samples by calculating the Euclidean distance.

Images