System for makgeolli brewing based on IoT and artificial intelligence

Abstract

An IoT-based AI makgeolli brewing system according to various embodiments of the present invention is disclosed. The system may include: a brewing tank that accommodates makgeolli raw materials and performs fermentation; an internal sensor installed inside the brewing tank to measure state data of the brewing liquid; an external sensor installed outside the brewing tank to measure external environment data; a temperature control unit installed to surround the brewing tank to control the internal temperature of the brewing tank; a stirring unit installed inside the brewing tank to stir the brewing liquid; and a computing device that receives data collected from the internal sensor and the external sensor, analyzes the received data using an AI model, and controls process conditions for each stage of the brewing process based on the analysis results.

Description

Technology Field

[0001] The present invention relates to an IoT-based AI makgeolli brewing system. Background Technology

[0003] Traditional fermented alcoholic beverages such as Makgeolli are produced using rice, nuruk, yeast, and water as main ingredients through a complex fermentation process in which saccharification and alcohol fermentation occur simultaneously. In this brewing process, the most important factors to manage for producing high-quality alcoholic beverages are maintaining the temperature and ensuring uniform mixing (stirring) of the ingredients. Generally, while yeast requires a warm temperature for activation during the initial culture phase, fermentation heat is generated due to rapid yeast proliferation during the main fermentation phase where alcohol production begins in earnest. If this heat is not properly cooled, problems such as excessive sourness or the formation of off-flavors (spoilage) can occur due to over-fermentation. Furthermore, after fermentation is complete, a process of low-temperature aging is essential to dissolve carbonation and stabilize the flavor.

[0004] Conventional brewing techniques primarily involved using traditional jars that relied on natural temperatures, or maintaining a set temperature using stainless steel fermentation tanks equipped with cooling jackets or heaters.

[0005] However, these conventional devices had limitations in that they could not reflect real-time changes in temperature or humidity (such as seasonal factors) of the external environment where the tank is installed, and were limited to mechanically controlling the temperature.

[0006] For example, if the ambient temperature rises or falls rapidly, heat may penetrate or be lost into the tank, causing the temperature of the fermentation liquid to deviate from the target range. This consequently leads to a decrease in yeast vitality or death, which is a major cause of reduced uniformity in the quality of the liquor. Furthermore, since the rice flour and solids used as ingredients for makgeolli tend to settle at the bottom of the tank over time, continuous stirring is required. However, conventional devices only perform the function of physically mixing using simple rotating blades; they lack precise control functions to adjust stirring speed according to fermentation conditions or viscosity changes, or to prevent overloading caused by sediment. Consequently, problems such as reduced energy efficiency and decreased equipment durability have been pointed out. In particular, given the modern brewing trend of adding various auxiliary ingredients, relying solely on constant speed operation and constant temperature maintenance has the disadvantage of making it difficult to optimize the complex fermentation mechanisms of microorganisms.

[0007] Therefore, there is a demand in the industry for advanced brewing systems capable of actively responding to changes in the external environment and precisely managing the state of the fermented liquid. In this regard, Korean Published Patent No. 10-2014-0057834 discloses a makgeolli manufacturing apparatus and a manufacturing method. The problem to be solved

[0009] The technical problem that the present invention aims to solve is to provide an IoT-based AI makgeolli brewing system.

[0010] The technical problems of the present invention are not limited to those mentioned above, and other unmentioned technical problems will be clearly understood by those skilled in the art from the description below. means of solving the problem

[0012] According to one embodiment of the present invention for solving the problem described above, an IoT-based AI makgeolli brewing system is disclosed. The system comprises: a brewing tank that accommodates makgeolli raw materials and performs fermentation; an internal sensor installed inside the brewing tank to measure state data of the brewing liquid; an external sensor installed outside the brewing tank to measure external environment data; a temperature control unit installed to surround the brewing tank to control the internal temperature of the brewing tank; and a stirring unit installed inside the brewing tank to stir the brewing liquid. The apparatus includes a computing device that receives data collected from the internal sensor and the external sensor, analyzes the received data using an AI model, and adjusts process conditions for each stage of the brewing process based on the analysis results; wherein the computing device adjusts process conditions including at least one of a temperature set value for a fermentation stage, a fermentation maintenance time, and a cooling time by reflecting data collected in real time from the internal sensor and the external sensor based on a preset brewing recipe; and wherein the preset brewing recipe may include a first section in which the internal temperature of the brewing tank is heated and maintained to a first target temperature for yeast activation, a second section in which the internal temperature is lowered and maintained to a second target temperature lower than the first target temperature for alcohol production, and a third section in which the internal temperature is rapidly cooled and maintained to a third target temperature lower than the second target temperature for fermentation cessation and maturation.

[0013] In an alternative embodiment, the internal sensor comprises: an internal temperature sensing sensor for detecting the temperature inside the brewing tank; a humidity sensing sensor for detecting the humidity inside the brewing tank; an alcohol content sensing sensor for detecting the alcohol content of the brewing liquid; a sugar content sensing sensor for detecting the sugar content of the brewing liquid; and a liquor quality sensing sensor for detecting liquor quality data of the brewing liquid; and the external sensor may comprise: an ambient temperature sensing sensor for detecting the ambient temperature; and an ambient humidity sensing sensor for detecting the ambient humidity.

[0014] In an alternative embodiment, the temperature control unit may include: a refrigerant circulation path installed to surround the outer surface of the brewing tank, through which a refrigerant circulates to perform heat exchange with the brewing tank; and a heating element disposed along the outer surface of the brewing tank intersecting or adjacent to the refrigerant circulation path, which generates heat by an electrical signal to heat the brewing tank.

[0015] In an alternative embodiment, the computing device maintains the internal temperature of the brewing tank at a first target temperature in the first section, which is the yeast activation stage after inoculation and yeast addition, according to the preset brewing recipe, lowers and maintains it at a second target temperature in the second section, which is the alcohol production stage, and cools it to a third target temperature in the third section, which is the aging and storage stage; wherein, if the amount of change in the ambient temperature detected by the external sensor exceeds a preset threshold change amount, the data on the rate of change of the ambient temperature is input into a pre-trained AI model to simulate the expected trend of change of the internal temperature of the brewing tank, and if the simulation result predicts that the internal temperature will deviate from the target temperature range, a preset fermentation stabilization buffer section is applied to control the output of the heating element or refrigerant circulation path in response to the rate of change before the internal temperature deviates from the target temperature range.

[0016] In an alternative embodiment, the computing device may be characterized by recognizing a fruit additive addition event when the initial sugar content data collected from the internal sensor exceeds the expected range of the preset brewing recipe, predicting the sugar decomposition rate and alcohol production rate by the fruit additive through the AI ​​model, shortening the maintenance time of the second section or lowering the second target temperature if the predicted sugar decomposition rate is faster than the preset alcohol production rate, and increasing the rotation speed of the stirring unit if the predicted sugar decomposition rate is slower than the preset sugar decomposition rate.

[0017] Other specific details of the present invention are included in the detailed description and drawings. Effects of the invention

[0019] The present invention enables the uniform production of high-quality makgeolli regardless of season or location by actively optimizing fermentation process conditions according to external environmental changes and the characteristics of input raw materials through the analysis of IoT sensor data and artificial intelligence (AI).

[0020] The effects of the present invention are not limited to those mentioned above, and other unmentioned effects will be clearly understood by a person skilled in the art from the description below. Brief explanation of the drawing

[0022] FIG. 1 is a drawing illustrating an IoT-based AI makgeolli brewing system according to one embodiment of the present invention. FIG. 2 is a hardware configuration diagram of an internal sensor and an external sensor according to an embodiment of the present invention. FIG. 3 is a hardware configuration diagram of a computing device according to one embodiment of the present invention. FIGS. 4 to 6 are drawings for explaining an IoT-based AI makgeolli brewing method according to an embodiment of the present invention. Specific details for implementing the invention

[0023] The advantages and features of the present disclosure and the methods for achieving them will become clear by referring to the embodiments described in detail together with the accompanying drawings. However, the present disclosure is not limited to the embodiments presented below, but can be implemented in various different forms and should be understood to include all modifications, equivalents, and substitutions that fall within the spirit and scope of the present disclosure. The embodiments presented below are provided to make the present disclosure complete and to fully inform those skilled in the art of the scope of the invention. In describing the present disclosure, detailed descriptions of related prior art are omitted where it is determined that such detailed descriptions may obscure the essence of the present invention.

[0024] The terms used herein are used merely to describe specific embodiments and are not intended to limit the disclosure. Unless otherwise defined, all terms used herein have the same meaning as generally understood by those skilled in the art to which this disclosure pertains.

[0025] In this specification, singular expressions include plural expressions unless the context clearly indicates otherwise. Furthermore, terms such as "comprising" or "having" are intended to specify the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0026] Additionally, terms including ordinal numbers, such as "first" or "second" as used herein, may be used to describe various components, but the components should not be limited by the terms. The terms are used solely for the purpose of distinguishing one component from another.

[0027] Phrases such as "in one embodiment," "according to one embodiment," "related to one embodiment," or "according to an implementation of one embodiment" in this specification do not necessarily refer to the same embodiment. Furthermore, throughout this specification, "examples" are arbitrary distinctions to facilitate the description of the present disclosure, and each embodiment does not need to be mutually exclusive. For example, configurations mentioned for the description of one embodiment may be applied and / or implemented in other embodiments, and may be modified and applied and / or implemented to the extent that they do not depart from the scope of the present disclosure.

[0028] Some embodiments of the present disclosure may be represented by functional block configurations and various processing steps. Some or all of these functional blocks may be implemented by various numbers of hardware and / or software configurations that perform specific functions. For example, the functional blocks of the present disclosure may be implemented by one or more microprocessors or by circuit configurations for a specific function.

[0029] Additionally, for example, the functional blocks of the present disclosure may be implemented in various programming or scripting languages. The functional blocks may be implemented as algorithms executed on one or more processors. Furthermore, the present disclosure may employ prior art for electronic configuration, signal processing, and / or data processing, etc. Terms such as "mechanism," "element," "means," and "configuration" may be used broadly and are not limited to mechanical and physical configurations. Additionally, terms such as "-part," "-module," etc. refer to a unit that processes at least one function or operation, which may be implemented in hardware or software, or as a combination of hardware and software.

[0030] Furthermore, the connecting lines or connecting members between the components depicted in the drawings are merely illustrative of functional connections and / or physical or circuit connections. In the actual device, connections between components may be represented by various alternative or added functional connections, physical connections, or circuit connections.

[0031] In addition, some components in the drawings may be depicted with their size or proportions slightly exaggerated. Also, components depicted in one drawing may not be depicted in another drawing.

[0032] The present disclosure will be described in detail below with reference to the attached drawings.

[0034] FIG. 1 is a diagram illustrating an IoT-based AI makgeolli brewing system according to an embodiment of the present invention. FIG. 2 is a hardware configuration diagram of an internal sensor and an external sensor according to an embodiment of the present invention.

[0035] Referring to FIG. 1, a brewing tank (10), an internal sensor (20), an external sensor (30), a temperature control unit (40), a refrigerant circulation path (41), a heating element (42), a stirring unit (50), a supply pipe (60), an outlet (70), and a computing device (100) according to one embodiment of the present invention may be included. The system illustrated in FIG. 1 is according to one embodiment, and its components are not limited to the embodiment illustrated in FIG. 1 and may be added, changed, or deleted as needed.

[0036] In one embodiment, the computing device (100) can perform IoT-based AI makgeolli brewing. For example, the computing device (100) can determine the current stage of fermentation by analyzing sugar content and alcohol content data received from an internal sensor (20), and create an optimal fermentation environment by controlling the temperature control unit (40) and the stirring unit (50) accordingly.

[0037] Specifically, the computing device (100) can adjust process conditions including at least one of a temperature setting value for a fermentation stage, a fermentation maintenance time, and a cooling time by reflecting data collected in real time from an internal sensor and an external sensor based on a preset brewing recipe.

[0038] Here, the pre-set brewing recipe may include a first section in which the internal temperature of the brewing tank (10) is heated and maintained to a first target temperature (e.g., 28°C) for yeast activation, a second section in which the internal temperature is lowered and maintained to a second target temperature (e.g., 25°C) lower than the first target temperature for alcohol production, and a third section in which the internal temperature is rapidly cooled and maintained to a third target temperature (e.g., 5.6°C) lower than the second target temperature for fermentation cessation and maturation. However, it is not limited thereto.

[0039] More specifically, when the computing device (100) reaches a point in time when it is determined that sufficient yeast proliferation has occurred in the first section (e.g., when the slope of the sugar content decrease reaches a preset threshold), it may enter the second section early and lower the temperature to the second target temperature (25°C) even if a preset time (e.g., 1 day) has not elapsed, thereby preventing over-fermentation. Additionally, when the computing device (100) detects the addition of fruit additives such as strawberries or bananas in the second section, it may perform flexible recipe control by further lowering the second target temperature from 25°C to 24°C or 23°C to suppress the rapid generation of fermentation heat caused by the sugars of the fruit, or by advancing the time of entry into the third section (aging stage).

[0040] Accordingly, the computing device (100) of the present invention can produce high-quality makgeolli of uniform quality by applying an optimal recipe in real time through artificial intelligence-based analysis, even if there are changes in the external environment or the characteristics of the raw materials being input (e.g., whether fruit is added).

[0041] Hereinafter, an example of a method in which a computing device (100) performs IoT-based AI makgeolli brewing will be described with reference to FIGS. 4 to 6.

[0042] In various embodiments, the computing device (100) may provide Web or Application-based services. However, it is not limited thereto.

[0043] The computing device (100) may include any type of computer system or computer device, such as, for example, a microprocessor, a mainframe computer, a digital processor, a portable device, and a device controller. However, it is not limited thereto.

[0044] Hereinafter, the hardware configuration of the computing device (100) will be described with reference to FIG. 3.

[0045] Meanwhile, the brewing tank (10) can accommodate makgeolli raw materials, and the accommodated makgeolli raw materials can be fermented within the brewing tank (10).

[0046] Specifically, the brewing tank (10) is formed of stainless steel material with excellent acid resistance and corrosion resistance to provide a hygienic fermentation environment, and may be formed with a double jacket structure or a structure including insulation to efficiently control heat exchange with the outside. In addition, the brewing tank (10) may include a sealable lid to form anaerobic fermentation conditions or may further include an air-lock configuration to discharge carbon dioxide generated during the fermentation process.

[0047] An internal sensor (20) is installed inside a brewing tank (10) to measure the state data of the brewing liquid. Here, the internal sensor (20) may include an internal temperature sensing sensor (21), a humidity sensing sensor (22), an alcohol content sensing sensor (23), a sugar sensing sensor (24), and a liquor quality sensing sensor (25). However, it is not limited thereto.

[0048] Specifically, the internal temperature sensing sensor (21) measures the core temperature of the brewing liquid (mash) in real time to provide an indicator for determining the active state of the yeast, and the humidity sensing sensor (22) measures the humidity of the headspace of the brewing tank (10) to detect whether condensation has occurred or the degree of fermentation gas saturation. In addition, the alcohol content sensing sensor (23) measures the concentration of ethanol produced as fermentation progresses, and the sugar sensing sensor (24) measures the refractive index or specific gravity to generate residual sugar (Brix) data, thereby providing basic data for the computing device (100) to calculate the fermentation rate. Furthermore, the liquor quality sensing sensor (25) may be configured to include a pH sensor or a conductivity sensor to detect changes in the acidity of the makgeolli or to detect signs of abnormal fermentation (such as acetic acid fermentation) at an early stage.

[0049] For example, the internal sensor (20) can detect when the sugar content inside the brewing tank (10) decreases from an initial 15 Brix to 5 Brix and the alcohol content reaches 6%, and transmit the data to the computing device (100), thereby providing data that the computing device (100) determines that the current fermentation stage should be switched from 'main fermentation' to 'aging stage'.

[0050] In an additional embodiment, the internal sensor (20) may further include a camera module or a vision sensor for recognizing the type and amount of additives introduced into the brewing tank (10).

[0051] In this case, the computing device (100) can analyze image data acquired from the camera module to identify the type of fruit introduced, such as strawberries, Korean melons, and bananas, and perform adaptive recipe control to automatically correct the fermentation recipe according to the sugar content characteristics of the identified fruit.

[0052] Additionally, if the internal sensor (20) includes a camera module or a vision sensor, the system may further include a lens cleaning nozzle for removing contaminants or condensation attached to the lens surface of the camera module or vision sensor.

[0053] Here, the lens cleaning nozzle may be configured to spray cleaning water (e.g., distilled water, purified water, or sterilized water). And, the computing device (100) may calculate the dilution ratio of the brewing liquid by reflecting the increased moisture content according to the amount of cleaning water sprayed by the lens cleaning nozzle, and perform precise concentration control to extend the fermentation time to maintain the target alcohol content or correct the predicted final alcohol content data.

[0054] An external sensor (30) can be installed outside the brewing tank (10) to measure external environment data. Here, the external sensor (30) may include an external temperature sensing sensor (31) and an external humidity sensing sensor (32). However, it is not limited thereto.

[0055] Specifically, the ambient temperature sensor (31) measures the ambient temperature of the space where the brewing system is installed and can provide ambient temperature change rate data, which is a variable for the computing device (100) to simulate changes in the internal temperature of the brewing tank (10). In addition, the ambient humidity sensor (32) measures ambient humidity and can be used as data to predict condensation that may occur when the cooling pipe is operated or to correct the insulation efficiency of the outside of the brewing tank (10).

[0056] For example, an external sensor (30) can detect a trend of the outside temperature dropping rapidly by more than 3°C per hour after sunset and transmit it to a computing device (100), and the computing device (100) that receives this can perform feedforward control to preemptively activate the heating element (42) before the internal temperature of the brewing tank (10) drops below a set value (e.g., 25°C).

[0057] In an additional embodiment, the external sensor (30) may further include a piezoelectric microphone or vibration sensor attached around the brewing tank (10) to collect carbon dioxide bubble generation frequency and burst acoustic data.

[0058] In this case, the computing device (100) can analyze the collected acoustic data to estimate the fluctuation of biological fermentation heat according to the actual activity level of the yeast in real time, and combine this with ambient temperature change data to enhance the variables of the temperature prediction simulation.

[0059] The temperature control unit (40) is installed to surround the brewing tank (10) and can control the internal temperature of the brewing tank (10).

[0060] Specifically, the temperature control unit (40) is connected to the refrigerant circulation path (41) and the heating element (42), and can control the supply amount (flow rate) of the refrigerant or the power application time of the heating element (42) (PID control, etc.) according to the control signal received from the computing device (100) to control the internal temperature of the brewing tank (10).

[0061] Here, the refrigerant circulation path (41) is installed to surround the outer surface of the brewing tank, and the refrigerant can circulate inside to perform heat exchange with the brewing tank (10).

[0062] Specifically, the refrigerant circulation path (41) acts as an evaporator of the refrigeration cycle and can perform the function of absorbing the fermentation heat generated during alcohol fermentation to prevent the temperature of the brewing liquid from overheating above a set range, or rapidly cooling the brewing liquid to a low temperature (e.g., 5.6°C) for aging after fermentation is completed.

[0063] A heating element (42) is positioned along the outer surface of the brewing tank (10) in an intersecting or adjacent manner with the refrigerant circulation path (41), and can heat the brewing tank (10) by generating heat through an electrical signal.

[0064] Specifically, the heating element (42) can be implemented as a heating wire or a film heater and can perform a heat retention function to raise the brewing liquid to an appropriate temperature (e.g., 28°C) to help activate yeast during the initial brewing stage, or to prevent the internal temperature of the brewing tank (10) from dropping below the appropriate fermentation temperature when the outside temperature drops sharply, such as during winter.

[0065] For example, the temperature control unit (40) can execute a temperature profile in which, under the control of the computing device (100), the heating element (42) is driven to maintain 28°C during the yeast activation section on day 1, the refrigerant circulation path (41) is PID controlled to maintain 25°C during the alcohol production section on day 2, and the refrigerant circulation path (41) is operated at maximum load to perform rapid cooling down to 5.6°C when entering the aging section on day 5.

[0066] The stirring unit (50) is installed inside the brewing tank (10) and can stir the brewing liquid.

[0067] Specifically, the stirring unit (50) includes a rotary motor (51) and a screw or blade connected thereto, and through rotational movement, it prevents solid matter such as rice flour or yeast from settling and sticking to the bottom of the brewing tank (10), maintains a uniform temperature of the entire brewing liquid, and smoothly supplies oxygen to the yeast to increase fermentation efficiency. For example, the rotation speed (RPM) of the stirring unit (50) can be variably adjusted according to the viscosity of the brewing liquid or the fermentation stage under the control of the computing device (100).

[0068] For example, the stirring unit (50) rotates at a high speed (e.g., 60 RPM) when the initial raw materials are introduced to uniformly mix the rice flour and yeast, and switches to a low speed (e.g., 10 RPM) or intermittent operation (e.g., 10 minutes of operation every hour) when stable fermentation is required, thereby preventing sedimentation of the lees without causing physical stress to the yeast and supplying appropriate oxygen.

[0069] The supply pipe (60) and the discharge port (70) represent a flow path configuration connecting the inside and outside of the brewing tank (10).

[0070] Specifically, the supply pipe (60) is provided on the upper or side of the brewing tank (10) and serves as a passage for introducing raw materials such as water, rice, yeast, and additives, or for supplying washing water, and the discharge port (70) is provided at the lower part of the brewing tank (10) and can be configured to discharge the fermented makgeolli to the outside or collect a tasting sample by operating a valve.

[0071] Therefore, the system of the present invention detects biological changes (sugar content, alcohol content, acidity) inside the brewing tank and external environmental changes (temperature, humidity) in real time using IoT sensors, and actively controls the temperature control and stirring processes by analyzing this data with an artificial intelligence (AI) model. This provides the effect of consistently producing high-quality Makgeolli regardless of operator skill level or seasonal factors. Furthermore, even if various additives such as fruit are added, the AI ​​recognizes them and automatically adjusts the recipe to the optimal level, thereby providing a brewing solution optimized for modern brewing trends that require multi-variety, small-batch production.

[0073] FIG. 3 is a hardware configuration diagram of a computing device according to one embodiment of the present invention.

[0074] Referring to FIG. 3, a computing device (100) according to one embodiment of the present invention may include one or more processors (110), a memory (120) for loading a computer program (151) executed by the processor (110), a bus (130), a communication interface (140), and a storage (150) for storing the computer program (151). Here, FIG. 3 illustrates only the components related to the embodiment of the present invention. Therefore, a person skilled in the art to which the present invention pertains will understand that other general-purpose components may be included in addition to the components illustrated in FIG. 3.

[0075] The processor (110) controls the overall operation of each component of the computing device (100). The processor (110) may be composed of one or more cores and may include processors for data analysis and deep learning, such as a central processing unit (CPU), a general purpose graphics processing unit (GPGPU), or a tensor processing unit (TPU) of the computing device. Alternatively, it may be configured to include any type of processor well known in the art of the present invention.

[0076] Additionally, the processor (110) can perform operations for at least one application or program for executing the method according to embodiments of the present invention, and the computing device (100) may have one or more processors.

[0077] In various embodiments, the processor (110) may further include RAM (Random Access Memory, not shown) and ROM (Read-Only Memory, not shown) for temporarily and / or permanently storing signals (or data) processed within the processor (110). Additionally, the processor (110) may be implemented in the form of a system-on-chip (SoC) comprising at least one of a graphics processing unit, RAM, and ROM.

[0078] Memory (120) stores various data, instructions and / or information. Memory (120) may load a computer program (151) from storage (150) to execute a method / operation according to various embodiments of the present invention. When the computer program (151) is loaded into memory (120), the processor (110) may perform the method / operation by executing one or more instructions constituting the computer program (151). Memory (120) may be implemented as volatile memory such as RAM, but the technical scope of the present invention is not limited thereto.

[0079] The bus (130) provides communication functions between components of the computing device (100). The bus (130) can be implemented as various types of buses, such as an address bus, a data bus, and a control bus.

[0080] The communication interface (140) supports wired and wireless internet communication of the computing device (100). Additionally, the communication interface (140) may support various communication methods other than internet communication. To this end, the communication interface (140) may be configured to include a communication module well known in the art of the present invention. In some embodiments, the communication interface (140) may be omitted.

[0081] Storage (150) can store a computer program (151) non-temporarily. When performing a process according to an embodiment of the present invention through a computing device (100), storage (150) can store various information necessary to perform a method according to the disclosed embodiment or to provide a service.

[0082] The storage (150) may be configured to include non-volatile memory such as ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), flash memory, a hard disk, a removable disk, or any form of computer-readable recording medium well known in the art to which the present invention belongs.

[0083] A computer program (151) may include one or more instructions that cause a processor (110) to perform a method / operation according to various embodiments of the present invention when loaded into memory (120). That is, the processor (110) may perform the method / operation according to various embodiments of the present invention by executing the one or more instructions.

[0084] In one embodiment, the computer program (151) may include one or more instructions to perform various methods related to various tasks related to learning a neural network model.

[0085] The steps of the method or algorithm described in connection with embodiments of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or implemented by a combination thereof. The software module may reside in RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), Flash Memory, a hard disk, a removable disk, a CD-ROM, or any form of computer-readable recording medium well known in the art to which the present invention belongs.

[0086] The components of the present invention may be implemented as a program (or application) and stored on a medium to be executed in combination with a computer, which is hardware. The components of the present invention may be implemented as software programming or software elements, and similarly, embodiments may be implemented in programming or scripting languages ​​such as C, C++, Java, assembler, etc., including various algorithms implemented as combinations of data structures, processes, routines, or other programming configurations. Functional aspects may be implemented as algorithms executed on one or more processors.

[0088] FIGS. 4 to 6 are drawings for explaining an IoT-based AI makgeolli brewing method according to an embodiment of the present invention.

[0089] In one embodiment, the computing device (100) can adjust process conditions including at least one of a temperature set value for a fermentation stage, a fermentation maintenance time, and a cooling time by reflecting data collected in real time from an internal sensor and an external sensor based on a preset brewing recipe.

[0090] Referring to FIG. 4, the computing device (100) can maintain the temperature of the brewing tank at a first target temperature in the first section, which is the yeast activation step after the introduction and yeast input, according to a preset brewing recipe (S110).

[0091] Specifically, the computing device (100) compares the temperature data of the brewing liquid (mold) received in real time from the internal temperature sensing sensor (21) of the internal sensor (20) with the first target temperature, and by controlling the refrigerant circulation path (41) and heating element (42) of the temperature control unit (40) in proportional-integral-derivative (PID) based on the deviation (Error), the internal environment of the brewing tank (10) can be precisely maintained at a constant temperature in a state optimized for the initial proliferation of yeast. Here, the first section may refer to the initial culture stage before alcohol fermentation begins, in which the introduced yeast exponentially increases the number of cells through aerobic respiration. And, the first target temperature may refer to the optimal growth temperature at which the yeast can be most active and proliferate.

[0092] Here, the computing device (100) not only controls the temperature but also drives the stirring unit (50) to ensure that the solid raw material (koji, rice flour, etc.) and the liquid raw material (water) are uniformly mixed, thereby securing the amount of dissolved oxygen in the brewing liquid and promoting smooth respiration and proliferation of the yeast.

[0093] For example, when the computing device (100) detects that a mixture of a co-enzyme (koji), koji, yeast, rice flour, and water is contained in the brewing tank (10) during the first day of the brewing process, or receives a process start signal from the user, it can set the first target temperature to 28°C according to a preset brewing recipe, and power the heating element (42) or control the opening and closing amount of the valve of the refrigerant circulation path (41) so that the internal temperature reaches and is maintained at 28°C.

[0094] In this process, the computing device (100) can prevent the raw material from clumping by setting the rotational torque of the stirring unit (50) high or controlling the rotation cycle to be short, taking into account that the initial state of the brewing liquid has high viscosity, and can induce the yeast to spread evenly throughout the brewing tank (10) and become activated.

[0096] In one embodiment, the computing device (100) can be lowered and maintained to a second target temperature during the second section, which is the alcohol generation step (S120).

[0097] Specifically, the computing device (100) can control the internal temperature of the brewing tank (10) to a second target temperature lower than the first target temperature in order to suppress the rapid fermentation heat generated inside the brewing liquid and to secure stable quality as full-scale anaerobic fermentation begins, in which yeast that has proliferated sufficiently through the first section breaks down sugars to produce ethanol and carbon dioxide. Here, the second section refers to the main fermentation stage in which the alcohol production metabolism of the yeast occurs most actively while the oxygen supply is cut off, and the second target temperature may refer to an appropriate control temperature that can maximize the alcohol yield while preventing a decrease in the vitality or death of the yeast due to high temperature, the occurrence of strong acidity (sour taste) due to over-fermentation, or spoilage caused by the proliferation of unwanted microorganisms.

[0098] To this end, the computing device (100) monitors the temperature rise trend of the brewing liquid due to fermentation heat in real time through an internal temperature sensing sensor (21), and when a sign is detected that the internal temperature is about to exceed a set second target temperature, it can create a comfortable fermentation environment by performing cooling control to release the fermentation heat to the outside by supplying a refrigerant to the refrigerant circulation path (41) of the temperature control unit (40).

[0099] For example, when the computing device (100) enters the second phase from the second to the fourth day after the first phase of the first day ends, it can set the second target temperature to 25°C according to a preset brewing recipe, gradually cool the inside of the brewing tank (10) from 28°C to 25°C, and then continuously maintain the temperature.

[0100] During the second period, the computing device (100) can cross-analyze in real time the saccharification rate at which rice flour (starch) is converted into sugar and the fermentation rate at which the generated sugar is converted into alcohol based on data collected from the sugar detection sensor (24) and the alcohol content detection sensor (23).

[0101] For example, if the computing device (100) determines that the alcohol production rate is excessively fast, exceeding a preset normal range along with a rapid decrease in residual sugar (Brix) as a result of analysis, it can produce makgeolli with a uniform taste and aroma by performing adaptive process control to suppress the physical metabolic activity of yeast by further finely lowering the second target temperature from 25°C to 24°C to stably delay the fermentation rate, or by reducing or stopping the driving speed of the stirring unit (50).

[0103] In one embodiment, the computing device (100) can be cooled to a third target temperature in a third section, which is a maturation and storage stage (S130).

[0104] Specifically, the computing device (100) can rapidly cool the internal temperature of the brewing tank (10) to a third target temperature that is significantly lower than the second target temperature in order to forcibly suppress the metabolic activity of yeast to prevent further fermentation (over-fermentation and spoilage) and to stabilize the quality of the liquor as the alcohol fermentation reaches the target alcohol level and the decomposition of residual sugars stabilizes. Here, the third stage refers to a low-temperature aging and storage stage in which alcohol and various aromatic components generated during the fermentation process blend together, and carbon dioxide gas generated naturally dissolves into the low-temperature brewing liquid to increase a unique refreshing sensation. The third target temperature may refer to a refrigerated storage temperature that can maintain the freshness and acidity of the makgeolli for a long period by inducing the yeast into a complete dormant state without freezing the brewing liquid.

[0105] To this end, when the computing device (100) determines the target fermentation end time through the internal sensor (20), it performs rapid cooling control by completely blocking the operation of the heating element (42) of the temperature control unit (40) and operating the refrigerant circulation path (41) at maximum load, thereby rapidly stopping the activity of remaining microorganisms and yeast and fixing the intended taste and flavor.

[0106] For example, when the computing device (100) enters the 5th day, which is determined to have completed the second stage of the main fermentation process up to the 4th day and reached a preset target value (e.g., 6% to 8%), it can lower the temperature inside the brewing tank (10) from 25°C to 5.6°C by setting the third target temperature to 5.6°C according to a preset brewing recipe. In this process, the computing device (100) can control the stirring unit (50) to operate continuously or intermittently at a low speed to prevent localized freezing on the inner wall of the brewing tank (10) around the refrigerant circulation channel (41) which acts as a cooling pipe and to maximize cooling efficiency, so that the core temperature of the entire brewing liquid reaches 5.6°C evenly.

[0107] Afterwards, the computing device (100) can continuously perform a storage mode that prevents the deterioration of the quality of the liquor by continuously maintaining the third target temperature of 5.6°C until the user extracts the aged makgeolli through the outlet (70) for tasting or transfers it to an external packaging process.

[0108] Accordingly, the computing device (100) of the present invention deviates from the conventional method of simply mechanically maintaining a designated temperature and monitors the changes in the state of the brewing liquid (sugar content, alcohol content, viscosity, etc.) in real time in accordance with the growth cycle of yeast (initial culture, alcohol fermentation, dormancy and aging), and organically and actively variably controls the temperature control unit (40) and the stirring unit (50) based on this, thereby preventing the risk of over-fermentation and spoilage and enabling consistent brewing of makgeolli.

[0110] Meanwhile, the computing device (100) can perform predictive control in parallel to prevent temperature disturbance of the brewing liquid caused by external environmental factors during the process of performing step-by-step temperature control according to a preset brewing recipe.

[0111] Referring to FIG. 5, when the amount of change in ambient temperature detected by an external sensor exceeds a preset threshold change amount, the computing device (100) can input the rate of change of ambient temperature data into a pre-trained AI model to simulate the expected change trend of the internal temperature of the brewing tank (S140).

[0112] Specifically, the computing device (100) can monitor ambient temperature data in real time around the brewing system installed through the ambient temperature sensing sensor (31) of the external sensor (30), and calculate the rate of increase or decrease of the ambient temperature per unit time to continuously determine whether the amount of change exceeds a preset threshold change amount. Here, the preset threshold change amount refers to an external temperature fluctuation range that can have a significant effect on the temperature of the internal brewing liquid when considering the insulation performance of the brewing tank (10), and the pre-trained AI model may refer to an AI Thermodynamics Model built through a machine learning algorithm based on the ambient temperature accumulated in past brewing processes, the surface temperature of the brewing tank (10), the core temperature of the brewing liquid, and the driving history data of the temperature control unit (40).

[0113] The computing device (100) can input data on the rate of change of the detected ambient temperature, along with the internal temperature of the current brewing tank (10), the volume of the brewing liquid, and the expected fermentation heat generated during the current fermentation stage, as multiple variables into the pre-trained AI model.

[0114] Through this, the computing device (100) can calculate a complex heat transfer mechanism in which external thermal energy penetrates into the brewing tank (10) or internal heat is lost to the outside, thereby producing an expected change trend curve indicating how the internal temperature of the brewing tank (10) will change after a specific time if the current state is maintained.

[0115] For example, while the computing device (100) is performing a second phase of maintaining the internal temperature of the brewing tank (10) at 25°C for alcohol fermentation, if it detects that the outside temperature detected by the outside temperature sensor (31) drops rapidly by more than 5°C within one hour due to factors such as sunset or sudden weather changes, it can recognize this as a state exceeding a preset threshold change amount.

[0116] In this case, the computing device (100) can input environmental change rate data such as "outside temperature drops by 5°C per hour" and fermentation heat data emitted by the current brewing liquid itself into the aforementioned pre-trained AI model.

[0117] The AI ​​model can calculate the thermal resistance due to the double jacket or insulation structure of the brewing tank (10) and provide the simulation result to the computing device (100) that although the internal temperature will be maintained initially despite the drop in the outside temperature, cold air will penetrate in earnest after about 30 minutes, causing the core temperature of the brewing liquid to gradually drop from the second target temperature of 25°C to around 23°C.

[0118] In an additional embodiment, the computing device (100) may further utilize the frequency of carbon dioxide bubble generation and burst acoustic data collected through a piezoelectric microphone or vibration sensor attached around the brewing tank (10) as input variables for the artificial intelligence thermodynamic model. Since the amount and frequency of bubbles generated by the metabolic activity of yeast are indicators representing the fluctuations in biological fermentation heat generated in real time within the brewing liquid, the computing device (100) can further enhance the simulation precision regarding the timing of internal temperature changes and the rate of departure by combining the predicted amount of physical heat loss or gain due to external weather changes with the amount of biological fermentation heat generated through the acoustic sensing.

[0119] In an additional embodiment, the computing device (100) may perform temporary non-linear micro-quenching control in conjunction with the result of synthesizing the simulation and acoustic sensing, when it is predicted that the internal temperature will steeply deviate upward due to a rapid increase in biological fermentation heat inside the brewing liquid or when it is determined that the fermentation peak time is imminent. That is, the computing device (100) may execute an advanced liquor quality prevention logic that preemptively suppresses only the activity of high-grade alcohol-producing enzymes, such as fusel oil that causes hangovers, without affecting the overall fermentation and ethanol yield, by applying a thermal shock that rapidly drops the brewing liquid to a preset temperature (e.g., 15°C) for a short period of time by operating the refrigerant circulation path (41) at maximum output just before the expected rapid temperature rise.

[0120] In one embodiment, when the internal temperature is predicted to deviate from the target temperature range as a result of a simulation, the computing device (100) can apply a preset fermentation stabilization buffer section to control the output of the heating element or refrigerant circulation path in response to the rate of change before the internal temperature deviates from the target temperature range (S150).

[0121] Specifically, the computing device (100) can determine the point in time when the temperature of the brewing liquid exceeds a preset allowable error range based on the predicted change trend curve derived from the preceding simulation, and perform feedforward control by preemptively operating the temperature control unit (40) before the brewing liquid receives an actual temperature shock when that point in time arrives. Here, the preset fermentation stabilization buffer section may refer to a preliminary temperature section or preliminary operating time section set to prevent rapid heating or cooling stress applied inside the brewing tank (10) when the temperature control unit (40) operates in response to a sudden change in the ambient temperature.

[0122] The computing device (100) can block biological disturbances that cause the metabolic activity of yeast to be inhibited or excessively activated by gradually adjusting the voltage applied to the heating element (42) or the refrigerant flow rate of the refrigerant circulation path (41) in proportion to the rate of change of the ambient temperature within this buffer section without waiting for the actual temperature measured by the internal temperature sensing sensor (21) to start to fall or rise.

[0123] For example, if the computing device (100) predicts that the core temperature of the brewing liquid will drop from the second target temperature of 25°C to 23°C after about 30 minutes from the simulation result, it may not wait until the subsequent point in time when the temperature of the brewing liquid actually drops below 25°C, but instead may set a point in time that is a certain time earlier than the predicted temperature deviation point (e.g., 15 minutes before the predicted temperature drop point) as the fermentation stabilization buffer period and activate the heating element (42) in advance. More specifically, for example, the computing device (100) may provide a compensation amount of heat by linearly increasing the heat output of the heating element (42) to correspond to a steep rate of change of the ambient temperature of 5°C per hour, and at the same time, perform complex control by rotating the stirring unit (50) at a low speed so that the heat energy supplied through the inner wall spreads uniformly throughout the entire interior of the brewing tank (10).

[0124] Accordingly, the computing device (100) of the present invention overcomes the response delay phenomenon of simple feedback control that detects and compensates for temperature changes after the fact, and predicts extreme weather changes in the external environment in advance using an artificial intelligence thermodynamic model and performs smooth and preemptive temperature compensation through a preset buffer interval, thereby minimizing stress and fermentation cessation phenomena that live microorganisms such as yeast may experience, so that consistent fermentation yield and liquor quality can be secured regardless of seasonal factors.

[0126] According to various embodiments of the present invention, a computing device (100) can perform brewing control by analyzing the characteristics of auxiliary materials introduced into a brewing tank (10) in real time through artificial intelligence and dynamically optimizing fermentation process parameters according to the results.

[0127] Referring to FIG. 6, the computing device (100) can recognize a fruit additive addition event when the initial sugar content data collected from an internal sensor exceeds the expected range of a preset brewing recipe (S210).

[0128] Specifically, the computing device (100) can collect real-time initial sugar content data by measuring the refractive index or specific gravity of the brewing liquid through a sugar detection sensor (24) included in an internal sensor (20) immediately after the brewing process is initiated or at the initial culture stage when the mixing of raw materials is completed. Here, the initial sugar content data refers to the initial concentration of pure sugar (glucose, etc.) generated by the decomposition of grain starch, such as rice flour, by the amylase enzyme in the coenzyme agent (koji) before the sugar is converted into alcohol by fermentation, and the estimated range of the pre-set brewing recipe may refer to the upper and lower tolerance range of the normal initial sugar content value calculated theoretically and experimentally when only basic raw materials such as rice, koji, and water are mixed at a standard ratio without auxiliary materials.

[0129] When the actual initial sugar content data collected is measured to be high, exceeding the upper limit of the expected range to a significant level, the computing device (100) can identify that a fruit additive addition event has occurred in which a sugary auxiliary ingredient, such as strawberries, Korean melons, bananas, etc., which contain a large amount of fructose or glucose in addition to the basic ingredient, is additionally added into the brewing tank (10).

[0130] For example, if the computing device (100) calculates that the actual sugar content of the brewing liquid measured by the sugar detection sensor (24) at the beginning of the first section of the first day, immediately after rice flour, koji, yeast, water, etc. are mixed according to a preset basic makgeolli brewing recipe, is 18 Brix, while the normal expected initial sugar content range according to the ratio of the basic ingredients is preset in the memory of the computing device (100) as 10 Brix to 12 Brix, this can be determined as a physicochemical sign of increased sugar content due to the addition of an external high-sugar content additive, rather than a temporary phenomenon caused by a simple sensor measurement error or uneven stirring.

[0131] In an additional embodiment, the computing device (100) can further acquire color data (e.g., an increase in redness due to the addition of strawberries or a change in turbidity due to the addition of bananas) or shape data of the mixture inside the brewing tank (10) through a camera module or vision sensor equipped in an internal sensor (20), and by cross-verifying this with the above-mentioned sugar content data, it can determine with greater accuracy and higher reliability which fruit additive has been added.

[0132] Specifically, the computing device (100) can extract pixel data in the RGB (Red, Green, Blue) or HSV (Hue, Saturation, Value) color space region from images of the inner surface of the brewing tank (10) and the brewing liquid captured through a camera module, and analyze the morphological features and dispersion state of the added auxiliary materials through an object recognition algorithm. Here, the object recognition algorithm may refer to a vision AI model based on a Convolutional Neural Network (CNN) that has been machine-learned in advance using a large-scale image dataset regarding the shape of various raw fruits, the texture of crushed or mashed fruits, and the unique color changes that occur when fruit juice is mixed into the brewing liquid.

[0133] The computing device (100) can classify whether the substance causing the increase in sugar content is simply refined sugars or fruit pulp such as strawberries, Korean melons, and bananas accompanied by unique colors and solid content by fusion analysis of the visual feature data derived from the vision sensor and the excess sugar content measured from the sugar detection sensor (24) on a multimodal basis using probabilistic weights.

[0134] For example, the computing device (100) can finally determine that the added additive is strawberry when, as in the previous example, the initial sugar content of the brewing liquid exceeds the reference range of 6 Brix or more, and the pixel data of the image acquired from the camera module is analyzed and it is confirmed that the brightness value of the red channel increases by more than a certain percentage compared to the reference value and that many fine patterns matching strawberry seeds or flesh tissue are distributed on the surface of the brewing liquid.

[0135] In another example, when a computing device (100) detects a high sugar level of 18 Brix but the image analysis results show a dominant distribution density of yellow pixels and signs of a rapid increase in the viscosity and turbidity of the liquid surface are captured in the image, it identifies this as a situation where bananas with different starch and sugar decomposition structures have been added, thereby overcoming the malfunction or limitations of a single sensor and accurately identifying the nature of the additive.

[0137] In one embodiment, the computing device (100) can predict the rate of sugar decomposition and alcohol production by the fruit additive through an AI model (S220).

[0138] Specifically, the computing device (100) can apply the types of fruit additives and excess sugar data identified in the preceding step as input variables to a pre-trained AI fermentation kinetics model to calculate the time-series trends of saccharification and alcohol production that will occur in the future fermentation process.

[0139] Here, the AI ​​fermentation kinetics model may refer to a machine-learned prediction algorithm based on weight data assigned to the fermentation rate curve, such as the effect of various auxiliary materials (fruit pulp, fruit juice, etc.) on the metabolic activity of yeast, namely the ratio of monosaccharides to polysaccharides unique to the auxiliary materials, acidity (pH), and mineral content. The computing device (100) can pre-simulate, through the AI ​​fermentation kinetics model, the rate at which yeast breaks down the sugars of a specific fruit (sugar decomposition rate) and the rate at which ethanol is produced in proportion to this (alcohol production rate) in the form of a trajectory graph over time.

[0140] For example, if the identified additive is identified as 'strawberry', the computing device (100) can reflect in the AI ​​fermentation kinetics model the characteristic that monosaccharides such as fructose and glucose, which are abundant in strawberries, are broken down by yeast much faster and more directly than the coarse starch particles of rice.

[0141] Accordingly, the computing device (100) can produce prediction data that, although it was set to take 72 hours to reach a target alcohol content of 6% in the second section at 25°C in the basic recipe, the sugar decomposition rate is accelerated by the addition of strawberries, so the target alcohol content is reached in 48 hours while maintaining the same temperature, and at the same time, the excess fermentation heat will increase rapidly.

[0143] In one embodiment, if the predicted alcohol production rate is faster than the preset alcohol production rate, the computing device (100) may shorten the maintenance time of the second section or lower the second target temperature (S230).

[0144] Specifically, the computing device (100) can perform adaptive recipe control by dynamically changing process parameters of the second section, which is the main fermentation stage, in order to prevent excessive acidity caused by over-fermentation and volatilization of fruit-specific aroma components, if it is determined that the predicted alcohol production rate derived through the simulation will exceed the preset reference alcohol production rate (normal fermentation trajectory on the basic recipe) and draw an excessively steep rising curve.

[0145] Here, adaptive recipe control refers to a control method that actively optimizes the fermentation environment in response to externally introduced variables (additives). That is, the computing device (100) can reconfigure the control logic to increase the cooling intensity of the refrigerant circulation path (41) by adjusting the second target temperature downward by a certain amount from the existing set value to physically slow down the growth and metabolic rate of the yeast, or to advance the entry point to the third section (aging and rapid cooling stage) by shortening the total maintenance time of the second section by the time predicted to reach the target alcohol content early.

[0146] For example, if the computing device (100) predicts in the preceding simulation that the alcohol production rate will be 1.5 times faster than the basic recipe due to the strawberry additive, it can control the temperature control unit (40) to lower the target temperature of the second section by 2°C from the existing 25°C to 23°C, thereby cooling the heat of the yeast and relaxing the fermentation rate to approach the normal trajectory.

[0147] Alternatively, the computing device (100) can maintain the temperature at around 25°C, reset the internal timer to shorten the fermentation maintenance time of the second phase, originally scheduled for 3 days (72 hours), to 2 days (48 hours), and immediately switch to a rapid cooling process of 5.6°C, the third phase, immediately after 48 hours have elapsed. Through this, the computing device (100) can perform advanced brewing control to prevent sugar depletion and yeast stress phenomena that commonly occur when fruit is added, and to fully preserve the optimized sweetness and fruit aroma of the fruit makgeolli.

[0149] In one embodiment, the computing device (100) can increase the rotation speed of the stirring unit if the predicted sugar decomposition rate is slower than the preset sugar decomposition rate (S240).

[0150] Specifically, the computing device (100) can perform adaptive recipe control to vary the operating conditions of the stirring unit (50) to physically promote the metabolic activity of yeast when it is determined that the overall fermentation process will be delayed because the predicted sugar decomposition rate derived through the simulation does not reach the preset reference sugar decomposition rate.

[0151] Here, a slow predicted sugar decomposition rate may mean that the rate at which sugar is extracted is reduced because the texture of the added fruit additive is hard or the particles are large, or that the fermentation efficiency is reduced because the viscosity of the entire brewing liquid increases rapidly, thereby reducing the contact area between the yeast and the sugar. To solve this, the computing device (100) can increase the rotational speed (RPM) of the stirring unit (50) above a reference value or shorten the stirring cycle, thereby continuously floating high-density solid matter in the form of fruit pulp or puree that may settle at the bottom of the brewing tank (10), and increase the mass transfer coefficient inside the brewing liquid to create a physical environment in which the yeast can smoothly consume the nutrient source (sugar).

[0152] For example, if the AI ​​model predicts that the saccharification and alcohol production rate will be delayed by more than 20% compared to the normal trajectory of the basic recipe due to the addition of a high-viscosity banana or a Korean melon raw material with hard flesh, the computing device (100) can generate a control signal to the rotation motor (51) of the temperature control unit (40) and the stirring unit (50) to increase the basic stirring speed of 10 RPM in the second section, which is the main fermentation section, to 30 RPM. When the screw of the stirring unit (50) is driven at the increased rotation speed, the yeasts that were stagnant around the sticky flesh due to viscous resistance are evenly redistributed throughout the brewing liquid by forced convection, and the degraded fermentation reaction is reactivated so that the target alcohol content and residual sugar content can be stably reached within the planned fermentation maintenance time of the second section.

[0153] Accordingly, the computing device (100) of the present invention, in accordance with the modern makgeolli brewing trend in which various types of auxiliary materials are added, identifies and predicts the type of additive and the resulting changes in biochemical fermentation kinetics through artificial intelligence-based vision sensing and data analysis, thereby performing three-dimensional and flexible process compensation control, such as lowering the temperature or shortening the time when fermentation is fast and increasing the stirring speed when fermentation is slow, so that fruit makgeolli with the best quality and flavor intended by the brewer can be uniformly produced regardless of which additive is added.

[0155] According to a further embodiment of the present invention, as the internal sensor (20) includes a camera module or a vision sensor, the system of the present invention may further include a lens cleaning nozzle for removing contaminants or condensation attached to the lens surface of the camera module or vision sensor. Here, the lens cleaning nozzle may be configured to spray cleaning water (e.g., distilled water, purified water, or sterilized water).

[0156] In one embodiment, when the sharpness or contrast of image data acquired from a camera module is reduced to below a preset threshold or when condensation on the lens surface is detected, the computing device (100) can secure the field of view of the lens by controlling the lens cleaning nozzle to spray cleaning water at a preset pressure and spraying time.

[0157] Specifically, the computing device (100) can quantify the degree of blurriness of an image by applying an edge detection algorithm or a blur analysis index to image data acquired periodically. Inside the brewing tank (10), condensation frequently occurs as water vapor condenses on the lens surface due to the temperature difference between the fermentation heat generated during the fermentation process and the upper space, or brewing liquid or lees particles scattered by the rotation of the stirring unit (50) may adhere to the lens, causing a physical obstruction of view.

[0158] When the computing device (100) determines that the calculated blur level index deviates from the normal range and that accurate visual recognition of internal additives and fermentation status is impossible, it can control the solenoid valve or small pump connected to the lens cleaning nozzle to spray cleaning water onto the lens surface in the form of high-pressure fine particles.

[0159] For example, the computing device (100) may determine that a lens cleaning event has occurred if it calculates that the contrast of the outline of a pixel in the image has decreased by more than a certain ratio compared to normal during the process of capturing the amount of bubbles on the surface of the brewing liquid, or if it detects through an internal humidity sensor (22) that the relative humidity at the top of the tank has exceeded 95% and the probability of condensation occurring has become extremely high.

[0160] Accordingly, the computing device (100) can open the lens cleaning nozzle in the form of two pulses for 2 seconds and spray 10 ml of sterile distilled water, thereby physically striking and removing foreign substances and water film from the lens surface and immediately restoring a clear monitoring field of view.

[0161] Meanwhile, the computing device (100) can calculate the dilution ratio of the brewing liquid by reflecting the increased moisture content according to the amount of washing water sprayed by the lens washing nozzle, and perform precise concentration control to extend the fermentation time or correct the predicted final alcohol content data to maintain the target alcohol content.

[0162] Specifically, the computing device (100) can count the cumulative flow rate data of the cleaning water sprayed whenever the lens cleaning nozzle is operated and store it in memory. Since water is the solvent that forms the basis of fermentation in the brewing process, even a small amount of water introduced for cleaning purposes can act as a disturbance factor that increases the total volume of the brewing liquid and, consequently, dilutes the concentration of ethanol produced by the breakdown of sugars by yeast if it accumulates continuously.

[0163] Accordingly, the computing device (100) calculates a real-time dilution ratio by summing the initial volume data of the total brewing liquid introduced into the initial brewing tank (10) and the additional volume data of the accumulated washing water, and inputs this as an updated parameter to the previously described pre-trained AI fermentation kinetics model, thereby performing compensation control to precisely inversely calculate the deviation of the final alcohol content expected to decrease due to the increase in water content.

[0164] For example, the computing device (100) can calculate that a total of 10 lens washes are performed during the second phase of the main fermentation for 4 days in a brewing tank (10) containing a total of 10 L of brewing liquid, and that 100 ml of distilled water is additionally introduced. If the computing device (100) simulates that the original target alcohol content of 6.0% will drop to 5.94% due to a 1% increase in volume, it can reset the timer to extend the 25°C fermentation maintenance time of the second phase by 2 hours compared to the existing set time to produce the missing 0.06% alcohol, thereby adjusting the alcohol content of the final produced makgeolli to 6.0%.

[0165] Accordingly, the computing device (100) of the present invention can ensure continuous and stable vision data collection by actively solving the problem of contamination and condensation of the optical sensor, which inevitably occurs due to the characteristics of a closed anaerobic fermentation environment, through a self-cleaning mechanism, and at the same time, dynamically supplement the fermentation recipe by replacing even the chemical variable of moisture inflow accompanying the physical cleaning process with a control factor of the AI ​​model, thereby enabling the initial set target liquor quality and alcohol content to be achieved without error.

[0167] According to an additional embodiment of the present invention, a computing device (100) can perform acoustic sensing-based fermentation monitoring by analyzing the sound of bubble generation inside the brewing liquid in a second section where alcohol production is active, thereby non-contactually identifying the progress of fermentation and the type of additive.

[0168] Specifically, the computing device (100) can collect signals of the frequency of generation, size distribution, and popping acoustic pattern of carbon dioxide bubbles generated by yeast metabolism inside the brewing liquid during fermentation in real time through a piezoelectric microphone or vibration sensor attached around the outer wall or inner airlock of the brewing tank (10).

[0169] And, the computing device (100) converts the collected analog acoustic signal into the frequency domain through a fast Fourier transform algorithm to filter out noise, and inputs it into a pre-trained AI acoustic analysis model to extract the unique burst frequency band of the bubbles, thereby enabling precise estimation of the current viscosity change and fermentation progress of the brewing liquid without direct contact with the sensor. Here, the pre-trained AI acoustic analysis model may refer to a deep learning-based time-series acoustic recognition model (e.g., RNN or 1D-CNN algorithm) that is machine-learned by mutually mapping acoustic spectrum data generated when carbon dioxide bubbles rise and burst in a fermentation liquid environment with various viscosities with actual measurement data regarding the actual residual sugar content, alcohol content, and type of additive at that time.

[0170] For example, the computing device (100) can determine the completion time of saccharification and fermentation without the help of a chemical sugar content sensor by detecting a characteristic acoustic spectrum change in which, as the second section progresses, sugar is converted into ethanol and the overall viscosity of the brewing liquid gradually decreases, the speed at which bubbles rise to the liquid surface increases and the proportion of the high-frequency band of bubble bursting sounds increases.

[0171] As another example, the computing device (100) can utilize the above-mentioned burst acoustic pattern as an auxiliary identification means to identify the type of fruit when a fruit additive is added. That is, in a brewing liquid with high viscosity banana puree added, bubbles rise slowly through sticky resistance and produce a large, dull low-frequency burst sound, and in a brewing liquid with low viscosity and high moisture content strawberry juice added, many small, light high-frequency burst sounds are produced. The AI ​​model can classify these acoustic characteristics to compensate for the visual limitations of the vision sensor (e.g., increased turbidity due to the addition of dark-colored ingredients).

[0172] Additionally, the computing device (100) can perform multi-modal based hybrid monitoring by cross-verifying the fermentation progress and viscosity data estimated through the acoustic sensing with chemical measurements obtained from the sugar detection sensor (24) and alcohol content detection sensor (23) of the internal sensor (20).

[0173] For example, even if a sensor measurement error occurs where solid matter (lees) inside the brewing liquid during fermentation adheres to the surface of the sugar detection sensor (24) and the residual sugar content remains abnormally high, if the computing device (100) identifies that the high-frequency proportion of bubble burst sounds extracted through an acoustic analysis model already meets the standard value corresponding to the fermentation completion stage, it may temporarily ignore the measurement value of the chemical sensor and prioritize the acoustic data to instruct entry into the third stage (aging and cooling), or induce the sensor to be cleaned by operating the lens cleaning nozzle.

[0174] Accordingly, the computing device (100) of the present invention overcomes physical contamination and measurement limitations that conventional optical or contact chemical sensors may have due to solids or dark colors within the brewing liquid through non-contact acoustic analysis technology, and can precisely determine the characteristics of additives and the progress of fermentation without error through cross-verification between heterogeneous sensors.

[0176] According to an additional embodiment of the present invention, a computing device (100) can perform temporary non-linear micro-quenching (thermal shock) control by predicting a specific critical point at which the fermentation heat increases rapidly in order to suppress the production of higher alcohols that cause hangovers.

[0177] Here, microquenching may refer to a short-cycle rapid cooling process that can selectively and temporarily paralyze only the activity of enzymes that produce fusel oil, a representative byproduct that causes headaches and hangovers when consumed, without completely killing the yeast.

[0178] Specifically, the computing device (100) can identify the fermentation peak time, when the metabolic activity of the yeast reaches its peak, in real time based on the temperature rise rate and sugar content decrease rate data collected through the internal sensor (20). Then, when the peak time arrives, the computing device (100) can apply a thermal shock by operating the refrigerant circulation path (41) of the temperature control unit (40) at maximum output to rapidly lower the internal temperature of the brewing tank (10) for a preset short period of time.

[0179] More specifically, the computing device (100) can predict the timing (i.e., peak time) when amino acid metabolism, which is a precursor of fusel oil, occurs explosively by utilizing a pre-trained AI fermentation kinetics model to derive an inflection point from the current fermentation rate curve of the brewing liquid. Then, the computing device (100) can prevent the accumulation of by-products that cause unpleasant flavors and hangovers without adversely affecting the normal ethanol yield or the entire fermentation period by applying a non-linear temperature profile that momentarily lowers the target temperature to match the peak time and then returns it to the original fermentation set temperature.

[0180] For example, the computing device (100) can determine the point in time when the alcohol production rate exceeds 0.5% per hour and the fermentation heat increases rapidly as a risk zone for fusel oil production (peak time) by analyzing the data from the internal temperature sensor (21) and the alcohol content sensor (23) while the second phase (e.g., 25°C maintenance phase) in which alcohol fermentation occurs vigorously. In this case, the computing device (100) can execute a control command to fully open the valve of the refrigerant circulation path (41) to rapidly quench the internal temperature of the brewing tank (10) from 25°C to 15°C for 15 minutes, and then, when the thermal shock period ends, drive the heating element (42) again to quickly return to the original target temperature of 25°C.

[0181] Additionally, during the 15-minute micro-quenching section in which the temperature drops rapidly to 15°C, the computing device (100) can temporarily increase the rotation speed of the stirring unit (50) from the existing 10 RPM to 40 RPM to induce the cold air delivered from the refrigerant circulation path (41) to penetrate uniformly and rapidly to the core of the entire brewing liquid, thereby enhancing the impact effect of the thermal shock. Furthermore, during the recovery section in which the temperature rises again to 25°C, the computing device (100) can induce a degassing effect by linking the phenomenon of reduced gas solubility due to the temperature rise with the physical flow of the stirring unit (50). Through this, the computing device (100) can improve the quality of the final produced makgeolli by controlling the internal off-flavor components to volatilize together through the airlock of the discharge port (70) when the micro-fermentation gas dissolved inside the brewing liquid is discharged to the outside.

[0182] Accordingly, the computing device (100) of the present invention can artificially tune the metabolic pathway of living microorganisms through instantaneous thermodynamic quenching control of the hardware, thereby producing makgeolli with reduced hangover-causing substances without separate chemical additives or complex physical purification filter processes.

[0184] Meanwhile, embodiments according to the present disclosure may be implemented in the form of a computer program that can be executed through various components on a computer, and such a computer program may be recorded on a computer-readable medium. In this case, the medium may include, but is not limited to, magnetic media such as hard disks, floppy disks, and magnetic tapes, optical recording media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, and flash memory.

[0185] Meanwhile, the above computer program may be one specifically designed and configured for the present disclosure or one known and available to those skilled in the art of computer software. Examples of computer programs may include machine code, such as that produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc.

[0186] According to one embodiment, the method according to various embodiments of the present disclosure may be provided by being included in a computer program product. The computer program product may be traded between a seller and a buyer as a product. The computer program product may be distributed in the form of a device-readable storage medium (e.g., compact disc read-only memory (CD-ROM)), or distributed online (e.g., download or upload) through an application store (e.g., Play Store™) or directly between two user devices. In the case of online distribution, at least a portion of the computer program product may be temporarily stored or temporarily created in a device-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or a relay server.

[0187] Unless explicitly stated otherwise, the steps constituting the method according to the present disclosure may be performed in a suitable order. The present disclosure is not necessarily limited by the order in which the steps are described. The use of any examples or exemplary terms (e.g., etc.) in the present disclosure is merely for the purpose of describing the present disclosure in detail and, unless limited by the claims, the scope of the present disclosure is not limited by such examples or exemplary terms. Furthermore, those skilled in the art will understand that various modifications, combinations, and changes may be made according to design conditions and factors within the scope of the claims or equivalents to which they are added.

[0188] Accordingly, the scope of the present disclosure should not be limited to the embodiments described above, and all scopes equivalent to or equivalently modified from the claims set forth below, as well as the claims set forth below, shall be considered to fall within the scope of the scope of the present disclosure.

Claims

Claim 1 In an IoT-based AI makgeolli brewing system, a brewing tank that accommodates makgeolli ingredients and performs fermentation; an internal sensor installed inside the brewing tank to measure state data of the brewing liquid; an external sensor installed outside the brewing tank to measure external environment data; a temperature control unit installed to surround the brewing tank to regulate the internal temperature of the brewing tank; and a stirring unit installed inside the brewing tank to stir the brewing liquid. A computing device that receives data collected from the internal sensor and the external sensor, analyzes the received data using an AI model, and adjusts process conditions for each stage of the brewing process based on the analysis results; wherein the computing device is characterized by adjusting process conditions including at least one of a temperature set value for a fermentation stage, a fermentation maintenance time, and a cooling time by reflecting data collected in real time from the internal sensor and the external sensor based on a preset brewing recipe; wherein the preset brewing recipe includes a first section in which the internal temperature of the brewing tank is heated and maintained to a first target temperature for yeast activation, a second section in which the internal temperature is lowered and maintained to a second target temperature lower than the first target temperature for alcohol production, and a third section in which the internal temperature is rapidly cooled and maintained to a third target temperature lower than the second target temperature for fermentation cessation and maturation; and wherein, according to the preset brewing recipe, the computing device maintains the internal temperature of the brewing tank at the first target temperature in the first section, which is the yeast activation stage after koji and yeast addition, and lowers and maintains the internal temperature to the second target temperature in the second section, which is the alcohol production stage. While maintaining, in the third section, which is the aging and storage stage, the temperature is cooled to a third target temperature; however, if the change in ambient temperature detected by the external sensor exceeds a preset threshold change amount, the change rate data of the ambient temperature is input into a pre-trained AI model to simulate the expected change trend of the internal temperature of the brewing tank.An IoT-based AI Makgeolli brewing system characterized by: applying a preset fermentation stabilization buffer section when the simulation result predicts that the internal temperature will deviate from the target temperature range to control the output of the heating element or refrigerant circulation path in response to the rate of change before the internal temperature deviates from the target temperature range; recognizing a fruit additive addition event when the initial sugar content data collected from the internal sensor exceeds the expected range of the preset brewing recipe; predicting the sugar decomposition rate and alcohol production rate by the fruit additive through the AI ​​model; shortening the maintenance time of the second section or lowering the second target temperature if the predicted alcohol production rate is faster than the preset alcohol production rate; and increasing the rotation speed of the stirring unit if the predicted sugar decomposition rate is slower than the preset sugar decomposition rate. Claim 2 An IoT-based AI Makgeolli brewing system according to claim 1, wherein the internal sensor comprises: an internal temperature sensing sensor for detecting the temperature inside the brewing tank; a humidity sensing sensor for detecting the humidity inside the brewing tank; an alcohol content sensing sensor for detecting the alcohol content of the brewing liquid; a sugar content sensing sensor for detecting the sugar content of the brewing liquid; and a liquor quality sensing sensor for detecting liquor quality data of the brewing liquid; and the external sensor comprises: an ambient temperature sensing sensor for detecting the ambient temperature; and an ambient humidity sensing sensor for detecting the ambient humidity. Claim 3 An IoT-based AI Makgeolli brewing system according to claim 1, wherein the temperature control unit comprises: a refrigerant circulation path installed to surround the outer surface of the brewing tank and through which a refrigerant circulates to perform heat exchange with the brewing tank; and a heating element disposed along the outer surface of the brewing tank intersecting or adjacent to the refrigerant circulation path and generating heat by an electrical signal to heat the brewing tank. Claim 4 delete Claim 5 delete

Images