Fermentation state management system, fermentation state management method and program

The fermentation state management system addresses the challenge of managing aroma components in fermented products by using multiple gas sensor channels to predict and visualize the fermentation state, enhancing quality control and adaptability.

JP2024157278AInactive Publication Date: 2024-11-07ASAHI KASEI KOGYO KABUSHIKI KAISHA
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
JP2023071543
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-04-25
Publication Date
2024-11-07
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing fermentation state management systems fail to effectively visualize and control the aroma components generated during the fermentation process, making it difficult to manage yeast function and achieve desired final product qualities in sake and other fermented products, as they are largely reliant on expert intuition and limited to managing alcohol and extract content.

Method used

A fermentation state management system utilizing a gas signal acquisition unit with multiple gas sensor channels sensitive to various gases, including aroma components, to predict and visualize the fermentation state by analyzing gas concentrations and environmental conditions, enabling real-time monitoring and control.

Benefits of technology

The system facilitates the design, control, and management of fermented products by accurately predicting and visualizing the fermentation state, allowing for improved quality control and adaptability to different raw materials and yeast strains.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024157278000001_ABST
    Figure 2024157278000001_ABST
Patent Text Reader

Abstract

To provide a fermentation state management system, fermentation state management method, and program that can visualize the state of a substance during fermentation (the activity of yeast) from multiple gases including aroma components generated during the fermentation process of fermented products, such as sake and food, produced by fermentation, and facilitate the design, control, and management of fermented products.SOLUTION: Provided is a fermentation state management system for managing the fermentation state of a fermentation substance, the fermentation state management system comprising: a gas signal acquisition unit having a plurality of gas sensor channels, at least one of which has sensitivity to the concentrations of two or more types of gas among a plurality of target gases; a measurement signal group including measurement signals output from the gas sensor channels in response to gas generated from the fermentation substance; a sensitivity information group including sensitivity information of the gas sensor channels for each of the plurality of target gases; and a prediction unit for outputting a predicted gas concentration indicating the concentration of at least one of the plurality of target gases, using the intensities of each measurement signal included in the measurement signal group, the intensity ratio between each measurement signal, and the sensitivity information group.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a fermentation state management system, a fermentation state management method, and a program for managing the fermentation state of sake, food, and the like produced by fermentation. [Background technology]

[0002] Conventionally, status management in the fermentation process of sake and food products (hereinafter sometimes referred to as "fermented products") has been largely dependent on the intuition and experience of experts, making it difficult to quantify and visualize the fermentation status. As a result, a large number of resources (man-hours, raw material costs, analysis costs, etc.) were required to properly manage the quality of fermented products. For example, in the fermentation (brewing) of sake, status management must be performed on holidays and late at night and early in the morning, and there was a demand for improvements in the working environment for workers. In addition, there was also the issue that it was difficult to transfer skills because it was difficult to quantify and visualize status management in the fermentation process for fermented products. To solve these problems, fermentation status management systems that utilize IoT (Internet of Things) technology have been developed in recent years.

[0003] Patent Document 1 proposes a system for calculating quality control indicators in the sake brewing process. By directly or indirectly measuring carbon dioxide and Brix values ​​using a sensor or the like, the system calculates the alcohol content, sake meter value, and extract content during fermentation of the mash, allowing users to check these indicators in real time. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent Publication No. 2023-2080 Summary of the Invention [Problem to be solved by the invention]

[0005] In the method of Patent Document 1, the objects to be managed are limited to the management of the alcohol content, sake meter value, extract content, etc. of sake. On the other hand, how to manage the activity of yeast is important for accurate management of the fermentation state, and for that, the aroma components generated by the activity of yeast must be managed. For example, in sake brewing, the aroma components are mainly managed by the nose of the master brewer. In addition, it is necessary to design and control the aroma for the state of sake that the producer ultimately wants to make, but in sake brewing, a huge variety of aroma components are generated, and even an expert (master brewer) finds it difficult to manage them. For this reason, there is a demand for the development of a technology that can visualize the fermentation state according to the aroma components generated according to the fermentation state of the fermented product so that the activity of yeast can be easily managed. In addition, even if the same sake is produced, the aroma components generated depending on the raw materials and yeast used are different, so for the purpose of managing the fermentation state of fermented products such as sake and food, a system that can only detect a specific gas is insufficient as a general-purpose system, even if it is compatible with multiple gases. As a general-purpose fermentation state management system that can handle various fermentation products, one that can detect gases to be managed even if they change is preferred.

[0006] In order to solve the above-mentioned problems, the present invention aims to provide a fermentation state management system, a fermentation state management method, and a program that can visualize the state of a substance during fermentation (the activity of yeast) from multiple gases containing aroma components that are generated during the fermentation process of fermented products such as sake and foods produced by fermentation, and facilitate the design, control, and management of fermented products. [Means for solving the problem]

[0007] <1> A fermentation state management system for managing the fermentation state of a fermentation substance, a gas signal acquiring unit having a plurality of gas sensor channels, at least one of the plurality of gas sensor channels having sensitivity to concentrations of two or more types of gas among a plurality of target gases; A group of measurement signals including measurement signals output from the gas sensor channels in response to gas generated from the fermentation substance; a sensitivity information group including sensitivity information of the gas sensor channel for each of the plurality of target gases; a prediction unit that outputs a predicted gas concentration indicating a concentration of at least one of the plurality of target gases using the intensity of each measurement signal included in the group of measurement signals and the intensity ratio between each measurement signal and the group of sensitivity information; A fermentation status control system having: <2> The prediction unit outputs fermentation state information indicating a fermentation state of the fermentation substance based on the predicted gas concentration. The above <1> The fermentation state management system according to claim 1. <3> At least one of the plurality of gas sensor channels is sensitive to concentrations of all of the target gases. The above <1> or <2> The fermentation state management system according to claim 1. <4> All of the plurality of gas sensor channels are sensitive to concentrations of two or more types of gases among the target gases. The above <1> or <2> The fermentation state management system according to claim 1. <5> At least one of the plurality of target gases is responsive to two or more of the plurality of gas sensor channels. The above <1> or <2> The fermentation state management system according to claim 1. <6> At least one of the plurality of target gases is responsive to all of the plurality of gas sensor channels. The above <1> or <2> The fermentation state management system according to claim 1. <7> all of the plurality of target gases are responsive to two or more of the plurality of gas sensor channels; The above <1> or <2> The fermentation state management system according to claim 1. <8> an environmental temperature acquisition unit that acquires a temperature of an external environment in which the gas signal acquisition unit exists or a temperature of an internal environment of the gas signal acquisition unit; The prediction unit corrects the group of measurement signals using a temperature of the external environment or a temperature of the internal environment. The above <1> ~ <7> The fermentation state control system according to any one of the preceding claims. <9> A fermentation substance temperature acquisition unit that acquires fermentation substance temperature information that can be converted into the temperature of the fermentation substance, The prediction unit predicts a concentration of a substance having the same component as the target gas in the fermentation substance based on the predicted gas concentration of a target gas which is at least one of the plurality of target gases and the fermentation substance temperature information. The above <1> ~ <8> The fermentation state management system according to claim 1. <10> The prediction unit uses at least one of the plurality of measurement signals as a reference signal, and corrects at least one of the other measurement signals, the predicted gas concentration of at least one of the plurality of target gases, and the concentration of the same component substance as the target gas in the fermentation substance using the reference signal. The above <9> The fermentation state management system according to claim 1. <11> The reference signal is a measurement signal output from a gas sensor channel having sensitivity to a concentration of only one type of gas among the plurality of target gases. The above <10> The fermentation state management system according to claim 1. <12> The prediction unit calibrates at least one of the measurement signals, the predicted gas concentration of at least one of the plurality of target gases, and the concentration in the fermentation material of a substance having the same component as a target gas that is at least one of the plurality of target gases, using other information not derived from the measurement signal from the gas sensor channel. The above <1> ~ <11> The fermentation state control system according to any one of the preceding claims. <13> The gas signal acquisition unit acquires the measurement signals in a non-contact state with the fermentation substance. The above <1> ~ <12> The fermentation state control system according to any one of the preceding claims. <14> The fermentation substance temperature acquisition unit acquires the fermentation substance temperature information in a non-contact state with the fermentation substance. The above <9> ~ <11> The fermentation state control system according to any one of the preceding claims. <15> The prediction unit notifies information on a deviation of at least one of the measured value and the predicted value from a preset range when the deviation is detected. The above <1> ~ <14> The fermentation state control system according to any one of the preceding claims. <16> the gas signal acquisition unit further acquires time information associated with a time at which the group of measurement signals was acquired; Notifying a user of information on a recommended action based on the predicted gas concentration of at least one of the plurality of target gases or the concentration in the fermentation material of a substance having the same component as the target gas that is at least one of the plurality of target gases, and the time information; The above <1> ~ <15> The fermentation state control system according to any one of the preceding claims. <17> At least one of the plurality of target gases is an aroma component generated by the action of yeast present in the fermented substance. The above <1> ~ <16> The fermentation state control system according to any one of the preceding claims. <18> The fermented material is moromi, The target gas includes at least one of water, ethanol, ethyl caproate, isoamyl acetate, isobutyl acetate, ethyl acetate, ethyl caprylate, ethyl caprate, ethyl laurate, phenylethyl acetate, ethyl lactate, acetaldehyde, isovaleraldehyde, diacetyl, isoamyl alcohol, phenethyl alcohol, 4-vinyl guaiacol, and 4-mercapto-4-methyl-2 pentanone; The above <1> ~ <17> The fermentation state control system according to any one of the preceding claims. <19> A fermentation state management method for managing the fermentation state of a fermentation substance, comprising: obtaining a set of sensitivity information including sensitivity information of the gas sensor channel for each of a plurality of target gases; A step of acquiring a group of measurement signals from a plurality of gas sensor channels, at least one of which has sensitivity to concentrations of two or more types of gas among the plurality of target gases, the group of measurement signals including measurement signals output from the gas sensor channels in response to the gas generated from the fermentation substance; predicting a predicted gas concentration indicating a concentration of at least one of the target gases using the intensity of each measurement signal included in the measurement signal group, the intensity ratio between each measurement signal, and the sensitivity information group, and outputting the predicted gas concentration; A fermentation state control method comprising the steps of: <20> The device, a sensitivity information group including sensitivity information of the gas sensor channel for each of a plurality of target gases; a plurality of gas sensor channels, at least one of which has sensitivity to concentrations of two or more types of gas among the plurality of target gases, and a measurement signal group including measurement signals acquired from the plurality of gas sensor channels and output from the gas sensor channels in response to a gas generated from a fermentation substance; Enter A program that executes a process of predicting a predicted gas concentration indicating the concentration of at least one of the multiple target gases using the intensity of each measurement signal and the intensity ratio between each measurement signal included in the measurement signal group, and the sensitivity information group, and outputting the predicted gas concentration. Effect of the Invention

[0008] According to the present invention, it is possible to provide a fermentation state management system that can visualize the state of a substance during fermentation (the activity of yeast) from a plurality of gases containing aroma components that are generated during the fermentation process of fermented products such as sake and foods produced by fermentation, and can facilitate the design, control, and management of fermented products. [Brief description of the drawings]

[0009] [Figure 1] 1 is a block diagram showing an overview of a fermentation state management system according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a schematic diagram for explaining the hardware configuration of each device constituting the fermentation state management system. [Diagram 3] FIG. 2 is a diagram showing an example of a flow of the fermentation state management system of the present embodiment. [Figure 4]FIG. 1 is a diagram showing an overview of a fermentation state management system of this embodiment when applied to sake brewing. [Diagram 5] FIG. 2 is a diagram showing an example of a sensitivity information group used in the fermentation state management system of the present embodiment. [Figure 6] 1A and 1B are diagrams showing a measurement signal group y, an environmental temperature corrected measurement signal group yc, and a reference signal corrected measurement signal group z, all of which are acquired by the fermentation state management system of this embodiment. [Figure 7] FIG. 2 is a diagram showing predicted gas concentrations of each target gas (analyte gas) predicted by the fermentation state management system of the present embodiment. [Figure 8] FIG. 2 is a diagram showing changes over time in predicted gas concentration of each target gas (analyte gas) predicted by the fermentation state management system of the present embodiment. [Figure 9] FIG. 2 is a diagram showing the change over time in the alcohol content in mash predicted by the fermentation state management system of this embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] The fermentation state management system and each of the components constituting the fermentation state management system will be described below with specific examples. However, these embodiments do not limit the present invention. Furthermore, it is not meant that all of the combinations of features described in the embodiments are essential to the solution of the present invention.

[0011] <Fermentation status management system> An example of the configuration of the fermentation state management system of this embodiment (hereinafter, sometimes referred to as "the system of this embodiment") will be described with reference to FIG. 1. FIG. 1 is a block diagram showing an outline of the fermentation state management system of this embodiment. As shown in FIG. 1, the fermentation state management system 100 is a management system that manages the fermentation state of a substance (fermented substance) undergoing fermentation, and includes at least a gas signal acquisition unit 10 and a prediction unit 20, and is capable of outputting a predicted gas concentration OP1 indicating the concentration of at least one gas among a plurality of target gases from the prediction unit 20 using the intensity of each measurement signal included in the measurement signal group (y)IN2 and the intensity ratio between the measurement signals, and the sensitivity information group IN1.

[0012] The term "target gas" refers to a gas that is the object of management in the fermentation state management system of this embodiment, and is not particularly limited, but may include, for example, aroma components generated by the action of yeast present in the fermentation substance. In addition, the term "target gas" in this embodiment refers to a gas for which a sensitivity information group is set. Furthermore, among the target gases, a gas whose predicted gas concentration is output by the prediction unit may be referred to as an "analysis gas". In this embodiment, the target gas and the analysis gas can be treated interchangeably in many cases, but to be precise, the target gas has a broader meaning. Furthermore, among the analysis gases, a gas whose concentration in the fermentation substance is predicted may be referred to as a "target gas". "Sensitivity" indicates the output change of the gas sensor channel with respect to the change in concentration of each target gas, and may be linear or nonlinear. "Sensitivity information" means information on the sensitivity of each gas sensor channel with respect to the concentration of the target gas, and is a numerical expression of the magnitude of the sensitivity. The sensitivity information may be, for example, the sensitivity of the gas sensor channel with respect to the concentration of a target gas when the target gas exists alone. Even if a background gas exists together with a target gas, the sensitivity of the gas sensor channel with respect to the concentration of the target gas under the background may be treated as the sensitivity information. That is, in any system, the information is not particularly limited as long as it represents the sensitivity of the gas sensor channel when only the concentration of a specific target gas changes. For example, when the sensitivity is linear, a numerical value indicating the slope of the output of the gas sensor channel with respect to the concentration of the target gas may be used as the sensitivity information. In this embodiment, a collection of the sensitivity information of each gas sensor channel with respect to each of a plurality of target gases is referred to as a "sensitivity information group". The sensitivity information group may be a measured value or a representative value measured in advance, and is stored in a prediction unit or a gas signal acquisition unit described later. The set of sensitivity information may be fine-tuned by, for example, relearning using the measured values ​​acquired by the gas signal acquisition unit at an early stage after the system of this embodiment is put into operation. A "measurement signal" is a signal (unit: mV, for example) output from a gas sensor channel in response to a mixed gas (gas generated from the fermentation substance) containing the target gas present in the fermentation section, and in this embodiment, a collection of the measurement signals is referred to as a "measurement signal group."

[0013] 1, the fermentation state management system 100 can include a fermentation unit 30 and a display unit 40 in addition to a gas signal acquisition unit 10 and a prediction unit 20. Note that the gas signal acquisition unit 10, the prediction unit 20, the fermentation unit 30, and the display unit 40 do not need to be physically connected to each other, and may be connected to each other via a wired or wireless network.

[0014] <Gas signal acquisition unit> The gas signal acquiring section 10 has a plurality of gas sensor channels. In the fermentation state management system 100 shown in FIG. 1, an embodiment having five gas sensor channels GSC1 to GSC5 is shown. Hereinafter, the plurality of gas sensor channels may be collectively referred to as "gas sensor channels (GSC)." The gas sensor channels (GSC) are sensors that output signals related to gas concentrations, and the signals output from each gas sensor channel are referred to as "measurement signals." In FIG. 1, measurement signals 1 to 5 are output from the gas sensor channels GSC1 to GSC5.

[0015] The gas signal acquiring unit 10 acquires a measurement signal group (y)IN2 responsive to a mixed gas composed of a plurality of gases present in the fermentation unit 30 by using these sensors. The "measurement signal group" includes measurement signals output from each gas sensor channel in response to gas generated (volatilized) from the fermentation substance in the fermentation unit 30, and is information summarizing output signals from a plurality of gas sensor channels (GSC1 to GSC5) possessed by the gas signal acquiring unit 10. The measurement signal group (y)IN2 may include signals from other than the gas sensor channels (e.g., signals output from a temperature sensor), and may be configured to include measurement signals from a plurality of types of sensors other than the gas sensor channels.

[0016] In the system of this embodiment, at least one of the gas sensor channels has sensitivity to the concentration of two or more types of gas among the multiple target gases to be managed. Specifically, the gas sensor channels GSC1 to GSC4 have sensitivity to the concentration of two or more types of gas among the multiple target gases. That is, the measurement signals 1 to 4 output from the gas sensor channels GSC1 to GSC4 do not correspond one-to-one to the multiple gases, but one measurement signal is associated with multiple gases and one gas is associated with multiple measurement signals.

[0017] In the fermentation state control system 100, for example, at least one of the multiple gas sensor channels may be sensitive to the concentrations of all the target gases, or all of the multiple gas sensor channels may be sensitive to the concentrations of two or more types of gas among the target gases. Also, for example, the gas signal acquisition unit 10 may use, as the gas sensor channels GSC1 to GSC4, sensors that are sensitive to the concentrations of all the target gases and output measurement signals 1 to 4 in response to all the target gases, respectively, and may use, as the gas sensor channel GSC5, a so-called one-to-one type sensor that responds only to the concentration of a specific gas species for a reference signal, as described later.

[0018] The gas sensor channel is mainly intended to measure gas concentration, but the format of the output signal is not important as long as it outputs a signal based on a voltage, current, resistance, capacitance, temperature, refractive index, fluorescence intensity, weight, frequency, or the like that can be converted into a gas concentration, in addition to a sensor that directly displays the gas concentration. As a gas sensor having multiple gas sensor channels in which one gas sensor channel responds to multiple gases, for example, a sensor that imitates the principle of the olfactory sense of living organisms, such as an MSS (Membrane-type Surface stress Sensor) olfactory sensor, can be cited. The olfactory sense of living organisms is a "many-to-many" mechanism in which one olfactory receptor recognizes various odor molecules (gases), and one odor molecule is recognized by various olfactory receptors, and the MSS olfactory sensor is a sensor that utilizes this "many-to-many" mechanism. This MSS olfactory sensor or the like can be applied to the gas signal acquisition unit 10.

[0019] During the fermentation process of a fermentation substance, many kinds of gases are generated for various purposes, so that it is necessary to manage the concentrations of multiple gases when managing the fermentation process of the fermentation substance. In addition, the target gas to be managed changes depending on the final state of the fermentation substance. As described above, the gas sensor channel (GSC) used in the gas signal acquisition unit 10 is not a sensor specialized for a specific gas (i.e., a sensor that corresponds one-to-one to a specific gas), but a sensor that can respond to multiple types of gases. For example, when the type of target gas to be managed is changed, such as when changing the fermentation substance to which the present system is applied, the gas to be managed can be set as the target gas by having the prediction unit 20, which will be described later, learn the sensitivity information group IN1 of the new gas to be managed before operating the system, and the predicted gas concentration of the gas can be predicted from the measurement signal group of the mixed gas containing the gas obtained by the gas signal acquisition unit 10.

[0020] In order to select the gas to be managed and to change or add the target gas, it is preferable that at least one of the gas sensor channels (GSC) has some sensitivity to many (two or more types) of gases, although the sensitivity may vary from strong to weak, and it is also preferable that all of the gas sensor channels have sensitivity to multiple gases. That is, at least one of the gas sensor channels may be sensitive to the concentrations of all of the target gases, and all of the gas sensor channels may be sensitive to the concentrations of two or more types of gases among the target gases.

[0021] As a combination of the multiple target gases, a combination may be selected in which at least one of the multiple target gases responds to two or more of the multiple gas sensor channels, a combination may be selected in which at least one of the multiple target gases responds to all of the multiple gas sensor channels, or a combination may be selected in which all of the multiple target gases respond to two or more of the multiple gas sensor channels.

[0022] As described above, the gas signal acquiring unit 10 may be provided with a separate "one-to-one" gas sensor channel in which one measurement signal (output signal from one gas sensor channel) corresponds to one gas. The measurement signal output from such a gas sensor channel corresponds one-to-one to one of the target gases, and can be treated as a "reference signal." As described above, the gas signal acquiring unit 10 uses a sensor that outputs a measurement signal in response to the concentration of only one gas among multiple target gases as the gas sensor channel GSC5, and can treat the measurement signal 5 as a reference signal.

[0023] It is preferable that the gas signal acquiring section 10 acquires a plurality of measurement signals (i.e., a group of measurement signals) at the same time in one measurement. The reason is that the predicted gas concentration of the target gas is predicted in the predicting section using the intensity of each measurement signal and the intensity ratio between each measurement signal, and the prediction accuracy decreases if data with a large timing difference is used. In the fermentation state control system 100, a difference of 10 seconds or less can be regarded as the same time. The reason is that fermentation is a task that takes a long time (even days in some cases), and the target to be detected is gas that volatilizes from the fermentation substance, so that changes in state do not occur suddenly. As shown in FIG. 1, in the fermentation state control system 100, the gas signal acquiring section 10 can acquire time information IN3 linked to the time when the group of measurement signals was acquired at the same time as acquiring a plurality of measurement signals. As described below, the prediction unit 20 can notify the user of information on recommended actions based on the predicted gas concentration OP1 of at least one of the multiple target gases, or the concentration in the fermentation substance of a substance having the same component as the target gas, which is at least one of the multiple target gases (concentration in the fermentation substance OP2), together with time information IN3.

[0024] The gas signal acquiring unit 10 may be installed inside the fermentation unit 30, or may be installed outside the fermentation unit 30 with a gas sampling means provided between the gas signal acquiring unit 10 and the fermentation unit 30. From the viewpoint of hygiene, it is preferable that the measurement signals 1 to 5 acquired by the gas signal acquiring unit 10 are acquired in a non-contact state with the fermentation substance in the fermentation unit 30. For this reason, it is preferable that the gas signal acquiring unit 10 and the fermentation substance temperature acquiring unit 32 acquire the measurement signal group (y) IN2 and the fermentation substance temperature information IN6 in a non-contact state with the fermentation substance in the fermentation unit 30.

[0025] 1, the gas signal acquisition unit 10 may further include an environmental temperature acquisition unit 12 that acquires the temperature of the external environment in which the gas signal acquisition unit 10 exists (external environmental temperature information IN5) or the temperature of the internal environment of the gas signal acquisition unit 10 (internal environmental temperature information IN4). The environmental temperature acquisition unit 12 may include, for example, a temperature sensor S1 that acquires the internal environmental temperature information IN4 of the gas signal acquisition unit 10 and a temperature sensor S2 that acquires the external environmental temperature information IN5 of the gas signal acquisition unit 10, and the temperature sensor S1 may be installed inside the gas signal acquisition unit and the temperature sensor S2 may be installed near the outside of the gas signal acquisition unit 10. As will be described later, the prediction unit 20 may correct the measurement signal group (y)IN2 using the external environmental temperature information IN5 or the internal environmental temperature information IN4 to obtain the measurement signal group (yc)OP3.

[0026] <Environmental temperature acquisition section> The environmental temperature acquiring unit 12 acquires the temperature of the external environment in which the gas signal acquiring unit 10 exists, or the temperature of the internal environment of the gas signal acquiring unit 10. The information acquired here can be used for temperature correction of the measurement signal acquired by the gas signal acquiring unit 10 as described above.

[0027] Generally, many sensors have temperature characteristics, and the measurement signal output from the gas sensor channel (GSC) of the gas signal acquiring unit 10 also has temperature dependency. Therefore, by correcting the measurement signal using the temperature of the environment in which the gas signal acquiring unit 10 exists, the prediction accuracy of the system of this embodiment is improved.

[0028] Furthermore, instead of using the environmental temperature (internal environmental temperature information IN4 or external environmental temperature information IN5) acquired by the environmental temperature acquisition unit 12 for a specific purpose such as "to perform temperature correction of the measurement signal", the environmental temperature may be treated as being included in part of the measurement signal group (y)IN2, and the predicted gas concentration OP1 may be predicted from the measurement signal group (y)IN2 including the environmental temperature in the prediction unit 20. Furthermore, when the environmental temperature is included in the measurement signal group (y)IN2, the environmental temperature may be treated as a reference signal for reference signal correction.

[0029] <Prediction Department> The prediction unit 20 outputs a predicted gas concentration OP1 indicating the concentration of at least one of the multiple target gases using the intensity of each measurement signal included in the measurement signal group (y) IN2, the intensity ratio between each measurement signal, and the sensitivity information group IN1. The prediction unit 20 may output fermentation state information indicating the fermentation state of the fermentation substance based on the predicted gas concentration. The prediction unit 20 predicts the concentration of the target gas in the fermentation substance (concentration OP2 in the fermentation substance) as fermentation state information in addition to the predicted gas concentration OP1 of each target gas present in the fermentation unit 30 using the sensitivity information group IN1 (information of output changes of multiple gas sensor channels with respect to concentration changes of each target gas) previously set before operating the system of this embodiment, and each information acquired by the gas signal acquisition unit 10, the environmental temperature acquisition unit 12, and the fermentation substance temperature acquisition unit 32 described later after the system is operated, and outputs the concentration of the target gas in the fermentation substance (concentration OP2 in the fermentation substance) to the display unit 40 to notify the user. The information acquired by the gas signal acquisition unit 10, the environmental temperature acquisition unit 12, and the fermentation substance temperature acquisition unit 32 is sent to the prediction unit 20 using any transmission means.

[0030] More specifically, the prediction unit 20 predicts the predicted gas concentration OP1 of the target gas present in the fermentation unit 30 at that time using the intensity of each measurement signal (measurement signals 1 to 5 in FIG. 1) acquired by the gas signal acquisition unit 10, the intensity ratio between each measurement signal, and a sensitivity information group IN1 set in advance. The prediction by the prediction unit 20 is not limited to a calculation algorithm, such as machine learning or deep learning, and is performed by adopting the composition ratio and concentration of the mixed gas that minimizes the error from the information learned at that time using an arbitrary algorithm. For example, by using multiple regression analysis on the measurement signal group (y)IN2 of a mixed gas composed of multiple target gases, the predicted gas concentration OP1 of each target gas constituting the mixed gas can be calculated from the measurement signal group (y)IN2 of the mixed gas. In the prediction process, individual restrictions may be used according to the actual usage environment in which the system is operated.

[0031] In the prediction unit 20, the sensitivity information group IN1 set in advance may be fine-tuned by re-learning using the measured values ​​acquired by the gas signal acquisition unit 10 at an early stage after the system of this embodiment is operated. The reason for fine-tuning the sensitivity information group IN1 is that the background gas environment is different between the environment assumed when setting in advance and the actual usage environment, so that the information set in advance may have a more appropriate value in the actual usage environment. The range of fine-tuning the sensitivity information group IN1 is preferably within ±50% of the value set in advance, and more preferably within ±30% of the value set in advance. The calculation algorithm for fine-tuning the sensitivity information group IN1 is not limited, and can be a value that can be matched by any method such as machine learning or deep learning.

[0032] In order to improve the prediction accuracy in the prediction unit 20, it is preferable to increase the number of measurement signals from the gas sensor channel (GSC) relative to the number of types of target gases to be predicted. However, if it is desired to reduce the number of measurement signals while maintaining high accuracy, it is preferable to make the number of types of target gases to be predicted equal to the number of measurement signals from the gas sensor channel (GSC).

[0033] When performing environmental temperature correction (temperature correction of the measurement signal from the gas sensor channel (GSC)), the prediction unit 20 may perform temperature correction on the measurement signal group obtained by the gas signal acquisition unit 10 using the environmental temperature (internal environmental temperature information IN4 and external environmental temperature information IN5) acquired by the environmental temperature acquisition unit 12 before predicting the predicted gas concentration OP of the target gas present in the fermentation unit 30. In addition, the environmental temperature acquired by the environmental temperature acquisition unit 12 may be included as part of the measurement signal group, and the prediction unit 20 may predict the predicted gas concentration OP1 while simultaneously handling the measurement signal group (y) IN2 acquired by the gas signal acquisition unit 10 and the environmental temperature acquired by the environmental temperature acquisition unit, thereby substantially performing temperature correction.

[0034] In cases where it is known in advance that a specific gas is present (or not present) in the fermentation section 30, or when it is desired to remove the influence of a gas that is a main component of the mixed gas present in the fermentation section 30 or a gas that generates noise, the remaining measurement signals may be corrected using a measurement signal (reference signal) that responds only to the specific gas (reference gas) to generate the measurement signal group (z) OP4. A "reference signal" is a signal that responds only to specific information (e.g., a specific gas), and some of the multiple measurement signals may be treated as a reference signal. When performing reference signal correction, it is preferable that the prediction section 20 performs reference signal correction before predicting the predicted gas concentration OP1 of the target gas present in the fermentation section 30. In addition, a measurement signal that responds only to a specific gas may not be treated as a reference signal, but may simply be treated as one signal of the measurement signal group like other measurement signals. For example, the prediction unit 20 can use at least one of the measurement signals (e.g., measurement signal 5) as a "reference signal" and use the reference signal to correct at least one of the other measurement signals (e.g., measurement signals 1 to 4) among the measurement signals 1 to 5, the predicted gas concentration OP1 of at least one of the target gases, and the concentration in the fermentation substance of a component substance having the same component as the target gas, which is at least one of the target gases (concentration OP2 in the fermentation substance).

[0035] Examples of gases for the reference signal include water and ethanol, and a humidity sensor or an alcohol sensor may be used to measure the humidity or ethanol gas concentration in the fermentation unit 30, and the information may be used to correct other measurement signals. Correction using the reference signal may be performed before correction using the environmental temperature.

[0036] The reference signal may be a signal for a gas present outside the fermentation unit 30. By using the signal for the gas present outside the fermentation unit 30 as the reference signal, background information for the location where the fermentation unit 30 is present can be obtained. By correcting the measurement signals using the background information, the prediction accuracy of the system of this embodiment can be improved even when the fermentation unit is located in a unique location.

[0037] As shown in Fig. 1, the fermentation state management system 100 can be configured to have a fermentation substance temperature acquisition unit 32 that acquires fermentation substance temperature information IN6 that can be converted into the temperature of the fermentation substance. The prediction unit 20 can further predict the concentration of a substance having the same component as the target gas in the fermentation substance (concentration OP2 in the fermentation substance) by assuming that each target gas present in the fermentation substance has volatilized from the fermentation substance, using a predicted gas concentration OP1 of a target gas, which is at least one of a plurality of target gases present in the fermentation unit 30 predicted by the prediction unit 20, and the fermentation substance temperature information IN6 acquired by the fermentation substance temperature acquisition unit 32.

[0038] The prediction unit 20 may calibrate each predicted value, etc. For example, the prediction unit 20 can calibrate each predicted value, etc., predicted by the prediction unit 20 using other information acquired by a means different from that of the system of this embodiment (a means not derived from the measurement signal of the gas sensor channel). This calibration is performed to match absolute values ​​of the measurement signal group (y) IN2 acquired by the gas signal acquisition unit 10, the measurement signal group (yc) OP3 after environmental temperature correction, the measurement signal group (z) OP4 after reference signal correction, the predicted gas concentration OP1, the predicted concentration value in the fermentation substance (concentration in the fermentation substance OP2), etc.

[0039] The prediction unit 20 may predict the fermentation state at that time point using the predicted gas concentration OP1 of each of the predicted target gases, the concentration OP2 of the target gas in the fermentation substance, and the like. Furthermore, as described above, the gas signal acquisition unit 10 can notify "information on the action recommended to the user" via the display unit 40 based on the predicted gas concentration OP1 of at least one of the multiple target gases or the concentration OP2 of the same component substance as the target gas, which is at least one of the multiple target gases, in the fermentation substance, and the time information IN3, using the time information IN3 linked to the time when the measurement signal group (y)IN2 was acquired. Specifically, the fermentation unit 30 may determine the elapsed time of fermentation from the time information IN3, and further predict the recommended action for the user of the system of this embodiment from the predicted fermentation state and the elapsed time of fermentation. The content of the recommended action is, for example, the temperature adjustment of the fermentation unit 30, the addition of additives to the fermentation unit 30, stirring of the fermentation unit 30, and the like.

[0040] In addition, the prediction unit 20 may have a preset range for at least one of the measured values ​​and predicted values ​​(the intensity of each measured signal, the intensity ratio between each measured signal, the predicted gas concentration OP1 of at least one of the target gases, the concentration OP2 of the same component substance in the fermentation substance as the target gas which is at least one of the target gases, the internal environment temperature information IN4, the external environment temperature information IN5, and the fermentation substance temperature information IN6), and may be configured to detect any deviation from the range as an abnormality. For example, when at least one of the measured values ​​and predicted values ​​deviates from the preset range, the prediction unit 20 can notify the user of information regarding the deviation of the measured value, the predicted value, etc. via the display unit 40.

[0041] <Fermentation Department> The fermentation section 30 is a place where raw materials such as sake and fermented foods are introduced together with microorganisms such as yeast, bacteria, and mold to ferment the fermentation substance. During the fermentation of the fermentation substance, multiple gases that volatilize from the fermenting substance, such as liquid, are also present in the fermentation section 30. For example, in the case of fermentation (brewing) of sake, the multiple gases include aroma components generated by the action of yeast, in addition to ethanol and water vapor. At least one of the multiple target gases is preferably an aroma component generated by the action of yeast present in the fermentation substance. This aroma component is derived from substances generated in the production and decomposition process of amino acids and fatty acids, and examples of such aroma components include ethyl caproate, isoamyl acetate, isobutyl acetate, ethyl acetate, ethyl caprylate, ethyl caprate, ethyl laurate, phenylethyl acetate, ethyl lactate, acetaldehyde, isovaleraldehyde, diacetyl, isoamyl alcohol, phenethyl alcohol, 4-vinyl guaiacol, and 4MMP (4-mercapto-4-methyl-2-pentanone).

[0042] The combination of the fermentation substance and the target gas is not particularly limited, but examples thereof include a combination in which the fermentation substance is mash and the target gas includes at least one of water, ethanol, ethyl caproate, isoamyl acetate, isobutyl acetate, ethyl acetate, ethyl caprylate, ethyl caprate, ethyl laurate, phenylethyl acetate, ethyl lactate, acetaldehyde, isovaleraldehyde, diacetyl, isoamyl alcohol, phenethyl alcohol, 4-vinyl guaiacol, and 4MMP.

[0043] It is preferable that the fermentation section 30 has a function of adjusting the temperature in order to control the activity of yeast. In order to operate the system of this embodiment with high accuracy, it is preferable that 50% or more of the gases present in the fermentation section 30, excluding nitrogen, oxygen, and carbon dioxide, are composed of the target gases registered in advance in the sensitivity information group IN1. More preferably, 70% or more of the gases are composed of the target gases registered in advance in the sensitivity information group IN1. The higher the content of the target gases registered in advance in the sensitivity information group IN1 among the gases present in the fermentation section 30, the higher the prediction accuracy of the system of this embodiment. The reason why nitrogen, oxygen, and carbon dioxide are excluded is that these three types of gases are the main gases that make up the atmosphere, and are often included in the background gas present in the fermentation section.

[0044] <Fermentation material temperature acquisition unit> As described above, the fermentation state management system 100 further has a fermentation substance temperature acquisition unit 32 that acquires fermentation substance temperature information IN6 that can be converted into the temperature of the fermentation substance, and the prediction unit 20 can predict the concentration in the fermentation substance of a substance having the same component as the target gas (concentration in the fermentation substance OP2) from the predicted gas concentration OP1 of the target gas, which is at least one of the multiple target gases, and the fermentation substance temperature information IN6. Specifically, the fermentation substance temperature acquisition unit 32 can acquire the temperature of the fermentation substance or information that can be converted into the temperature of the fermentation substance. The fermentation substance temperature acquisition unit 32 may not only acquire the fermentation substance temperature information IN6, but also have a mechanism for controlling the temperature of the fermentation substance.

[0045] From the viewpoint of hygiene, it is preferable that the fermentation substance temperature information IN6 is acquired in a non-contact state by the fermentation substance temperature acquisition unit 32. For example, an infrared sensor can be used as a method for acquiring the temperature in a non-contact manner. Here, if the members forming the fermentation unit 30 have a temperature measurement function or a temperature adjustment function, it is the members forming the fermentation unit 30 itself that are in contact with the fermentation substance, and the fermentation substance temperature acquisition unit 32 itself can be considered not to be in contact with the fermentation substance.

[0046] The fermentation substance temperature information IN6 acquired by the fermentation substance temperature acquisition unit 32 is used not only to measure and control the temperature of the fermentation substance, but also to convert the predicted gas concentration OP1 of the gas present in the fermentation unit predicted by the prediction unit 20 into the concentration in the fermentation substance (i.e., the concentration OP2 in the fermentation substance). Since the concentration of the gas volatilized from the fermentation substance depends on the fermentation substance temperature, by using the temperature information, the concentration of the same substance in the state before volatilization can be predicted from the concentration of the volatilized gas.

[0047] <Display> As shown in FIG. 1, the fermentation state management system 100 may have a display unit 40 as a means for notifying a user. The display unit 40 can notify a user of the state in the fermentation unit 30 in real time by displaying the results of prediction and analysis in the prediction unit 20 (each measured value, each predicted value, fermentation state, recommended actions, etc.) on a display or the like. The notification method may be any method that can inform the user of the fact, such as sound, display, or email, and may have a function of issuing an alarm to notify the user when an abnormality is detected or when work is required. The display unit 40 may be a display or the like installed in the gas signal acquisition unit 10, the prediction unit 20, or the fermentation unit 30 in the fermentation state management system, or may be a device that can acquire and display output information from the prediction unit 20 by wireless communication, such as a communication terminal.

[0048] In addition, the fermentation status management system 100 may not only display and notify, but also be a system that automatically performs recommended actions (such as adjusting the temperature of the fermentation section, adding additives, stirring, etc.) in conjunction with fermentation equipment and actuators.

[0049] <Hardware configuration> 2, the hardware configuration of each device constituting the fermentation state management system, such as the gas signal acquisition unit 10, the prediction unit 20, and the display unit 40, will be described. For example, like the device 200, the gas signal acquisition unit 10, the prediction unit 20, and the display unit 40 include a processor 210, a communication interface 220, an input / output interface 230, a memory 240, a storage 250, and one or more communication buses 260 for interconnecting these components. However, the hardware configuration of each device constituting the fermentation state management system, such as the gas signal acquisition unit 10 and the prediction unit 20, is not limited to the configuration shown in FIG. 2.

[0050] The processor 210 executes processes, functions, or methods realized by codes or instructions included in a program stored in the storage 250. The processor 210 may include, for example and without limitation, one or more central processing units (CPUs), micro processing units (MPUs), graphics processing units (GPUs), microprocessors, processor cores, multiprocessors, application-specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), etc., and may realize the processes, functions, or methods disclosed in each embodiment by logic circuits (hardware) or dedicated circuits formed in integrated circuits (ICs (Integrated Circuits) chips, LSIs (Large Scale Integration)), etc.

[0051] The communication interface 220 transmits and receives various data to and from other devices via a network. The communication may be performed either wired or wirelessly, and any communication protocol may be used as long as the devices can communicate with each other. For example, the communication interface 220 is implemented as hardware such as a network adapter, various communication software, or a combination of these.

[0052] The network N may be, but is not limited to, an ad hoc network, an intranet, an extranet, a virtual private network (VPN), a local area network (LAN), a wireless LAN (WLAN), a wide area network (WAN), a wireless WAN (WWAN), a metropolitan area network (MAN), a part of the Internet, a part of the public switched telephone network (PSTN), a mobile phone network, ISDNs (Integrated Service Digital Networks), wireless LANs, LTE (Long Term Evolution), CDMA (Code Division Multiple Access), Bluetooth (registered trademark), satellite communication, etc., or a combination of these. The network may include one or more networks.

[0053] The input / output interface 230 includes an input device for inputting various operations to the device 200, and an output device for outputting processing results processed by the device 200. For example, the input / output interface 230 includes information input devices such as a keyboard, a mouse, and a touch panel, and an information output device such as a display. Note that the device 200 may receive a predetermined input or execute a predetermined output by connecting an external input / output interface 230.

[0054] For example, the device 200 may be connected to a sensor or another device 200 (for example, the device 200 is the gas signal acquisition unit 10 and the other device 200 is the prediction unit 20) as an external input / output interface 230 via a wired or wireless network N. The other device 200 may be a server or the like that provides information to the device 200 upon request.

[0055] The memory 240 temporarily stores the programs loaded from the storage 250 and provides a working area for the processor 210. The memory 240 also temporarily stores various data generated while the processor 210 is executing the programs. The memory 240 may be, for example, a high-speed random access memory such as a DRAM, an SRAM, a DDR RAM, or another random access solid-state storage device, or may be a combination of these.

[0056] Storage 250 stores programs, each functional unit, and various data. Storage 250 may be, for example, one or more magnetic disk storage devices, optical disk storage devices, flash memory devices, or nonvolatile memories such as other nonvolatile solid-state storage devices, or may be a combination of these. Another example of storage 250 may be one or more storage devices installed remotely from processor 210.

[0057] In one embodiment of the present invention, the storage 250 stores programs, functional units, and data structures, or a subset thereof. The device 200 is configured to function as, for example, the gas signal acquisition unit 10, the prediction unit 20, etc., as shown in FIG. 1, by the processor 210 executing instructions included in the programs stored in the storage 250.

[0058] Operating system 251, for example, handles various basic system services and contains procedures for performing tasks using the hardware.

[0059] The network communications module 252 may be used, for example, to connect the device 200 to other computers via the communications interface 220 and one or more communications networks, such as the Internet, other wide area networks, local area networks, metropolitan area networks, etc.

[0060] <How to control fermentation status> The fermentation state management method of this embodiment is a fermentation state management method for managing the fermentation state of a fermentation substance, and includes the steps of: obtaining a sensitivity information group including sensitivity information of the gas sensor channel for each of a plurality of target gases; acquiring a measurement signal group from a plurality of gas sensor channels, at least one of which has sensitivity to the concentrations of two or more types of gas among a plurality of target gases, including measurement signals output from the gas sensor channels in response to gas generated from the fermentation substance; and predicting a predicted gas concentration indicating the concentration of at least one of the plurality of target gases using the intensity of each measurement signal included in the measurement signal group and the intensity ratio between the measurement signals, and the sensitivity information group, and outputting the predicted gas concentration.

[0061] A flow of carrying out the fermentation state management method of this embodiment using the fermentation state management system of this embodiment will be described with reference to Fig. 3. Fig. 3 is a diagram showing an example of the flow of the fermentation state management system of this embodiment. However, the flow of the fermentation state management system and the fermentation state management method of this embodiment is not limited to the following aspects.

[0062] First, before the system of this embodiment is operated as a fermentation state management system, for example, a set of sensitivity information of a plurality of gas sensor channels is set in advance for each of a plurality of gases (target gases) to be present in the fermentation section (step S1). Although not shown in Fig. 3, this set of sensitivity information may be fine-tuned by re-learning using the measured values ​​acquired by the gas signal acquisition section at an early stage after the system of this embodiment is operated.

[0063] Next, the user of the system of this embodiment operates the system and starts analysis with the gas signal acquisition unit installed so that the gas present in the fermentation unit can be detected (step S2).

[0064] When the system of this embodiment is operated, the gas signal acquisition unit acquires a measurement signal group y including a plurality of measurement signals for the mixed gas present in the fermentation unit (step S3). The output signal from each gas sensor channel becomes a measurement signal, and the measurement signals are collected to become a "measurement signal group."

[0065] Next, it is determined whether or not to perform environmental temperature correction on the measurement signal group acquired by the gas signal acquisition unit (step S4). If environmental temperature correction is to be performed (step S4: Yes), the process proceeds to step S5. If environmental temperature correction is not to be performed (step S4: No), the process proceeds to step S6. When each measurement signal is to be corrected using the environmental temperature acquired by the environmental temperature acquisition unit, a measurement signal group yc is generated from the measurement signal group y (step S5). Note that even if the environmental temperature is acquired, environmental temperature correction is not performed, and the environmental temperature may be included as part of the measurement signal group.

[0066] Next, it is determined whether or not the measurement signal group acquired by the gas signal acquisition unit is corrected using a reference signal (step S6). If the reference signal correction is to be performed (step S6: Yes), the process proceeds to step S7. If the reference signal correction is not to be performed (step S6: No), the process proceeds to step S8. As described above, a "reference signal" is a signal that responds only to specific information (e.g., a specific gas), and a part of a plurality of measurement signals may be treated as a reference signal. When the reference signal correction is performed using a measurement signal corresponding only to a specific gas as a reference signal, the measurement signal group z is generated by correcting the remaining measurement signals of the measurement signal group y or the measurement signal group yc (step S7). Note that the reference signal correction may be performed before the environmental temperature correction (step S5), and when the environmental temperature is included in the measurement signal group, the environmental temperature may be treated as a reference signal for the reference signal correction.

[0067] Next, the prediction unit predicts the concentration (predicted gas concentration) of each target gas present in the fermentation unit using the intensity of each measurement signal and the intensity ratio between each measurement signal of the measurement signal group y acquired by the gas signal acquisition unit (if correction has been performed, the measurement signal group yc after environmental temperature correction or the measurement signal group z after reference signal correction), as well as the sensitivity information group (step S8).

[0068] The prediction unit predicts the concentration of the target gas in the fermentation substance (concentration in the fermentation substance) from the predicted gas concentration of the target gas, which is at least one of the target gases present in the fermentation unit, using the predicted gas concentration and the temperature information of the fermentation substance acquired by the fermentation substance temperature acquisition unit (step S9).

[0069] Next, it is determined whether or not to calibrate the predicted value (step S10). If the predicted value is to be calibrated, the process proceeds to step S11. If the predicted value is not to be calibrated, the process proceeds to step S12. If the predicted value is to be calibrated (step S11), the value predicted by the prediction unit is calibrated using information acquired by a means other than the gas signal acquisition unit. The value predicted by the prediction unit may deviate significantly from the actual state due to the influence of the environment in which the system is operated (e.g., the shape of the fermentation unit, etc.), so the predicted value is calibrated to calibrate it. Note that, although calibration is performed on the concentration in the fermentation substance in FIG. 3, this calibration may be performed on any of the values ​​of the measurement signal group y acquired by the gas signal acquisition unit, the measurement signal group yc after correction for the environmental temperature, the measurement signal group z after correction for the reference signal, and the predicted gas concentration.

[0070] Next, the display unit notifies the user of the status in real time by displaying the prediction results (each measured value, each predicted value, fermentation state, recommended actions, etc.) in the prediction unit on a display or the like (step S12).

[0071] The system of this embodiment repeats the above operations until fermentation is complete, and if fermentation is not complete (No in step S13), returns to step S3, making it a system capable of managing the fermentation state in real time. Furthermore, the system of this embodiment ends the analysis when it is determined that fermentation is complete (step S13). The intervals at which steps S3 to S13 are repeated may be equal or variable. When the intervals at which steps S3 to S13 are repeated are variable, the intervals may be changed using the prediction results from the prediction unit. For example, when the prediction results from the prediction unit fluctuate significantly, the system may repeat the operations at a high frequency, and when the fluctuations are gentle, the system may repeat the operations at a low frequency.

[0072] In addition, the system of this embodiment may be configured, for example, when it is determined in step S13 that management of the fermentation state should be continued, to return to step S1 instead of step S3 based on the results up to step S12, and to increase, decrease or change the type of analysis gas to continue analysis of the fermentation state.

[0073] <Program> The program of this embodiment causes the apparatus to execute the steps of inputting a sensitivity information group including sensitivity information of the gas sensor channel for each of a plurality of target gases, and a plurality of gas sensor channels, at least one of which has sensitivity to the concentrations of two or more types of gas among a plurality of target gases, and a measurement signal group including measurement signals acquired from the plurality of gas sensor channels and output from the gas sensor channel in response to gas generated from a fermentation substance, predicting a predicted gas concentration indicating the concentration of at least one of the plurality of target gases using the intensity of each measurement signal included in the measurement signal group and the intensity ratio between each measurement signal, and the sensitivity information group, and outputting the predicted gas concentration.

[0074] The program of the present embodiment may be provided in a state stored in a computer-readable storage medium. Here, the storage medium can store the program in a "non-transitory tangible medium." The program includes, but is not limited to, a software program or a computer program.

[0075] The present disclosure is not limited to the above-described embodiment, and various modifications are possible within the scope of the gist of the present disclosure. In other words, the above-described embodiment is merely an example in every respect, and should not be interpreted as being restrictive. For example, the above-described processing steps can be arbitrarily changed in order or executed in parallel within the scope of not causing any inconsistency in the processing contents.

[0076] As described above, according to the fermentation state management system, fermentation state management method, and program of the present embodiment, it is possible to visualize fermentation state information showing the fermentation state of a fermentation substance, such as the predicted gas of the target gas and the concentration in the substance, by using a group of measurement signals and a group of sensitivity information, and to notify the user. Furthermore, according to the system of the present embodiment, by using a group of gas sensor channels that have sensitivity to a plurality of target gases and can transmit measurement signals, and by setting a group of sensitivity information, it is not necessary to replace the sensor with a dedicated sensor, and it is possible to change the target gas to be managed depending on the fermentation substance and the desired fermentation state, and it is versatile for various fermentation substances such as sake.

[0077] [Embodiment] The fermentation state management system according to the embodiment of the present invention will be described with reference to Figs. 4 to 9, taking sake brewing management as an example.

[0078] FIG. 4 is a diagram showing an overview of the fermentation state management system of this embodiment when applied to sake brewing. In FIG. 4, the fermentation state management system 300 includes a gas signal acquisition unit 10, a prediction unit 20, a fermentation unit 30, and a display unit 40. In the brewing tank that serves as the fermentation unit 30, there is moromi, which is the raw material for sake. Moromi is a liquid before it becomes sake (Japanese sake), and is a mixture of yeast starter (yeast), koji, steamed rice, and water. A number of target gases (water, ethanol, ethyl caproate, isoamyl acetate, and acetaldehyde in this embodiment) are volatilized from the fermenting moromi. The gas in the brewing tank is drawn into the gas signal acquisition unit 10 by gas sampling means (tube and pump) 14 so that the gas in the brewing tank can be detected by the gas signal acquisition unit 10 installed outside the brewing tank. The group of measurement signals acquired by the gas signal acquisition unit 10 is sent to the prediction unit (cloud) 20 via the Internet or the like, and the prediction unit 20 performs various calculations and predictions, and notifies the user of the results by displaying them on a display unit 40 such as a smartphone.

[0079] The operation of the system of this embodiment when it is used in the manner shown in FIG. 4 will be described below with reference to an example. First, before operating the fermentation state management system 300 on-site, a set of sensitivity information for each target gas to be managed is set. The set of sensitivity information is, for example, sensitivity information of multiple gas sensor channels of the gas signal acquisition unit for the gas concentration when the target gas exists alone, such as a value measured in advance or a representative value, and the number of gas sensor channels of the gas signal acquisition unit is preferably equal to or greater than the number of types of target gas.

[0080] In this embodiment, for the sake of simplicity, the target gas and the analysis gas are five types, namely, water, ethanol, ethyl caproate, isoamyl acetate, and acetaldehyde, and the gas signal acquisition unit has five gas sensor channels. Ethyl caproate and isoamyl acetate are typical gases that are aroma components of sake, and are generated by the activity of yeast. By detecting the trend of the concentration of such aroma components, the activity of yeast can be monitored. Note that the types of target gas and the gas sensor channels in actual use are not limited to five, and can be increased or decreased. The relationship between the gas species that react with each gas sensor channel (channels 1 to 5) in the fermentation state management system 300 shown in FIG. 4 and the channel number that reacts with each gas species is shown below.

[0081] [Table 1]

[0082] [Table 2]

[0083] For example, in this embodiment, the target gases are the five types mentioned above, but even in the case of managing sake, the aroma components generated are diverse. The aroma of sake is mainly divided into three types: ginjo aroma, raw material aroma, and mature aroma. In this embodiment, esters (ethyl caproate and isoamyl acetate), which are representative aroma components of ginjo aroma that are strongly related to the activity of yeast, are included in the target gas. However, although ginjo aroma is an aroma that is considered important in ginjo sake, the target gas may be changed if the sake to be finally made changes. For example, in the case of pure rice sake, the raw material aroma, which is the aroma from the raw material rice and rice koji, can be set as the target gas, and even if the target gas changes, the system of this embodiment can be operated by resetting the sensitivity information group. In addition, the system of this embodiment can be used for component management during storage after the brewing of sake is completed. Storage may be performed in a bottled state, but the brewing tank used for fermentation may also be used as a storage tank for storage, and the storage state can also be managed using the fermentation state management system of the present invention. In that case, it is preferable to include mature aromas such as sotolon and polysulfides in the target gas.

[0084] FIG. 5 shows an example of a set of sensitivity information (a collective term for a plurality of pieces of sensitivity information) that have been set in advance. FIG. 5 shows the sensitivity (sensor output intensity per unit concentration) of each gas sensor channel to the gas concentration when the concentration of only one target gas changes for each of the target gases water, ethanol, ethyl caproate, isoamyl acetate, and acetaldehyde. In this embodiment, an MSS olfactory sensor is used as the sensor, and the output is output in voltage, so the unit of the sensitivity information is mV / ppm. In this embodiment, the output of each gas sensor channel is assumed to change linearly with respect to the gas concentration, but a nonlinear model may be constructed. The information of this set of sensitivity information (the intensity of each piece of sensitivity information and the intensity ratio between each piece of sensitivity information) is stored in advance in the prediction unit of this fermentation state management system.

[0085] In this embodiment, four of the five gas sensor channels (channels 1 to 4) are sensitive to all target gases, and the remaining one (channel 5) is a channel sensitive only to water. That is, water responds to all the gas sensor channels, and target gases other than water (ethanol, ethyl caproate, isoamyl acetate, and acetaldehyde) respond only to the four gas sensor channels (channels 1 to 4).

[0086] The sensitivity information group may be adjusted using the measurement signal group at the initial stage after the system is operated. In addition, in the system of this embodiment, the prediction unit predicts the predicted gas concentration of each target gas at the time when the measurement signal group is measured using the information of the sensitivity information group and the measurement signal group, but if the information of the sensitivity information group is significantly different from the actual situation, the correct prediction result is not output. For this reason, when the prediction result is obviously abnormal, when there is no continuity in the prediction result after multiple measurements, when there is an influence of background gas, etc., it is preferable to re-learn the information of the sensitivity information group by setting a limit within ±50% of the value of the sensitivity information group set in advance. The calculation algorithm for re-learning is not limited, and it is possible to drive the regression error to a value that is small by setting an arbitrary limit using any method such as machine learning.

[0087] Next, a user of the system of this embodiment installs the gas signal acquisition unit 10 in the sake brewing tank 30, which serves as the fermentation unit, and operates the system in a form in which the gas signal acquisition unit 10 can detect the gas in the brewing tank. When the system is operated, the gas signal acquisition unit 10 measures a measurement signal group y derived from a mixed gas containing the target gas present in the brewing tank 30. Figure 6 shows the measurement signal group y acquired by the fermentation state management system of this embodiment, the measurement signal group yc corrected for environmental temperature, and the measurement signal group z corrected for reference signals. The solid line in Figure 6 shows the measurement signal group y measured by the five gas sensor channels in one measurement. All measurements in each gas sensor channel are performed simultaneously.

[0088] Next, the measurement signal group y is corrected for environmental temperature. The gas signal acquisition unit 10 includes an environmental temperature acquisition unit 12 that measures the internal or external environmental temperature of the gas signal acquisition unit, and the prediction unit 20 performs temperature correction of the measurement signal group y based on the environmental temperature acquired by the environmental temperature acquisition unit 12. The purpose of the temperature correction here is to correct the temperature dependency of the gas sensor channel of the gas signal acquisition unit 10. This temperature correction is performed by setting information on the temperature dependency of each gas sensor channel (change in gas sensor channel output with respect to temperature change) in advance. The environmental temperature correction is a general method of adjusting the output of each gas sensor channel with respect to the measured environmental temperature, so details will be omitted. The dashed line in FIG. 6 indicates the measurement signal group yc after environmental temperature correction.

[0089] Here, each gas sensor channel of the gas signal acquisition unit 10 outputs measurement signals y1, y2, y3, y4, and y5 of the mixed gas according to a linear model such as the following formula 1. Note that only formula 1 corresponding to the measurement signal y1 output from the gas sensor channel 1 will be shown below, and formulas corresponding to the gas sensor channels 2 to 5 (measurement signals y2 to y5) will be omitted. Also, the measurement signals y1, y2, y3, y4, and y5 will be collectively referred to as a measurement signal group y. Note that the subscripts refer to the numbers of the gas sensor channels, and y, a, b, c, d, and e are different values ​​for each gas sensor channel.

[0090]

number

[0091] The predicted gas concentration (i.e., A, B, C, D, E) of each target gas may be predicted in the prediction unit using the measurement signal group y and the measurement signal group yc as they are. However, in order to improve the prediction accuracy, in this embodiment, the signal of the gas sensor channel 5 is treated as a reference signal, and the remaining four signals are corrected in the prediction unit. Here, since the gas sensor channel 5 is a channel that is sensitive only to water, the measurement signal from this channel indicates the amount of water vapor in the brewing tank. Therefore, since the water concentration E can be obtained before machine learning is performed, the remaining four gas sensor channels generate a measurement signal group z composed of measurement signals in which the contribution of water has been removed from the measurement signal of the mixed gas (for example, the relationship z1=y1-e1E for the gas sensor channel 1). That is, the model formulas for the measurement signals z1, z2, z3, and z4 of the remaining four gas sensor channels are as shown in Equations 2 to 5. Moreover, the measurement signals of each gas sensor channel of the measurement signal group z are values ​​as shown by the thick solid line in FIG. 6. Here, the contribution of water is treated as known from the result of the gas sensor channel 5, so the measurement signal z5 is eliminated.

[0092]

number

[0093] Next, the predicted gas concentration of each target gas present in the brewing tank 30 is predicted by the prediction unit 20 using the measurement signal group z and the sensitivity information group. This prediction is not limited to a calculation algorithm, and may be performed using machine learning, deep learning, or the like, and is performed by adopting the composition ratio and concentration of the mixed gas that minimizes the error from the information learned at that time using any algorithm. In this embodiment, the predicted gas concentrations A, B, C, and D that minimize the error of Equations 2 to 5 for gas sensor channels 1 to 4 are calculated by regression. Figure 7 shows the predicted gas concentrations of each target gas (the predicted gas concentrations of each target gas predicted by the fermentation state management system of this embodiment). In this embodiment, all calculations are performed using a linear model, but a nonlinear model may also be used. Note that the concentration of water in Figure 7 shows the value calculated from y5 or yc5.

[0094] Next, the concentration of the gas substance in the mash (i.e., the concentration in the fermentation substance) is predicted from the predicted gas concentration of at least one target gas (target gas) using, for example, fermentation substance temperature information of the mash measured or predicted by the fermentation substance temperature acquisition unit 32 installed in the brewing tank 30. In this embodiment, the target gas is ethanol.

[0095] We assume that the concentration of ethanol gas evaporating from ethanol in the mash (liquid) can be expressed using the liquid concentration and ethanol vapor pressure, for example, as in Equation 6. Ethanol vapor pressure is a function of liquid temperature, and can be expressed, for example, using the Antoine formula, as in Equation 7.

[0096]

number

[0097] Coefficient f and Antoine constant A ant , B ant , C ant Since the value of varies depending on the units used and the type of the system in which the system is operated, an appropriate value may be set in advance, or an arbitrary value may be set first, and the appropriate value may be obtained by subsequent calibration. In addition, since the above-mentioned formulas 6 and 7 are empirical formulas and examples of models, an original model formula may be used according to the system in which the system is operated.

[0098] Under the Liquor Tax Act, the ethanol concentration (in the liquid) at a temperature of 15°C is called the alcohol content, so the alcohol content is the calculated ethanol concentration in the mash converted to the value for a temperature of 15°C.

[0099] The obtained alcohol content may also be calibrated. In this calibration, the alcohol content predicted by the prediction unit is calibrated using information obtained by a means other than the gas signal acquisition unit. The different means can be, for example, sampling the mash and measuring it using a buoy or a vibration density meter after distillation. If there is a difference between the alcohol content obtained by the different means and the alcohol content predicted by the prediction unit, the alcohol content predicted by the prediction unit can be calibrated by applying an offset to cancel the difference, or by applying a gain to cancel the slope difference using two or more measurement results by different means.

[0100] In FIG. 4, the display unit 40 notifies the user of the status in real time by displaying the results of the prediction analysis in the prediction unit (each measured value, each predicted value, fermentation state, recommended action) on the user's smartphone. The contents to be notified may include the predicted gas concentration of each target gas, the alcohol content, and the sake meter value related to the alcohol content, and may further express the state of yeast activity and the fermentation state. The yeast activity and fermentation state may express, for example, the induction phase, logarithmic growth phase, stop phase, death phase, etc. of yeast, and may also express the early stage of fermentation, insufficient fermentation, appropriate fermentation, over-fermentation, etc. In addition, there may be recommended actions for the user, such as instructions for temperature adjustment, instructions for adding water, and stirring the mash.

[0101] By repeating the above-mentioned operation at any interval until fermentation is completed, the system can manage the fermentation state in real time. Figure 8 shows the time-dependent changes in each target gas predicted by the prediction unit from the start to the end of operation of the system of this embodiment. Figure 9 also shows the predicted time-dependent changes in the alcohol content in the mash.

[0102] Furthermore, according to the system of this embodiment, it is possible to further predict the sake meter value from the predicted alcohol content by using the relational expression between the alcohol content and the Baume degree, called the ab line, shown in Equation 8. Note that the sake meter value is the value obtained by multiplying the Baume degree value by -10, and when predicting the sake meter value, constants α and β are set in the prediction unit in advance.

[0103]

number

[0104] As described above, according to the embodiment of the present invention, the state of the mash (the activity of yeast) can be visualized from a plurality of gases containing aroma components generated during sake brewing, making it easier to manage it. [Explanation of symbols]

[0105] 100, 300: Fermentation state management system, 10: Gas signal acquisition unit, 12: Environmental temperature acquisition unit, 20: Prediction unit, 30: Fermentation unit, 32: Fermentation substance temperature acquisition unit, 40: Display unit, IN1: Sensitivity information group, IN2: Measurement signal group (y), IN3: Time information, IN4: Internal environmental temperature information, IN5: External environmental temperature information, IN6: Fermentation substance temperature information, IN7: Other information, OP1: Predicted gas concentration, OP2: Concentration inside the fermentation substance, OP3: Measurement signal group (yc), OP4: Measurement signal group (z)

Claims

1. A fermentation state management system for managing the fermentation state of a fermentation substance, a gas signal acquiring unit having a plurality of gas sensor channels, at least one of the plurality of gas sensor channels having sensitivity to concentrations of two or more types of gas among a plurality of target gases; A group of measurement signals including measurement signals output from the gas sensor channels in response to gas generated from the fermentation substance; a sensitivity information group including sensitivity information of the gas sensor channel for each of the plurality of target gases; a prediction unit that outputs a predicted gas concentration indicating a concentration of at least one of the plurality of target gases using the intensity of each measurement signal included in the group of measurement signals and the intensity ratio between the respective measurement signals and the group of sensitivity information; A fermentation status control system having:

2. The prediction unit outputs fermentation state information indicating a fermentation state of the fermentation substance based on the predicted gas concentration. The fermentation state management system according to claim 1 .

3. At least one of the plurality of gas sensor channels is sensitive to concentrations of all of the target gases. The fermentation state management system according to claim 1 .

4. All of the plurality of gas sensor channels are sensitive to concentrations of two or more types of gases among the target gases. The fermentation state management system according to claim 1 .

5. At least one of the plurality of target gases is responsive to two or more of the plurality of gas sensor channels. The fermentation state management system according to claim 1 .

6. At least one of the plurality of target gases is responsive to all of the plurality of gas sensor channels. The fermentation state management system according to claim 1 .

7. all of the plurality of target gases are responsive to two or more of the plurality of gas sensor channels; The fermentation state management system according to claim 1 .

8. an environmental temperature acquisition unit that acquires a temperature of an external environment in which the gas signal acquisition unit exists or a temperature of an internal environment of the gas signal acquisition unit; The prediction unit corrects the group of measurement signals using a temperature of the external environment or a temperature of the internal environment. The fermentation state management system according to claim 1 .

9. A fermentation substance temperature acquisition unit that acquires fermentation substance temperature information that can be converted into the temperature of the fermentation substance, The prediction unit predicts a concentration of a substance having the same component as the target gas in the fermentation substance based on the predicted gas concentration of a target gas which is at least one of the plurality of target gases and the fermentation substance temperature information. The fermentation state management system according to claim 1 .

10. The prediction unit uses at least one of the plurality of measurement signals as a reference signal, and corrects at least one of the other measurement signals, the predicted gas concentration of at least one of the plurality of target gases, and the concentration of the same component substance as the target gas in the fermentation substance using the reference signal. The fermentation state management system according to claim 9.

11. The reference signal is a measurement signal output from a gas sensor channel having sensitivity to a concentration of only one type of gas among the plurality of target gases. The fermentation state management system according to claim 10.

12. The prediction unit calibrates at least one of the measurement signals, the predicted gas concentration of at least one of the plurality of target gases, and the concentration in the fermentation material of a substance having the same component as a target gas that is at least one of the plurality of target gases, using other information not derived from the measurement signal from the gas sensor channel. The fermentation state management system according to claim 1 .

13. The gas signal acquisition unit acquires the measurement signals in a non-contact state with the fermentation substance. The fermentation state management system according to claim 1 .

14. The fermentation substance temperature acquisition unit acquires the fermentation substance temperature information in a non-contact state with the fermentation substance. The fermentation state management system according to claim 9.

15. The prediction unit notifies information on a deviation of at least one of the measured value and the predicted value from a preset range when the deviation is detected. The fermentation state management system according to claim 1 .

16. the gas signal acquisition unit further acquires time information associated with a time at which the group of measurement signals was acquired; Notifying a user of information on a recommended action based on the predicted gas concentration of at least one of the plurality of target gases or the concentration in the fermentation material of a substance having the same component as the target gas that is at least one of the plurality of target gases, and the time information; The fermentation state management system according to claim 1 .

17. At least one of the plurality of target gases is an aroma component generated by the action of yeast present in the fermented substance. The fermentation state management system according to claim 1 .

18. The fermented material is moromi, The target gas includes at least one of water, ethanol, ethyl caproate, isoamyl acetate, isobutyl acetate, ethyl acetate, ethyl caprylate, ethyl caprate, ethyl laurate, phenylethyl acetate, ethyl lactate, acetaldehyde, isovaleraldehyde, diacetyl, isoamyl alcohol, phenethyl alcohol, 4-vinyl guaiacol, and 4-mercapto-4-methyl-2-pentanone; The fermentation state management system according to claim 1 .

19. A fermentation state management method for managing the fermentation state of a fermentation substance, comprising: obtaining a set of sensitivity information including sensitivity information of the gas sensor channel for each of a plurality of target gases; A step of acquiring a group of measurement signals from a plurality of gas sensor channels, at least one of which has sensitivity to concentrations of two or more types of gas among the plurality of target gases, the group of measurement signals including measurement signals output from the gas sensor channels in response to the gas generated from the fermentation substance; predicting a predicted gas concentration indicating a concentration of at least one of the target gases using the intensity of each measurement signal included in the group of measurement signals and the intensity ratio between each measurement signal and the group of sensitivity information, and outputting the predicted gas concentration; A fermentation state control method comprising the steps of:

20. The device, a sensitivity information group including sensitivity information of the gas sensor channel for each of a plurality of target gases; a plurality of gas sensor channels, at least one of which has sensitivity to concentrations of two or more types of gas among the plurality of target gases, and a measurement signal group including measurement signals acquired from the plurality of gas sensor channels and output from the gas sensor channels in response to a gas generated from a fermentation substance; Enter A program that executes a process of predicting a predicted gas concentration indicating the concentration of at least one of the multiple target gases using the intensity of each measurement signal included in the group of measurement signals and the intensity ratio between each measurement signal, and the group of sensitivity information, and outputting the predicted gas concentration.

Citation Information

Patent Citations

  • Index calculation system and index calculation method

    JP2023002080A