Odor identification method and odor identification system

Abstract

This odor identification method includes: (a) a step for maintaining the temperature of an odor sensor (8) at a first temperature; (b) a step for acquiring a first signal output from the odor sensor (8) when the odor sensor (8) maintained at the first temperature is exposed to a sample gas; (c) a step for maintaining the temperature of the odor sensor (8) at a second temperature different from the first temperature; (d) a step for acquiring a second signal output from the odor sensor (8) when the odor sensor (8) maintained at the second temperature is exposed to the sample gas; and (e) a step for identifying the odor of the sample gas by using at least the first signal and the second signal.

Description

Odor identification method and odor identification system

[0001] The present disclosure relates to an odor identification method and an odor identification system for identifying the odor of a sample gas.

[0002] 2. Description of the Related Art There is known an odor identification system that identifies the odor of a sample gas by detecting odor substances contained in the sample gas using an odor sensor (see, for example, Patent Document 1).

[0003] Japanese Patent Application Publication No. 9-159594

[0004] In the conventional odor identification systems described above, in order to accurately identify the odors of a wide variety of sample gases, it is necessary to use high-precision dedicated sensors, which poses the problem of increased costs.

[0005] Therefore, the present disclosure provides an odor identification method and odor identification system that can identify the odor of a sample gas at low cost and with high accuracy.

[0006] An odor identification method according to one aspect of the present disclosure is an odor identification method for identifying the odor of a sample gas using an odor sensor that outputs a signal corresponding to the adsorption concentration of odor molecules contained in the sample gas, and includes the steps of: (a) maintaining the temperature of the odor sensor at a first temperature by heating or cooling the odor sensor; (b) acquiring a first signal output from the odor sensor when the odor sensor maintained at the first temperature is exposed to the sample gas; (c) maintaining the temperature of the odor sensor at a second temperature different from the first temperature by heating or cooling the odor sensor; (d) acquiring a second signal output from the odor sensor when the odor sensor maintained at the second temperature is exposed to the sample gas; and (e) identifying the odor of the sample gas using at least the first signal and the second signal.

[0007] In addition, an odor identification system according to one aspect of the present disclosure includes an odor sensor that outputs a signal corresponding to the adsorption concentration of odor molecules contained in a sample gas; a temperature control element that maintains the temperature of the odor sensor at a first temperature and a second temperature by heating or cooling the odor sensor; an acquisition unit that acquires a first signal output from the odor sensor when the odor sensor maintained at the first temperature by the temperature control element is exposed to the sample gas, and acquires a second signal output from the odor sensor when the odor sensor maintained at the second temperature by the temperature control element is exposed to the sample gas; and an identification unit that identifies the odor of the sample gas using at least the first signal and the second signal.

[0008] These comprehensive or specific aspects may be realized by a system, a method, an integrated circuit, a computer program, or a recording medium such as a computer-readable CD-ROM (Compact Disc-Read Only Memory), or may be realized by any combination of a system, a method, an integrated circuit, a computer program, and a recording medium.

[0009] According to an odor identification method and the like according to one aspect of the present disclosure, the odor of a sample gas can be identified at low cost and with high accuracy.

[0010] FIG. 1 is a block diagram showing the configuration of an odor identification system according to embodiment 1. FIG. 2 is a schematic diagram showing the configuration of an exposure unit of the odor identification system according to embodiment 1. FIG. 3 is a graph showing an example of a signal output from an odor sensor of the odor identification system according to embodiment 1. FIG. 4 is a plan view showing an odor sensor of the odor identification system according to embodiment 1. FIG. 5 is a flowchart showing the operation flow of the odor identification system according to embodiment 1. FIG. 6 is a graph showing an example of a signal output from an odor sensor of a conventional odor identification system. FIG. 7 is a graph showing the experimental results of Experiment 1 and Experiment 2. FIG. 8 is a block diagram showing the configuration of an odor identification system according to embodiment 2. FIG. 9 is a schematic diagram showing the configuration of an exposure unit of the odor identification system according to embodiment 2. FIG. 10 is a flowchart showing the operation flow of the odor identification system according to embodiment 2.

[0011] (Technology 1) An odor identification method for identifying the odor of a sample gas using an odor sensor that outputs a signal corresponding to the adsorption concentration of odor molecules contained in the sample gas, the odor identification method comprising: (a) a step of maintaining the temperature of the odor sensor at a first temperature by heating or cooling the odor sensor; (b) a step of acquiring a first signal output from the odor sensor when the odor sensor maintained at the first temperature is exposed to the sample gas; (c) a step of maintaining the temperature of the odor sensor at a second temperature different from the first temperature by heating or cooling the odor sensor; (d) a step of acquiring a second signal output from the odor sensor when the odor sensor maintained at the second temperature is exposed to the sample gas; and (e) a step of identifying the odor of the sample gas using at least the first signal and the second signal.

[0012] According to Technology 1, the odor of a sample gas is identified using at least a first signal output from an odor sensor maintained at a first temperature and a second signal output from an odor sensor maintained at a second temperature. This allows the odor of the sample gas to be identified using more information, making it possible to identify the odor of the sample gas at low cost and with high accuracy even when using an existing odor sensor.

[0013] (Technology 2) The odor identification method according to Technology 1, wherein the odor sensor includes one specific odor sensor, and in (a), the temperature of the specific odor sensor is maintained at the first temperature, and in (b), after (a), the specific odor sensor maintained at the first temperature is exposed to the sample gas, and the first signal output from the specific odor sensor is acquired, and in (c), after (b), the temperature of the specific odor sensor is maintained at the second temperature, and in (d), after (c), the specific odor sensor maintained at the second temperature is exposed to the sample gas, and the second signal output from the specific odor sensor is acquired.

[0014] According to Technique 2, the odor of a sample gas can be identified at low cost and with high accuracy by using one specific odor sensor.

[0015] (Technology 3) The odor identification method according to Technology 1, wherein the odor sensor includes a first odor sensor and a second odor sensor, and in (a) and (c), the temperature of the first odor sensor is maintained at the first temperature, and simultaneously the temperature of the second odor sensor is maintained at the second temperature, and in (b) and (d), the first signal output from the first odor sensor is acquired when the first odor sensor maintained at the first temperature is exposed to the sample gas, and simultaneously the second signal output from the second odor sensor is acquired when the second odor sensor maintained at the second temperature is exposed to the sample gas.

[0016] According to Technique 3, by using the first odor sensor and the second odor sensor, the odor of the sample gas can be identified in a relatively short time, at low cost, and with high accuracy.

[0017] (Technology 4) An odor identification system comprising: an odor sensor that outputs a signal corresponding to the adsorption concentration of odor molecules contained in a sample gas; a temperature control element that maintains the temperature of the odor sensor at a first temperature and a second temperature by heating or cooling the odor sensor; an acquisition unit that acquires a first signal output from the odor sensor when the odor sensor maintained at the first temperature by the temperature control element is exposed to the sample gas, and acquires a second signal output from the odor sensor when the odor sensor maintained at the second temperature by the temperature control element is exposed to the sample gas; and an identification unit that identifies the odor of the sample gas using at least the first signal and the second signal.

[0018] According to Technology 4, the identification unit identifies the odor of the sample gas using at least a first signal output from the odor sensor maintained at a first temperature and a second signal output from the odor sensor maintained at a second temperature. This allows the odor of the sample gas to be identified using more information, making it possible to identify the odor of the sample gas at low cost and with high accuracy even when using an existing odor sensor.

[0019] These comprehensive or specific aspects may be realized as a system, a method, an integrated circuit, a computer program, or a non-transitory recording medium such as a computer-readable CD-ROM, or may be realized as any combination of a system, a method, an integrated circuit, a computer program, or a recording medium.

[0020] Hereinafter, the embodiments will be specifically described with reference to the drawings.

[0021] The embodiments described below are all comprehensive or specific examples. The numerical values, shapes, materials, components, component placement and connection configurations, steps, and step order shown in the following embodiments are merely examples and are not intended to limit the present disclosure. Furthermore, among the components in the following embodiments, components that are not described in the independent claims that represent the highest concepts are described as optional components.

[0022] (Embodiment 1) [1-1. Configuration of odor identification system] The configuration of an odor identification system 2 according to embodiment 1 will be described with reference to FIGS. 1 to 4. FIG. 1 is a block diagram showing the configuration of an odor identification system 2 according to embodiment 1. FIG. 2 is a schematic diagram showing the configuration of an exposure unit 4 of the odor identification system 2 according to embodiment 1. FIG. 3 is a graph showing an example of a signal output from the odor sensor 8 of the odor identification system 2 according to embodiment 1. FIG. 4 is a plan view showing the odor sensor 8 of the odor identification system 2 according to embodiment 1.

[0023] The odor identification system 2 is a system for identifying the odor of a sample gas. That is, the odor identification system 2 identifies which of a plurality of odor molecules is contained in the sample gas. The sample gas may be, for example, gas collected from food, exhaled breath collected from a human body, air surrounding a human body, or air collected from a room in a building.

[0024] As shown in FIG. 1 , the odor identification system 2 includes an exposure unit 4 , a control unit 6 , an odor sensor 8 (an example of one specific odor sensor), an acquisition unit 10 , and an identification unit 12 .

[0025] The exposure unit 4 is a mechanism for exposing the odor sensor 8 to a sample gas or a reference gas (described later). Specifically, as shown in FIG. 2 , the exposure unit 4 includes a container 14, a switching valve 16, a pump 18, a heater 20 (an example of a temperature control element), and a temperature sensor 22.

[0026] The container 14 is, for example, a box-shaped container for housing the odor sensor 8, the heater 20, and the temperature sensor 22. A sample gas or a reference gas is introduced into the container 14, as will be described later.

[0027] The switching valve 16 is a three-way solenoid valve for switching the gas introduced into the container 14, and is driven by the control unit 6 (see FIG. 1 ). The switching valve 16 has a first input port 24, a second input port 26, and an output port 28. The switching valve 16 is switched from one of a first state and a second state to the other. In the first state, the first input port 24 and the output port 28 are connected, and the second input port 26 is closed. On the other hand, in the second state, the second input port 26 and the output port 28 are connected, and the first input port 24 is closed.

[0028] The first input port 24 is connected via a pipe 30 to a sample gas supply source 32 for supplying a sample gas. The second input port 26 is connected via a pipe 34 to a reference gas supply source 36 for supplying a reference gas. The reference gas is a gas used to desorb odor molecules from the odor sensor 8, and may be, for example, an inert gas such as nitrogen gas that does not contain odor molecules, or a gas from which humidity, volatile components, etc. contained in air have been removed by passing the air through a hollow fiber filter. The output port 28 is connected via a pipe 38 to the interior of the container 14.

[0029] The pump 18 is an intake pump driven by the control unit 6. The intake side of the pump 18 is connected to the inside of the container 14 via piping 40, and the exhaust side of the pump 18 is connected to an exhaust duct (not shown) via piping (not shown). The pump 18 introduces a sample gas or a reference gas into the inside of the container 14, and discharges the sample gas or the reference gas introduced into the inside of the container 14 to the exhaust duct.

[0030] The heater 20 is, for example, an electric heater that generates heat by electrical resistance to supplied power, and is driven by the control unit 6. The heater 20 is housed inside the container 14 and is arranged so as to be in contact with the odor sensor 8. When the heater 20 is energized, heat from the heater 20 is transferred to the odor sensor 8, thereby heating the odor sensor 8. The heater 20 heats the odor sensor 8, thereby sequentially maintaining the temperature of the odor sensor 8 at three different temperatures, namely, a first temperature (e.g., 30°C), a second temperature (e.g., 40°C), and a third temperature (e.g., 50°C).

[0031] In this specification, "maintaining the temperature of the odor sensor 8 at 30°C" does not only mean maintaining the temperature strictly constant at 30°C, but also includes maintaining the temperature approximately constant within a range of, for example, 30°C ± 0.2°C. This also applies when the temperature of the odor sensor 8 is maintained at 40°C or 50°C.

[0032] Furthermore, in this embodiment, the heater 20 maintains the temperature of the odor sensor 8 at 30°C, 40°C, and 50°C, but is not limited thereto and may maintain any temperature in the range of room temperature to 100°C. Furthermore, in this embodiment, the heater 20 maintains the temperature of the odor sensor 8 at three different temperatures, but is not limited thereto and may maintain two different temperatures, or four or more different temperatures. Furthermore, in this embodiment, the heater 20 is configured as an electric heater for heating the odor sensor 8, but is not limited thereto and may be configured as, for example, a Peltier element for heating and / or cooling the odor sensor 8.

[0033] The temperature sensor 22 is housed inside the container 14 and is disposed near the odor sensor 8. The temperature sensor 22 detects the temperature of the odor sensor 8 and outputs temperature data indicating the detected temperature to the control unit 6.

[0034] 3, the exposure unit 4 exposes the odor sensor 8 to the sample gas during the first period T1 of the period Tm, which is made up of a first period T1 and a second period T2 following the first period T1. Furthermore, the exposure unit 4 exposes the odor sensor 8 to the reference gas during the second period T2 of the period Tm.

[0035] More specifically, during a first period T1 in the period Tm, the switching valve 16 is switched to a first state while the pump 18 is operating. In this first state, the sample gas supplied from the sample gas supply source 32 is introduced into the container 14 via the first input port 24 and the output port 28 of the switching valve 16. This exposes the odor sensor 8 to the sample gas introduced into the container 14, and odor molecules contained in the sample gas are adsorbed onto the odor sensor 8. The sample gas introduced into the container 14 is then exhausted to the exhaust duct via the pump 18.

[0036] During a second period T2 within the period Tm, the switching valve 16 is switched to the second state while the pump 18 is operating. In this second state, the reference gas supplied from the reference gas supply source 36 is introduced into the container 14 via the second input port 26 and the output port 28 of the switching valve 16. At this time, the sample gas remaining in the container 14 is discharged to the exhaust duct via the pump 18 by the flow of the reference gas introduced into the container 14. As a result, the odor sensor 8 is exposed to the reference gas introduced into the container 14, and odor molecules adsorbed to the odor sensor 8 are desorbed by the flow of the reference gas. The reference gas introduced into the container 14 is also discharged to the exhaust duct via the pump 18.

[0037] The control unit 6 controls the operation of the switching valve 16 and the pump 18 of the exposure unit 4. Specifically, in a first period T1 within the period Tm, the control unit 6 drives the pump 18 and switches the switching valve 16 to the first state. In addition, in a second period T2 within the period Tm, the control unit 6 drives the pump 18 and switches the switching valve 16 to the second state. The control unit 6 then repeatedly switches the switching valve 16 between the first state and the second state for each period Tm.

[0038] Furthermore, the control unit 6 controls the driving of the heater 20 so as to maintain a constant temperature of the odor sensor 8 based on the temperature data from the temperature sensor 22. Specifically, during the first period Tm (see FIG. 3 ), the control unit 6 performs feedback control to match the temperature data from the temperature sensor 22 with the first reference temperature (=30° C.) by driving the heater 20 to heat the odor sensor 8 based on a comparison of the temperature data from the temperature sensor 22 with the first reference temperature. As a result, during the first period Tm, the control unit 6 controls the driving of the heater 20 so as to maintain the temperature of the odor sensor 8 at 30° C.

[0039] Next, during a second period Tm (see FIG. 3 ) following the first period Tm, the control unit 6 performs feedback control to match the temperature data from the temperature sensor 22 with the second reference temperature (=40° C.) by driving the heater 20 to further heat the odor sensor 8 based on a comparison of the temperature data from the temperature sensor 22 with the second reference temperature (=40° C.). As a result, during the second period Tm, the control unit 6 controls the driving of the heater 20 so as to maintain the temperature of the odor sensor 8 at 40° C.

[0040] Next, during a third period Tm (see FIG. 3 ) following the second period Tm, the control unit 6 performs feedback control to match the temperature data from the temperature sensor 22 with the third reference temperature (=50° C.) by driving the heater 20 to further heat the odor sensor 8 based on a comparison of the temperature data from the temperature sensor 22 with the third reference temperature (=50° C.). As a result, during the third period Tm, the control unit 6 controls the driving of the heater 20 so as to maintain the temperature of the odor sensor 8 at 50° C.

[0041] The above-mentioned feedback control may be, for example, any of ON / OFF control, proportional control, and PID (Proportional-Integral-Differential) control.

[0042] The odor sensor 8 is housed within the container 14 and, upon exposure to sample gas introduced into the container 14, outputs a signal corresponding to the adsorption concentration of odor molecules contained in the sample gas. As shown in FIG. 4 , the odor sensor 8 is configured, for example, as an electrical resistance-type odor sensor and has multiple (e.g., a total of 16, CH1 to CH16) sensitive elements 42 with different sensing characteristics. Each of the multiple sensitive elements 42 has a sensing unit 44 formed of a sensitive film and a pair of electrodes 46, 48 electrically connected to the sensing unit 44. The electrical resistance value of the sensing unit 44 changes depending on the adsorption concentration of odor molecules in the sample gas to the sensing unit 44. Each of the multiple sensitive elements 42 outputs a signal corresponding to the electrical resistance value of the sensing unit 44 as a voltage signal or a current signal via the pair of electrodes 46, 48 to the acquisition unit 10.

[0043] FIG. 3 is a graph schematically illustrating the change over time in the signal output from the odor sensor 8. The horizontal axis of the graph in FIG. 3 represents time, and the vertical axis represents signal strength (signal value). Note that the signal shown in FIG. 3 is a signal output from one sensor element 42 (e.g., the sensor element of CH1) of the odor sensor 8. As shown in FIG. 3, the signal strength of the signal output from the odor sensor 8 changes over time over a period Tm. Specifically, during a first period T1 in which the odor sensor 8 is exposed to a sample gas, the sensing unit 44 of the odor sensor 8 adsorbs odor molecules contained in the sample gas, thereby increasing the signal strength of the signal output from the odor sensor 8. Thereafter, during a second period T2 in which the odor sensor 8 is exposed to a reference gas, the signal strength of the signal output from the odor sensor 8 decreases due to desorption of odor molecules from the sensing unit 44 of the odor sensor 8.

[0044] Hereinafter, the signals output from the odor sensor 8 maintained at 30° C., 40° C., and 50° C. will be referred to as the "first signal," the "second signal," and the "third signal," respectively. As the temperature of the odor sensor 8 increases, the affinity between odor molecules and the sensing unit 44 of the odor sensor 8 increases, and therefore, as shown in FIG. 3 , the signal intensities of the first signal, the second signal, and the third signal increase in that order.

[0045] The acquisition unit 10 acquires signals (first signal, second signal, and third signal) output from the odor sensor 8 during a first period T1 for each period Tm. The acquisition unit 10 then outputs the acquired first, second, and third signals to the identification unit 12. Specifically, during the first period T1 in the first period Tm, the acquisition unit 10 acquires a first signal output from the odor sensor 8 held at 30°C and outputs the acquired first signal to the identification unit 12. Next, during the first period T1 in the second period Tm, the acquisition unit 10 acquires a second signal output from the odor sensor 8 held at 40°C and outputs the acquired second signal to the identification unit 12. Next, during the first period T1 in the third period Tm, the acquisition unit 10 acquires a third signal output from the odor sensor 8 held at 50°C and outputs the acquired third signal to the identification unit 12. The acquiring unit 10 may acquire the signals (the first signal, the second signal, and the third signal) output from the odor sensor 8 during the second period T2 for each period Tm.

[0046] The identification unit 12 identifies the odor of the sample gas based on the first signal, the second signal, and the third signal acquired by the acquisition unit 10. Specifically, the identification unit 12 calculates the maximum signal intensity (signal sensitivity) and / or the signal slope (amount of change in signal intensity per unit time) as features for each of the first signal, the second signal, and the third signal acquired by the acquisition unit 10. The identification unit 12 then uses a trained model to identify the odor of the sample gas based on the calculated features. Specifically, the identification unit 12 inputs the calculated features into the trained model to identify which type of odor molecule is contained in the sample gas among multiple types of odor molecules. The identification unit 12 also outputs information indicating the identification result to the outside of the odor identification system 2. Note that the identification unit 12 may identify the odor of the sample gas based on at least two signals (e.g., the first signal and the second signal) acquired by the acquisition unit 10.

[0047] Here, the trained model is constructed by performing machine learning using, as training data, known odor molecules and feature quantities calculated from first, second, and third signals output from the odor sensor 8 exposed to a sample gas containing the known odor molecules and maintained at 30° C., 40° C., and 50° C., respectively. To construct a logical model in machine learning, for example, a neural network, a random forest, a support vector machine, a self-organizing map, or a Light Gradient Boosting Machine (LightGBM) is used.

[0048] [1-2. Operation of Odor Identification System] Next, the operation of the odor identification system 2 according to the first embodiment will be described with reference to Fig. 5. Fig. 5 is a flowchart showing the flow of the operation of the odor identification system 2 according to the first embodiment.

[0049] As shown in FIG. 5, first, during the first period Tm (see FIG. 3), the heater 20 heats the odor sensor 8 to maintain the temperature of the odor sensor 8 at 30° C. (S101).

[0050] Next, in the first period Tm, the exposure unit 4 exposes the odor sensor 8 to the sample gas (S102), and then exposes the odor sensor 8 to the reference gas (S103).

[0051] Next, during the first period Tm, the acquisition unit 10 acquires the first signal output from the odor sensor 8 maintained at 30°C (S104) and outputs the acquired first signal to the identification unit 12.

[0052] Thereafter, during a second period Tm (see FIG. 3) following the first period Tm, the heater 20 further heats the odor sensor 8, thereby maintaining the temperature of the odor sensor 8 at 40° C. (S105).

[0053] Next, in the second period Tm, the exposure unit 4 exposes the odor sensor 8 to the sample gas (S106), and then exposes the odor sensor 8 to the reference gas (S107).

[0054] Next, during the second period Tm, the acquisition unit 10 acquires a second signal output from the odor sensor 8 maintained at 40°C (S108) and outputs the acquired second signal to the identification unit 12.

[0055] Thereafter, during a third period Tm (see FIG. 3) following the second period Tm, the heater 20 further heats the odor sensor 8, thereby maintaining the temperature of the odor sensor 8 at 50° C. (S109).

[0056] Next, in the third period Tm, the exposure unit 4 exposes the odor sensor 8 to the sample gas (S110), and then exposes the odor sensor 8 to the reference gas (S111).

[0057] Next, during the third period Tm, the acquisition unit 10 acquires a third signal output from the odor sensor 8 maintained at 50°C (S112) and outputs the acquired third signal to the identification unit 12.

[0058] Next, the identifying unit 12 identifies the odor of the sample gas based on the first signal, the second signal, and the third signal acquired by the acquiring unit 10 (S113).

[0059] [1-3. Effects] Figure 6 is a graph showing an example of a signal output from an odor sensor in a conventional odor identification system. In conventional odor identification systems, the odor sensor is maintained at a single temperature (e.g., only 30°C). Therefore, as shown in Figure 6, only one type of signal is output from the odor sensor, and the odor of the sample gas is identified based on this single type of signal.

[0060] In contrast, in the odor identification system 2 according to the first embodiment, the odor sensor 8 is sequentially maintained at three different temperatures, for example, 30° C., 40° C., and 50° C. As a result, the odor sensor 8 outputs three types of signals (first signal, second signal, and third signal), and the odor of the sample gas is identified based on these three types of signals.

[0061] As a result, the odor identification system 2 of embodiment 1 can identify the odor of the sample gas using more information than the conventional odor identification system described above, and therefore can identify the odor of the sample gas at low cost and with high accuracy even when using an existing odor sensor 8.

[0062] [1-4. Experiments] [1-4-1. Experiment 1] In order to confirm the above-mentioned effects, the following experiment 1 was carried out.

[0063] In Experiment 1, the following sample gases A, B, and C (A to C) were used. Each of sample gases A to C was a mixture of the blueberry scent and the following three compounds: Sample gas A: Diacetyl Sample gas B: Butyric acid Sample gas C: Hexanal

[0064] In Comparative Example 1, an odor sensor maintained at a single temperature of 30°C was exposed to sample gases A to C, and the signals output from the odor sensor were acquired 25 times for each. The features calculated from these signals were input into a trained model, and a test to identify the sample gas was conducted. The machine learning algorithm used in the trained model was Random Forest.

[0065] In Comparative Example 2, an odor sensor maintained at a single temperature of 40°C was exposed to sample gases A to C, and the signals output from the odor sensor were acquired 25 times for each. The features calculated from these signals were input into a trained model, and a test to identify the sample gas was conducted. The machine learning algorithm used in the trained model was Random Forest.

[0066] In Comparative Example 3, an odor sensor maintained at a single temperature of 50°C was exposed to sample gases A to C, and the signals output from the odor sensor were acquired 25 times for each. The features calculated from these signals were input into a trained model, and a test to identify the sample gas was conducted. The machine learning algorithm used in the trained model was Random Forest.

[0067] In Example 1, an odor sensor was sequentially held at three different temperatures: 30°C, 40°C, and 50°C, and then exposed to sample gases A to C, respectively. The first signal, second signal, and third signal output from the odor sensor were acquired 25 times each. Feature quantities calculated from these first signal, second signal, and third signal were input into a trained model, and a test for identifying the sample gas was conducted. The machine learning algorithm of the trained model was Random Forest.

[0068] The experimental results of Experiment 1 were as shown in (a) of Figure 7. (a) of Figure 7 is a graph showing the experimental results of Experiment 1. As shown in (a) of Figure 7, the accuracy rate of the trained model in Comparative Example 1 was 90%, the accuracy rate of the trained model in Comparative Example 2 was 88%, and the accuracy rate of the trained model in Comparative Example 3 was 77%. On the other hand, the accuracy rate of the trained model in Example 1 was 92%.

[0069] From the above, it was confirmed that the accuracy rate of the trained model in Example 1 was improved compared to Comparative Examples 1 to 3.

[0070] [1-4-2. Experiment 2] Furthermore, in order to confirm the above-mentioned effects, the following experiment 2 was carried out.

[0071] In Experiment 2, the above-mentioned sample gases A to C were used as sample gases, as in Experiment 1.

[0072] In Comparative Example 4, an odor sensor maintained at a single temperature of 30°C was exposed to sample gases A to C, and the signals output from the odor sensor were acquired 25 times for each. The feature values ​​calculated from these signals were input into a trained model, and a test to identify the sample gas was conducted. The machine learning algorithm used in the trained model was LightGBM.

[0073] In Comparative Example 5, an odor sensor maintained at a single temperature of 40°C was exposed to sample gases A to C, and the signals output from the odor sensor were acquired 25 times for each. The features calculated from these signals were input into a trained model, and a test was conducted to identify the sample gas. The machine learning algorithm used in the trained model was LightGBM.

[0074] In Comparative Example 6, an odor sensor maintained at a single temperature of 50°C was exposed to sample gases A to C, and the signals output from the odor sensor were acquired 25 times for each. The features calculated from these signals were input into a trained model, and a test was conducted to identify the sample gas. The machine learning algorithm used in the trained model was LightGBM.

[0075] In Example 2, an odor sensor was sequentially held at three different temperatures: 30°C, 40°C, and 50°C, and then exposed to sample gases A to C. The first, second, and third signals output from the odor sensor were acquired 25 times each. Feature quantities calculated from these first, second, and third signals were input into a trained model, and a test for identifying the sample gas was conducted. The machine learning algorithm of the trained model was LightGBM.

[0076] The experimental results of Experiment 2 are shown in (b) of Figure 7. (b) of Figure 7 is a graph showing the experimental results of Experiment 2. As shown in (b) of Figure 7, the accuracy rate of the trained model in Comparative Example 4 was 85%, the accuracy rate of the trained model in Comparative Example 5 was 91%, and the accuracy rate of the trained model in Comparative Example 6 was 80%. On the other hand, the accuracy rate of the trained model in Example 2 was 93%.

[0077] From the above, it was confirmed that in Example 2, even when the machine learning algorithm of the trained model was changed from Experiment 1, the accuracy rate of the trained model was improved compared to Comparative Examples 4 to 6.

[0078] (Embodiment 2) [2-1. Configuration of odor identification system] Next, the configuration of an odor identification system 2A according to embodiment 2 will be described with reference to Figures 8 and 9. Figure 8 is a block diagram showing the configuration of the odor identification system 2A according to embodiment 2. Figure 9 is a schematic diagram showing the configuration of an exposure unit 4A of the odor identification system 2A according to embodiment 2. Note that in this embodiment, the same components as those in embodiment 1 above are assigned the same reference numerals, and their description will be omitted.

[0079] 8 and 9, in the odor identification system 2A according to embodiment 2, the configurations of the exposure unit 4A and the control unit 6A are different from those of embodiment 1. Furthermore, while the odor identification system 2 according to embodiment 1 includes one odor sensor 8, the odor identification system 2A according to embodiment 2 includes a first odor sensor 8a, a second odor sensor 8b, and a third odor sensor 8c (an example of an odor sensor).

[0080] The exposure section 4A has a first container 14a, a first switching valve 16a, a first heater 20a (an example of a temperature control element), a first temperature sensor 22a, a second container 14b, a second switching valve 16b, a second heater 20b (an example of a temperature control element), a second temperature sensor 22b, a third container 14c, a third switching valve 16c, a third heater 20c (an example of a temperature control element), a third temperature sensor 22c, and a pump 18.

[0081] The first container 14a accommodates a first odor sensor 8a, a first heater 20a, and a first temperature sensor 22a.

[0082] The first switching valve 16a is a three-way solenoid valve for switching the gas introduced into the first container 14a, and has a first input port 24a, a second input port 26a, and an output port 28a. The first input port 24a is connected to a sample gas supply source 32 via a pipe 30a. The second input port 26a is connected to a reference gas supply source 36 via a pipe 34a. The output port 28a is connected to the interior of the first container 14a via a pipe 38a.

[0083] The first odor sensor 8a is exposed to the sample gas introduced into the first container 14a, and outputs a first signal corresponding to the adsorption concentration of odor molecules contained in the sample gas.

[0084] The first heater 20a is disposed so as to be in contact with the first odor sensor 8a. When the first heater 20a is energized, heat from the first heater 20a is transferred to the first odor sensor 8a, thereby heating the first odor sensor 8a. By heating the first odor sensor 8a, the first heater 20a maintains the temperature of the first odor sensor 8a at a first temperature (e.g., 30°C).

[0085] The first temperature sensor 22a is disposed near the first odor sensor 8a. The first temperature sensor 22a detects the temperature of the first odor sensor 8a and outputs temperature data indicating the detected temperature to the control unit 6A.

[0086] The second container 14b accommodates a second odor sensor 8b, a second heater 20b, and a second temperature sensor 22b.

[0087] The second switching valve 16b is a three-way solenoid valve for switching the gas introduced into the second container 14b, and has a first input port 24b, a second input port 26b, and an output port 28b. The first input port 24b is connected to a sample gas supply source 32 via a pipe 30b. The second input port 26b is connected to a reference gas supply source 36 via a pipe 34b. The output port 28b is connected to the interior of the second container 14b via a pipe 38b.

[0088] The second odor sensor 8b is exposed to the sample gas introduced into the second container 14b, and outputs a second signal corresponding to the adsorption concentration of odor molecules contained in the sample gas.

[0089] The second heater 20b is disposed so as to be in contact with the second odor sensor 8b. When the second heater 20b is energized, heat from the second heater 20b is transferred to the second odor sensor 8b, thereby heating the second odor sensor 8b. By heating the second odor sensor 8b, the second heater 20b maintains the temperature of the second odor sensor 8b at a second temperature (e.g., 40°C).

[0090] The second temperature sensor 22b is disposed near the second odor sensor 8b. The second temperature sensor 22b detects the temperature of the second odor sensor 8b and outputs temperature data indicating the detected temperature to the control unit 6A.

[0091] The third container 14c accommodates a third odor sensor 8c, a third heater 20c, and a third temperature sensor 22c.

[0092] The third switching valve 16c is a three-way solenoid valve for switching the gas introduced into the third container 14c, and has a first input port 24c, a second input port 26c, and an output port 28c. The first input port 24c is connected to a sample gas supply source 32 via a pipe 30c. The second input port 26c is connected to a reference gas supply source 36 via a pipe 34c. The output port 28c is connected to the interior of the third container 14c via a pipe 38c.

[0093] The third odor sensor 8c is exposed to the sample gas introduced into the third container 14c, and outputs a third signal corresponding to the adsorption concentration of odor molecules contained in the sample gas.

[0094] The third heater 20c is disposed so as to be in contact with the third odor sensor 8c. Thus, when the third heater 20c is energized, heat from the third heater 20c is transferred to the third odor sensor 8c, thereby heating the third odor sensor 8c. By heating the third odor sensor 8c, the third heater 20c maintains the temperature of the third odor sensor 8c at a third temperature (e.g., 50°C).

[0095] The third temperature sensor 22c is disposed near the third odor sensor 8c. The third temperature sensor 22c detects the temperature of the third odor sensor 8c and outputs temperature data indicating the detected temperature to the control unit 6A.

[0096] The intake side of the pump 18 is connected to the interiors of the first container 14a, the second container 14b, and the third container 14c via pipes 40a, 40b, and 40c, respectively. The exhaust side of the pump 18 is connected to an exhaust duct (not shown) via a pipe (not shown). The pump 18 introduces a sample gas or a reference gas into the first container 14a and discharges the sample gas or reference gas introduced into the first container 14a to the exhaust duct. The pump 18 also introduces a sample gas or a reference gas into the second container 14b and discharges the sample gas or reference gas introduced into the second container 14b to the exhaust duct. The pump 18 also introduces a sample gas or a reference gas into the third container 14c and discharges the sample gas or reference gas introduced into the third container 14c to the exhaust duct.

[0097] The exposure unit 4A simultaneously exposes the first odor sensor 8a, the second odor sensor 8b, and the third odor sensor 8c to the sample gas during the first period T1 of a period Tm (not shown) consisting of a first period T1 and a second period T2 following the first period T1. Furthermore, the exposure unit 4 simultaneously exposes the first odor sensor 8a, the second odor sensor 8b, and the third odor sensor 8c to the reference gas during the second period T2 of the period Tm.

[0098] The control unit 6A controls the operation of the first selector valve 16a, the second selector valve 16b, the third selector valve 16c, and the pump 18 of the exposure unit 4A. Specifically, during a first period T1 within the period Tm, the control unit 6A operates the pump 18 and simultaneously switches the first selector valve 16a, the second selector valve 16b, and the third selector valve 16c to the first state. During a second period T2 within the period Tm, the control unit 6A operates the pump 18 and simultaneously switches the first selector valve 16a, the second selector valve 16b, and the third selector valve 16c to the second state.

[0099] Furthermore, the control unit 6A controls the driving of the first heater 20a based on the temperature data from the first temperature sensor 22a so as to maintain the temperature of the first odor sensor 8a at 30° C. At the same time, the control unit 6A controls the driving of the second heater 20b based on the temperature data from the second temperature sensor 22b so as to maintain the temperature of the second odor sensor 8b at 40° C. At the same time, the control unit 6A controls the driving of the third heater 20c based on the temperature data from the third temperature sensor 22c so as to maintain the temperature of the third odor sensor 8c at 50° C.

[0100] [2-2. Operation of Odor Identification System] Next, the operation of the odor identification system 2A according to the second embodiment will be described with reference to Fig. 10. Fig. 10 is a flowchart showing the flow of the operation of the odor identification system 2A according to the second embodiment.

[0101] 10 , first, during the period Tm, the first heater 20a heats the first odor sensor 8a, thereby maintaining the temperature of the first odor sensor 8a at 30° C. (S201). At the same time, during the period Tm, the second heater 20b heats the second odor sensor 8b, thereby maintaining the temperature of the second odor sensor 8b at 40° C. (S201). At the same time, during the period Tm, the third heater 20c heats the third odor sensor 8c, thereby maintaining the temperature of the third odor sensor 8c at 50° C. (S201).

[0102] Next, during the period Tm, the exposure unit 4A simultaneously exposes the first odor sensor 8a, the second odor sensor 8b, and the third odor sensor 8c to the sample gas (S202), and then simultaneously exposes the first odor sensor 8a, the second odor sensor 8b, and the third odor sensor 8c to the reference gas (S203).

[0103] Next, during the period Tm, the acquisition unit 10 simultaneously acquires the first signal output from the first odor sensor 8a held at 30°C, the second signal output from the second odor sensor 8b held at 40°C, and the third signal output from the third odor sensor 8c held at 50°C (S204), and outputs the acquired first signal, second signal, and third signal to the identification unit 12.

[0104] Next, the identifying unit 12 identifies the odor of the sample gas based on the first signal, the second signal, and the third signal acquired by the acquiring unit 10 (S205).

[0105] [2-3. Effects] In the odor identification system 2A according to the second embodiment, the first heater 20a, the second heater 20b, and the third heater 20c simultaneously maintain the temperatures of the first odor sensor 8a, the second odor sensor 8b, and the third odor sensor 8c at 30°C, 40°C, and 50°C, respectively.

[0106] This allows the acquisition unit 10 to simultaneously acquire the first signal, the second signal, and the third signal output from the first odor sensor 8 a, the second odor sensor 8 b, and the third odor sensor 8 c, respectively, thereby enabling the odor of the sample gas to be identified in a relatively short time.

[0107] While the odor identification system according to one or more aspects has been described above based on the above-described embodiments, the present disclosure is not limited to the above-described embodiments. As long as it does not deviate from the spirit of the present disclosure, various modifications conceivable by a person skilled in the art to the above-described embodiments, or configurations constructed by combining components of different embodiments, may also be included within the scope of one or more aspects.

[0108] In each of the above embodiments, each component may be configured with dedicated hardware, or may be realized by executing a software program suitable for each component. Each component may be realized by a program execution unit such as a CPU or processor reading and executing a software program recorded on a recording medium such as a hard disk or semiconductor memory.

[0109] Furthermore, some or all of the functions of the odor identification system according to each of the above embodiments may be realized by a processor such as a CPU executing a program.

[0110] Some or all of the components constituting each of the above devices may be configured as an IC card or a standalone module that can be attached to each device. The IC card or module is a computer system composed of a microprocessor, ROM, RAM, etc. The IC card or module may include the above-mentioned ultra-multifunctional LSI. The IC card or module achieves its functions when the microprocessor operates according to a computer program. The IC card or module may be tamper-resistant.

[0111] The odor identification system according to the present disclosure is useful, for example, in a system for inspecting food for off-flavors on a food production line.

[0112] 2, 2A Odor identification system 4, 4A Exposure unit 6, 6A Control unit 8 Odor sensor 8a First odor sensor 8b Second odor sensor 8c Third odor sensor 10 Acquisition unit 12 Identification unit 14 Container 14a First container 14b Second container 14c Third container 16 Switching valve 16a First switching valve 16b Second switching valve 16c Third switching valve 18 Pump 20 Heater 20a First heater 20b Second heater 20c Third heater 22 Temperature sensor 22a First temperature sensor 22b Second temperature sensor 22c Third temperature sensor 24, 24a, 24b, 24c First input port 26, 26a, 26b, 26c Second input port 28, 28a, 28b, 28c: output ports 30, 30a, 30b, 30c, 34, 34a, 34b, 34c, 38, 38a, 38b, 38c, 40, 40a, 40b, 40c: piping 32: sample gas supply source 36: reference gas supply source 42: sensitive element 44: sensing section 46, 48: electrodes

Claims

1. An odor identification method for identifying the odor of a sample gas using an odor sensor that outputs a signal corresponding to the adsorption concentration of odor molecules contained in the sample gas, comprising: (a) a step of maintaining the temperature of the odor sensor at a first temperature by heating or cooling the odor sensor; (b) a step of acquiring a first signal output from the odor sensor when the odor sensor maintained at the first temperature is exposed to the sample gas; (c) a step of maintaining the temperature of the odor sensor at a second temperature different from the first temperature by heating or cooling the odor sensor; (d) a step of acquiring a second signal output from the odor sensor when the odor sensor maintained at the second temperature is exposed to the sample gas; and (e) a step of identifying the odor of the sample gas using at least the first signal and the second signal.

2. The odor identification method of claim 1, wherein the odor sensor includes one specific odor sensor, and in (a), the temperature of the specific odor sensor is maintained at the first temperature, and in (b), after (a), the specific odor sensor maintained at the first temperature is exposed to the sample gas, and the first signal output from the specific odor sensor is obtained, and in (c), after (b), the temperature of the specific odor sensor is maintained at the second temperature, and in (d), after (c), the specific odor sensor maintained at the second temperature is exposed to the sample gas, and the second signal output from the specific odor sensor is obtained.

3. The odor identification method according to claim 1, wherein the odor sensor includes a first odor sensor and a second odor sensor, and in (a) and (c), the temperature of the first odor sensor is maintained at the first temperature, and simultaneously the temperature of the second odor sensor is maintained at the second temperature, and in (b) and (d), the first signal output from the first odor sensor is acquired when the first odor sensor maintained at the first temperature is exposed to the sample gas, and simultaneously the second signal output from the second odor sensor is acquired when the second odor sensor maintained at the second temperature is exposed to the sample gas.

4. An odor identification system comprising: an odor sensor that outputs a signal corresponding to the adsorption concentration of odor molecules contained in a sample gas; a temperature control element that maintains the temperature of the odor sensor at a first temperature and a second temperature by heating or cooling the odor sensor; an acquisition unit that acquires a first signal output from the odor sensor when the odor sensor maintained at the first temperature by the temperature control element is exposed to the sample gas, and acquires a second signal output from the odor sensor when the odor sensor maintained at the second temperature by the temperature control element is exposed to the sample gas; and an identification unit that identifies the odor of the sample gas using at least the first signal and the second signal.

Images