Odor measurement device and odor measurement method

Abstract

The present invention makes it possible to obtain more detailed odor information about the distribution, size, density, or the like of clumps of an odor substance in a sample gas that conventional odor measurement devices have not been able to measure. An odor measurement device according to the present invention comprises a flow passage that has an inlet and an outlet, a flow rate control means that introduces a sample gas into the flow passage through the inlet, causes the sample gas to be discharged through the outlet, and controls the flow rate of the sample gas inside the flow passage, at least one type of odor sensor that is provided inside the flow passage, and a computation unit that derives or calculates vibration information that indicates a vibration state for a detection value from the odor sensor and uses the vibration information as a parameter to calculate odor information that is information about an odor.

Description

Odor measuring device and odor measuring method

[0001] The present invention relates to an odor measuring device and an odor measuring method.

[0002] For example, when measuring odors in the air, air is passed through a chamber in which one or more types of odor sensors are placed, and the type and concentration (intensity of odor) of odor substances contained in the introduced air are measured based on the detection values ​​of the odor sensors.

[0003] Conventionally, it has been thought that odor substances are dispersed evenly throughout the atmosphere, and that when air containing these odor substances is passed through a chamber, the output value of the odor sensor changes very slowly over time.

[0004] Therefore, if the detected value of the odor sensor vibrates at a frequency higher than expected during odor measurement, this has traditionally been considered noise, and vibration cancellation processing has been performed using hardware or software (such as installing a low-pass filter or performing smoothing calculations).

[0005] Japanese Patent Application Laid-Open No. 2000-019137

[0006] However, the present inventors discovered for the first time that vibrations in the detection values ​​of odor sensors, which had previously been considered noise, contribute to odor measurement, and have completed the present invention.

[0007] That is, the odor measurement device and odor measurement method according to the present invention are as follows: [1] An odor measurement device comprising: a flow passage having an inlet and an outlet; a flow control means for introducing a sample gas through the inlet of the flow passage and discharging it through the outlet, and for controlling the flow rate of the sample gas within the flow passage; one or more types of odor sensors disposed within the flow passage; and a calculation unit for extracting or calculating vibration information indicating the vibration pattern of the detected value of the odor sensor, and calculating odor information, which is information related to the odor, using the vibration information as a parameter. [2] The odor measurement device according to [1], wherein the vibration information is a value indicating the peak and / or width of each or any oscillation, or a value calculated from this value. [3] The odor measurement device according to [1] or [2], wherein the odor information includes one or more of the distribution, size, and density of odor substance clusters in the sample gas. [4] The odor measurement device according to any one of [1] to [3], wherein the flow control means is disposed downstream of the odor sensor and draws the sample gas through the inlet. [5] The odor measurement device according to any one of [1] to [4], wherein the flow path is cylindrical, having the inlet at one end and the outlet at the other end, and the cross-sectional contour shape of the flow path and the contour shape of the inlet (and the contour shape of the outlet) are set to be approximately equal. [6] The odor measurement device according to any one of [1] to [5], wherein the odor sensor is disposed along the inner wall surface of the flow path. [7] The odor measurement device according to any one of [1] to [6], wherein the flow rate control means controls the flow velocity of the sample gas flowing inside the flow path to a constant value of 0.01 m / sec or more and 10 m / sec or less. [8] The odor measurement device according to any one of [1] to [7], wherein the cross-sectional contour shape of the flow path and the cross-sectional contour shape of the outlet have a length dimension in the short dimension direction of 1 mm or more. [9] The odor measuring device according to any one of [1] to [8], further comprising a sheet member having a large number of holes formed in the inlet, the diameter of the holes in the sheet member being 0.1 mm or more.

[10] The odor measurement device according to any one of [1] to [9], wherein the calculation unit generates a model that associates vibration information with odor information through machine learning, and calculates the odor information from the vibration information obtained from the sample gas to be measured based on this model.

[11] The odor measurement device according to any one of [1] to

[10] , further comprising a removal filter unit that removes frequency components outside a predetermined range from time-series changes in the detection values ​​of the odor sensor, and the calculation unit extracts the vibration pattern from the time-series changes in the detection values ​​of the odor sensor that have passed through the removal filter unit.

[12] An odor measurement method comprising: circulating a sample gas through a flow path having an inlet and an outlet; acquiring detection values ​​from one or more types of odor sensors arranged in the flow path; extracting vibration information that indicates the vibration pattern of the detection values; and calculating odor information that is information related to the odor using this vibration information as a parameter.

[0008] According to the present invention, it is possible to obtain more detailed odor information such as the distribution, size, or density of odorant clumps in a sample gas, which could not be measured with conventional odor measuring devices, thereby determining the type and intensity of the odor.

[0009] 1 is a schematic diagram of an overall odor measurement device according to one embodiment of the present invention. 2 is a schematic diagram showing the overall structure of the odor measurement device according to this embodiment. 3 is an example of a signal value from a sensor obtained by the odor measurement device according to this embodiment. 4 is a schematic diagram showing an example of an odor sensor provided in the odor measurement device according to this embodiment. 5 is a schematic diagram showing the structure of an odor measurement device according to another embodiment of the present invention. 6 is a schematic diagram showing the structure of an odor measurement device according to another embodiment of the present invention. 7 is a graph showing the distribution of indices related to odor measurement according to one embodiment of the present invention. 8 is a graph showing the distribution of indices related to odor measurement according to one embodiment of the present invention.

[0010] An embodiment of the present invention will be described below with reference to the drawings.

[0011] 1 , the odor measurement device 100 according to the present embodiment includes a flow channel 1 having an inlet 1 a and an outlet 1 b, a flow control means 2 that introduces a sample gas through the inlet 1 a of the flow channel 1 and discharges it through the outlet 1 b, and controls the flow rate of the sample gas within the flow channel 1, one or more types of odor sensors 3 disposed within the flow channel 1, and a calculation unit 4 that calculates odor information, which is information related to odors, based on detection values ​​from the odor sensors 3. The odor measurement device 100 according to the present embodiment is a tabletop-sized device, for example, in the shape of a cube or rectangular parallelepiped with each side measuring 20 cm or less, that can be carried by a user in one hand and used outdoors to measure odors present in the environment.

[0012] The flow channel 1 accommodates one or more odor sensors 3 inside, and in this embodiment, as shown in Figure 2, a large number of odor sensors 3 can be arranged inside in a matrix, and is, for example, cylindrical with an inlet 1a at one end and an outlet 1b at the other end.

[0013] In this embodiment, the flow path 1 is formed inside the rectangular parallelepiped housing C so as to connect an inlet 1a and an outlet 1b formed on each side of the housing C, as shown in Fig. 1 as an example. In this embodiment, the contour shape of the inlet 1a and the cross-sectional contour shape of the flow path 1 from the inlet 1a to the outlet 1b are set to be approximately equal to each other so as to prevent turbulence in the flow of the sample gas flowing inside the flow path 1 from breaking up the odorant clumps contained in the sample gas, and to measure the odorant clumps while maintaining their presence in the atmosphere as much as possible. In this way, by making the contour shape of the inlet 1a and the cross-sectional contour shape of the flow path 1 from the inlet 1a to the outlet 1b approximately equal, the flow of air flowing inside the flow path 1 becomes laminar, making it less likely that large turbulence will break up the odorant clumps inside the flow path 1.

[0014] Although research is still ongoing into the size of the clumps of odorant contained in the sample gas, one theory is that they are small enough to come into contact with the antennae of an insect for 10 to 100 milliseconds, which is thought to be significantly smaller than 1 mm. Therefore, in order to prevent these clumps of odorant from colliding with the edges of the inlet 1a, the inner wall of the flow path 1, the edges of the outlet 1b, etc. and breaking up, it is preferable that the short-side length dimensions of the outline shape of the inlet 1a, the cross-sectional outline shape of the flow path 1, and the outline shape of the outlet 1b be 1 mm or more.

[0015] In order to avoid breaking up the clumps of odorous substances contained in the sample gas as much as possible, it is preferable that the odor sensors 3 do not protrude significantly from the inner wall surface of the flow passage 1 toward the inside of the flow passage 1, and it is even more preferable that each odor sensor 3 is arranged as flat as possible, for example, so that it is aligned along the inner wall surface of the flow passage 1, so that the surface of the odor sensor 3 is flush with the inner wall surface.

[0016] The flow rate control means 2 controls the flow rate of the sample gas in the flow passage 1, for example, by promoting or suppressing the flow of the sample gas. The flow rate control means 2 may be provided inside or outside the flow passage 1.

[0017] In this embodiment, a fan disposed inside the flow path 1 functions as a flow control means 2, and the rotation speed of the fan controls the flow rate of the sample gas flowing through the flow path 1. In order to detect the odor substance clumps contained in the sample gas as intact as possible using the odor sensor, it is preferable that the flow control means 2 is provided downstream of the odor sensor 3 in the flow path 1 and draws the sample gas into the flow path 1 from the inlet 1a.

[0018] If the flow rate of the sample gas flowing through the flow passage 1 is too fast, there is a risk that the clumps of odorous substances contained in the sample gas will be broken down. Therefore, it is preferable that the flow control means 2 controls the flow rate of the sample gas flowing inside the flow passage 1 to a constant value of 0.01 m / sec or more and 10 m / sec or less, and it is more preferable to set it in the range of 0.1 m / sec or more and 1 m / sec or less.

[0019] The system may further include a flow control unit 5 that controls the operation of the flow control means 2. This flow control unit 5 is physically one or more general-purpose computers COM that have analog electrical circuits including buffers, amplifiers, etc., digital electrical circuits including a CPU, memory, DSP, etc., and an A / D converter or the like interposed between them, and this computer COM is configured to function as the flow control unit 5 by the CPU and its peripheral devices working together in accordance with a predetermined program stored in the memory.

[0020] The odor sensor 3 is not particularly limited as long as it can detect odor substances, and may be a semiconductor gas sensor, a gas sensor using an organic polymer, an alcohol detection device, etc. For example, the odor sensor 3 may be a commercially available gas sensor such as those listed below. MICS6814 (manufactured by SGX SENSORTECH), MICS5914 (manufactured by SGX SENSORTECH), MICS5524 (manufactured by SGX SENSORTECH), MQ-137 (manufactured by Winsen), SGP40 (manufactured by Sensirion), TGS2620 (manufactured by Figaro Giken), IR12GM (manufactured by SGX SENSORTECH), EC4-10-ETO (manufactured by SGX SENSORTECH), FECS44-100 (manufactured by Figaro Giken), MP7227 (manufactured by SGX SENSORTECH), VQ31MB (manufactured by SGX SENSORTECH), PS1HCH5 (manufactured by SGX SENSORTECH). The odor sensor may include only one of the above-described odor sensors, or multiple of one type. Alternatively, multiple types of odor sensors with different responses to odorants may be included. In this embodiment, as an example of such a sensor, a gas sensor is used that includes an odorant-receiving layer whose electrical conductivity changes when an odorant is adsorbed. Details of this gas sensor will be described later.

[0021] The calculation unit 4 calculates odor information based on the detection value of the odor sensor 3, and is configured, for example, such that the aforementioned computer COM, in cooperation with a CPU and its peripheral devices, executes the functions of the calculation unit 4 in accordance with a predetermined program stored in memory. The calculation unit 4 may be located entirely outside the housing C, or partly or entirely within the housing C.

[0022] Here, a graph showing the detection values ​​output over time from one odor sensor 3 arranged in the flow path 1 is characterized by large fluctuations in the detection values, as shown in Figure 3. The inventors believe that the reason such a graph is obtained is that odor substances exist as clumps in the sample gas, such as the atmosphere, and these clumps of odor components approach the odor sensor in sequence and are detected by the odor sensor.

[0023] Therefore, in this embodiment, the calculation unit 4 extracts or calculates vibration information that indicates the vibration pattern of the detection value of the odor sensor 3, and calculates odor information, which is information related to the odor, using this vibration information as a parameter.

[0024] <Odor Measurement Method Using the Odor Measurement Device According to the Present Embodiment> A method for measuring an odor using the odor measurement device 100 configured as described above will be described below. A sample gas is circulated through a flow path 1 having an inlet 1a and an outlet 1b, and detection values ​​of one or more types of odor sensors 3 arranged in the flow path 1 are obtained.

[0025] The calculation unit 4 generates a model that associates vibration information with odor information, for example, by machine learning, and calculates the odor information from the vibration information obtained from the sample gas that is the measurement target, based on this model.

[0026] The above-mentioned model is created, for example, by the following method. First, a model gas containing a predetermined concentration of a known type of odor substance is supplied to the flow path 1 of the odor measurement device 100 according to this embodiment, and a graph is obtained with the detection value by the odor sensor at that time on the vertical axis and time on the horizontal axis. As described above, this graph has a complex waveform, as shown in FIG. 3 . The calculation unit 4 does not average or smooth this graph waveform, but instead uses the vibration pattern of the graph waveform as is to calculate vibration information representing this vibration pattern from this vibration pattern. This vibration information is, for example, one or more of values ​​indicating the peak and / or amplitude of each or any of the vibrations included in the graph waveform of the detection value as shown in FIG. 3 , or values ​​calculated from these values.

[0027] A set of data (teacher data) is generated that links information equivalent to odor information about the sample gas used when this vibration information was obtained (information regarding the type of (known) odor substance contained in the sample gas, the concentration of the odor substance, the distance between the source of the odor substance and the odor sensor 3, the direction of the source relative to the odor sensor 3, etc.) with the vibration information (also called known vibration information), and is stored in a memory unit (not shown).

[0028] Once the training data has been accumulated in the memory unit as described above, a machine learning model generation unit (not shown) generates a machine learning model for calculating odor information based on vibration information, based on the training data accumulated in the memory unit.

[0029] After the machine learning model is generated in this manner, a detection value is obtained from the odor sensor for the sample gas, which is the target for which the odor is actually to be measured, and the calculation unit 4 calculates the vibration information (also called unknown vibration information) based on this detection value, and calculates the odor information for the sample gas based on the calculated unknown vibration information and the machine learning model.

[0030] The calculation unit 4 may further calculate the type and intensity of the odor based on the odor information obtained as described above, and the information indicating the type and intensity of the odor calculated by the calculation unit 4 may be output, for example, to a presentation unit (not shown), which may then present this information to the user.

[0031] <Gas sensor> Here, the odor sensor 3 used in this embodiment is a gas sensor that includes a sensor substrate 31, an odorant receiving layer 32 formed on the sensor substrate 31, and metal wiring 33 for electrically connecting the odorant receiving layer 32 to, for example, a voltmeter (not shown) that detects the potential change of the odorant receiving layer 32 as a signal value, as shown in Figure 4, for example.

[0032] [Substrate] A wide variety of substrates commonly used in electronic circuits can be used as the sensor substrate 31. Specifically, a substrate made of one or more materials selected from the group consisting of glass epoxy, paper, and glass cloth can be used.

[0033] [Odorant Receiving Layer] The odorant receiving layer 32 preferably contains, for example, a resin composition whose electrical conductivity differs when odorant a is adsorbed and when odorant b, a substance different from odorant a, is adsorbed, and is made of this resin composition.

[0034] The resin composition contains, for example, a resin (A) and a conductive carbon material (B).

[0035] (Resin (A)) The resin (A) contained in the resin composition according to one embodiment of the present invention is not particularly limited, but preferably contains at least one selected from the group consisting of urethane resins, polyalkylene oxides, acrylic resins, fluorine-containing resins, vinyl polymer resins, silicone resins, polyamide resins, polyester resins, epoxy resins, phenolic resins, phenylene oxides, and polyimides.

[0036] (Conductive Carbon Material (B)) The conductive carbon material (B) is, for example, a carbon material having a volume resistivity of 0.1 Ω cm or less. This conductive carbon material (B) is dispersed in the resin composition, and the conductive carbon materials (B) come into contact with each other to form conductive paths, thereby imparting conductivity to the resin composition.

[0037] Specific examples of the conductive carbon material (B) include carbon black, carbon nanotubes, and graphene.

[0038] The conductive carbon material (B) is preferably in the form of fibers or spheres.

[0039] When the conductive carbon material (B) is fibrous, the fiber diameter is preferably 0.1 μm or more and 10 μm or less, more preferably 0.1 μm or more and 5 μm or less. When the conductive carbon material (B) is fibrous, the fiber length is preferably 0.1 μm or more and 10 μm or less, more preferably 1 μm or more and 10 μm or less.

[0040] When the conductive carbon material (B) is spherical, the primary particle diameter is preferably 10 nm or more and 200 nm or less, more preferably 20 nm or more and 150 nm or less, and more preferably 100 nm or less, since this can further improve the conductivity and sensor sensitivity in the resin composition.

[0041] The primary particle diameter of the conductive carbon material (B) can be measured, for example, by observing the material with a transmission electron microscope (TEM) and analyzing the image using an image processing device (for example, a digital microscope VHX-700F manufactured by Keyence Corporation). The primary particle diameter of the conductive carbon material (B) can also be determined by other known methods. Furthermore, when the conductive carbon material (B) is a known material or a commercially available product, the primary particle diameter may be a literature value, a catalog value, or the like.

[0042] The content of the conductive carbon material (B) is preferably 5% by weight or more and 60% by weight or less, relative to 100% by weight of the total of the resin (A) and the conductive carbon material (B), from the viewpoint of ensuring that the sensor element formed from the resin composition exhibits sufficient conductivity as an odor sensor and sufficient sensitivity as the odor sensor.

[0043] The resin composition may contain, for example, a surfactant in addition to the resin (A) and conductive carbon material (B) described above, as long as the effects of the present invention are obtained. The surfactant preferably acts as a dispersant for the conductive carbon material (B), and one or more surfactants may be appropriately selected from anionic surfactants, cationic surfactants, amphoteric surfactants, and nonionic surfactants. The surfactant may be either compatible or incompatible with the resin (A). The resin composition may further contain components other than the components described above, and the other components may be suitably used as long as both the effects of the present invention and the effects of the other components are obtained.

[0044] [Metal Wiring] The metal wiring 33 is arranged so as to be in contact with the odorant receiving layer 32 described above, and includes, for example, a first metal wiring and a second metal wiring.

[0045] These first and second metal wirings are preferably made of copper, gold, or the like, and preferably have a flat cross-sectional shape.

[0046] The width of each of the first and second metal wirings as viewed perpendicular to the surface of the substrate is preferably 10 μm to 2 mm, more preferably 10 μm to 1 mm, and the height, i.e., thickness, of each of the first and second metal wirings as viewed parallel to the surface of the substrate is preferably 1 μm to 100 μm, more preferably 10 μm to 50 μm.

[0047] It is preferable that the first metal wiring and the second metal wiring are not in direct contact with each other and are arranged substantially parallel to each other.

[0048] As described above, the distance between the first metal wiring and the second metal wiring arranged approximately in parallel is preferably 1 μm to 3 mm, more preferably 1 μm to 1.5 mm.

[0049] It is desirable that the distance between the first metal wiring and the second metal wiring be less than a predetermined distance (e.g., 500 μm) when the electrical conductivity of the odorant receiving layer 32 (i.e., the electrical conductivity of the sensor element 3) is low.

[0050] The length of the portion of the first metal wiring and the second metal wiring that is in contact with the odorant receiving layer is preferably 100 μm or more and 50 mm or less, and more preferably 500 μm or more and 30 mm or less.

[0051] <Effects of this embodiment> The odor measurement device and odor measurement method of this embodiment can obtain not only information such as the type and intensity of odor that can be measured with conventional odor measurement devices, but also odor information that is closer to the sensations that humans and animals have when sensing odors in their environment (such as what kind of odor is coming from which direction).

[0052] Other Embodiments of the Invention The present invention is not limited to the above-described embodiments. For example, in the above-described embodiments, the shape of the flow path of the odor measuring device is shown as an example, but the shape may be bent at a different angle, or the flow path 1 may be formed linearly as shown in FIG. 5 .

[0053] As described above, in order to measure the odorant clumps as they are without breaking them down as much as possible, it is preferable to take in sample gas such as the atmosphere directly through the inlet as in the above embodiment, but for example, as shown in Figure 6, a sheet member 6 with multiple holes formed in the inlet may be further provided to prevent the intrusion of foreign matter through the inlet. The material and shape of the sheet member 6 are not particularly limited as long as they do not impede the flow of sample gas from the inlet into the flow channel, but if this sheet member 6 is provided, it is preferable that the diameter of the holes formed in the sheet member 6 be 0.1 mm or more so that the odorant clumps can be taken in as they are as much as possible.

[0054] The apparatus may further include a elimination filter unit that removes frequency components outside a predetermined range from the time-series change in the detection value of the odor sensor. In this case, the calculation unit may extract the vibration information from the time-series change in the detection value of the odor sensor that has passed through the elimination filter unit.

[0055] The calculation unit may also acquire output values ​​from measurement condition detection sensors 7, such as thermometers and hygrometers, disposed within the flow passage that detect measurement conditions for odor measurement, and appropriately correct the signal values ​​from the odor sensor based on the information obtained from these measurement condition detection sensors 7. These measurement condition detection sensors 7 may be disposed upstream or downstream of the odor sensor within the flow passage. Furthermore, some of the calculation and other functions performed by the calculation unit (e.g., determining the type of odor based on the odor information) may be performed manually by a human user. In addition, some or all of the above-described embodiments and modified embodiments may be combined as appropriate, and it goes without saying that various modifications are possible within the spirit and scope of the present invention.

[0056] The present invention will be described in more detail below using specific examples, but it goes without saying that the present invention is not limited to these. Here, as examples, the following experiments were conducted in cases where various types of odor substances (hereinafter also referred to as analytes) were present in the air. The analytes used in these examples were hexane, ethyl acetate, methanol, diethyl carbonate, and toluene.

[0057] For odor measurement, an odor measurement device was used, as described in Figure 1, which is equipped with an inlet for introducing air containing a sample (odor substance) and a fan for directing the sample from the inlet to the odor sensor at a constant speed. The sensor used was the sensor shown in Figure 4, which was described in the previous embodiment. In this example, a stainless steel mesh plate was used as the sheet member placed at the inlet.

[0058] A 2 cm filter paper was impregnated with 0.2 ml of sample, and the filter paper was placed 10 cm away from the inlet. Measurement was immediately started. The measurement time was 100 seconds, and the sample-impregnated filter paper was removed after 100 seconds. The voltage was measured at 1-second intervals. This odor measurement was repeated 100 times under the same conditions.

[0059] The electrical resistance R was calculated according to Ohm's law using the voltage measured at each time and the current value supplied from the constant current power supply. A graph was obtained by plotting the electrical resistance R versus time. In all of these graphs, the waveforms fluctuated significantly over time, as shown in Figure 3.

[0060] Next, vibration information indicating the vibration pattern was extracted from this graph waveform. In this example, values ​​indicating the peak and width of the vibration for each or any of the times, or values ​​calculated from these values, were calculated as the vibration information. Specifically, as the values ​​indicating the peak and width of the vibration for each or any of the times, a waveform index, which is an index indicating the shape of the graph waveform over a certain period of time, was calculated by directly using the vibration pattern of the graph waveform without averaging or smoothing the graph waveform based on the detection values ​​obtained from the values ​​calculated as described above. Furthermore, as the value calculated from the values ​​indicating the peak and width of the vibration for each or any of the times, a probability distribution index, which is an index based on the shape of the graph waveform over a certain period of time, was calculated.

[0061] As specific examples of the waveform index, one or more types selected from the group consisting of fractal dimensions such as Df and Lf, Hurst exponents, Lyapunov indices, autocorrelation functions, etc. can be suitably used.

[0062] Specific examples of the probability distribution index include one or more selected from the group consisting of Kullback-Leibler (KL) divergence, Jensen-Shannon (JS) divergence, Earth Mover's Distance (EMD), Hellinger distance, Bhattacharya distance, Maximum Mean Discrepancy (MMD), Shannon Entropy, Cross Entropy, mutual information, and Wasserstein distance.

[0063] In this example, the responsiveness of the sensor element to each sample was analyzed using the waveform index and probability distribution index described above using the k-nearest neighbor method, but this is not limited to this. Specifically, the odor sensors used in the odor measurement device were sensor elements (E-1) to (E-107) and comparative sensor elements (E'-1) to (E'-28). Data corresponding to the example measured using the sensor elements (E-1) to (E-107) and data corresponding to the comparative example measured using the comparative sensor elements (E'-1) to (E'-28) were obtained 100 times each for 5 samples, for a total of 500 times. Based on these data, the waveform index and probability distribution index were calculated using the method described above. Examples of the distributions of these indices are shown in Figures 7 to 9. As shown in Figures 7 to 9, it can be seen that the distributions of the waveform index and probability distribution index for each sample are clearly different.

[0064] The distribution data of the indices obtained in this way was randomly divided so that the number of training data: the number of test data = 80:20, and a classifier (learning model) was created for the training data using the k-nearest neighbor method.

[0065] The accuracy rates when classifying test data using classifiers created based on each example and comparative example as described above are shown in Table 1 as Examples 1 to 8. In this example, this accuracy rate was used as a performance index for the odor measurement device, and the higher the accuracy rate, the higher the performance of the odor measurement device was determined to be.

[0066]

[0067] As shown in Table 1, the accuracy rate was also checked when the configuration of the odor measurement device was changed in various ways. For example, when comparing Examples 3, 5, and 8, it was found that a accuracy rate of over 50% was achieved even when the diameter of the holes in the sheet member placed at the inlet of the odor measurement device was 0.04 mm, and that the accuracy rate tended to be further improved as the diameter was increased. Furthermore, according to the results of Examples 1 to 4, 6, and 7, a accuracy rate of approximately 50% was achieved even when the flow velocity of the analyte-containing air (sample gas) in the flow passage was 0.005 m / sec or 20 m / sec, but a velocity of 0.01 m / sec or higher was found to be more preferable.

[0068] According to the present invention, it is possible to obtain more detailed odor information such as the distribution, size, or density of odorant clumps in a sample gas, which could not be measured with conventional odor measuring devices, thereby determining the type and intensity of the odor.

[0069] REFERENCE SIGNS LIST 100: Odor measuring device 1: Flow path 1a: Inlet 1b: Outlet 2: Flow rate control means 3: Odor sensor 4: Calculation unit 5: Sheet member

Claims

1. An odor measuring device comprising: a flow passage having an inlet and an outlet; a flow control means for introducing sample gas into the flow passage through the inlet and discharging it from the outlet, and for controlling the flow rate of the sample gas within the flow passage; one or more types of odor sensors arranged within the flow passage; and a calculation unit for extracting or calculating vibration information that indicates the vibration pattern of the detected value of the odor sensor, and for calculating odor information, which is information related to the odor, using this vibration information as a parameter.

2. The odor measuring device according to claim 1, wherein the vibration information is a value indicating the peak and / or amplitude of the vibration at each or any one of the times, or a value calculated from this value.

3. The odor measuring device according to claim 1, wherein the odor information includes one or more of the distribution, size, and density of odor substance clusters in the sample gas.

4. The odor measuring device according to claim 1, wherein the flow rate control means is provided downstream of the odor sensor and draws in the sample gas through the inlet.

5. The odor measuring device described in claim 1, wherein the flow passage is cylindrical with the inlet at one end and the outlet at the other end, and the cross-sectional contour shape of the flow passage and the contour shape of the inlet (and the contour shape of the outlet) are set to be approximately equal.

6. The odor measuring device according to claim 1, wherein the odor sensor is disposed along the inner wall surface of the flow path.

7. An odor measuring device according to claim 1, wherein the flow rate control means controls the flow rate of the sample gas flowing inside the flow path to a constant value of 0.01 m / sec or more and 10 m / sec or less.

8. The odor measuring device according to claim 1, wherein the cross-sectional contour shape of the flow path and the cross-sectional contour shape of the outlet have a length dimension in the shorter direction of 1 mm or more.

9. The odor measuring device according to claim 1, further comprising a sheet member having a large number of holes formed at the inlet, the holes of the sheet member having a diameter of 0.1 mm or more.

10. The odor measurement device described in claim 1, wherein the calculation unit generates a model that associates vibration information with odor information through machine learning, and calculates the odor information from the vibration information obtained from the sample gas to be measured based on this model.

11. An odor measuring device as described in claim 1, further comprising an elimination filter unit that removes frequency components outside a predetermined range from the time-series changes in the detection value of the odor sensor, and the calculation unit extracts the vibration pattern from the time-series changes in the detection value of the odor sensor that have passed through the elimination filter unit.

12. An odor measurement method comprising: circulating a sample gas through a flow passage having an inlet and an outlet; obtaining detection values ​​from one or more types of odor sensors arranged within said flow passage; extracting vibration information indicating the vibration pattern of said detection values; and calculating odor information, which is information relating to the odor, using this vibration information as a parameter.

Images