JP4935668B2 - Odor measuring device - Google Patents

Description

The present invention relates to an odor measuring apparatus, and more particularly to an odor measuring apparatus that can evaluate the odor of a measurement sample by sensory expression.

Conventionally, the identification and evaluation of odors are generally performed using human sense of smell. However, since it is necessary to assume that the olfactory sensation varies according to the individual differences of the person who actually smells the odor (panelists) and the physical condition of the day, in order to obtain the measurement results with a high degree of accuracy, Also, sufficient consideration should be given to the environment of the test site. Therefore, it took enormous effort and time. Furthermore, even if done in this way, it is difficult to always make a definitive judgment based on a certain standard because human olfaction adapts to odor.

Therefore, in recent years, an odor measuring device using an odor sensor which is a kind of gas sensor having a response characteristic with respect to an odor substance has been developed (for example, see Patent Documents 1 and 2). In such an odor measuring device, based on detection signals acquired by a plurality of odor sensors, various multivariate analysis processing such as cluster analysis and principal component analysis, nonlinear analysis processing using a neural network, etc. The intensity of odor can be calculated.

Furthermore, an odor measuring device is also disclosed that can determine similarities with respect to a plurality of types of standard odors based on detection signals acquired by a plurality of odor sensors for the odor of the measurement sample (for example, Patent Document 3). reference). In such an odor measuring apparatus, when a measurement sample is measured by, for example, m odor sensors, the intensity signal is output from each odor sensor, so m measurement data are generated. These measurement data are represented mathematically by one point in the m-dimensional space (odor space). Therefore, when a plurality of samples are prepared so that the concentrations of the measurement samples are different and measured respectively, the point in the odor space moves in a certain direction from the origin as the concentration changes, so the points are connected. One line can be considered. This single line has a specific orientation corresponding to the type of odor of the measurement sample. Therefore, if the direction of the measurement result of the measurement sample and the measurement result of the standard sample are similar by storing the measurement result of the standard sample having the standard odor in advance in the odor space, On the other hand, if the directions are completely different, on the other hand, if the directions are completely different, the two are determined to be distant odors.

Specifically, in the odor space, when the measurement odor vector and the standard odor vector are created as the measurement results of the measurement sample and the standard sample, respectively, the angle formed by the measurement odor vector and the standard odor vector is calculated. . As a result, the similarity of the odor of the measurement sample to the standard odor based on the calculated angle is set to 100 when the angle is 0 °, and is set to 0 when the angle is greater than or equal to a predetermined value. , By setting in the range of 0-100. Further, an orthographic projection of the measured odor vector with respect to the standard odor vector is taken, and based on the length of the orthographic projection, the intensity of the reference odor included in the odor of the measurement sample is calculated.
Japanese Patent Laid-Open No. 11-352088 JP 2002-22692 A JP 2003-315298 A

In the odor measuring device described above, as a standard sample, for example, toluene as an aromatic odor, n-butanol as an alcohol odor, heptane as a hydrocarbon odor, ethyl acetate as an ester odor, etc. Trimethylamine as an amine odor, butyraldehyde as an aldehyde odor, methyl mercaptan as a sulfur odor, butyric acid as an organic acid odor, and the like have been used. Therefore, the similarity and intensity of the odor of the measurement sample were evaluated not by sensory expression but by gas component concentration display.

On the other hand, in the odor measuring device, for example, the evaluation is performed with a sensory expression such as a smoking odor instead of a gas component concentration display such as the similarity 50 to the aromatic odor and the intensity 50 of the aromatic odor. However, smoke odors and the like determined by sensory tests have not been put into practical use because they do not show a specific direction from the origin in the odor space.
Then, an object of this invention is to provide the odor measuring apparatus which can evaluate the odor of a measurement sample by sensory expression.

In order to solve the above problems, the inventors have prepared and measured a plurality of samples so that the intensity of the standard odor such as smoking odor by the sensory test is different, and in the odor space along with the intensity change We examined in which direction the point moves. Therefore, it has been found that the reference odor such as smoking odor determined by the sensory test moves in a certain direction from the origin in the odor space as the intensity changes.

In other words, the odor measuring apparatus of the present invention includes m (m is an integer of 2 or more) odor sensors having different response characteristics, and an m-dimensional space formed to represent the measurement results of the m odor sensors. A standard sample measuring unit for creating and storing n standard odor vectors based on the measurement results of n (n is an integer of 2 or more) standard samples having different types of standard odors, and the m-dimensional space The odor measuring device comprises a reference sample measuring unit that creates an odor vector based on a measurement result of a reference sample in which the strength of the reference odor is determined by a sensory test, wherein the reference sample measuring unit includes the reference odor A plurality of reference samples having different intensities are measured, a reference odor direction vector is created based on the positional relationship of the plurality of odor vectors, and the n standard odor vectors and the reference odor direction vector are generated. So that, characterized in that it comprises a calculation unit for storing the positional relationship between Torr.

According to the odor measuring apparatus of the present invention, an odor vector is created in an m-dimensional space by measuring a reference sample in which the strength of the reference odor is predetermined by a sensory test. Odor vectors are created based on the measurement results of a reference sample in which various intensities of the reference odor are determined by a sensory test. Thereby, in the m-dimensional space, for example, a reference odor direction vector indicating a change in the angle (direction) of the odor vector is created and stored.
Here, the “standard odor” includes, for example, a smoke odor, an odor of rotten eggs, and the like.

According to the odor measuring apparatus of the present invention, since the reference odor direction vector associated with the direction of the intensity of the reference odor by the sensory test is stored, by comparing the reference odor vector and the measurement result of the measurement sample, The odor of the measurement sample can be evaluated by sensory expression using the strength of the reference odor. It is also possible to evaluate small differences that could not be determined using human olfaction.

(Means and effects for solving other problems)
Further, in the odor measuring apparatus according to the present invention, when the standard sample measuring unit updates and stores n standard odor vectors, the positions of the stored n standard odor vectors and reference odor direction vectors are stored. A recalculation unit that calculates and stores the positional relationship between the updated n standard odor vectors and the reference odor direction vector based on the relationship may be provided.
According to the odor measuring apparatus of the present invention, the relationship between the standard odor vector indicating each type of standard odor and the reference odor direction vector indicating the intensity of the reference odor is constant. Therefore, when the standard sample measurement unit updates n standard odor vectors, the updated n number of standard odor vectors based on the stored positional relationship between the n standard odor vectors and the reference odor direction vector. The positional relationship between the standard odor vector and the reference odor direction vector is calculated and stored. This eliminates the need to measure the reference sample by the reference sample measurement unit every time the standard sample measurement unit updates n standard odor vectors.

Further, the odor measuring apparatus of the present invention may use the plurality of reference samples adjusted so that the lengths of the odor vectors are equal in the m-dimensional space.
In the odor measuring apparatus of the present invention, the calculation unit may create a reference odor direction vector indicating a change in an angle of an odor vector having a different reference odor intensity.
In the odor measuring apparatus of the present invention, the calculation unit takes an orthogonal projection with respect to each standard odor vector for an odor vector having a different intensity of the reference odor, and changes the length of the orthogonal projection to change the reference odor direction vector. You may make it create.
Further, in the odor measuring apparatus of the present invention, the standard odor is aromatic odor, alcohol odor, hydrocarbon odor, ester odor, amine odor, aldehyde odor, sulfur odor, and organic acid odor. You may make it be at least 2 thing selected from the group which consists of an odor.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. In addition, this invention is not limited to the following embodiment, It cannot be overemphasized that various aspects are included in the range which does not deviate from the meaning of this invention.

FIG. 1 is a block diagram showing a configuration of an odor measuring apparatus according to an embodiment of the present invention.
The odor measuring device includes an inlet 1 for sucking a sample (standard sample, reference sample, measurement sample), a pre-processing unit (collecting tube) 2 for pre-processing, and six odor sensors 31 having different response characteristics. To the sensor cell 3, the pump 4 for drawing the sample into the sensor cell 3, the gas tank 11 for storing the inert gas, and the detection signals from the odor sensors 31 to 36 are analyzed and the operation of the entire apparatus is controlled. Control unit 5, display unit 6 for displaying the output from control unit 5 on the display screen, and operation unit 8 having an input device such as a keyboard.

In addition, the reference sample or the measurement sample sucked from the suction port 1 is sealed in a gas cylinder or the like in the case of gas at normal temperature, and a certain amount is taken out and used. In the case of a liquid, it is put in a glass container or the like and used by being kept at a predetermined temperature or being bubbled with nitrogen gas or the like. Further, in the case of a solid, it is put in a glass container or the like and used at a predetermined temperature.

The pretreatment unit (collection tube) 2 performs removal of moisture contained in the sample, concentration / dilution of the sample, removal of interfering gas, and the like. Therefore, in the pretreatment unit 2, for example, by diluting the standard sample with an inert gas or the like, the concentration of the standard sample is divided into a plurality of levels (for example, no dilution, 1/3, 1/10, inert) It is also possible to produce a sample that has been changed to four stages of gas only.

The odor sensors 31 to 36 are oxide semiconductor sensors each having a different response characteristic in which the electrical resistance value changes according to the intensity of the odor. The sensor by other detection methods, such as the sensor which formed the gas adsorption film, may be sufficient. Further, the number of odor sensors is not particularly limited. When a sample comes into contact with such odor sensors 31 to 36, detection signals are output from the odor sensors 31 to 36 to the control unit 5 in parallel.

The control unit 5 is configured around a personal computer, and by executing a predetermined program on the computer, the sample processing unit 20, the standard sample measurement unit 21, the reference sample measurement unit 22, the calculation unit 23, It functions as the measurement sample measurement unit 24, the determination unit 25, and the recalculation unit 26. The control unit 5 is connected to a storage device (memory, HDD, etc.) 7 for storing data.

The sample processing unit 20 controls the drawing of the sample into the sensor cell 3 by operating the pump 4 based on an operation signal from the operation unit 8 or the like.
For example, when an operation signal (standard odor setting signal) for performing a standard odor setting operation is received, a sample obtained by diluting the standard sample with an inert gas is prepared by the preprocessing unit 2 and drawn into the sensor cell 3. At this time, for example, a sample in which the concentration of the standard sample is changed into four stages of no dilution, 1/3 dilution, 1/10 dilution, and inert gas alone is prepared and sequentially drawn into the sensor cell 3. Further, when an output signal (reference odor setting signal) for performing a setting operation for the reference odor is received, the sample is directly drawn into the sensor cell 3. Furthermore, when an output signal (measurement signal) for measuring a measurement sample is received, the sample is directly pulled into the sensor cell 3.

The standard sample measurement unit 21 is a six-dimensional space (odor) formed by six odor sensors 31 to 36 based on measurement results obtained by measuring samples obtained by changing the mixing ratio of the standard sample and the inert gas into a plurality of stages. Standard odor vectors S1 to S7 are created in the space) and stored in the storage device 7.
Specifically, six measurement data DS1, DS2, DS3, DS4, DS5, and DS6 are obtained from the odor sensors 31 to 36 for one sample. Here, considering the six-dimensional space (odor space) formed by using the six measurement data obtained from the odor sensors 31 to 36 as axes in different directions, the six measurement data DS1, DS2, DS3, DS4. , DS5, DS6 can be represented by one point (DS1, DS2, DS3, DS4, DS5, DS6) in the odor space. The measurement results obtained by measuring the standard sample concentration in four stages of undiluted, diluted to 1/3, diluted to 1/10, and inert gas only are shown in the odor space. Since the points are shifted almost linearly in a specific fixed direction, this is regarded as a standard odor vector in the odor space.
Here, since it is difficult to illustrate a 6-dimensional space (odor space), here, in order to facilitate understanding, it is formed by two measurement data DS1 and DS2 obtained from the two odor sensors 31 and 32. Consider a simplified two-dimensional space. As shown in FIG. 2, in the two-dimensional space, the two measurement data DS1 and DS2 for the undiluted sample of the standard sample are represented by one measurement point P1. The measurement points for the sample diluted as described above are represented as P2, P3, and P4, respectively. Thereby, the standard odor vector S1 can be drawn by connecting P4, P3, P2, and P1 in a straight line. That is, the standard odor vector S1 associated with the standard odor of the standard sample is created.

Therefore, as seven kinds of standard samples, for example, toluene as an aromatic odor, heptane as a hydrocarbon odor, ethyl acetate as an ester odor, trimethylamine as an amine odor, aldehyde odor, etc. When measurement is performed using butyraldehyde, methyl mercaptan as a sulfur odor, butyric acid as an organic acid odor, etc., seven standard odor vectors S1 to S7 can be created.

The reference sample measurement unit 22 performs control for creating an odor vector SX in a six-dimensional space (odor space) and storing the odor vector SX in the storage device 7 based on the measurement result of the reference sample in which the strength of the reference odor is determined by a sensory test. Is what you do.
Specifically, as described above, six measurement data DS1, DS2, DS3, DS4, DS5, and DS6 are obtained from the odor sensors 31 to 36 for one sample. The six measurement data DS1, DS2, DS3, DS4, DS5, DS6 can be represented by one point (DS1, DS2, DS3, DS4, DS5, DS6) in the odor space (6-dimensional space). Thereby, what connected this point and the origin P4 memorize | stored in the memory | storage device 7 will be caught as an odor vector in odor space.
Here, as shown in FIG. 2 as described above, in the two-dimensional space, the two measurement data DS1 and DS2 for the reference sample are represented by one measurement point PX. Thereby, the odor vector SX corresponding to the intensity | strength of reference | standard odor can be drawn by connecting P4 and PX with a straight line.
In addition, the reference odor determined by the sensory test does not show a certain direction from the origin corresponding to the type of the reference odor in the odor space, and the direction in which the point is seen from the origin in accordance with the intensity change is a certain direction. Therefore, for example, as shown in Table 1, six reference samples having different smoking odor strengths were measured by a sensory test, and a 6-dimensional space (odor space) as shown in FIG. Individual odor vectors SX1 to SX6 are created. At this time, as the six reference samples, it is preferable to use samples adjusted so that the lengths of the odor vectors SX1 to SX6 are equal.

The calculation unit 23 calculates a similarity α between the standard odor and the reference odor intensity based on the angles θ1 to θ6 of the odor vectors SX1 to SX6 with respect to the standard odor vector, thereby indicating a reference odor direction vector indicating a change in the similarity α. SA is created, and the storage device 7 is controlled to store the positional relationship between the standard odor vectors S1 to S7 and the reference odor direction vector SA.
For example, first, the similarity α of each of the odor vectors SX1 to SX6 with respect to the standard odor vector S1 is calculated.
Here, when the direction of the odor vector SX is similar to the direction of the standard odor vector S1, it can be determined that the intensity of the reference odor is a kind of odor similar to the standard odor, If the directions are completely different, it can be determined that the odor is a distant odor. Therefore, as an index for determining the similarity between the odor vector SX and the standard odor vector S1, the similarity α between the standard odor and the standard odor intensity is calculated based on the angle θ formed by the standard odor vector S1 and the odor vector SX. Will be calculated.
Therefore, as shown in FIG. 4, when the angle θ formed by the standard odor vector S1 and the odor vector SX is 0 °, the similarity α is 100, and the angle θ is equal to or greater than the threshold θth. In some cases, the relationship between the angle θ and the similarity α is set as a function in the storage device 7 so as to determine that the similarity α is zero. Accordingly, the similarity α between the standard odor and the standard odor intensity is represented by a numerical value of 0 to 100.

Specifically, for the odor vectors SX1 to SX6 indicating the intensity of different smoking odors as shown in FIG. 3, the similarity α is calculated by the respective angles θ1 to θ6 with respect to the standard odor vector S1 indicating the sulfur odor, The result shown in FIG. 5 is obtained. Similarly, when the similarity α is calculated from the angle θ with respect to the standard odor vectors S2 to S7, results as shown in FIGS. 6 to 11 are obtained.
Next, a reference odor direction vector SA indicating a change in similarity α with respect to the standard odor vectors S1 to S7 is created with reference to FIGS. For example, as shown in FIG. 5, the odor direction vector SA is similarly created based on FIGS. 6 to 11 so that the intensity of the smoke odor increases as the similarity α to the standard odor vector S1 increases.
Thereafter, the positional relationship between the standard odor vectors S1 to S7 and the reference odor direction vector SA is stored in the storage device 7. The reference odor direction vector SA can be positioned even in a six-dimensional space (odor space) because the positional relationship with the seven standard odor vectors S1 to S7 is known.

The measurement sample measurement unit 24 performs control to create a measurement odor vector SY based on the measurement result of an unknown measurement sample in a six-dimensional space (odor space).
Specifically, as described above, six measurement data DS1, DS2, DS3, DS4, DS5, and DS6 are obtained from the odor sensors 31 to 36 for one sample. The six measurement data DS1, DS2, DS3, DS4, DS5, DS6 can be represented by one point (DS1, DS2, DS3, DS4, DS5, DS6) in the odor space (6-dimensional space). Thereby, what connected this point and the origin P4 memorize | stored in the memory | storage device 7 will be caught as a measurement odor vector in odor space.
Here, as shown in FIG. 12, the two measurement data DS1 and DS2 for the measurement sample are represented by one measurement point PY in the two-dimensional space, as described above. Thereby, the measurement odor vector SY corresponding to a measurement sample can be drawn by connecting P4 and PY in a straight line.

The determination unit 25 performs control to calculate the intensity of the reference odor contained in the measurement sample according to the direction indicated by the measurement odor vector SY using the reference odor direction vector SA. At this time, it is preferable to use a measurement sample adjusted to be equal to the length of the odor vectors SX1 to SX6.

When the standard sample measurement unit 21 updates and stores the seven standard odor vectors S1 ′ to S7 ′, the recalculation unit 26 stores the seven standard odor vectors S1 to S7 stored in the storage device 7. Control based on the positional relationship between the reference odor direction vector SA and the updated seven standard odor vectors S1 ′ to S7 ′ and the reference odor direction vector SA ′ and stored in the storage device 7 Is to do.
That is, since the relationship between the standard odor vectors S1 to S7 indicating the respective types of standard odor and the reference odor direction vector SA indicating the intensity of the reference odor is constant, for example, as shown in FIG. Since the responsiveness of 31 to 36 changes, when the standard sample measuring unit 21 updates the seven standard odor vectors S1 ′ to S7 ′, the seven standard odor vectors S1 to S7 stored in the storage device 7 are used. And calculating the positional relationship between the updated seven standard odor vectors S1 ′ to S7 ′ and the reference odor direction vector SA ′ based on the positional relationship between the reference odor direction vector SA and the reference odor direction vector SA. become. This eliminates the need for the reference sample measurement unit 22 to measure the reference sample every time the standard sample measurement unit 21 updates the seven standard odor vectors S1 to S7.

Next, a method of using the odor measuring device will be described. Here, the case where the intensity | strength of the smoking odor contained in the unknown measurement sample is measured is demonstrated. FIG.14 and FIG.15 is a flowchart for demonstrating an example of the usage method of an odor measuring apparatus.
(1) Standard odor setting operation First, in the process of step S101, when the angle θ between the standard odor vectors S1 to S7 and the odor vector SX is 0 °, the similarity α is assumed to be 100. When θ is greater than or equal to the threshold θth, the relationship between the angle θ and the similarity α is set as a function in the storage device 7 so as to determine that the similarity α is 0 (see FIG. 4).
Next, in the process of step S102, an operation signal (standard odor setting signal) for performing standard odor setting work is transmitted from the operation unit 8.
Next, in the process of step S103, the number of standard odor vectors n = 0 is stored in the storage device 7 as a function.

Next, in the process of step S <b> 104, the sample processing unit 20 creates a sample obtained by diluting the standard sample with an inert gas in the preprocessing unit 2, and draws it into the sensor cell 3.
Next, in the process of step S105, the standard sample measuring unit 21 is formed by six odor sensors based on the measurement results obtained by measuring the samples in which the mixing ratio of the standard sample and the inert gas is changed to a plurality of levels. A standard odor vector is created in a six-dimensional space (odor space) and stored in the storage device 7 (see FIG. 2).

Next, in the process of step S106, it is determined whether n = 6. If not n = 6, the number of standard odor vectors n = n + 1 is stored in the storage device 7 as a function in the process of step S107, and the process returns to step S104.
(2) Setting operation of reference odor On the other hand, when n = 6, in the process of step S108, six pieces of smoke odor intensity different from those shown in Table 1 determined by the sensory test are shown. Prepare a reference sample. At this time, the lengths of the odor vectors SX are adjusted to be equal in each reference sample.
Next, in the process of step S109, an operation signal (reference odor setting signal) for performing a setting operation of the reference odor is transmitted from the operation unit 8.

Next, in the process of step S <b> 110, the sample processing unit 20 draws the reference sample from the suction port 1 into the sensor cell 3.
Next, in the process of step S111, the reference sample measurement unit 22 creates an odor vector SX in a six-dimensional space (odor space) based on the measurement result of the reference sample, and stores it in the storage device 7 (see FIG. 3). .

Next, in the process of step S112, it is determined whether or not the reference sample is still measured. When the reference sample is still measured, the process returns to step S110.
On the other hand, when the reference sample is no longer measured, in the process of step S113, the calculation unit 23 determines the similarity between the standard odor and the intensity of the reference odor based on the angles θ1 to θ6 of the odor vectors SX1 to SX6 with respect to the standard odor vector. By calculating α, a reference odor direction vector SA indicating a change in the similarity α is created, and the positional relationship between the standard odor vector and the reference odor direction vector SA is stored in the storage device 7 (FIGS. 5 to 11). reference).
And when the process of step S113 is complete | finished, this flowchart will be complete | finished.

(3) Measurement work of measurement sample First, in the process of step S201, the type of the operation signal received from the operation unit 8 is determined. When an operation signal (measurement signal) for measuring a measurement sample is received, the sample processing unit 20 draws the measurement sample from the suction port 1 into the sensor cell 3 in the process of step S202.
Next, in the process of step S203, the measurement sample measurement unit 24 creates a measurement odor vector SY in a six-dimensional space (odor space) based on the measurement result of the measurement sample (see FIG. 12).

Next, in the process of step S204, the determination unit 25 calculates the intensity of the reference odor included in the measurement sample based on the direction indicated by the measurement odor vector SY using the reference odor direction vector SA.
And when the process of step S204 is complete | finished, this flowchart will be complete | finished.

On the other hand, when an operation signal (standard sample setting signal) for measuring a standard sample is received in the process of step S201, the number of standard odor vectors n = 0 is stored in the storage device 7 as a function in the process of step S205. Is done.

Next, in the process of step S <b> 206, the sample processing unit 20 creates a sample obtained by diluting the standard sample with an inert gas in the preprocessing unit 2, and draws it into the sensor cell 3.
Next, in the process of step S207, the standard sample measurement unit 21 determines the six odor sensors 31 to 36 based on the measurement results obtained by measuring the samples obtained by changing the mixing ratio of the standard sample and the inert gas into a plurality of stages. The standard odor vector is created in the 6-dimensional space (odor space) formed by the above and stored in the storage device 7.

Next, in the process of step S208, it is determined whether n = 6. When not n = 6, the number of standard odor vectors n = n + 1 is stored in the storage device 7 as a function in the process of step S209, and the process returns to step S206.
On the other hand, when n = 6, in the process of step S210, the recalculation unit 26 determines the positional relationship between the standard odor vectors S1 to S7 stored in the storage device 7 and the reference odor direction vector SA. Based on this, the positional relationship between the updated standard odor vectors S1 ′ to S7 ′ and the reference odor direction vector SA ′ is calculated and stored in the storage device 7.
And when the process of step S210 is complete | finished, it returns to the process of step S202.

As described above, according to the odor measuring apparatus of the present invention, six odor vectors are created by measuring six reference samples in which the smoke odor intensity is determined in advance by a sensory test. Thereby, the reference | standard odor direction vector SA which shows the change of the angle (theta) of the six odor vectors SX is created and memorize | stored. Therefore, since the reference odor direction vector SA in which the direction of the smoke odor intensity by the sensory test is associated is stored, the measurement is performed by comparing the reference odor direction vector SA with the measurement odor vector SY of the measurement sample. The odor of the measured sample can be evaluated by the intensity of the smoking odor. It is also possible to evaluate small differences that could not be determined using human olfaction.

(Other embodiments)
(1) In the odor measuring apparatus described above, seven standard odor vectors S1 to S7 are set in a six-dimensional space (odor space) formed by six odor sensors based on measurement results obtained by measuring seven types of standard samples. For example, 10 standard odor vectors are created in a 9-dimensional space (odor space) formed by 9 odor sensors based on measurement results obtained by measuring 10 types of standard samples. The number of types of standard samples and the number of odor sensors are not particularly limited.
(2) In the odor measuring apparatus described above, the reference sample measuring unit 22 generates one reference odor direction vector SA based on the measurement result of only the reference sample whose smoke odor intensity is determined by the sensory test. In addition, the positional relationship between the standard odor vectors S1 to S7 and one reference odor direction vector SA is stored. For example, a reference sample in which the strength of the reference odor different from the smoke odor is determined by a sensory test. Based on the measurement result, another reference odor direction vector may be created to store the positional relationship between the standard odor vector and the other reference odor direction vector. The number of reference odor direction vectors to be stored Is not particularly limited.

The vector is usually a straight line, but the vector in the present invention should be interpreted to include a curve obtained by changing the concentration of the same odor.

INDUSTRIAL APPLICABILITY The present invention can be used for an odor measuring apparatus that evaluates the odor of a measurement sample by sensory expression.

It is a block diagram which shows the structure of the odor measuring apparatus which is one Embodiment of this invention. It is explanatory drawing which shows the standard odor vector and odor vector in an odor space. It is a figure which shows angle (theta) of the odor vector with respect to a standard odor vector. It is a figure which shows the relationship between angle (theta) and similarity degree (alpha). It is a graph which shows the relationship between the similarity degree (alpha) with respect to sulfur type odor, and the intensity | strength of a smoking odor. It is a graph which shows the relationship between the similarity degree (alpha) with respect to amine odor, and the intensity | strength of a smoking odor. It is a graph which shows the relationship between the similarity degree (alpha) with respect to organic acid type | system | group odor, and the intensity | strength of a smoking odor. It is a graph which shows the relationship between the similarity degree (alpha) with respect to an aldehyde type | system | group odor, and the intensity | strength of a smoking odor. It is a graph which shows the relationship between the similarity degree (alpha) with respect to ester type | system | group odor, and the intensity | strength of a smoking odor. It is a graph which shows the relationship between the similarity degree (alpha) with respect to an aromatic odor, and the intensity | strength of a smoking odor. It is a graph which shows the relationship between the similarity degree (alpha) with respect to a hydrocarbon type odor, and the intensity | strength of a smoking odor. It is explanatory drawing which shows the reference | standard odor direction vector in a odor space, and a measurement odor vector. It is explanatory drawing which shows the relationship between the reference | standard odor direction vector in a odor space, and a standard odor vector. It is a flowchart for demonstrating an example of the usage method of an odor measuring apparatus. It is a flowchart for demonstrating an example of the usage method of an odor measuring apparatus.

Explanation of symbols

11: Gas tank 21: Standard sample measurement unit 22: Reference sample measurement unit 23: Calculation units 31 to 36: Odor sensor

Claims (6)

M (m is an integer of 2 or more) odor sensors having different response characteristics;
Based on the measurement results of n (n is an integer of 2 or more) standard samples having different types of standard odors in an m-dimensional space formed to represent the measurement results of the m odor sensors. A standard sample measurement unit for creating and storing a standard odor vector of
In the m-dimensional space, an odor measuring device comprising a reference sample measuring unit that creates an odor vector based on a measurement result of a reference sample in which the strength of the reference odor is determined by a sensory test,
The reference sample measurement unit measures a plurality of reference samples having different reference odor intensities,
Furthermore, a calculation unit is provided that creates a reference odor direction vector based on the positional relationship between the plurality of odor vectors and stores the positional relationship between the n standard odor vectors and the reference odor direction vector. Odor measuring device. When the standard sample measurement unit updates and stores n standard odor vectors, the updated n number based on the positional relationship between the stored n standard odor vectors and the reference odor direction vector. The odor measuring apparatus according to claim 1 , further comprising a recalculation unit that calculates and stores a positional relationship between the standard odor vector and the reference odor direction vector. 3. The odor measuring apparatus according to claim 1, wherein, in the plurality of reference samples, ones adjusted so that the lengths of the odor vectors are equal in an m-dimensional space are used. The odor measuring apparatus according to any one of claims 1 to 3, wherein the calculation unit creates a reference odor direction vector indicating a change in an angle of an odor vector having a different reference odor intensity. The calculation unit, for odor vectors having different reference odor intensities, takes an orthogonal projection with respect to each standard odor vector, and creates a reference odor direction vector by changing the length of the orthogonal projection. 3. The odor measuring device according to 3. The standard odor is at least 2 selected from the group consisting of aromatic odor, alcohol odor, hydrocarbon odor, ester odor, amine odor, aldehyde odor, sulfur odor, and organic acid odor. The odor measuring device according to claim 1, wherein the odor measuring device is one.