JP4446718B2 - Odor measuring device - Google Patents

Description

  The present invention relates to an odor measuring apparatus that measures odors using a plurality of odor sensors having different sensitivity characteristics to odors.

The odor sensor used for conventional odor measurement uses the fact that the odorous substance contained in the gas to be measured adsorbs or approaches the detection surface and changes the physical and chemical characteristics of the sensor, thereby changing the odor intensity. taking measurement. As this odor sensor, there is known an odor sensor using a semiconductor material such as SnO ₂ or ZnO whose electrical resistance increases or decreases conversely when an odor substance is adsorbed or approaches. This type of odor sensor changes its odor sensitivity characteristics depending on the type of additive substance such as a catalyst and the operating temperature of the sensor material. That is, the intensity of the detection signal of the sensor when the odor is detected changes depending on the type of odor. Therefore, the present applicant pays attention to the sensitivity characteristics of such an odor sensor, and an odor measuring apparatus capable of measuring both the amount (intensity) and type (fragrance) of odors using two odor sensors having different sensitivity characteristics. (See Patent Document 1). This odor measuring apparatus has the magnitude of a vector obtained when the intensities of detection signals of two odor sensors having different sensitivity characteristics are taken on different X-axis and Y-axis of the orthogonal coordinate system, and the coordinate axis of the vector. The intensity and type (fragrance) of the odor of the measurement object are measured based on the inclination with respect to.

Registered Utility Model No. 3074494

However, in the measurement using the above two conventional odor sensors, there is a problem that it is difficult to obtain measurement results close to human olfactory characteristics with respect to odors having various odor components in all measurement objects.
For example, an odor (bad odor) generated from a facility such as a garbage disposal plant is strongly stimulated by a variety of odor odor components contained in the odor, but the above two odor sensors can identify odor odor components. Since accuracy cannot be achieved, the odor odor intensity cannot be corrected due to the difference in the odor odor component composition ratio, and it is difficult to obtain a measurement result close to human olfactory characteristics depending on the type of garbage in the garbage disposal site. Monitoring the odor of the surroundings of facilities such as this garbage disposal plant is important in order to maintain the facilities so that the surrounding residents can maintain a comfortable living environment by always keeping the surrounding odor environment below a certain standard. Therefore, in Japan, regulations on odor environmental standards to be observed by the Odor Control Law are being implemented. In the early days of the enactment of this law, a method for measuring the component concentration (ppm) of a specific odor gas with an analytical instrument such as a gas chromatograph was proposed. However, the odor component of odor is rarely composed of a single molecule, it is composed of complex components, the component composition is diverse, and the odor sensation of human olfaction and the chemical composition of odors. There is still no established correlation. Therefore, the official method expresses the strength of the composite odor by the “odor index” which is 10 times the logarithm of the dilution factor until the odor disappears by measuring the dilution concentration by the olfaction of a person (panelist). This person (paneler) olfactory method has a problem in that there is an error due to individual differences in the olfactory sense of the paneler, and there is a possibility that the intensity and type (fragrance) of the odor cannot be accurately measured. Furthermore, this method also has problems such as the cost of having to collect six panelists, and the risk of inhaling dangerous gases and harming health.

  The present invention has been made in view of the above problems. The purpose is to measure odors that can measure the intensity and type (fragrance) of a wider variety of odors more accurately than when using two odor sensors or measuring with the olfaction of a person (panelist). Is to provide a device.

In order to achieve the above object, the invention of claim 1 is an odor measuring apparatus for measuring the odor of a gas to be measured using a plurality of odor sensors having different sensitivity characteristics with respect to the odor. Three or more types of odor sensors, and based on the output values of the plurality of odor sensors for the odorless reference gas, the output values of the plurality of odor sensors for the gas to be measured are corrected, and the sensor output correction values Data for processing the sensor output correction value data so as to obtain element odor vectors defined in each of a plurality of quadrants of the coordinate system by taking each on different coordinate axes extending from the origin of the coordinate system of two or more dimensions. A processing means is provided.
According to a second aspect of the present invention, in the odor measuring apparatus according to the first aspect, the coordinate system is a two-dimensional plane orthogonal coordinate system, and the data processing means outputs each of the plurality of sensor output correction values to the plane orthogonal coordinate system. The sensor output correction value data is processed so as to obtain the element odor vector on different coordinate axes extending from the origin of the sensor.
The invention of claim 3 is the odor measuring apparatus according to claim 1 or 2, further comprising display means for displaying a plurality of element odor vectors obtained by the data processing means on the coordinate system. It is.
According to a fourth aspect of the present invention, in the odor measuring apparatus according to any one of the first to third aspects, the data storage means for storing data is provided, wherein the data processing means is an average value of the magnitudes of the plurality of element odor vectors. The value obtained by dividing the size of each element odor vector is calculated as element odor intensity rate data, and the angle formed by the plurality of element odor vectors and at least one coordinate axis in the quadrant where the element odor vector exists is The data storage unit calculates a data set obtained by combining the plurality of element odor intensity data and the plurality of element fragrance data as a total odor data set for the odor of the gas to be measured. The data processing is performed so as to store the data.
According to a fifth aspect of the present invention, in the odor measuring apparatus according to any one of the first to fourth aspects, the data processing means uses the comprehensive odor data set when the intensity and type (fragrance) are measured for a known odor. The comprehensive odor data set when the measurement is performed for the gas to be measured, the odor intensity and type (fragrance) stored in the data storage means in association with the odor intensity and type (fragrance), Comparing with a comprehensive odor data set of known odors stored in the data storage means, and data processing is performed so as to identify the type (fragrance) of the unknown odor based on the comparison result It is what.
According to a sixth aspect of the present invention, in the odor measurement apparatus according to any one of the first to fifth aspects, the data processing means squares the sum of the plurality of sensor output correction values and adds the square root value to the total odor intensity. Data processing is performed such that the data is calculated as the data.
According to a seventh aspect of the present invention, in the odor measuring apparatus according to the sixth aspect, the data processing means adds the element intensity rate data and the element fragrance data or the odor type (fragrance) to the total odor intensity value. The data processing is performed so as to calculate a numerical value of the odor intensity approximated to the human olfactory characteristic by multiplying by a constant determined by the quality data.

In the odor measuring apparatus according to the first aspect, three or more types of odor sensors having different sensitivity characteristics with respect to the odor are used. Using these plural odor sensors, the measurement with respect to the odorless reference gas is performed before the measurement with respect to the gas to be measured. Based on the output values of the plurality of odor sensors for the odorless reference gas, the output values of the plurality of odor sensors for the gas to be measured are corrected. By this correction, the influences of fluctuations in environmental conditions such as temperature and sensor characteristics are removed from the output values of the plurality of odor sensors for the gas to be measured. Then, each of the plurality of sensor output correction values is set on different coordinate axes extending from the origin of the coordinate system of two or more dimensions, and two or more element odor vectors defined in the plurality of quadrants of the coordinate system are obtained. . The magnitudes and directions of the two or more elemental odor vectors are obtained from three or more types of odor sensors having different sensitivity characteristics to the odor, and therefore depending on the intensity and type (fragrance) of the odor of the gas to be measured. Show different changes. By using the data of the magnitudes of two or more element odor vectors that change in this way and the angle with the coordinate axis, compared to the case of using one odor vector obtained from two conventional odor sensors, It is possible to measure the intensity and type (fragrance) of more various types of odors. Moreover, unlike the case of measuring by human (paneler) olfaction, there is no error due to individual differences in human (paneler) olfaction, so it is possible to more accurately measure the intensity and type (fragrance) of various types of odors. .
In the odor measuring apparatus according to the second aspect, the sensor output correction values of the plurality of odor sensors are set on different coordinate axes extending from the origin of the plane orthogonal coordinate system. When there are five or more types of odor sensors, the sensor output correction values of three or four types of odor sensors selected from the odor sensors are set on different coordinate axes. Thereby, the element odor vector defined in at least two quadrants among the four quadrants is obtained. Since data processing is performed so as to obtain the element odor vector on a two-dimensional planar orthogonal coordinate system that facilitates vector calculation in this way, data processing is performed compared to the case of using a three-dimensional or higher coordinate system. Becomes simple.
In the odor measuring apparatus according to claim 3, by displaying the plurality of element odor vectors on a coordinate system, the magnitude of each element odor vector corresponding to the intensity and type of the odor of the gas to be measured and the angle between the coordinate axes (Tilt) can be visually confirmed.
The odor measuring apparatus according to claim 4 calculates the element odor intensity rate data obtained by dividing the size of each element odor vector by the average value of the sizes of the plurality of element odor vectors. The data of each element odor intensity rate is standardized by dividing the size of each element odor vector by the average value, so what ratio is the overall size of each element odor vector? The feature of the element odor vector related to the odor type (fragrance) is easily extracted. Further, an angle formed by the plurality of element odor vectors and at least one coordinate axis in a quadrant in which the element odor vector exists is calculated as element fragrance data. The value indicated by the element fragrance data indicates the magnitude relationship between the output values of any two odor sensors having different sensitivity characteristics to odors, and the odor type (fragrance) It is a value related to. Data storage means as a comprehensive odor data set for the odor of the gas to be measured by combining a data set combining a plurality of element odor intensity rate data and a plurality of element fragrance data related to these odor types (fragrances) To remember. By using this comprehensive odor data set, the odor type (fragrance) of the gas to be measured can be specified with high accuracy.
6. The odor measuring apparatus according to claim 5, wherein the comprehensive odor data set when measuring an odor having a known intensity and type (fragrance) is stored in the data storage means in association with the odor intensity and fragrance data. To do. Then, the total odor data set when measuring the measurement target gas whose odor intensity and type (fragrance) are unknown is compared with the total odor data set of known odors stored in the data storage means. Based on the comparison result, the odor type (fragrance) of the gas to be measured can be specified from the degree of coincidence of each data of the comprehensive odor data set. Here, by increasing the types of the total odor data set of known odors stored and accumulated in the data storage means, the accuracy of identifying the odor type (fragrance) of the gas to be measured can be improved.
In the odor measuring apparatus according to the sixth aspect, a square root value obtained by squaring and adding each of the plurality of sensor output correction values is calculated as total odor intensity data. Since this total odor intensity data changes according to the magnitudes of all sensor output correction values, it can be treated as a value indicating the total intensity of the odor of the gas to be measured. In addition, even if there is interference between odor sensors with respect to odor sensitivity characteristics, interference is obtained compared to adding each sensor output correction value as it is by taking the square root of the sum of squared sensor output correction values. The effect of.
In the odor measuring apparatus according to claim 7, the value of the total odor intensity is multiplied by a constant determined by element intensity rate data and element fragrance data or odor type (fragrance) data. An index value of the odor intensity approximated to the olfactory characteristic is calculated. From this index value, the odor intensity of the odor of the gas to be measured approximates the human olfactory characteristics.

According to the first to seventh aspects of the present invention, a plurality of element odor vectors on a two-dimensional or higher coordinate system are obtained from sensor output correction values of three or more types of odor sensors having different sensitivity characteristics to odors. . By using the data of the size of the element odor vector and the angle with the coordinate axis, compared to the case of using one odor vector obtained from two conventional odor sensors, the intensity and type of odors of a wider variety of types (Fragrance) can be measured. Moreover, unlike the case of measuring by human (paneler) olfaction, there is no error due to individual differences in human (paneler) olfaction, so it is possible to more accurately measure the intensity and type (fragrance) of various types of odors. There is an effect.
In particular, according to the invention of claim 2, since the data processing is performed so as to obtain the element odor vector on a two-dimensional plane orthogonal coordinate system in which vector calculation processing is easy, a three-dimensional or higher coordinate system is used. There is an effect that data processing is simplified as compared with the case.
In particular, according to the third aspect of the invention, there is an effect that the size of each element odor vector corresponding to the intensity and type of the odor of the gas to be measured and the angle (inclination) between the odor vector and the coordinate axis can be visually confirmed.
In particular, according to the fourth aspect of the present invention, a data set combining a plurality of element odor intensity rate data and a plurality of element fragrance data related to the type of odor (fragrance) is used as the odor of the gas to be measured. By storing and using as a comprehensive odor data set for, there is an effect that the odor type (fragrance) of the gas to be measured can be specified with high accuracy.
In particular, according to the invention of claim 5, it is possible to improve the identification accuracy of the odor type (fragrance) of the gas to be measured by storing and storing the comprehensive odor data set of known odors in the data storage means. There is an effect that can be done.
In particular, according to the invention of claim 6, the value indicating the total intensity of the odor of the gas to be measured based on the data of the total odor intensity which is the square root value obtained by squaring and adding each of the plurality of sensor output correction values. Can be treated as Moreover, even if there is interference between the odor sensors with respect to the odor sensitivity characteristics, there is an effect that the influence of the interference is reduced as compared with the case where the sensor output correction values are added as they are.
In particular, according to the invention of claim 7, the effect that the odor intensity approximated to the human olfactory characteristics can be measured for the odor of the gas to be measured by the index value calculated using the value of the total odor intensity. There is.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
First, the overall configuration of the odor measuring apparatus according to the embodiment of the present invention will be described. FIG. 1 is an overall configuration diagram of an odor measuring apparatus according to an embodiment of the present invention. This odor measuring apparatus includes a measurement chamber 20 that can be sealed as a measurement gas storage unit having four types of odor sensors 10A, 10B, 10C, and 10D, and a gas supply for introducing the measurement gas into the measurement chamber 20. Means 30. The gas supply means 30 is configured using a supply pipe 31 that forms a gas introduction path and an electromagnetic valve 32 that opens and closes the gas introduction path. The measurement chamber 20 is also provided with two spare gas supply means 30 ′ and 30 ″ configured using a supply pipe 31 and a solenoid valve 32 having the same configuration.
In addition, the odor measuring device includes a gas discharge unit configured using a pump unit 40 for exhausting the gas in the measurement chamber 20 and exhaust pipes 41 and 42 forming a gas exhaust path. As the pump unit 40, for example, an air pump or a diaphragm type micro pump can be used.
By controlling the electromagnetic valve 32 and the pump unit 40, gas can be supplied into the measurement chamber 20 or gas can be discharged from the measurement chamber 20. For example, when the solenoid valve 32 is opened and the pump unit 40 is turned on, the gas to be measured sucked from the suction port of the introduction supply pipe 31 while the gas in the measurement chamber 20 is discharged from the exhaust port of the exhaust pipe 42. Can be introduced. Then, in a state where the measurement chamber 20 is filled with the gas to be measured, the pump unit 40 is turned off and the electromagnetic valve 32 is closed, whereby the inside of the measurement chamber 20 becomes a sealed space, and the odor measurement of the gas to be measured is performed. Can start.

  In addition, the odor measuring apparatus of the present embodiment introduces an odorless standard gas generated by a standard gas generating means (not shown), and can perform measurement data correction processing based on the output value of the odor sensor at that time. Yes. The standard gas generation means can be configured to generate odorless standard gas by, for example, sucking ambient air and passing activated carbon, or generate odorless standard gas from a commercially available odorless gas cylinder.

The odor sensors 10A to 10D have different sensitivity characteristics with respect to odors, and are selected according to the odor to be measured. As the odor sensors 10A to 10D, an odor sensor made of a metal oxide semiconductor to which a catalyst is added, an odor sensor combining a synthetic film and a crystal resonator, a biosensor, or the like can be used. Among these odor sensors, an odor sensor using a metal oxide semiconductor has good durability, environmental resistance, reproducibility, and response speed, and a wide measurement concentration range. The odor sensor using the metal oxide semiconductor has a sensitivity characteristic close to a logarithmic function with respect to the concentration of odor molecules (odor gas concentration). That is, the sensor output signal voltage (sensitive output) Y with respect to the odor gas concentration X of the odor sensor using the metal oxide semiconductor can be approximated by a logarithmic function as in the following equation (1). Here, K in the equation is a constant that varies depending on the type of odor.

In addition, as the metal oxide semiconductor of the odor sensor, sintered SnO ₂ , ZnO, or the like is used, and there is a large difference in sensitivity to odor molecules constituting the odor depending on the type of material, the type of catalyst, the structure, and the like. Arise. These metal oxide semiconductors such as SnO ₂ and ZnO are subjected to basic gas selection for odorous gases as shown in Table 1 according to the difference in additive catalyst, temperature, and pore size of the porous sintered structure. Can do. The pores are classified into macropores (50 nm or more), mesopores (50 to 2.0 nm), and micropores (2.0 nm or less) according to their diameters. Table 2 shows a control example of the gas selectivity to be detected by the added catalyst, the pore diameter, and the temperature.

  However, the sensitivity selectivity with respect to the odor gas species of the metal oxide semiconductor is not sharp, and each odor sensor has a property of reacting to various odors. That is, the sensitivity selectivity of the odor sensor interferes with odor gases that are different from each other. If the tendency of the sensitivity selectivity is schematically illustrated with three kinds of odor sensors, it is as shown in FIG.

  Moreover, the odor measuring apparatus of this embodiment is provided with the sensor signal processing means 50 which processes the output signal of each odor sensor 10A-10D. The sensor signal processing means 50 includes a sensor circuit unit 51 and an AD converter 52. The sensor circuit unit 51 is configured by a circuit that supplies a predetermined drive voltage to each of the odor sensors 10A to 10D and passes a sensor output signal (analog signal) from each of the odor sensors 10A to 10D to the next stage. The AD converter 52 converts the sensor output signal (analog signal) received from the sensor circuit unit 51 into output value data (digital signal) of the odor sensor.

  FIG. 3 is an explanatory diagram showing an example of a specific circuit configuration of the sensor circuit unit 51. Although FIG. 3 shows the case of the odor sensor 10A, a similar circuit configuration can be adopted for the other odor sensors 10B to 10D. In the odor sensor 10A, a metal oxide semiconductor 11A and a platinum thin film 12A are integrally formed. Each platinum thin film 12A is supplied with a pulse current generated by the switching element 510 and generates heat. By heating the metal oxide semiconductor 11A to a high temperature around 400 ° C. with this platinum thin film 11B, the influence of ambient temperature change and moisture can be reduced, and odor molecules attached to the sensor can be easily washed away with clean air. I have to. ON / OFF of the switching element 510 is controlled by inputting a control pulse signal Vp generated by a pulse generator (not shown) to the switching element 510 via the input resistor 511. The output unit from which the detection signal of the odor sensor 10A is output includes an integrated voltage generation circuit including a resistor 512 and a capacitor 513 connected in series to the metal oxide semiconductor 11A. A sensor output signal (detection signal) is output from a connection point between the integrated voltage generation circuit and the odor sensor 10.

  In addition, the odor measuring apparatus according to the present embodiment includes an arithmetic control unit 60 as data processing means for processing data on the output values of the odor sensors 10A to 10D converted into digital signals by the AD converter 52. The arithmetic control unit 60 has a function of controlling the entire odor measuring device including the electromagnetic valve 32 and the pump unit 40 in addition to the function of data processing of sensor output values. The arithmetic control unit 60 may be configured using a digital logic circuit designed for the odor measuring apparatus, or may be configured using a general-purpose microcomputer including a CPU, a RAM, a ROM, and the like.

  Furthermore, the odor measuring apparatus of the present embodiment includes a display unit 70 including a liquid crystal display as a display unit, and a communication control unit 80. The display unit 70 displays the odor measurement results such as the output values of the odor sensors 10A to 10D and the odor intensity and type (fragrance) obtained by the data processing. The communication control unit 80 controls communication for data transmission / reception with the external computer 90 so that the data of the odor measurement result can be used by the external computer 90.

Next, a measurement procedure and data processing when measuring an odor (bad odor) using the odor measuring apparatus having the above configuration will be described.
First, prior to the measurement of the odor to be measured, an odorless standard gas is introduced into the measurement chamber 20, and the output signals of the odor sensors 10A to 10D are measured. The measured value data is stored in a memory as data storage means in the arithmetic control unit 60, and is used for correcting the measured value of the odor of the measurement target thereafter. By this correction processing, it is possible to eliminate influences such as variations in environmental conditions such as temperature and variations in sensor characteristics. Hereinafter, the corrected measurement value is referred to as “sensor output correction value”.

  Next, a gas to be measured containing odor is introduced into the measurement chamber 20, and output signals of the odor sensors 10A to 10D are measured. This output signal is converted into digital data by the AD converter 52 and input to the arithmetic control unit 60, and predetermined data processing is executed.

  Here, the odor component that humans feel in daily life is rarely a single component (chemical substance), and in most cases is a plurality of odor components. The odor of the mixed gas containing a plurality of odor components is felt as a so-called composite odor. Representative examples of the odor odor molecule that is the object of odor measurement according to this embodiment include methyl mercaptan, trimethylamine, hydrogen sulfide, toluene, aldehyde, acetic acid, alcohol, and ammonia. The odor of a mixed gas containing a plurality of these odor components is felt as an odor (bad odor) of a composite odor.

[Total odor intensity]
The “total odor intensity” S indicating the overall odor intensity of the odor (bad odor) of the composite odor is a sensor output correction value corresponding to the four odor sensors 10A to 10D as shown in the following equation (2). The square root of the sum of squares a to d is obtained.

When the total odor intensity S is expressed by the above equation (2), the degree of interference of the sensitivity characteristics of the odor sensors 10A to 10D can be reduced as follows. Here, it is assumed that measurement is performed on a gas having a single odor molecule having sensitivity selectivity in the odor sensor 10A. When the above (2) is modified, the following equation (3) is obtained.

(B / a), (c / a), and (d / a) in the above equation (3) indicate the degree of interference of the sensitivity characteristics of the odor molecules in the measurement gas with the other odor sensors 10B to 10D. It can be referred to as the “interference rate” shown. If all the interference rates are about 0.2, the sum of the squares of the interference rates in the equation (3) is 0.12, and the total odor intensity S is a · (1.12) ^1/2. = A × 1.06. Therefore, the influence of interference on the other odor sensors 10B to 10D can be reduced as compared with a value (= a + b + c + d = a × 1.6) when the total odor intensity is simply obtained by addition.
Further, when measurement is performed on a gas having a single odor molecule having sensitivity selectivity in the odor sensor 10A, the sensor output correction values a to d corresponding to the odor sensors 10A to 10D are all the same value. However, the total odor intensity S of the above equation (2) is a · (6) ^1/2 = a × 2.5. That is, it is not 6 times as in the value (= a + b + c + d = a × 6) when the total odor intensity is simply obtained by addition, but it is only 2.5 times, and sensitivity selectivity is given to the odor sensor 10A. The influence of interference on the other odor sensors 10B to 10D with a single odor molecule is reduced.
As described above, it is reasonable to express the overall odor intensity of the odor (odor) of the composite odor by the overall odor intensity S expressed by the above equation (2).

[Odor intensity approximated by human olfactory characteristics]
When the overall odor intensity of the complex odor (bad odor) is expressed by the overall odor intensity S of the above formula (2), depending on the type of odor (fragrance), the odor intensity felt by human olfaction May be different from Causes of this divergence include the fact that the odor sensitivity characteristics of the odor sensors do not completely match the human olfactory characteristics, and the odor sensitivity characteristics interfere with each other.
Therefore, as shown in the following equation (4), it is preferable to calculate the total odor intensity St by multiplying a predetermined constant Ω determined for each odor type (fragrance). This constant Ω is measured for a plurality of odors (bad odors), each of which is known for each type of odor (fragrance) and the odor intensity by human sense of smell, and the measured value Ω is calculated by the calculation control unit 60. Save to memory.

Here, when the odor sensor has a characteristic very close to human olfaction, the total odor intensity St may be calculated by setting the constant Ω to 1. However, for example, human olfaction with a high concentration of sulfur components is extremely high even at low concentrations, and conversely, human olfaction with those containing ammonia, toluene, etc. has a relatively insensitive property to low concentrations. Therefore, the sensitive characteristics of the odor sensor may not faithfully reflect the human olfactory characteristics. In this case, the constant Ω is set for each odor type (fragrance) so as to match the human olfactory characteristics. In other words, the constant Ω is weighted to the total odor intensity measured by the odor measuring device according to the odor type (fragrance).
In the above formula (4), the sensor output correction values a to d are weighted by multiplying the square root values by adding one constant Ω to the whole square root value. Each squared value may be multiplied by a different constant Ωa to Ωd and weighted.

[Specification of odor type (fragrance)]
Identification of the odor type (fragrance) using the odor measuring apparatus having the above-described configuration can be performed as follows. That is, the sensor output correction values of the odor sensors 10A to 10D are taken on different coordinate axes extending from the origin of the plane orthogonal coordinate system, and “element odor vectors” defined in the four quadrants of the plane orthogonal coordinate system are used. Can be done.

4 to 7 are graphs showing the relationship between the sensor output values [V] of the four odor sensors 10A to 10D used in the present embodiment and the concentration [ppm] of the odor component in the gas to be measured. These graphs are obtained by measuring a measurement gas for test measurement in which each odor component is contained in air in a predetermined concentration. In particular, the odor sensor 10C is a sensor sensitive to ammonia, and the odor sensor 10D is a sensor sensitive to hydrogen sulfide.
Of these four odor sensors 10A to 10D, for example, if the sensor output correction values a and b of the two odor sensors 10A and 10B are taken in the X axis plus region and the Y axis plus region of the plane orthogonal coordinate system, respectively. A vector extending from the origin can be defined in the first quadrant of the plane orthogonal coordinate system. Hereinafter, this vector is referred to as an “element odor vector”. The length of the element odor vector becomes longer as the concentration of the odor component that the odor sensors 10A and 10B are sensitive to increases (the intensity of the odor increases). The angle formed by the element odor vector and the X axis varies depending on the odor type.

FIG. 8 is a graph obtained by setting the sensor output voltage values a and b of the odor sensors 10A and 10B measured by changing the concentrations of various odor components on the horizontal axis (X axis) and the vertical axis (Y axis), respectively. . In FIG. 8, the “element odor vector” is obtained by connecting the origin and each measurement data point. Here, the size | Se _ab | of the element odor vector represented by the following expression (5) is defined as “element strength”, and the angle θ _ab of the element odor vector represented by the following expression (6) is defined as “element It is defined as “fragrance”.

Also, other odor sensor combinations can be defined in the same manner, and the number of combinations is six. When the total number of odor sensors is n and the number of odor sensors to be combined is r, the general formula of the number P of combinations is as shown in the following equation (7).

Accordingly, six element odor vectors are obtained for each measurement of one gas to be measured, and the six element intensities | Se _ab |, | Se _bc |, | Se _cd |, | Se _da |, | Se _A data set consisting of _db |, | Se _ca | and six element fragrances θ _ab , θ _bc , θ _cd , θ _da , θ _db , θ _ca is obtained. That is, six sets of element strength and element fragrance (| Se _ab |, θ _ab ), (| Se _bc |, θ _bc ), (| Se _cd |, θ _cd ), (| Se _da |, θ _A data set consisting of _da ), (| Se _db |, θ _db ), (| Se _ca |, θ _ca ) is obtained. Based on these data sets, it is possible to specify the odor type (fragrance) of the gas to be measured.
In addition, in order to specify the odor type (fragrance) of the gas to be measured, instead of the element strength, each element strength is expressed by the average value Sx of the element strengths as exemplified by the following equations (8) and (9). It is preferable to use “element strength ratios” SR _ab , SR _bc , SR _cd , SR _da , SR _db , SR _ca divided by and normalized. When this element intensity ratio is used, the influence of the overall odor intensity of the gas to be measured is reduced, and the odor type (fragrance) can be accurately identified.

The type of odor (fragrance) can be specified, for example, as follows. First, odor measurement is performed for a plurality of types of known complex odors (odors) including a plurality of odor components at a predetermined ratio. Six elements odor vector obtained in this odor measurement, data of six sets of elements strength index and element fragrance substance _{(SR ab, θ ab),} (SR bc, θ bc), (Se cd, θ cd), A data set composed of (Se _da , θ _da ), (Se _db , θ _db ), and (Se _ca , θ _ca ) is calculated, stored in a memory in the arithmetic control unit 60 and stored. Then, the same odor measurement is performed on the gas to be measured including the complex odor (odor) to be measured, and six sets of element intensity ratio and element fragrance data are calculated from the six element odor vectors. Compare the element strength rate and element fragrance data of the gas to be measured with the element strength rate and element fragrance data of a plurality of known complex odors measured in advance, and from the degree of agreement between the two data, The type (flavor) of the complex odor of the gas to be measured can be specified. It is also possible to specify the ratio of the odor component contained in the composite odor of the gas to be measured based on the result of the comparison process of the element intensity rate and element fragrance data.

Note that the data sets (SR _ab , θ _ab ), (SR _bc , θ _bc ), (Se _cd , θ _cd ), (Se _da , θ _da ), (Se _db , θ _db ), (Se _ca , θ The standard deviation of the numerical values of the six element strength ratios and element fragrances of _ca ) can be calculated by accumulating data. By performing the data set comparison process in consideration of this standard deviation, it is possible to increase the specific accuracy of the type of complex odor (fragrance).

  Further, the odor sensor 10D reacts sensitively when there are many odorous components containing sulfur such as hydrogen sulfide in the functional group, and the odor sensor 10C reacts sensitively when there is much ammonia. Therefore, it is possible to easily identify the type (fragrance) of the composite odor from the numerical values in the data set. In this case, the type (fragrance) of the composite odor can be identified and specified with a certain degree of accuracy by a figure connecting the end points of the element odor vectors displayed on the coordinate system described later. Then, the constant Ω in the total odor intensity St represented by the above-described equation (4) is changed depending on the type (fragrance) of the composite odor. Thereby, the relationship between the intensity of the odor due to the human olfaction and the measured value of the total odor intensity St is corrected, and the measurement value can be corrected for magnification so that the total odor intensity St is close to the sensitivity characteristic of the human olfaction. .

  FIG. 9 is a graph showing the relationship between the measured value before correction of the total odor intensity St measured for various odor components and the concentration [ppm] of the odor component. As can be seen from this graph, the relationship between the value of the total odor intensity St and the concentration differs depending on the type of odor component. Therefore, by correcting the constant Ω in the above equation (4) for each type of odor component, the same total odor intensity St can be obtained for any odor component at the same concentration. . Furthermore, if the relationship between the concentration and the intensity of odor due to human olfaction is known for each odor component, the total odor close to the sensitivity characteristic of human olfaction is obtained by correcting the total odor intensity St in consideration of the relationship. Strength St can be obtained.

[Display of element odor vector]
Element odor vectors obtained from the sensor output correction value data of the four odor sensors 10A to 10D can be displayed on a two-dimensional plane orthogonal coordinate system. The element odor vector may be displayed on the display unit 70 of the odor measuring apparatus, or may be displayed by sending data to an external computer 90. By visually confirming the behavior of the element odor vector, it is possible to intuitively determine the type (fragrance) of the complex odor of the gas to be measured to some extent.

FIG. 10 is an explanatory diagram of element odor vectors obtained when the sensor output correction value data of the four odor sensors 10A to 10D are taken on each coordinate axis of the orthogonal coordinate system. In FIG. 10, the sensor output correction value data of the odor sensor 10A is on the positive side of the X axis, and the sensor output correction value data of the odor sensor 10B is on the positive side of the Y axis. Further, the sensor output correction value data of the odor sensor 10C is on the negative side of the X axis, and the sensor output correction value data of the odor sensor 10D is on the negative side of the Y axis. A circle indicated by a dotted line in the figure is a circle having a radius of an average value Sx of sensor output correction values of the odor sensors with the origin at the center.
In FIG. 10, four element odor vectors Se _ab , Se _bc , Se _cd , and Se _da among the above six element odor vectors can be displayed. Each element strength | Se _ab |, | Se _bc |, | Se _cd |, and | Se _da | can be confirmed as the lengths of these element odor vectors. The element fragrances θ _ab , θ _bc , θ _cd , and θ _da can be confirmed as angles from each coordinate axis to the element odor vector in the counterclockwise direction in each quadrant.

FIG. 11 is an explanatory diagram of element odor vectors obtained by switching the coordinate axes corresponding to the sensor output correction values of the four odor sensors 10A to 10D. In FIG. 11, unlike FIG. 10, the sensor output correction value data of the odor sensor 10B is placed on the negative side of the X axis. Further, the sensor output correction value data of the odor sensor 10C is on the negative side of the Y axis, and the sensor output correction value data of the odor sensor 10D is on the positive side of the Y axis. By taking the sensor output correction value in this manner, the element odor vectors Se _db and Se _ca that could not be displayed in FIG. 10 are displayed, and the remaining element intensities | Se _db | and | Se _{ca are calculated} from these element odor vectors. | And the element fragrances θ _db and θ _ca can be visually confirmed.

The measurement result measured by the odor measuring apparatus may be sent to an external computer 90 to be displayed on a display, or may be processed using a predetermined program incorporated in the computer 90.
FIG. 12 is an explanatory diagram of a display screen of the computer 90 displaying the sensor output correction values A to D of the odor sensors 10A to 10D, the element intensity of each element odor vector, the element fragrance, and the like.
13, 14, 15, and 16 show the element odor vectors when the odor component in the gas to be measured is 4 ppm of hydrogen sulfide, 18 ppm of ammonia, 0.1 ppm of hydrogen sulfide, and 1.8 ppm of ammonia, respectively. It is explanatory drawing of the screen of a display. A partial diagram (a) in each figure is a screen displaying element odor vectors Se _ab , Se _bc , Se _cd , Se _da , and a partial diagram (b) is an element odor vector Se _ad , Se _db , Se _bc. , _Seca is displayed.
A program for specifying the type (fragrance) of the composite odor in the external computer 90 is based on a general numerical pattern matching method as a basic method of the numerical pattern of the element intensity ratio and the elemental fragrance data set. What was created to perform the matching process can be used. In addition, for the variation in the element intensity ratio and element fragrance data, a range based on the standard deviation may be set to determine the type (fragrance) of the complex odor of the gas to be measured.

[Measurement of odor concentration and odor index]
There are “odor concentration” and “odor index” as public expression units of odor intensity by human olfaction. The odor concentration is defined as a dilution factor at which the odor is not felt for the first time when a gas (air) having an odor to be measured is diluted with an odorless gas (odorless air). The odor index is a value obtained by multiplying the common logarithm of the odor concentration by 10 times. In order to measure the odor concentration and odor index of the gas to be measured using the odor measuring apparatus of the present embodiment, measurement is performed on a plurality of types of sample gases including odors whose odor concentration and odor index are known in advance. A calibration curve is obtained from the measurement result, and calibration curve data is stored and stored in the arithmetic control unit 60.

FIG. 17 is a graph of a calibration curve measured for a gas to be measured including three types of odor components related to paint. The horizontal axis of this graph is the odor index measured by a conventional panel method, and the vertical axis is the value of the total odor intensity St measured by the odor measuring apparatus of this embodiment. In the figure, the scent component of “spray 1” is an acrylic lacquer resin-based paint for spray coating, and the scent component of “spray 2” is spray paint thinner. Also, “baking” in the figure is an acrylic resin-based baking paint. “Before treatment” and “after treatment” in the figure indicate before and after deodorization treatment, respectively. The detailed composition of each odor component is shown in Table 3.

  By using with reference to the calibration curve of FIG. 17, the odor index and odor concentration of the paint can be obtained from the value of the total odor intensity St of the various paints measured by the odor measuring apparatus of the present embodiment.

  FIG. 18 is a graph of a calibration curve measured for a plurality of measured gases whose odor component is ammonia and a measured gas whose odor component is hydrogen sulfide. The horizontal axis of this graph is the odor concentration obtained from the odor index measured by the conventional paneler method, and the vertical axis is the value of the total odor intensity St measured by the odor measuring apparatus of this embodiment. By using with reference to the calibration curve of FIG. 18, the odor index and odor concentration of the gas containing ammonia and hydrogen sulfide can be calculated from the value of the total odor intensity St of ammonia and hydrogen sulfide measured by the odor measuring apparatus of this embodiment. Can be requested.

[Correlation between element strength rate SR and odor type (fragrance)]
FIG. 19 is a diagram (hereinafter referred to as “scent spectrum diagram”) showing values of element intensity ratios SR measured for a measurement gas whose odor component is 4 ppm of ammonia and a measurement gas whose odor component is 18 ppm of hydrogen sulfide. . The element strength rate SR in the figure is calculated using the formulas exemplified in the above-described formulas (8) and (9). The first screen on the left side in the figure is a coordinate screen when the sensor output correction values of the odor sensors 10A to 10D are taken on the coordinate axes shown in FIG. On the other hand, the second screen on the right side in the drawing is a coordinate screen when the sensor output correction values of the odor sensors 10A to 10D are taken on the coordinate axes shown in FIG. In the second screen, the element strength rate data overlapping the first screen is omitted. Further, the horizontal axis of the odor spectrum diagram of FIG. 19 takes the value of the angle (fragrance) θ in the counterclockwise direction of the element odor vector with respect to the X axis (plus region) for each screen. The odor type (fragrance) may be identified and specified by a numerical pattern indicating the correlation between the element intensity rate SR and the angle (element fragrance) θ shown in the odor spectrum diagram. For example, paying attention to the angular position (element angle θ) of the element odor vector whose element intensity ratio exceeds the average value (= 1) in FIG. 19, it is visually determined what kind of odor component is large. can do. Further, the numerical pattern of the gas to be measured shown in FIG. 19 is analyzed by computer processing so as to be compared with a plurality of numerical patterns previously obtained for known odors and stored in the memory, and the odor type of the gas to be measured is determined. You may make it identify (fragrance). For example, when a numerical pattern close to ammonia appears, the total odor intensity St is corrected by setting the constant Ω to 0.8, and when a numerical pattern close to hydrogen sulfide appears, the constant Ω is set to 1.2 and the total odor intensity. St is corrected.

FIG. 20 is a feature extraction diagram in which a black mark is attached to the part of the cell corresponding to the angle of the element odor vector in which the element intensity rate SR exceeds the average value (= 1) in the odor spectrum diagram of FIG. . In this feature extraction diagram, for each odor component, a mark corresponding to the number of counts is added to the grid created in an angle unit of 30 degrees. For example, if the count number within the applicable angle range is 1, a mark is added to one part of the quadrant of the cell, and if the count is 4, the mark is applied to all four parts of the cell. Attached. The odor type (fragrance) of the gas to be measured can be specified by the mark distribution pattern indicating how the mark is distributed with respect to the angle.
Further, a feature extraction diagram of known odors may be stored in a computer as a database. In this case, a feature extraction diagram is generated from the measurement data of the gas to be measured including an unknown odor component, and the measured feature extraction diagram pattern is compared with the feature extraction diagram pattern stored in the computer as a database. As a result of this comparison, if the two patterns are completely or almost coincident with each other, it can be determined that the gas to be measured contains the coincident odor component. When the pattern of the feature extraction diagram measured for the gas to be measured does not match any of the patterns of the feature extraction diagrams of known odor components, a new odor type name is assigned and added to the database.
Although visually expressed in a tabular form as shown in FIG. 20, the numerical value of the element intensity may be recorded as an integer in 30 degree units for each odor component for actual calculation. In this case, it is also possible to subtract the numerical value of the feature extraction diagram of the unknown odor and identify and specify the odor type based on the result of the subtraction.

As described above, according to the present embodiment, a plurality of elements on the two-dimensional plane orthogonal coordinate system are obtained from the sensor output correction values a, b, c, d of the four types of odor sensors 10A to 10D having different sensitivity characteristics with respect to the odor. The odor vectors Se _ab , Se _bc , Se _cd , Se _da , Se _db , Se _ca are obtained. The size of this element odor vector (element strength) | Se _ab |, | Se _bc |, | Se _cd |, | Se _da |, | Se _db |, | Se _ca | and the angle with the coordinate axis (element fragrance) By using the data of θ _ab , θ _bc , θ _cd , θ _da , θ _db , θ _ca , various types of data can be used as compared with the case of using one odor vector obtained from two conventional odor sensors. The intensity and type (fragrance) of odor can be measured. Moreover, unlike the case of measuring by human (paneler) olfaction, there is no error due to individual differences in human (paneler) olfaction, so it is possible to more accurately measure the intensity and type (fragrance) of various types of odors. .
In particular, according to the present embodiment, data processing is performed so as to obtain an element odor vector on a two-dimensional plane orthogonal coordinate system that facilitates vector calculation processing. Therefore, when a three-dimensional or higher coordinate system is used. Data processing is simpler than that.
In particular, according to this embodiment, the magnitude of each element odor vector corresponding to the intensity and type of the odor of the gas to be measured and the angle (inclination) between the coordinate axes can be visually confirmed on the display.
In particular, according to the present embodiment, a data set combining a plurality of elemental odor intensity rate SR data and a plurality of elemental fragrance θ data related to the type of odor (fragrance) is obtained. By storing and using the comprehensive odor data set for the odor in the memory of the arithmetic control unit 60, the odor type (fragrance) of the gas to be measured can be specified with high accuracy.
In particular, according to the present embodiment, the accuracy of identifying the odor type (fragrance) of the gas to be measured is improved by storing and storing the total odor data set of known odors in the memory of the arithmetic control unit 60. be able to.
In particular, according to the present embodiment, the value indicating the total intensity of the odor of the gas to be measured based on the data of the total odor intensity St, which is a square root value obtained by squaring and adding each of the plurality of sensor output correction values. Can be treated as Moreover, even if there is interference between the odor sensors with respect to the odor sensitivity characteristics, the influence of the interference is smaller than when the sensor output correction values are added as they are.
In particular, according to the present embodiment, the odor intensity approximated to the human olfactory characteristics is measured for the odor of the gas to be measured by the index value such as the odor index and the odor concentration calculated using the value of the total odor intensity St. Can be measured.

  In the above embodiment, the case where four types of odor sensors are used has been described. However, the present invention can also be applied to a case where three types of odor sensors are used or five or more types of odor sensors.

  Further, in the above-described embodiment, measurement of odor that is a bad odor such as ammonia has been described. However, the present invention can be similarly applied to measurement of not only such odor but also other types of odor, and similarly. The effect is obtained.

1 is an overall configuration diagram of an odor measuring apparatus according to an embodiment of the present invention. The schematic diagram which shows the tendency of the sensitivity selectivity of an odor sensor. Explanatory drawing of the peripheral circuit of the odor sensor arrange | positioned inside the measurement chamber of the odor measuring apparatus. The graph which shows the relationship between the sensor output value of odor sensor 10A, and the density | concentration of the odor component in to-be-measured gas. The graph which shows the relationship between the sensor output value of the odor sensor 10B, and the density | concentration of the odor component in to-be-measured gas. The graph which shows the relationship between the sensor output value of odor sensor 10C, and the density | concentration of the odor component in to-be-measured gas. The graph which shows the relationship between the sensor output value of odor sensor 10D, and the density | concentration of the odor component in to-be-measured gas. The graph obtained by making sensor output value a and b of odor sensor 10A, 10B measured by changing the density | concentration of various odor components on the horizontal axis (X-axis) and the vertical axis (Y-axis), respectively. The graph which shows the relationship between the measured value before correction | amendment of the total odor intensity | strength St measured about various odor components, and the density | concentration of an odor component. Explanatory drawing of the element odor vector obtained when the sensor output correction value data of each odor sensor are taken on each coordinate axis of the orthogonal coordinate system. Explanatory drawing of the element odor vector obtained by replacing the coordinate axis corresponding to the sensor output correction value of each odor sensor. Explanatory drawing of the screen of the computer display which displayed sensor output correction value AD of each odor sensor, element intensity | strength of each element odor vector, element fragrance | flavor, etc. FIG. (A) And (b) is explanatory drawing of the screen of the display which displayed the element odor vector in case the odor component in to-be-measured gas is hydrogen sulfide 4ppm. (A) And (b) is explanatory drawing of the screen of the display which displayed the element odor vector in case the odor component in to-be-measured gas is ammonia 18ppm. (A) And (b) is explanatory drawing of the screen of the display which displayed the element odor vector in case the odor component in to-be-measured gas is hydrogen sulfide 0.1ppm. (A) And (b) is explanatory drawing of the screen of the display which displayed the element odor vector in case the odor component in to-be-measured gas is ammonia 1.8ppm. The graph of the calibration curve measured about the to-be-measured gas containing three types of odor components related to paint. The graph of the calibration curve measured about the to-be-measured gas whose odor component is hydrogen sulfide and the to-be-measured gas whose odor component is hydrogen sulfide. The odor spectrum figure which displayed the value of the element intensity rate SR measured about the to-be-measured gas whose odor component is 4 ppm of ammonia, and the to-be-measured gas whose odor component is 18 ppm of hydrogen sulfide. The feature extraction figure which attached the black mark to the part of the mesh corresponding to the angle of the element odor vector in which element intensity rate SR exceeds the average value (= 1).

Explanation of symbols

10A, 10B, 10C, 10D Odor sensor 20 Measurement chamber 30 Gas supply means 40 Pump part 50 Sensor signal processing means 51 Sensor circuit part 52 AD converter 60 Arithmetic control part 70 Display part 80 Communication control part

Claims (4)

An odor measuring device that measures the odor of a gas to be measured using a plurality of odor sensors having different sensitivity characteristics to odors ,
Three or more types of odor sensors as the plurality of odor sensors;
Based on the output values of the plurality of odor sensors with respect to the odorless reference gas, the output values of the plurality of odor sensors with respect to the gas to be measured are corrected, and each of the plurality of sensor output correction values is expressed in a coordinate system of two or more dimensions. Data processing means for processing the sensor output correction value data so as to obtain element odor vectors defined on each of a plurality of quadrants of the coordinate system on different coordinate axes extending from the origin;
Data storage means for storing data;
With
The data processing means calculates a value obtained by dividing the size of each element odor vector by the average value of the sizes of the plurality of element odor vectors as element odor intensity rate data, and the plurality of element odor vectors and their An angle formed by at least one coordinate axis in the quadrant where the element odor vector exists is calculated as element fragrance data, and data obtained by combining the plurality of element odor intensity rate data and the plurality of element fragrance data Data processing is performed so that the set is stored in the data storage means as a comprehensive odor data set for the odor of the gas to be measured.
An odor measuring device characterized by that. The odor measuring device according to claim 1 ,
The data processing means stores the comprehensive odor data set obtained by measuring an odor having a known intensity and type (fragrance) in the data storage means in association with the odor intensity and type (fragrance) data. The above-mentioned comprehensive odor data set when measuring the measured gas whose odor intensity and type (fragrance) are unknown is compared with the comprehensive odor data set of known odors stored in the data storage means. An odor measuring apparatus that performs data processing so as to specify the type of the unknown odor (fragrance) based on the comparison result. The odor measuring apparatus according to any one of claims 1 to 2 ,
The odor measuring apparatus, wherein the data processing means performs data processing so as to calculate a square root value obtained by squaring and adding each of the plurality of sensor output correction values as data of total odor intensity. The odor measuring device according to claim 3 ,
The data processing means multiplies the value of the total odor intensity by a constant determined by the element intensity rate data and the element fragrance data or the odor type (fragrance) data, thereby An odor measuring apparatus that performs data processing so as to calculate a numerical value of an odor intensity approximate to a characteristic.

Images