Estimation system, its control method, and control program

Abstract

This system estimates the degree to which odors originating from areas that could be odor sources are influencing odors in a specific area. [Solution] An estimation system (100) according to one embodiment of the present invention comprises: a first acquisition unit (11) that acquires target data output from an odor measuring device (30) that measures the odor of a target gas; and an estimation unit (13) that uses an estimation model (21) that has learned the characteristics of the gas of interest to estimate the degree to which the odor of the gas of interest influences the odor of the target gas from target input data including the target data.

Description

Technical Field

[0006] , ,

[0001] The present invention relates to an estimation system, a control method thereof, and a control program.

Background Art

[0002] In recent years, devices for evaluating a target based on the odor in a specific target have been developed. For example, in Patent Document 1, a specific device is described for identifying the source and cause of an unknown odor in a steel mill or the like, using coke odor and tar odor, which are odors collected from odor generation sources assumed in a steel mill, as reference odors.

[0003] The odor generated at an odor source may diffuse into various surrounding areas by air currents or the like. When such an odor diffuses into a surrounding residential area, it may not only cause discomfort to residents but also pose a risk of health damage to the residents. In such a case, it is necessary to identify the odor source.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, when there are multiple areas that can be odor sources in the vicinity of a residential area, it has been difficult to identify which area the odor felt in the house originated from and what caused the odor. Therefore, for example, when there are multiple factories in the vicinity of a residential area, there is a possibility that complaints related to the odor are conveyed from the residents living in the residential area to factories that are not odor sources.

[0006] One aspect of the present invention aims to realize an estimation system that estimates the degree to which odors originating from an area that may be an odor source are affecting odors in a specific area. [Means for solving the problem]

[0007] An estimation system according to one aspect of the present invention includes: a first acquisition unit that acquires target data output from an odor measuring device when the odor of a target gas, which is a gas collected in a target region, is measured using an odor measuring device equipped with one or more odor sensor elements capable of interacting with odor substances via a contact surface that comes into contact with the gas containing odor substances; and an estimation unit that estimates the degree to which the odor of the target gas is influencing the odor of the target gas from target input data including the target data, using an estimation model that has learned the characteristics of the target gas, which is a gas emitted from a region of interest.

[0008] A control method for an estimation system according to one aspect of the present invention is a control method for an estimation system executed by one or more computers, comprising: a first acquisition step of acquiring target data output from an odor measuring device when the odor of a target gas, which is a gas collected in a target area, is measured using an odor measuring device equipped with one or more odor sensor elements capable of interacting with odor substances via a contact surface that comes into contact with the gas containing odor substances; and an estimation step of estimating the degree to which the odor of the gas of interest is influencing the odor of the target gas from target input data including the target data, using an estimation model that has learned the characteristics of the gas of interest. [Effects of the Invention]

[0009] According to one aspect of the present invention, an estimation system can be realized that estimates the degree to which odors originating from an area that may be an odor source are affecting odors in a specific area. [Brief explanation of the drawing]

[0010] [Figure 1] This is a schematic diagram showing an example of the configuration of an estimation system according to one embodiment of the present invention. [Figure 2] This is a functional block diagram showing an example of an estimation system including an odor measuring device according to one embodiment of the present invention. [Figure 3] This is a top view showing an example of the configuration of an odor sensor element. [Figure 4] This is a functional block diagram showing an example of an estimation system according to one embodiment of the present invention. [Figure 5] This flowchart shows an example of the process flow for estimating the degree of influence of odor on a target gas using an estimation system according to one embodiment of the present invention. [Figure 6] This is a schematic diagram showing an example of the configuration of an estimation system according to one embodiment of the present invention. [Figure 7] This is a functional block diagram showing an example of an estimation system according to one embodiment of the present invention. [Figure 8] This flowchart shows an example of the process flow for estimating the degree of influence of odor on a target gas using an estimation system according to one embodiment of the present invention. [Modes for carrying out the invention]

[0011] One embodiment of the present invention is described below, but the present invention is not limited thereto. Unless otherwise specified in this specification, "A to B" representing a numerical range means "A or greater and B or less".

[0012] [Embodiment 1] (Estimated System 100) First, an overview of the estimation system 100 according to one embodiment of the present invention will be described using Figure 1. Figure 1 is a schematic diagram showing an example of the configuration of the estimation system 100 according to one embodiment of the present invention.

[0013] As shown in Figure 1, the estimation system 100 includes an odor measuring device 30 equipped with an odor sensor element 31, and an estimation device 10 that, when the odor of a target gas is measured using the odor measuring device 30, uses the target data output from the odor measuring device 30 to estimate the degree to which the odor of the gas of interest is influencing the odor of the target gas. As shown in Figure 1, the estimation system 100 may also have the odor measuring device 30 and the estimation device 10 connected via a wide-area communication network 40.

[0014] In this specification, "gas of interest" means a gas released from a region of interest. The gas of interest is not particularly limited as long as it has an odor. Examples of odors that the gas of interest may have include odors from the manufacture of recycled rubber, odors that may be generated during food manufacturing, the smell of burnt organic matter, the smell of decaying organic matter, the chemical smell of organic solvents, the smell of waste treatment, and the smell of processing oils (such as lubricants). The gas of interest may have multiple types of odors. In this specification, "region of interest" means a region that can be a source of odor. The region of interest may contain elements that can be sources of odor, such as buildings such as factories and waste treatment plants, environments such as lakes and marshes, fields and farms, garbage collection sites, and habitats of wild animals.

[0015] Preferably, the gas of interest is collected within the area of ​​interest and / or near the area of ​​interest. Furthermore, the gas of interest may be collected at only one location or at multiple locations. In one embodiment, if gas is emitted from multiple locations in a facility within the area of ​​interest, the gas of interest may be collected from each of these locations. Specifically, if a factory is located in the area of ​​interest, for example, the gas of interest may be collected from the factory's chimney, exhaust vent, around the manufacturing equipment, and inside the manufacturing room. If there are multiple elements in the area of ​​interest that could be sources of odor, the gas of interest may be collected from each of them. If the gas of interest emitted in the area of ​​interest changes depending on the time of day, the gas of interest may be collected from the same area of ​​interest at different times. According to the above configuration, it becomes easy to identify which element in the area of ​​interest is the cause of the odor.

[0016] In this specification, the "target gas" means the gas emitted from the target area. The target gas is not particularly limited as long as it is a gas having the odor that the above-mentioned target gas may have. In this specification, the "target area" is not particularly limited as long as it is an area affected by odor damage. Examples of areas that can be the target area include houses, public facilities, commercial facilities, and the like. The target gas is preferably collected in at least one of the target area and the vicinity of the target area. Also, the position where the target gas is collected may be only one location, or the target gas may be collected at a plurality of locations.

[0017] In this specification, the "degree to which the odor of the target gas affects the odor of the target gas" means the coincidence rate of the odor of the target gas with respect to the odor of the target gas. More specifically, it means the coincidence rate between the measurement signal obtained by measuring the target gas by the odor measuring device 30 and the measurement signal of the target gas. The degree may be represented by a numerical value of, for example, 0 to 100%, or may be represented in a plurality of stages such as low, medium, and high. The coincidence rate is output by the estimation device 10 based on the measurement signal output by the odor measuring device 30.

[0018] In this specification, the "odor measuring device 30" means a measuring device including one or more odor sensor elements 31 capable of interacting with the odor substance through a contact surface that contacts the gas containing the odor substance. Note that the gas sensor described later is a sensor that detects a predetermined component and does not correspond to the odor measuring device 30.

[0019] In this specification, the data obtained when the odor of the target gas is measured by the odor measuring device 300 is referred to as the target data, and the data obtained when the odor of the target gas is measured by the odor measuring device 30 is referred to as the target data.

[0020] According to the above configuration, in the estimation system 100, the estimation device 10 can estimate the degree to which the odor of the gas of interest is influencing the odor of the target gas, based on the information transmitted from the odor measuring device 30. This makes it possible to estimate the degree to which the odor in the area of ​​interest is influencing the odor in the target area.

[0021] The number of odor measuring devices 30 included in the estimation system 100 according to one embodiment of the present invention is not particularly limited and may be one or more, three or more, five or more, or ten or more. Furthermore, the measurement by the odor measuring device 30 may be performed by installing the odor measuring device 30 in the area of ​​interest and the target area, respectively, or by measuring the gas collected in the area of ​​interest and the target area using the odor measuring device 30.

[0022] The wide-area communication network 40 is not particularly limited and may include the Internet, telephone lines, mobile communication networks, CATV communication networks, satellite communication networks, etc. Furthermore, the wide-area communication network 40 may be connected to a cloud server that stores information transmitted by the user terminal 50 and estimated values ​​output by the estimation device 10. If the wide-area communication network 40 is connected to a cloud server, the estimation device 10 may be implemented as the cloud server. The user terminal 50 and the estimation device 10 may be connected via a local area network connection, LTE communication, etc., without going through an internet service provider.

[0023] Furthermore, the estimation device 10 may transmit the estimation results to the user terminal 50. The user terminal may be the owner of the estimation device 10 or a user of the estimation system 100. Examples of user terminals 50 include laptop computers, smartphones, and tablet devices. The estimation device 10 may also display the estimation results on a web page or the like accessible by the user terminal 50.

[0024] The estimation system 100a may further include sensors other than the odor measuring device 30. Examples of such sensors include temperature sensors, humidity sensors, wind direction sensors, wind speed sensors, and atmospheric pressure sensors.

[0025] According to the above configuration, it is possible to estimate the degree to which the odor of the gas of interest is influencing the odor of the target gas, and based on the estimation results, it is possible to determine whether the facility in the area of ​​interest is emitting a gas that is causing odor in the target area.

[0026] (Odor measuring device 30) The following describes the overview and effects of the odor measuring device 30 to which the odor sensor element 31 is applied, using Figure 2. Figure 2 is a functional block diagram showing an example of the configuration of an estimation system 100 that includes the odor measuring device 30 to which the odor sensor element 31 is applied. The odor measuring device 30 includes an odor sensor element 31 for detecting odor substances, a power supply 32 (power supply), a clock 33 (timer), a control unit 34, and a communication unit 35. As described above, the odor measuring device 30 may be connected to a wide-area communication network 40.

[0027] The power supply 32 is a power source for supplying power to the odor sensor element 31. The power supply 32 may be a constant voltage power supply, a constant current power supply, or an AC power supply. If the power supply 32 is a constant voltage power supply, it supplies current (for example, a DC current of 1 μA to 10 mA) to the odor sensor element 31 via lead wires. The voltage value supplied by the power supply 32 is, for example, 0.01 V to 10 V, more specifically 2.5 V or 5.0 V.

[0028] Clock 33 measures the time. Clock 33 transmits the measured time to the control unit 34. Clock 33 may be a clock whose time is set by the user, or it may be a radio-controlled clock.

[0029] The control unit 34 controls all parts of the odor measuring device 30. The control unit 34 also outputs the odor detected by the odor sensor element 31 as a measurement signal. The control unit 34 may also output the odor measurement signal according to the time measured by the clock 33.

[0030] The communication unit 35 transmits the measurement signal output by the control unit 34. The communication unit 35 transmits the measurement signal to the wide-area communication network 40, and the transmitted measurement signal is acquired by the estimation device 10.

[0031] The odor measuring device 30 may also include a housing, although this is not a mandatory component. The housing is a container capable of enclosing air containing odor substances. If a housing is included, the odor sensor element 31 is installed inside the housing.

[0032] The odor measuring device 30 outputs a measurement signal that shows the change in the electrical conductivity of the odor sensor element 31 over time, before and after odor substances are adsorbed onto the odor sensor element 31. This makes it possible to detect and identify various odor substances.

[0033] <Odor sensor element 31> Figure 3 is a top view showing an example of the configuration of the odor sensor element 31. The odor sensor element 31 comprises an odor substance receiving layer 315 containing the resin composition described above, a first metal wiring 313A, and a second metal wiring 313B. In the following, when the first metal wiring 313A and the second metal wiring 313B are not distinguished, they may be referred to simply as metal wiring 313.

[0034] The first metal wiring 313A and the second metal wiring 313B are metal wirings that function as electrodes for measuring changes in the electrical conductivity of the odor substance receiving layer 315 (i.e., the resin composition). That is, the first metal wiring 313A and the second metal wiring 313B are spaced apart from each other, and the odor substance receiving layer 315 is in contact with at least a portion of the first metal wiring and at least a portion of the second metal wiring. In one example, the first metal wiring 313A and the second metal wiring 313B are metal wirings that are not in direct contact with each other, and may be substantially parallel to each other, as shown in Figure 3.

[0035] As shown in Figure 3, the metal wiring 313, including the first metal wiring 313A and the second metal wiring 313B, may be arranged on a substrate 311. The substrate 311 may be a substrate such as glass epoxy, which is commonly used in electronic circuits. The metal wiring 313 may be made of copper or gold. The thickness of the first metal wiring 313A and the second metal wiring 313B, as viewed from a direction perpendicular to the surface of the substrate, may be, for example, 10 μm to 2 mm.

[0036] The odor substance receiving layer 315 may be in contact with at least a portion of the first metal wiring 313A and at least a portion of the second metal wiring 313B. The odor substance receiving layer 315 may be arranged to fill the region sandwiched between the first metal wiring 313A and the second metal wiring 313B, for example, as shown in Figure 3.

[0037] If the electrical conductivity of the odor substance receiving layer 315 (i.e., the electrical conductivity of the odor sensor element 31) is low, it is desirable that the distance between the first metal wiring 313A and the second metal wiring 313B be less than or equal to a predetermined distance (for example, 500 μm).

[0038] The odor substance receiving layer may include a resin composition. The resin composition may include a resin and further include one or more selected from a surfactant and a filler (e.g., a conductive carbon material). In this specification, "odor substance receiving layer" means a layer that adsorbs odor substances to be identified. The odor substance receiving layer 315 is formed from the above-described resin composition. The odor substance receiving layer 315 may be provided as part of the odor sensor element 31. The electrical resistance of this odor substance receiving layer 315 changes in response to the adsorption of odor substances, etc. That is, the odor sensor element 31 is an odor detection device equipped with such an odor substance receiving layer 315, and the odor measurement method of the odor sensor element 31 may be a chemi-resistor type.

[0039] When the odor sensor element 31 is a chemisistor type containing a resin composition, the change in electrical conductivity over time differs depending on whether odor substance A is adsorbed or odor substance B, which is different from odor substance A, is adsorbed. Therefore, it is possible to detect and identify various odor substances. In the odor measuring device 30 described later, multiple odor sensor elements 31 are arranged, each having a substrate 311 equipped with a configuration for detecting odor substances (metal wiring 313 and odor substance receiving layer 315). Each substrate 311 has multiple sets of odor substance receiving layers 315, each containing odor substances that can be measured differently from each other. Each of the multiple odor sensor elements 31 may also be equipped with a constant voltage power supply and a voltmeter. In the odor measuring device 30, one configuration for detecting odor substances (metal wiring 313 and odor substance receiving layer 315) may be arranged on each substrate 311. Alternatively, in the odor measuring device 30, multiple sets of configurations for detecting odor substances (metal wiring 313 and odor substance receiving layer 315) may be arranged on a single substrate 311. In the latter case, a constant voltage power supply and a voltmeter are connected to each of the sets provided on the circuit board 311.

[0040] The resin compositions contained in the odor substance receiving layers 315 of the multiple odor sensor elements 31 in the odor measuring device 30 may be the same or different. If the odor substance receiving layers 315 contained in the multiple odor sensor elements 31 have the same composition, each of the multiple odor substance receiving layers 315 can detect the same odor substance. If the multiple odor sensor elements 31 each contain odor substance receiving layers 315 with different compositions, each of the multiple odor substance receiving layers 315 will respond differently to the odor substance. In this way, by providing multiple sets of configurations for detecting odor substances, the accuracy of odor substance identification in the odor measuring device 30 can be improved.

[0041] The odor measuring device 30 described above can output the change in the electrical conductivity of the odor sensor element 31 over time for each odor substance when various odor substances are adsorbed onto the odor sensor element 31. By applying this odor measuring device 30, it is possible to compare the change in the electrical conductivity of the odor sensor element 31 over time when odor substance A is adsorbed onto the odor sensor element 31 with the change in the electrical conductivity of the odor sensor element 31 over time when odor substance B is adsorbed onto the odor sensor element 31. Based on such comparison results, an estimation device 10 can be realized that can estimate the degree to which odors in a region of interest have an influence on odors in a target region, based on the odor substances adsorbed onto the odor sensor element 31.

[0042] (Estimation device 10) The following describes the overview and effects of the estimation device 10. The estimation device 10 is a device that estimates the degree of influence that odors in the area of ​​interest have on odors in the target area, based on the measurement signals output from the odor measuring device 30 described above. The estimation device 10 uses an estimation model that has already learned the characteristics of the gas of interest.

[0043] Figure 4 is a functional block diagram showing an example of the configuration of the estimation device 10. The estimation device 10 includes a control unit 1 that controls all parts of the estimation device 10, and a storage unit 2 that stores various data used by the estimation device 10, but is not limited to this configuration. For example, the storage unit 2 may be an external device attached to the estimation device 10. Also, as described above, the estimation device 10 may be connected to the wide-area communication network 40.

[0044] <Control Unit 1> First, let's describe the control unit 1. The control unit 1 includes a first acquisition unit 11, an extraction unit 12, an estimation unit 13, and an output control unit 14. Furthermore, some of the blocks included in the control unit 1 may be delegated to other devices capable of communicating with the estimation device 10, and those blocks may be omitted from the control unit 1. For example, the function of the output control unit 14 may be delegated to another device. In this case, the estimation device 10 may output the results estimated by the estimation unit 13 through the other device.

[0045] The first acquisition unit 11 acquires target data, including the measurement signal output from the odor measuring device 30, via the wide-area communication network 40. The first acquisition unit 11 may be configured to acquire the measurement signal output from the odor measuring device 30 in real time, or it may be configured to acquire the measurement signal stored on the wide-area communication network 40 at regular intervals. The first acquisition unit 11 may store the target input data, including the acquired target data, in the storage unit 2. In one embodiment, when the odor of the gas of interest is measured by the odor measuring device 30, the first acquisition unit may further acquire the data of interest output from the odor measuring device 30 and the measurement data measured by sensors other than the odor measuring device 30.

[0046] The extraction unit 12 extracts characteristic quantities that represent the characteristics of the target gas from the measurement signal acquired by the first acquisition unit 11. The characteristic quantities extracted by the extraction unit 12 may be values ​​relating to at least one of the following: the amount of change in the measurement signal, the rate of change in the measurement signal, and the waveform (frequency change) of the measurement signal. The extraction unit 12 may further extract principal components as characteristic quantities from the results of the principal component analysis acquired by the first acquisition unit 11.

[0047] The estimation unit 13 inputs the target data or the target input data including features extracted from the target data into the estimation model 21 to estimate the degree to which the odor of the gas of interest influences the odor of the target gas. The estimation unit 13 may also store the estimation results in the storage unit 2.

[0048] The estimation unit 13 may output a numerical value based on the agreement rate or similarity between the gas of interest and the target gas as an estimation result, for example, as a value in the range of 0 to 100%, or it may output two or more results such as agreement or mismatch based on the estimated agreement rate or similarity value. In one embodiment, the numerical value may be output as unpleasantness level, claim risk level, etc. Specifically, the estimation unit 13 may output an estimation result of mismatch if the agreement rate or similarity value is less than 50%, and agreement if it is 50% or more.

[0049] If the first acquisition unit 11 acquires further data of interest, the estimation unit 13 may input this data of interest into the estimation model 21. This allows the error between the data of interest included in the training data of the estimation model 21 and the data of interest actually measured to be estimated. With the above configuration, the degree can be estimated more accurately.

[0050] In one embodiment, the estimation unit 13 may perform estimation without inputting the target input data into the estimation model 21. Such methods include, for example, performing estimation by comparing the data of interest with the target input data.

[0051] The output control unit 14 outputs the estimation results output by the estimation unit 13 to the user terminal 50 via the wide-area communication network 40. The output control unit 14 may further output the target input data and data of interest acquired by the first acquisition unit 11 to the user terminal 50. The output control unit 14 may directly transmit the estimation results to the user terminal 50, or it may transmit the estimated values ​​to a web page or the like on the wide-area communication network 40 that is accessible to the user terminal 50. Alternatively, the output control unit 14 may be another device that has output control functionality, which is capable of acquiring estimated values ​​from the estimation device 10. In this case, the estimation device 10 may transmit the results estimated by the estimation unit 13 via the other device. In one embodiment, the output control unit 14 may further include a communication unit function.

[0052] <Storage section 2> Next, we will describe the memory unit 2. The memory unit 2 stores the estimation model 21, the target input data 22, and the estimation result 23.

[0053] The estimation model 21 is trained by machine learning using training data. The training data includes, as explanatory variables, (1) data of interest output from the odor measuring device 30 when the odor of the gas of interest is measured using the odor measuring device 30. Preferably, the training data further includes, as explanatory variables, (2) data of non-interest output from the odor measuring device 30 when the odor of non-interest gases, which are gases emitted from each of the one or more non-interest regions different from the region of interest, is measured using the odor measuring device 30.

[0054] In the machine learning of the estimation model 21, further data preprocessing and feature extraction may be performed. Furthermore, machine learning algorithms may be used for machine learning to generate the estimation model 21.

[0055] (Feature extraction) The training data for generating the estimation model 21 by machine learning may be the measured values ​​themselves or features extracted from the measured values. Features may be, for example, statistics, differential and integral values, peak detection values, or autocorrelation values. Examples of statistics include the mean, variance, maximum value, minimum value, the difference between the maximum and minimum values, and the standard deviation. Examples of differential and integral values ​​include the derivative (the slope of a graph showing the change in measured values ​​over time) and the integral (the area of ​​the region defined by the curve showing the change in measured values ​​in a graph showing the change in measured values ​​over time and the horizontal axis (e.g., the time axis)). Examples of peak detection values ​​include the number and height of peaks in the change in measured values ​​(e.g., change over time). Examples of autocorrelation values ​​include the difference in the change in measured values ​​(e.g., change over time). Extraction of these features from measured values ​​can be carried out based on known methods.

[0056] (Pre-treatment method) The training data for generating the estimation model 21 by machine learning may be used for machine learning without preprocessing, or it may be used after predetermined preprocessing as necessary. Furthermore, if preprocessing is performed, it may be performed before feature extraction, after feature extraction, or both before and after feature extraction. Preprocessing may be performed by known methods. Known methods include correction, denoising, standardization, data transformation, smoothing, and data augmentation. Examples of correction include integration, addition, subtraction, and division based on the output ratio, independent component analysis (ICA), or statistics, based on the measurement results of a standard gas by multiple sensor elements or commercially available sensors (e.g., temperature sensors or humidity sensors). Examples of denoising include removal of outliers or white noise. Examples of standardization include normalization and regularization of features. Examples of data transformation include trend removal, frequency transformation, and logarithmic transformation. Examples of smoothing include obtaining the moving average of the data and obtaining the difference. Data augmentation methods include, for example, adding the same sample data (e.g., adding data assuming a normal distribution) or adding new sample data (e.g., adding data related to the mixing ratio of vectors).

[0057] (Machine learning algorithms) Machine learning algorithms that can be used to create the estimated model 21 include regression analysis, classification, trees, time series analysis, neural networks, and clustering. Examples of regression analysis include logistic regression, Lasso regression, elastic network regression, support vector regression (SVR), linear regression, Ridge regression, and ensemble regression. Examples of classification include k-nearest neighbor method, support vector classification (SVC), Naibe Bayes classifier, stochastic gradient descent (SGD), and kernel approximation. Examples of trees include decision trees, regression trees, random forests, boosting (lightGBM, XGboost), and stacking. Examples of time series include AR, MA, ARIMA, and state space. Examples of neural networks include multilayer perceptrons (MLPs), convolutional neural networks (CNNs), recurrent neural networks (RNNs), residue neural networks (ResNets), transformers, and graph neural networks (GNNs). Examples of clustering methods include Gaussian mixture models (GMMs), k-means, mini k-means, variational Gaussian mixture models (VBGMMs), and kernel approximation. In one embodiment, the estimation model 21 may further perform machine learning using the data of interest acquired by the first acquisition unit 11 as training data.

[0058] The target input data 22 includes target data, which is the measurement signal data output from the odor measuring device 30. The target input data 22 may also include the location where the target data was measured, environmental conditions such as temperature and humidity on the day of measurement, and the identification number of the odor measuring device 30. In one embodiment, the target input data may also include feature quantities extracted from the target data.

[0059] Estimation result 23 is the result of estimating the degree to which the odor of the gas of interest influences the odor of the target gas, output from the estimation unit 13. Estimation result 23 may be the agreement rate between the odor of the target gas and the odor of the gas of interest, or it may be the similarity between the odor of the target gas and the odor of the gas of interest. (Processing performed by estimated system 100) A control method for the estimation system 100 according to one embodiment of the present invention will be described with reference to Figure 5. Figure 5 is a flowchart showing an overview of the control method for the estimation system 100.

[0060] In step S1, the first acquisition unit 11 acquires target data output from the odor measuring device 30 when the odor of the target gas, which is a gas collected in the target area, is measured using the odor measuring device 30 (first acquisition step). In step S1, the first acquisition unit 11 may further acquire target data output from the odor measuring device 30 when the odor of the gas of interest is measured by the odor measuring device 30.

[0061] In step S2, the extraction unit 12 extracts features from the target data acquired in step S1. The features extracted by the extraction unit 12 may be values ​​relating to at least one of the following: the amount of change in the measurement signal, the rate of change in the measurement signal, and the waveform (frequency change) of the measurement signal. The extraction unit 12 may further extract principal components as features from the results of the principal component analysis acquired by the first acquisition unit 11.

[0062] In step S3, the target data, or target input data including features extracted from the target data, is input to the estimation model 21 to estimate the degree to which the odor of the gas of interest influences the odor of the target gas (estimation step). If the first acquisition unit 11 has acquired the data of interest in step S11, the data of interest may be input to the estimation model 21 to estimate the degree, from the viewpoint of improving estimation accuracy.

[0063] In step S4, the output control unit 14 transmits (outputs) the estimation result output by the estimation unit 13 to the user terminal 50 via the wide-area communication network 40.

[0064] Conventionally, it has been difficult to determine whether an odor at a specific location is affecting an odor at a different location. With the estimation system 100, the estimation unit 13 makes an estimation based on the measurement results of the odor measuring device 30, making it possible to estimate the degree of influence of the odor.

[0065] [Embodiment 2] The outline of an estimation system 100a according to another embodiment of the present invention will be described below with reference to Figure 6. Figure 6 is a schematic diagram showing an example of the configuration of an estimation system 100a different from that of Embodiment 1. Matters that have already been explained will be omitted.

[0066] The estimation system 100a comprises an estimation device 10a, an odor measuring device 30, a gas sensor 60, and a user terminal 50. The estimation system 100a may also include a wide-area communication network 40 as needed. In the estimation system 100a, the estimation device 10a, the gas sensor 60, and the user terminal 50 may be connected via the wide-area communication network 40. Alternatively, in the estimation system 100a, the odor measuring device 30, the gas sensor 60, and the user terminal 50 may be connected via a local area network (LAN) connection, LTE communication, etc., without going through an internet service provider.

[0067] Since the estimation system 100a is equipped with a gas sensor 60 in addition to the odor measuring device 30, it can perform estimations with higher accuracy.

[0068] The gas sensor is preferably one or more selected from, for example, odor sensors, volatile organic compound (VOC) sensors, and optical sensors. More specifically, examples include ammonia sensors, hydrogen sulfide sensors, trimethylamine sensors, methyl mercaptan sensors, ethyl mercaptan sensors, dimethyl disulfide sensors, styrene sensors, carbon disulfide sensors, dimethyl trisulfide sensors, and other alcohol sensors.

[0069] (Estimation device 10a) The following describes an overview of an estimation device 10a according to another embodiment of the present invention, based on Figure 7. Figure 7 is a functional block diagram showing an example of the configuration of the estimation device 10a. The estimation device 10a includes a control unit 1a that controls all parts of the estimation device 10a, and a storage unit 2a that stores various data used by the estimation device 10a, but is not limited to this configuration. The control unit 1a includes a first acquisition unit 11, an extraction unit 12, an estimation unit 13, an output control unit 14, and a second acquisition unit 15. The storage unit 2a stores the estimation model 21a, the target input data 22a, and the estimation result 23.

[0070] <Control Unit 1a> The second acquisition unit 15 acquires target detection data output from the gas sensor 60 when the target gas is measured using the gas sensor 60. In one embodiment, the second acquisition unit 15 may further acquire target detection data output from the gas sensor 60 when the gas of interest is measured using the gas sensor 60.

[0071] The estimation unit 13 inputs the target data and target detection data, or target input data including features extracted from these data, into the estimation model 21a to estimate the degree to which the odor of the gas of interest is influencing the odor of the target gas. If the second acquisition unit 15 has further acquired the target detection data, the estimation unit 13 may input the target data and the target detection data into the estimation model 21a. With the above configuration, the degree can be estimated more accurately.

[0072] <Storage section 2a> The estimation model 21a is trained by machine learning using training data. The training data includes, as explanatory variables: (1) focus data output from the odor measuring device 30 when the odor of the gas of interest is measured using the odor measuring device 30; (2) focus detection data output from the gas sensor 60 when the gas of interest is measured using the gas sensor 60 capable of detecting predetermined components contained in the gas; (3) non-focus data output from the odor measuring device 30 when the odor of non-focus gases, which are gases emitted from one or more non-focus regions different from the region of interest, is measured using the odor measuring device 30; and (4) non-focus detection data output from the gas sensor when the non-focus gas is measured using the gas sensor capable of detecting predetermined components contained in the gas.

[0073] The target input data 22a further includes target detection data in addition to the data described above. In one embodiment, the target input data may include features extracted from the target data and the target detection data.

[0074] With the above configuration, the measurement results from the gas sensor 60, in addition to the odor measuring device 30, are also used in the estimation, thus improving the accuracy of the estimation.

[0075] (Processing performed by estimated system 100a) A control method for the estimation system 100a according to one embodiment of the present invention will be described with reference to Figure 8. Figure 8 is a flowchart showing an overview of the control method for the estimation system 100a.

[0076] In step S12, the second acquisition unit 15 acquires the target detection data output from the odor measuring device 30 when it measures the odor of the target gas, which is a gas collected in the target area, using the gas sensor 60 (second acquisition step).

[0077] In step S13, the extraction unit 12 extracts features from the target data and target detection data acquired in steps S1 and S12. The features extracted by the extraction unit 12 may be values ​​relating to at least one of the following: the amount of change in the measurement signal, the rate of change in the measurement signal, and the waveform (frequency change) of the measurement signal. The extraction unit 12 may further extract principal components as features from the results of the principal component analysis acquired by the first acquisition unit 11.

[0078] In step S14, the target data and target detection data, or target input data including features extracted from the target data and target detection data, are input to the estimation model 21 to estimate the degree to which the odor of the gas of interest influences the odor of the target gas (estimation step). If the first acquisition unit 11 acquires the data of interest in step S1 and the second acquisition unit 15 acquires the data of interest detection in step S12, these data may be input to the estimation model 21a to estimate the degree, from the viewpoint of improving estimation accuracy.

[0079] With the above configuration, since the measurement results from the gas sensor 60, as well as the odor measuring device 30, are used for estimation, estimation can be performed with higher accuracy.

[0080] [Examples of implementation using software] The functions of the estimation devices 10 and 10a (hereinafter referred to as "devices") are programs that cause a computer to function as the device, and these can be realized by programs that cause a computer to function as each control block of the device (especially each part included in the control unit 1).

[0081] In this case, the device includes a computer having at least one control device (e.g., a processor) and at least one storage device (e.g., memory) as hardware for executing the program. By executing the program using this control device and storage device, the functions described in each of the embodiments are realized.

[0082] The above program may be recorded on one or more computer-readable recording media, not temporary ones. These recording media may or may not be provided by the above device. In the latter case, the program may be supplied to the above device via any wired or wireless transmission medium.

[0083] Furthermore, some or all of the functions of each of the above control blocks can also be realized by logic circuits. For example, an integrated circuit in which logic circuits functioning as each of the above control blocks are formed is also included in the scope of the present invention. In addition, it is also possible to realize the functions of each of the above control blocks by, for example, a quantum computer.

[0084] Furthermore, each process described in the above embodiments may be performed by AI (Artificial Intelligence). In this case, the AI ​​may operate on the control device described above, or it may operate on other devices (for example, an edge computer or a cloud server).

[0085] The present invention is not limited to the embodiments described above, and various modifications are possible within the scope of the claims. Embodiments obtained by appropriately combining the technical means disclosed in different embodiments are also included in the technical scope of the present invention.

[0086] 〔summary〕 An estimation system according to embodiment 1 of the present invention includes a first acquisition unit that acquires target data output from an odor measuring device when the odor of a target gas, which is a gas collected in a target area, is measured using an odor measuring device equipped with one or more odor sensor elements capable of interacting with the odor substance via a contact surface that comes into contact with the gas containing the odor substance, The system includes an estimation unit that uses a pre-trained estimation model to learn the characteristics of a gas of interest, which is a gas emitted from a region of interest, to estimate the degree to which the odor of the gas of interest influences the odor of the target gas, based on target input data including the target data.

[0087] In the estimation system according to aspect 2 of the present invention, the estimation model in aspect 1 may be generated by machine learning using training data that includes as explanatory variables: (1) data of interest output from the odor measuring device when the odor of the gas of interest is measured using the odor measuring device, and (2) data of non-interest output from the odor measuring device when the odor of non-interest gases, which are gases emitted from each of the one or more non-interest regions different from the region of interest, is measured using the odor measuring device.

[0088] In the estimation system according to embodiment 3 of the present invention, in embodiment 1 or 2, the gas of interest may be collected in at least one of the areas of interest and in the vicinity of the area of ​​interest.

[0089] The estimation system according to embodiment 4 of the present invention may further include a second acquisition unit that acquires target detection data output from the gas sensor when the target gas is measured using the gas sensor, in any of embodiments 1 to 3. The estimation unit, The degree may be estimated using the estimation model described above, from the target input data, which includes the target data and the target detection data.

[0090] The estimation system according to aspect 5 of the present invention, in aspect 4, the gas sensor is The sensor may be one or more selected from odor sensors, volatile organic compound sensors, and optical sensors.

[0091] A control method for an estimation system according to aspect 6 of the present invention is a control method for an estimation system executed by one or more computers, A first acquisition step involves obtaining target data output from an odor measuring device when the odor of a target gas, which is a gas collected in a target area, is measured using an odor measuring device equipped with one or more odor sensor elements capable of interacting with the odor substance via a contact surface that comes into contact with the gas containing the odor substance, and the odor of the target gas, which is a target gas, is measured using the odor measuring device. The method includes an estimation step of using an estimation model that has learned the characteristics of the gas of interest to estimate the degree to which the odor of the gas of interest influences the odor of the target gas from target input data including the target data.

[0092] A control program according to aspect 7 of the present invention is a control program for causing a computer to function as an estimation system described in any of aspects 1 to 6, wherein the computer functions as the first acquisition unit and the estimation unit. [Explanation of symbols]

[0093] 10, 10a Estimation device 11 First acquisition part 13 Estimation part 30 Odor measuring device 100, 100a Estimation System

Claims

1. A first acquisition unit acquires target data output from an odor measuring device when the odor of a target gas, which is a gas collected in a target area, is measured using an odor measuring device equipped with one or more odor sensor elements capable of interacting with the odor substance via a contact surface that comes into contact with the gas containing the odor substance. The system includes an estimation unit that uses a pre-trained estimation model to determine the degree to which the odor of the target gas influences the odor of the target gas, based on target input data including the target data, using the characteristics of the target gas, which is a gas emitted from a region of interest. Estimation system.

2. The estimation model is generated by machine learning using training data that includes as explanatory variables: (1) data of interest output from the odor measuring device when the odor of the gas of interest is measured using the odor measuring device, and (2) data of non-interest output from the odor measuring device when the odor of non-interest gases, which are gases emitted from each of the one or more non-interest regions different from the region of interest, is measured using the odor measuring device. The estimation system according to claim 1.

3. The gas of interest is collected in at least one of the areas of interest and in the vicinity of the area of ​​interest. The estimation system according to claim 1.

4. The system further includes a second acquisition unit that acquires target detection data output from the gas sensor when the target gas is measured using the gas sensor. The estimation unit, Using the estimation model, the degree is estimated from the target input data, which includes the target data and the target detection data. The estimation system according to claim 1.

5. The aforementioned gas sensor is It is one or more types selected from odor sensors, volatile organic compound sensors, and optical sensors. The estimation system according to claim 4.

6. A method for controlling an estimation system run by one or more computers, A first acquisition step involves obtaining target data output from an odor measuring device when the odor of a target gas, which is a gas collected in a target area, is measured using an odor measuring device equipped with one or more odor sensor elements capable of interacting with the odor substance via a contact surface that comes into contact with the gas containing the odor substance. The method includes an estimation step of using an estimation model that has learned the characteristics of the gas of interest to estimate the degree to which the odor of the gas of interest influences the odor of the target gas from target input data including the target data. A method for controlling an estimation system.

7. A control program for causing a computer to function as the estimation system described in claim 1, wherein the computer functions as the first acquisition unit and the estimation unit.

Images