CN111426801A - Electronic nose learning and domesticating method and equipment thereof - Google Patents
Electronic nose learning and domesticating method and equipment thereof Download PDFInfo
- Publication number
- CN111426801A CN111426801A CN202010384970.1A CN202010384970A CN111426801A CN 111426801 A CN111426801 A CN 111426801A CN 202010384970 A CN202010384970 A CN 202010384970A CN 111426801 A CN111426801 A CN 111426801A
- Authority
- CN
- China
- Prior art keywords
- data
- olfactory
- smell
- electronic
- information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/0001—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00 by organoleptic means
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Engineering & Computer Science (AREA)
- Food Science & Technology (AREA)
- Medicinal Chemistry (AREA)
- Physics & Mathematics (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The electronic nose learning domestication method comprises the steps of obtaining olfactory smell data and electronic nose test data sent by electronic terminal equipment, searching olfactory smell data corresponding to the geographic position of an electronic nose in the olfactory smell data according to the geographic position information of the electronic nose and a set position range, fusing the olfactory smell data with the electronic nose test data according to time information to generate sample data for learning domestication of the electronic nose, and performing relational model parameter fitting on the electronic nose by using the sample data. The invention provides olfactory smell data electronic terminal equipment which comprises a geographic position information module, a clock module, an olfactory smell data input module and an information transmission communication module. The invention saves the sniffer sniffing time, improves the sniffing efficiency, overcomes the limitations of space and places, enables the data given by the electronic nose to be closer to the olfactory sensation of people, and enables the smell data of the electronic nose and the olfactory sensation of the nose to be visual and visualized.
Description
Technical Field
The invention relates to an ambient air quality monitoring technology, in particular to a method and equipment for learning and domesticating an electronic nose by using olfactory smell data.
Background
Environmental air pollution seriously affects the physical health of people, and in intelligent environmental information engineering, the monitoring of air quality is an important technical means, wherein the monitoring of the odor and peculiar smell which directly affect the olfactory perception of people
In the prior art, the determination of the air quality malodor is especially important by a sniffer according to the national standard. (GBT 14675-1993) determination of air quality malodor by three-point comparison odor bag method. The method has strict requirements on the olfactory discrimination environment through the processes of sampling, storing, olfactory discrimination and the like, so that an olfactory discriminator can obtain a correct olfactory discrimination conclusion, and daily ambient air quality monitoring is difficult to meet. The ambient air detection needs to be monitored in real time and continuously. With the progress of electronic technology, electronic equipment, namely an electronic nose, is provided, which can simulate the function of the human nose, adopts various smell sensing sensors and is provided with an artificial intelligence program (AI program) to detect the type and the intensity of the smell. However, in the existing electronic nose working mode, in an area with a large region range and large odor change, the given odor intensity and type information are greatly deviated from the olfactory sensation of a general person.
In this specification, the relation model refers to a transformation function or program for providing odor intensity values and odor type data from sensor data in the electronic nose. The general person refers to a person whose olfactory sensation represents the average sensation of the general population, and statistically, the given odor intensity is the arithmetic mean value, or the median value, or the most significant value, or the highest-density value of the odor intensities given by the general population, which is representative in any case. The sniffer refers to a person who has been trained in sniffing and has performed sniffing for the intensity and type of smell through formal or informal examination, and if a person uses the sniffer to perform sniffing, the statistical average or the most value of the given data should be close to the data given by the sniffer.
Disclosure of Invention
The technical problem is as follows: aiming at the defects of the prior art, the invention provides an electronic nose learning and domesticating method and application equipment thereof.
The technical scheme is as follows: the invention firstly provides an e-nose learning and domesticating method, which comprises the following steps:
obtaining olfactory smell data sent by electronic terminal equipment, wherein the olfactory smell data comprise an ID (identity) of the electronic terminal equipment, geographical position information of a place where the electronic terminal equipment is located, olfactory smell intensity information or smell intensity information and olfactory smell type information input by a terminal user, and time information when the terminal equipment sends the olfactory smell data;
obtaining electronic nose test data, wherein the electronic nose test data comprises an ID (identity) of each electronic nose, geographical position information of a place, electronic nose sensor data, odor intensity information or odor intensity information and odor type information generated according to a relation model, and time information;
searching olfactory smell data corresponding to the geographical position of the electronic nose in the olfactory smell data according to the geographical position information of the electronic nose and a set position range, and fusing the olfactory smell data with the electronic nose test data according to time information to generate sample data for learning and domesticating the electronic nose;
and performing relational model parameter fitting on the electronic nose by using the sample data.
The electronic nose test data or olfactory smell data can be stored in the electronic nose and can be read and written in the storage device, and can be conveniently obtained as long as needed, and the data information of the electronic nose and olfactory smell data can be stored in the server.
Generally, the electronic nose test data is data with a time sequence at equal intervals, and the olfactory smell data is relatively random in time and is input by an end user or input prompted by a terminal device.
In a better implementation mode, the method can further comprise the step of judging whether to update the relation model parameters of the electronic nose according to the fitting of the current relation model parameters and the coincidence degree of the parameter fitting generated according to the new fitting.
The electronic nose relation model is generally of two types, one type is about the classification of the odor type, the classification model at this time is generally selected according to the required estimation information by using Discriminant Factor Analysis (DFA), linear discriminant analysis (L DA), a Neural Network (NN) and the like, and the other type is about the odor intensity, particularly the odor degree or the fragrance degree of the odor, and the partial least squares regression (P L S), the Neural Network (NN) and the like.
And determining whether to update the relation model parameter according to whether the newly obtained parameter can better approach the human olfactory sensation, namely, the difference between the odor intensity and classification data given by the electronic nose data and the olfactory odor information is smaller. The parameters of the relational model and the time and version of the update are retained in memory as historical data that can be used to trace back or return previous parameters as necessary. Usually, the quality of the fit degree is judged, which may be the deviation between the odor intensity data and/or the odor type data estimated by the electronic nose and the human olfactory sensation, and usually, the deviation may be evaluated by means of variance, and when the variance of the new fitting parameter estimation result becomes smaller, the fit degree is judged to be good, and the update is performed.
It should be noted that, according to the electronic nose of the present invention, the terminal device for collecting odor information in a certain area is not necessarily connected online, nor in a system, as long as the two collected data have time record marks in the neighboring area, and the two data can be merged together according to time and geographic location, of course, both the two data are better connected to a server, such as a server of a control center. That is, as long as the information data exists, no matter what medium and location are stored, it can be obtained as needed, for example, by using a storage medium such as a hard disk, a usb disk, an optical disk, a ROM, or the like, by a mobile copy method in non-real time, preferably by a wired or wireless network connection in real time, and more preferably stored together for data fusion.
In one embodiment of the invention, when the sample data is generated by fusion, the method comprises the steps of selecting data segments with olfactory smell data and within a time interval range of a preset value of the variation range of the electronic nose test data according to time sequence of the electronic nose test data and the olfactory smell data, and taking a representative value of each data segment as a group of numbers of the sample data to form the sample data; the parameter fitting is to fit the relation parameters of the electronic nose by using the sample data; the representative value is one of an arithmetic mean, a mode, or a statistical average.
In another embodiment of the invention, when sample data is generated by fusion, when olfactory odor data is found in a region near the geographical position of the electronic nose, according to the time information of the olfactory odor data, electronic nose test data within a time interval range under a preset value is found in the electronic nose test data, and according to data within a preset variation range, a table value is replaced, and a group number of the sample data is formed by the electronic nose test data and the olfactory odor data; the parameter fitting is to fit the relation parameters of the electronic nose by using the sample data; the representation is one of an arithmetic mean, a mode, or a statistical average.
The collection of the olfactory smell information can be carried out when the electronic terminal equipment is turned on, when a terminal user selects the olfactory smell information, or when the server urges to input the olfactory smell information. The input mode of the odor intensity and the odor type can adopt a number or character input mode or a menu selection mode. Particularly, when the odor type is input, the user can select the odor type by a menu, so that the data collection and statistics are convenient.
Further, according to the electronic nose data information, marking the electronic nose smell information on an electronic map to form an electronic nose smell map; similarly, according to the olfactory smell information, marking the olfactory smell information on an electronic map to form an olfactory smell map; and when a request for sending the smell map is received, pushing the corresponding smell map to the requesting party.
The electronic map may be obtained from data stored in the system, or may be obtained from a public application such as a hundredth map app. One embodiment of the invention is realized by utilizing an app interface of the Baidu map to render the odor information on the Baidu map.
Furthermore, the invention can push various odor information according to the requested geographic position, for example, a time-varying trend array of olfactory odor information at the position or a time-varying trend array of electronic nose odor information at the position can be pushed, a two-dimensional odor map is displayed in the electronic terminal equipment, the time-varying trend of the odor is displayed, and the user can know the general trend of the change of the surrounding odor.
The invention also provides an olfactory smell data electronic terminal device for realizing the method, which comprises the following steps: the geographic position information module is used for acquiring the geographic position information of the electronic terminal equipment; the clock module is used for acquiring time information when olfactory smell data are sent; the olfactory smell data input module is used for inputting olfactory smell data by a user of the electronic terminal equipment, wherein the olfactory smell data comprise smell intensity data or smell intensity data and smell type data; and the information transmission communication module receives or sends the information when receiving the information receiving or sending instruction. The olfactory smell data input module can also input information such as wind direction, temperature, humidity and the like.
When the olfactory smell data electronic terminal equipment is turned on and activated, inputting smell type data sensed by olfactory sense, such as malodor, aroma and the like, and further inputting intensity data; and (4) pressing a submission button, and transmitting the geographical position information, the time information, the odor type information and the odor intensity information to a server through an information transmission communication module.
Furthermore, the olfactory smell data electronic terminal device also comprises a smell information map display module which displays the smell map and/or the smell change trend chart when receiving the smell map data sent by the server.
The olfactory smell data electronic terminal equipment further comprises an electronic nose part, air is tested in real time, and test data are uploaded to form electronic nose test data.
The electronic terminal equipment for olfactory smell data can collect the mass olfactory smell data to be acquired by the invention at least as long as the olfactory smell data can be uploaded, and the electronic nose learning domestication step is completed. The olfactory smell data electronic terminal device may be a dedicated electronic terminal device, more conveniently a mobile communication electronic device, such as a smartphone. The mode of uploading the smell information to the server can be realized by adopting a webpage http protocol, and can also be a special smell information app, and better is electronic equipment loaded with an electronic nose module, such as a smart phone loaded with the electronic nose module.
According to one embodiment of the disclosure, the electronic nose is provided with a communication module, and the detection data of the electronic nose is automatically uploaded to an upper server. After the electronic nose is embedded into the olfactory smell data uploading unit, the olfactory smell data and the electronic nose detection data can be uploaded to a server to form the olfactory smell data electronic terminal device. The ID of the electronic terminal equipment for olfactory smell data corresponds to the ID of the portable electronic nose one by one, and the acquired olfactory smell data corresponds to the gas test data acquired by the electronic nose one by one to form learning training sample data, so that parameter fitting, namely learning domestication, of the electronic nose relation model can be performed. The smell data value output by the electronic nose has individual characteristics and approaches the smell sense of the smell of the terminal user. That is to say, this end user's electron nose will test the air in real time, and the smell data that gives is very close to user's sense of smell sensation, has artificial intelligence's characteristics more.
The invention provides a server for realizing learning and domestication method of an electronic nose, which comprises
The information transmission communication module is used for receiving olfactory smell data, electronic nose test data and a request of electronic terminal equipment, and sending electronic map information, electronic smell map information and smell change trend information;
the data memory is used for storing the ID of the electronic nose equipment, the ID of the electronic terminal equipment, olfactory smell data, electronic nose test data and electronic map information;
the relational model parameter fitting and updating module comprises: selecting and fusing electronic nose test data and olfactory smell data to generate sample data for learning and domestication; a relational model parameter fitting unit which uses the learning and domestication sample data to fit a relational model in the electronic nose to obtain new relational model parameters; the relation model updating unit is used for estimating the test data of the electronic nose according to the current relation model parameters and the new relation model parameters to give a result of odor intensity and odor type, comparing and judging the coincidence degree, updating the relation model parameters of the electronic nose if the coincidence degree becomes high after the new parameters are used, and storing the relation model, parameter information and time in a memory; if the degree of coincidence does not become high, no update is performed.
The server can also comprise an odor map generation module which marks the collected electronic nose test data and olfactory odor data on an electronic map to generate the electronic odor map in real time or intermittently or according to a request, and pushes the electronic odor map near the geographic position of the terminal equipment to the terminal equipment when receiving the request of the electronic terminal equipment with the olfactory odor data.
The electronic nose can be an independent portable electronic nose, and can also be an odor detection and analysis unit loaded on the olfactory odor data electronic terminal device.
In the invention, in an area where an electronic nose is arranged for monitoring, such as an industrial park, wherein the electronic nose is fixedly arranged and the electronic terminal equipment which is used for collecting olfactory smell data and moves or is fixed in the area, when a natural person suffers from the change of smell and needs to upload smell information, the smell strength and/or smell type information is input to the electronic terminal equipment for uploading. The olfactory smell data can be input by a trained sniffer or a general person. The olfactor can reach the position near the set electronic nose for olfactive identification and input the olfactive identification result. In the collected olfactory smell data, which data are from trained olfactory discriminators and which data are from ordinary persons can be identified, and the data are respectively used for fitting a domesticated sample to a relation model of the electronic nose according to different sources of the olfactory smell data.
Furthermore, in a wider area, if the type and the intensity of the electronic nose sensor have obvious difference according to the smell information map, a segmented relation model and/or parameters associated with the geographic position can be formed, and the electronic nose sensor test data is converted into the smell intensity and the smell type by using the relation model associated with the geographic position, so that the estimation data of the electronic nose is closer to and more accurate with the human olfactory sensation.
As a better embodiment, the electronic terminal device for collecting odor information includes an odor test sensor array, or further includes a sensor array, a data conditioning module, a relationship model and an MCU for parameter fitting of the relationship model, and an olfactory sensation information collecting module can also be installed on a general portable electronic nose.
The server updates the smell map in real time or intermittently according to the collected smell information. The server is loaded with various relation models, the relation models in the electronic nose are fitted according to the collected information of the smell of the natural human and the test information of the electronic nose, and when the new fitting result is better than the current result in goodness of fit, the parameters obtained by the new fitting are substituted into the relation models.
For odors requiring to distinguish different regions and different time, classification models such as Principal Component Analysis (PCA), component Discrimination (DFA), Neural Network (NN) and the like are generally selected, and for providing temporal and spatial changes of odor intensity, a relationship model of odor intensity can be connected by linear partial least squares regression (P L S), neural network NN and the like.
Has the advantages that: the electronic nose learning and domesticating method, the olfactory smell data electronic terminal equipment and the server utilize olfactory smell data to learn and domesticate the electronic nose in operation, save the olfactory distinguishing time of a olfactory person, improve the olfactory distinguishing efficiency, overcome the limitation of space and places, enable the data given by the electronic nose to be closer to the olfactory feelings of people, particularly to the olfactory feelings of people living and working in the environmental area, and are suitable for automatic learning and domestication of the electronic nose in a monitoring network with high density and multiple number arrangement. Meanwhile, the electronic nose odor map and the olfactory odor map provided by the invention enable the odor data of the electronic nose and the nose to be visual and visual.
Drawings
FIG. 1: the invention is a schematic diagram of a use scene.
FIG. 2: the flow chart of the natural person uploading smell information data is shown schematically.
FIG. 3: the invention discloses a flow diagram for marking and sharing a smell map.
FIG. 4: the invention discloses a flow diagram of learning and domestication of an electronic nose.
FIG. 5: the invention relates to a specific learning and domesticating flow chart.
FIG. 6: a schematic diagram of a scent information electronic terminal device of an embodiment.
FIG. 7: the server of one embodiment of the invention is schematically constructed.
FIG. 8: in one embodiment, the electronic nose odor intensity is plotted against olfactory odor intensity before fitting.
FIG. 9: figure 8 is a graph of the fitted smell intensity of the electronic nose of the example versus the smell intensity of the smell.
FIG. 10: the electronic nose of the invention detects the real map of the odor type.
Detailed Description
The present invention will be further described with reference to the following examples and the accompanying drawings.
The electronic terminal device for collecting olfactory smell data according to the embodiment of the present invention may include, but is not limited to, a mobile phone, a Personal Digital Assistant (PDA), a wireless handheld device, a tablet Computer (tablet Computer), a Personal Computer (PC), a wearable device (e.g., smart glasses, smart watches, smart bracelets, etc.), and may be a device that can exchange information through a wired/wireless/network.
Fig. 1 illustrates a usage scenario 100 of the present invention, in which a five-pointed star symbol 101 represents an electronic nose, and in a monitoring area such as an industrial park, electronic noses are fixedly arranged in a grid manner, each electronic nose has a unique ID number and geographical location information of a location, air quality in the area is detected, and the detected data is uploaded to a server; a human-shaped symbol 102 represents a terminal of an electronic device for collecting odor information, the device is carried by a natural person to move or stay in an area, the natural person can send information such as olfactory sensation intensity and type, geographical position information, time information and the like to a server through the device, and odor map information is obtained from the server and displayed.
In a certain area, for example, an urban block, an industrial park, a refuse landfill, a river, a sewage treatment area, a petrochemical industry area and other areas needing to control air quality, an air quality monitoring device is arranged for monitoring. The air quality detection device can be various, in the embodiment, the air quality detection device is an electronic nose which is generally composed of a plurality of gas sensors, a data conditioning and control communication component, monitors the smell in the air, particularly the odor concentration in real time, is provided with a data set of the sensors, and provides an estimation odor value (OU value) according to a relation model in the electronic nose. Some people living and working in the area carry with them an electronic smell information collecting terminal device capable of reporting the sensed air malodor degree, i.e. smell information. Electronic noses on the market, such as RQ Box electronic nose of French alpha MOS, Germany AIRSENSE PEN3 electronic nose and the like, can be used as the electronic nose for monitoring the air quality, can output sensor data and odor intensity and/or odor type data according to a relation model, wherein the relation model parameter fitting is realized by using the learning and domesticating method.
Fig. 2 shows a flow of uploading smell information data by a natural person holder using an electronic terminal device carrying collected smell information. S102 activates an electronic terminal device, which does not need to be limited, and it is a better choice to use the smart phone, and at this time, the device may be an application program (app) in the smart phone, and the activation may be to open the app by using a web page transmission method, or to open a web page by using a web page transmission method.
S104, automatically uploading the geographical position information, receiving and displaying a server smell information map, and automatically uploading the geographical position information of the terminal after the terminal is activated and opened, wherein the geographical position information is obtained in a plurality of ways and can be conveniently obtained from a GPS of the smart phone. If server scent map data is received, a scent map is displayed. If no corresponding smell map information exists in the server, no smell information exists, and then the map is not displayed, or the 'no smell map information in the area' and the like are displayed.
S106, displaying an odor information input interface, and prompting the terminal to input odor information, wherein the simplest method is to input odor intensity information, such as data of odor grade 0, grade 1 and the like. The input for the preferred embodiment is a menu option, e.g. no odour, very slight odour, clearly odour, very odour etc. corresponding to the rating data 1, 2, 3, 4, 5 etc. After the server receives the geographical position information of the terminal, if the electronic terminal device is found to be lack of olfactory smell information or needs to collect more olfactory smell information, an instruction can be sent to the electronic terminal device, and the electronic terminal device encourages and prompts a holder to upload the olfactory smell information in the modes of prompting, voice and the like. The geographical position can be near the set point of the electronic nose, can be a blank section of olfactory smell information, or an electronic nose with abnormal test data, and the like, and the electronic terminal equipment in the area can be used for prompting the holder to upload the olfactory smell information. Inputting the odor type by a menu, such as options of smelling odor of rotten eggs, smell of burning plastic, smell of rotten vegetables, smell of hydrochloric acid, smell of toluene, smell of aromatic hydrocarbon, smell of rubber, smell of sewage, and the like.
And S108, pressing a submit button, and uploading the submitted button, the geographical location information and the time information to the server. And a plurality of electronic terminal devices at mobile or fixed positions send olfactory smell data to be collected and stored in the server, so that the collection of the olfactory smell data of the air is realized, and an olfactory smell database is formed.
Fig. 3 shows a process of sharing olfactory odor data collected by the server according to the present invention by marking the olfactory odor data in a map.
S202, collecting current geographic position information and prompting a user to upload olfactory smell data. At this time, according to the geographical position of the electronic terminal device, the server can search to know whether the olfactory smell data of the geographical position exists or not and how much the olfactory smell data exists, and prompt information can be sent according to the situation, for example, prompt input of the air smell information in a mode of 'the smell information provided by you will fill the blank of the current position', 'the smell here is heavy, please evaluate the air quality here', and the like.
S204, according to the acquired current geographic position information, the odor information is marked in the geographic position of the current map. The current position refers to the geographical position of the electronic terminal equipment, the current map refers to a map with the geographical position as the center, and the odor information data is marked in the geographical position of the current map, namely the odor information data is marked on the current map according to the geographical position of the odor information data.
S206, sharing the map indicating the current position of the odor information, and displaying prompt information if the map has no odor data. Of course, the odor map may also be pushed when the electronic terminal device does not request it.
Fig. 4 shows an embodiment of the present invention for performing electronic nose learning domestication by using mass olfactory smell data. According to a preplanning, or when certain set conditions are met, starting to fit and update the relation model parameters of a certain electronic nose, wherein the flow is as follows:
s302, receiving the electronic nose gas detection information set in a gridding mode, and storing the electronic nose gas detection information in a server.
S304, the electronic nose smell information is rendered and marked on the map to form a smell map (smell map E).
S306, judging whether to perform fitting updating of the relational model data: whether enough new olfactory smell data exist near the geographical position of the electronic nose can be judged, such as 2 groups or 5 groups, 10 groups and the like; the geographical position is set as required and according to actual conditions, for example, a range of 20 meters, 50 meters, 100 meters, 200 meters, or 1 kilometer from the electronic nose. Overall, the smell data is better if there is more recent smell.
S308, fitting and updating the parameters of the electronic nose relation model: and entering a fitting updating process when judging that enough new olfactory smell data exist. In the process, firstly, the test data of the electronic nose and the olfactory data within a certain position range are fused according to time to generate domestication sample data, the sample data is used for fitting the relational model parameters, and the new parameters are substituted for the current parameters.
Fig. 5 shows a specific learning and domestication process of an electronic nose according to the present invention:
s402, the ID and the geographical position of the electronic nose to be domesticated are obtained.
S404, according to a preset geographical position range, whether corresponding olfactory smell data exist or not is searched, and if yes, the electronic nose sensor data and the olfactory smell data are matched and combined into a training sample according to time information.
The generation of sample data for domestication can be carried out by various methods, and factors to be considered are the geographic position of olfactory smell data and the distance from the geographic position of the electronic nose, and if the factors such as wind direction are better, the factors are considered. And taking representative values of the data, namely representative values of the measured values of the sensors of the electronic nose and representative values of smell data at certain same time intervals.
More specifically, the electronic nose test data information is usually performed at preset time intervals, so that the generated data is arranged in a time-equally spaced sequence, and the data amount is large. On the other hand, the olfactory smell data of the embodiment is not equally spaced but random in time sequence due to the randomness of collecting smell information by the electronic terminal device, and some time periods have large data quantity, some time periods have small data quantity, some geographic position data quantity is large, and some geographic position data quantity is small. Generating sample data, wherein for the sample data needing olfactory smell data, the electronic nose test data and the olfactory smell sample data are required to be in one-to-one correspondence. The actual data and the data are different in time, the geographical positions are different in distance, and the data volume of the olfactory smell is uncertain. Therefore, when using the smell data, it is necessary to set a certain time interval and a certain geographical location interval, and calculate a representative value, such as an arithmetic average value, a weighted average value, etc., from the smell data corresponding to the set time interval and geographical location interval by a certain rule. A representative value, such as an arithmetic mean, a statistical median, or the like, is calculated for the electronic nose test data. Are preset according to the required precision.
Since the amount of olfactory smell information data is usually small and is relatively random in time space, from the selected olfactory smell information data set and electronic nose test data set, if there is an olfactory smell information data collected at 08 o ' clock, 26 o ' clock, 34 o ' clock, 2019, 11 o ' clock, 26 o ' clock, for example, it can be seen whether the variation of olfactory smell data and electronic nose sensor data before and after the time point, for example, 3 minutes, is within a certain allowable range, for example, 5% or 10%, and if it is within a set range, the variation of mass component of air is not large for several minutes before and after the time point, and these data are averaged to generate sample data. By repeating the matching process, multiple sets of sample data can be generated.
S406 determines whether the sample data set is suitable for fitting, and generally, at least 5 sets are good, and determines again whether the time interval between the sample data set and the previous fitting time is equal to or more than a preset interval. The preset interval may be defined in an electronic nose acclimation plan.
And S408, fitting the parameters of the electronic nose relation model when the sample data group and the actual time interval accord with the preset requirement to obtain new parameters.
S409 evaluates whether the fitted new parameters are more consistent than the currently used parameters, and may generally compare the fitted values with the actual values to calculate the variance, where a smaller variance is more consistent.
And S410, if the odor estimation is more consistent, updating the relation model parameters, recalculating the odor estimation output by the electronic nose, and recording the relation model parameters.
Fig. 6 is a schematic view of an olfactory scent data electronic terminal device 200 for use in accordance with the present invention, the device comprising:
the olfactory smell information input module 201 is used for inputting smell information by the electronic terminal equipment;
a geographic location information module 202, typically using the Global Positioning System (GPS), for obtaining a geographic location of when olfactory scent data is input;
a clock module 203 for obtaining the date and time when the smell information is input;
and the odor map information display module 204 is used for displaying the odor map information. The odor map can have various marking modes, such as marking points on an electronic map, marking odor clouds, and odor information change trend of a certain position.
The information transmission communication module 205 is configured to perform data exchange and command transmission with an external device, for example, a server, send data input by the olfactory smell information input module 201, together with the time information of the clock module 203 and the geographic information of the geographic position information module 202, to the server, receive smell map information from the server, and transmit the smell map information to the smell map information display module 204 for display.
Fig. 7 is a schematic diagram showing a server configuration according to an embodiment of the disclosure, where the server 300 includes:
the data storage 310 is used for storing information, sending shared information, and providing information to the odor map generation module 320, the relationship model parameter fitting update module 330, and the information transmission communication module 340, and includes: the stored information includes electronic nose ID information and electronic terminal device ID information 311, electronic nose detection data 312, olfactory smell information data 313, and electronic map information 314. The electronic nose is usually fixedly arranged or can be moved, so that the electronic nose detection data 312 includes electronic nose ID, geographical position information, sensor test data, and odor intensity and odor type data calculated according to the relationship model; the olfactory smell information data 313 collected by the electronic terminal device includes an electronic terminal device ID, geographical location information, smell intensity and/or smell type information.
And an odor map generating module 320, configured to generate odor map information using the electronic nose detection data 312 and the olfactory odor information data.
The relation model parameter fitting updating module 330 performs fitting updating on the relation model parameters of the electronic nose by using the collected data, and comprises the following steps: the electronic nose odor information data and terminal data selecting and fusing unit 331 performs data selecting and fusing according to the position information of the electronic nose and a preset position range, and the time information in the electronic nose detection data 312 and the time information of the olfactory odor information data 313 to generate domestication sample data used by the relational model parameter fitting unit 332.
The relational model parameter fitting unit 332 fits the electronic nose relational model parameters using the domestication sample data.
And a relation model parameter determining and updating unit 333, which compares the fitted new parameter with the current parameter, and updates the current parameter to the new parameter or retains the current parameter according to a preset condition.
And the information transmission communication module 340 is used for carrying out transmission communication of data and instruction information between the server and the electronic nose and between the server and the electronic terminal equipment according to the request.
Fig. 8 and 9 show fitting of relational model data, here, fitting of odor intensity data (OU values), based on olfactory odor data and electronic nose sensor test data, with the abscissa being the estimated odor intensity OU value of the electronic nose calculated from the sensor test data based on the fitting model and the ordinate being the human olfactory odor intensity OU value. The correspondence between the odor intensity estimated by the electronic nose in fig. 8 and the olfactory odor intensity is relatively discrete, and the correspondence between the two is relatively concentrated and more linear in fig. 9. Along with the increase of olfactory smell data accumulation, fitting optimization is continuously carried out, fitting is better and better, dispersion is smaller and smaller, and linearity and consistency are better and better.
Fig. 10 is a diagram showing the detection of odor types by an electronic nose installed beside a petrochemical wastewater treatment tank, wherein the electronic nose classifies gases according to test data, estimates the types of the gases, and learns and acclimates the parameters of a relation model of the electronic nose by using artificial olfaction odor type data to optimize the parameters. The electronic nose is an alphaMOS RQ-box electronic nose arranged in the petrochemical plant area, olfactory smell data come from patrolmen in the plant area to patrol and upload smell types and smell intensity information, and typical smells near the plant area comprise a sewage treatment pool, an ethylene workshop, an aromatic hydrocarbon workshop, a chlor-alkali workshop and the like. There are 3 typical odors in the sewage treatment tank, which reflect the sewage source, and are respectively marked as odor a, odor b and odor c, and the judgment of the odor category of the electronic nose is closer to that of the artificial odor category. The electronic nose can distinguish odor types well using PCA-DFA (fig. 10) or using neural network NN (not shown) relational model.
With the increase of data accumulation, the accuracy of judgment by using the relational model is better and better. For example, in 2019 between 10 and 11 months, the accuracy of the recorded artificial olfactory smell data and the electronic nose data is estimated as follows:
according to another aspect of the disclosure, for the fixedly or movably arranged electronic nose, a person holding the electronic terminal device for collecting olfactory smell information, such as a trained inspector, a sniffer, etc., can patrol the vicinity of the electronic nose, send olfactory smell data (which may be called as sniffing data when the sniffer performs sniffing), perform special domestication on the electronic nose, so that the relation model output of the electronic nose is more approximate to olfactory perception information, and provide an odor map approximate to olfactory perception of a general person.
According to an aspect of the embodiment of the disclosure, the electronic nose and the server are loaded with various types of relation models, the relation models can be selected according to needs, and required information such as odor type, odor intensity and particularly malodor level can be output.
The electronic nose relation model is generally selected according to required information, one type is classification about odor types, the classification model at the moment is generally classified by discriminant factor analysis (DFA, &lTtTtranslation = L "&gTtL &lTt/T &gTtDA), Network Nerves (NN) and the like, and the other type is learning domestication of the electronic nose, namely, sample data for domestication is used for parameter fitting of the models to determine parameter values in the relation models, wherein partial least squares regression (P L S) and Network Nerves (NN) are generally used for odor intensity, particularly odor degree or fragrance degree of the odor.
Whether the relation model parameters need to be updated or not is judged according to whether the newly obtained parameters can better approach the olfactory sensation of people or not, namely, the better approach is that the difference between the odor intensity and classification data given by the electronic nose data and the olfactory odor information data is smaller. The parameters of the relational model and the time and version of the update are retained in memory as historical data that can be used to trace back or return previous parameters as necessary.
The electronic smell map can be obtained from data stored in the system, and in one embodiment of the disclosure, the smell information is rendered on the Baidu map by using an app interface of the Baidu map. The way in which the rendering is achieved may be varied, for example by representing the scent type by color, the intensity of the scent by shade of color, or contours of different colors. In the case of the lack of the data amount of the odor information, the odor information is represented at one or a few points by only one balloon, and the odor intensity and the odor type are expressed. In the area where the electronic nose is arranged for monitoring, the coverage area of the odor information data is wide, and the expression is more visual by using forms such as cloud pictures and the like. Since there is data information that changes over time, the odor map can be expressed in a time-varying form.
Claims (9)
1. An electronic rhinology domestication method comprises the following steps:
obtaining olfactory smell data sent by electronic terminal equipment, wherein the olfactory smell data comprise an ID (identity) of the electronic terminal equipment, geographical position information of a place where the electronic terminal equipment is located, olfactory smell intensity information or smell intensity information and olfactory smell type information input by a user, and time information when the electronic terminal equipment sends the olfactory smell data;
obtaining electronic nose test data, wherein the electronic nose test data comprises an ID (identity) of each electronic nose, geographical position information of a place, electronic nose sensor data, odor intensity information or odor intensity information and odor type information generated according to a relation model, and time information;
searching olfactory smell data corresponding to the geographical position of the electronic nose in the olfactory smell data according to the geographical position information of the electronic nose and a set position range, and fusing the olfactory smell data with the electronic nose test data according to time information to generate sample data for learning and domesticating the electronic nose;
and performing relational model parameter fitting on the electronic nose by using the sample data.
2. The e-nose learning and domesticating method of claim 1, wherein the sample data is generated by time information fusion in the following way: selecting data segments with olfactory smell data and within a time interval range of the variation range of the electronic nose test data under a preset value according to time sequence of the electronic nose test data and the olfactory smell data, and taking a representative value of each data segment as a group of numbers of sample data to form the sample data; the parameter fitting is to fit the relation parameters of the electronic nose by using the sample data; the representative value is one of an arithmetic mean, a mode, or a statistical average.
3. An electronic nose learning and domesticating method according to claim 1, wherein the sample data is generated by fusing the following ways according to time information: when olfactory odor data are found in a region near the geographical position of the electronic nose, electronic nose test data within a time interval range under a preset value are found in the electronic nose test data according to time information of the olfactory odor data, a table value is replaced according to data within a preset variation range, and a group of numbers of sample data are formed by the table value and the olfactory odor data; the parameter fitting is to fit the relation parameters of the electronic nose by using the sample data; the representation is one of an arithmetic mean, a mode, or a statistical average.
4. The learning and domestication method for the electronic nose according to claim 1, wherein the smell strength information and the smell type information of the olfactory sensation inputted by the terminal user are any one of numbers, characters or menu-type options.
5. An olfactory scent data electronic terminal device comprising:
the geographic position information module is used for acquiring the geographic position information of the electronic terminal equipment;
the clock module is used for acquiring time information when olfactory smell data are sent;
the olfactory smell data input module is used for inputting olfactory smell data by a user of the electronic terminal equipment, wherein the olfactory smell data comprise smell intensity data or smell intensity data and smell type data;
and the information transmission communication module receives or sends the information when receiving the information receiving or sending instruction.
6. The olfactory scent data electronic terminal device of claim 5, further comprising a scent information map display module for displaying the scent map and/or the scent change trend map when receiving olfactory scent data from the server.
7. The olfactory odor data electronic terminal device of claim 5, further comprising an electronic nose unit to test air in real time and upload test data to form electronic nose test data.
8. A server for realizing an electronic nose learning domestication method comprises the following steps:
the information transmission communication module is used for receiving olfactory smell data, electronic nose test data and a request of electronic terminal equipment, and sending electronic map information, electronic smell map information and smell change trend information;
the data memory is used for storing the ID of the electronic nose equipment, the ID of the electronic terminal equipment, olfactory smell data, electronic nose test data and electronic map information;
the relational model parameter fitting and updating module comprises: selecting and fusing electronic nose test data and olfactory smell data to generate sample data for learning and domestication;
a relational model parameter fitting unit which uses the learning and domestication sample data to fit a relational model in the electronic nose to obtain new relational model parameters;
the relation model updating unit is used for estimating the test data of the electronic nose according to the current relation model parameters and the new relation model parameters to give a result of odor intensity and odor type, comparing and judging the coincidence degree, updating the relation model parameters of the electronic nose if the coincidence degree becomes high after the new parameters are used, and storing the relation model, parameter information and time in a memory; if the degree of coincidence does not become high, no update is performed.
9. The server of claim 8, further comprising in the server: and the odor map generation module marks the collected electronic nose test data and olfactory odor data on the electronic map to generate an electronic odor map in real time or intermittently according to the request, and pushes the electronic odor map close to the geographical position of the terminal equipment to the terminal equipment when receiving the request of the electronic terminal equipment with the olfactory odor data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010384970.1A CN111426801B (en) | 2020-05-09 | 2020-05-09 | Electronic nose learning and domesticating method and equipment thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010384970.1A CN111426801B (en) | 2020-05-09 | 2020-05-09 | Electronic nose learning and domesticating method and equipment thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111426801A true CN111426801A (en) | 2020-07-17 |
| CN111426801B CN111426801B (en) | 2022-08-02 |
Family
ID=71558713
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010384970.1A Active CN111426801B (en) | 2020-05-09 | 2020-05-09 | Electronic nose learning and domesticating method and equipment thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111426801B (en) |
Citations (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6411905B1 (en) * | 2000-07-18 | 2002-06-25 | The Governors Of The University Of Alberta | Method and apparatus for estimating odor concentration using an electronic nose |
| CN1801136A (en) * | 2006-01-10 | 2006-07-12 | 华东理工大学 | Method for simultaneously determining smell kind and strength by machine olfaction |
| AR061120A1 (en) * | 2007-05-24 | 2008-08-06 | Comision Nac De En Atomica | METHOD FOR THE DETERMINATION OF ODOR INTENSITY AND ELECTRONIC NOSE TO PERFORM IT |
| CN101692053A (en) * | 2009-10-09 | 2010-04-07 | 江苏大学 | Multi-sensing information fusion based instrumental intelligent evaluation method for quality of famous tea |
| CN102876816A (en) * | 2012-07-23 | 2013-01-16 | 江苏大学 | Fermentation process statue monitoring and controlling method based on multi-sensor information fusion |
| CN103472197A (en) * | 2013-09-10 | 2013-12-25 | 江苏大学 | Cross-perception information interaction sensing fusion method in intelligent bionic evaluation for food |
| CN104007240A (en) * | 2014-06-13 | 2014-08-27 | 重庆大学 | Fusion positioning technology based on binocular recognition and electronic nose network gas detection |
| CN104007763A (en) * | 2014-06-13 | 2014-08-27 | 重庆大学 | Method for fixed electronic nose nodes and mobile robot to search for smell source cooperatively |
| CN105424840A (en) * | 2015-12-28 | 2016-03-23 | 周俊杰 | On-line continuous environmental air quality automatic monitoring system and peculiar smell source tracing method |
| CN105486812A (en) * | 2015-12-28 | 2016-04-13 | 周俊杰 | Electronic nose stink grade assignment method in continuous environment air quality monitoring process and application |
| CN205246124U (en) * | 2015-12-16 | 2016-05-18 | 上海宁和环境科技发展有限公司 | Environmental information monitoring devices |
| KR20160071000A (en) * | 2014-12-11 | 2016-06-21 | 현대자동차주식회사 | Apparatus for judging sense of smell and method for the same |
| CN105911219A (en) * | 2016-04-08 | 2016-08-31 | 北京盈盛恒泰科技有限责任公司 | Monitoring and early warning system and method for pollution gas |
| US20170131253A1 (en) * | 2015-11-10 | 2017-05-11 | Nuxtu S.A.S | Electronic nose and tongue device for real-time monitoring and analysis of liquid and gaseous substances |
| KR101852074B1 (en) * | 2016-11-29 | 2018-04-25 | 단국대학교 산학협력단 | Electronic Nose System and Method for Gas Classification |
| US20180120277A1 (en) * | 2016-10-31 | 2018-05-03 | Electronics And Telecommunications Research Institute | Apparatus and method for generation of olfactory information capable of calibration based on pattern recognition model |
| JP2018072310A (en) * | 2016-10-31 | 2018-05-10 | 韓國電子通信研究院Ｅｌｅｃｔｒｏｎｉｃｓ ａｎｄ Ｔｅｌｅｃｏｍｍｕｎｉｃａｔｉｏｎｓ Ｒｅｓｅａｒｃｈ Ｉｎｓｔｉｔｕｔｅ | Olfactory information generation device and generation method capable of detecting direction and position of aroma |
| CN108645971A (en) * | 2018-05-11 | 2018-10-12 | 浙江工商大学 | A kind of air peculiar smell strength grade detection method based on electronic nose |
| CN108709955A (en) * | 2018-05-17 | 2018-10-26 | 华东理工大学 | A kind of stench electronic nose instrument and foul gas multiple spot centralization on-line monitoring method |
| CN108896706A (en) * | 2018-05-17 | 2018-11-27 | 华东理工大学 | The foul gas multiple spot centralization electronic nose instrument on-line analysis of big data driving |
| US20180373840A1 (en) * | 2017-06-27 | 2018-12-27 | International Business Machines Corporation | Olfactory Cognitive Diagnosis |
| CN109799269A (en) * | 2019-01-24 | 2019-05-24 | 山东工商学院 | Electronic nose gas sensor array optimization method based on behavioral characteristics different degree |
| CN110163247A (en) * | 2019-04-08 | 2019-08-23 | 广东工业大学 | Design methods and increment smell classification method based on the nearest class mean value of depth |
| CN110687257A (en) * | 2019-11-04 | 2020-01-14 | 河北先河环保科技股份有限公司 | Tracing method based on malodor online monitoring system |
| CN110794090A (en) * | 2019-10-22 | 2020-02-14 | 天津大学 | Emotion electronic nose implementation method |
| US20200088702A1 (en) * | 2017-03-03 | 2020-03-19 | Commissariat A L'energie Atomique Et Aux Energies Alternatives | Method of calibrating an electronic nose |
| CN110907611A (en) * | 2019-12-26 | 2020-03-24 | 浙江省环境科技有限公司 | Detection control system for regional odor pollution |
| CN110974161A (en) * | 2019-12-03 | 2020-04-10 | 中国科学院心理研究所 | Olfactory function evaluation system based on human olfactory detection |
-
2020
- 2020-05-09 CN CN202010384970.1A patent/CN111426801B/en active Active
Patent Citations (28)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6411905B1 (en) * | 2000-07-18 | 2002-06-25 | The Governors Of The University Of Alberta | Method and apparatus for estimating odor concentration using an electronic nose |
| CN1801136A (en) * | 2006-01-10 | 2006-07-12 | 华东理工大学 | Method for simultaneously determining smell kind and strength by machine olfaction |
| AR061120A1 (en) * | 2007-05-24 | 2008-08-06 | Comision Nac De En Atomica | METHOD FOR THE DETERMINATION OF ODOR INTENSITY AND ELECTRONIC NOSE TO PERFORM IT |
| CN101692053A (en) * | 2009-10-09 | 2010-04-07 | 江苏大学 | Multi-sensing information fusion based instrumental intelligent evaluation method for quality of famous tea |
| CN102876816A (en) * | 2012-07-23 | 2013-01-16 | 江苏大学 | Fermentation process statue monitoring and controlling method based on multi-sensor information fusion |
| CN103472197A (en) * | 2013-09-10 | 2013-12-25 | 江苏大学 | Cross-perception information interaction sensing fusion method in intelligent bionic evaluation for food |
| CN104007240A (en) * | 2014-06-13 | 2014-08-27 | 重庆大学 | Fusion positioning technology based on binocular recognition and electronic nose network gas detection |
| CN104007763A (en) * | 2014-06-13 | 2014-08-27 | 重庆大学 | Method for fixed electronic nose nodes and mobile robot to search for smell source cooperatively |
| KR20160071000A (en) * | 2014-12-11 | 2016-06-21 | 현대자동차주식회사 | Apparatus for judging sense of smell and method for the same |
| US20170131253A1 (en) * | 2015-11-10 | 2017-05-11 | Nuxtu S.A.S | Electronic nose and tongue device for real-time monitoring and analysis of liquid and gaseous substances |
| CN205246124U (en) * | 2015-12-16 | 2016-05-18 | 上海宁和环境科技发展有限公司 | Environmental information monitoring devices |
| CN105424840A (en) * | 2015-12-28 | 2016-03-23 | 周俊杰 | On-line continuous environmental air quality automatic monitoring system and peculiar smell source tracing method |
| CN105486812A (en) * | 2015-12-28 | 2016-04-13 | 周俊杰 | Electronic nose stink grade assignment method in continuous environment air quality monitoring process and application |
| CN105911219A (en) * | 2016-04-08 | 2016-08-31 | 北京盈盛恒泰科技有限责任公司 | Monitoring and early warning system and method for pollution gas |
| JP2018072310A (en) * | 2016-10-31 | 2018-05-10 | 韓國電子通信研究院Ｅｌｅｃｔｒｏｎｉｃｓ ａｎｄ Ｔｅｌｅｃｏｍｍｕｎｉｃａｔｉｏｎｓ Ｒｅｓｅａｒｃｈ Ｉｎｓｔｉｔｕｔｅ | Olfactory information generation device and generation method capable of detecting direction and position of aroma |
| US20180120277A1 (en) * | 2016-10-31 | 2018-05-03 | Electronics And Telecommunications Research Institute | Apparatus and method for generation of olfactory information capable of calibration based on pattern recognition model |
| KR101852074B1 (en) * | 2016-11-29 | 2018-04-25 | 단국대학교 산학협력단 | Electronic Nose System and Method for Gas Classification |
| US20200088702A1 (en) * | 2017-03-03 | 2020-03-19 | Commissariat A L'energie Atomique Et Aux Energies Alternatives | Method of calibrating an electronic nose |
| US20180373840A1 (en) * | 2017-06-27 | 2018-12-27 | International Business Machines Corporation | Olfactory Cognitive Diagnosis |
| CN108645971A (en) * | 2018-05-11 | 2018-10-12 | 浙江工商大学 | A kind of air peculiar smell strength grade detection method based on electronic nose |
| CN108709955A (en) * | 2018-05-17 | 2018-10-26 | 华东理工大学 | A kind of stench electronic nose instrument and foul gas multiple spot centralization on-line monitoring method |
| CN108896706A (en) * | 2018-05-17 | 2018-11-27 | 华东理工大学 | The foul gas multiple spot centralization electronic nose instrument on-line analysis of big data driving |
| CN109799269A (en) * | 2019-01-24 | 2019-05-24 | 山东工商学院 | Electronic nose gas sensor array optimization method based on behavioral characteristics different degree |
| CN110163247A (en) * | 2019-04-08 | 2019-08-23 | 广东工业大学 | Design methods and increment smell classification method based on the nearest class mean value of depth |
| CN110794090A (en) * | 2019-10-22 | 2020-02-14 | 天津大学 | Emotion electronic nose implementation method |
| CN110687257A (en) * | 2019-11-04 | 2020-01-14 | 河北先河环保科技股份有限公司 | Tracing method based on malodor online monitoring system |
| CN110974161A (en) * | 2019-12-03 | 2020-04-10 | 中国科学院心理研究所 | Olfactory function evaluation system based on human olfactory detection |
| CN110907611A (en) * | 2019-12-26 | 2020-03-24 | 浙江省环境科技有限公司 | Detection control system for regional odor pollution |
Non-Patent Citations (4)
| Title |
|---|
| XUE JIANG等: "A novel electronic nose learning technique based on active learning:EQBC-RBFNN", 《SENSORS AND ACTUATORS B: CHEMICAL》 * |
| 梁子跃: "基于电子鼻的VOCs和恶臭检测***研究与应用", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 * |
| 苏爱华等: "电子鼻在垃圾填埋场的恶臭在线监测***中的应用", 《环境与可持续发展》 * |
| 苗加成: "基于信息挖掘技术的人工嗅觉***研究", 《中国优秀博硕士学位论文全文数据库(博士) 信息科技辑》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111426801B (en) | 2022-08-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kormi et al. | Estimation of fugitive landfill methane emissions using surface emission monitoring and Genetic Algorithms optimization | |
| Nong et al. | Urban growth pattern modeling using logistic regression | |
| WO2018214060A1 (en) | Small-scale air quality index prediction method and system for city | |
| CN101297337B (en) | Methods for predicting destinations from partial trajectories employing open-and closed-world modeling methods | |
| CN110428104A (en) | A kind of genes' contamination ratio determines method, apparatus, electronic equipment and storage medium | |
| CN116186566A (en) | Diffusion prediction method and system based on deep learning | |
| CN116384754B (en) | Deep learning-based environmental pollution risk assessment method in chemical industry park | |
| CN112069281B (en) | Method and device for determining position of interest point, computer equipment and storage medium | |
| CN110346517A (en) | A kind of smart city industrial air pollution visualization method for early warning and its system | |
| CN110738354B (en) | Method and device for predicting particulate matter concentration, storage medium and electronic equipment | |
| Atwal et al. | Predicting building types using OpenStreetMap | |
| CN116859006B (en) | Air pollution monitoring system and method based on atmospheric diffusion mechanism | |
| CN116485202B (en) | Industrial pollution real-time monitoring method and system based on Internet of things | |
| CN116437291A (en) | Cultural circle planning method and system based on mobile phone signaling | |
| Straka et al. | Analysis of energy consumption at slow charging infrastructure for electric vehicles | |
| CN117538503A (en) | Real-time intelligent soil pollution monitoring system and method | |
| Yamagata et al. | Urban carbon mapping with spatial BigData | |
| CN114997646A (en) | Atmospheric pollution health risk real-time dynamic evaluation method based on mobile phone signaling data | |
| CN111426801B (en) | Electronic nose learning and domesticating method and equipment thereof | |
| Wang et al. | Hyperlocal environmental data with a mobile platform in urban environments | |
| CN116233759B (en) | Resident travel track investigation method and system | |
| CN111782973A (en) | Interest point state prediction method and device, electronic equipment and storage medium | |
| Devi et al. | Live demonstration on smart water quality monitoring system using wireless sensor networks | |
| CN106124699A (en) | A kind of intelligent air quality Real-Time Evaluation device and control method thereof | |
| CN111461163B (en) | Urban interior PM2.5 concentration simulation and population exposure evaluation method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |