CN113484312B - Method for establishing temperature compensation model based on olfactory visualization technology - Google Patents
Method for establishing temperature compensation model based on olfactory visualization technology Download PDFInfo
- Publication number
- CN113484312B CN113484312B CN202110817323.XA CN202110817323A CN113484312B CN 113484312 B CN113484312 B CN 113484312B CN 202110817323 A CN202110817323 A CN 202110817323A CN 113484312 B CN113484312 B CN 113484312B
- Authority
- CN
- China
- Prior art keywords
- data
- gas
- reaction device
- temperature
- olfactory
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/75—Systems in which material is subjected to a chemical reaction, the progress or the result of the reaction being investigated
- G01N21/77—Systems in which material is subjected to a chemical reaction, the progress or the result of the reaction being investigated by observing the effect on a chemical indicator
- G01N21/78—Systems in which material is subjected to a chemical reaction, the progress or the result of the reaction being investigated by observing the effect on a chemical indicator producing a change of colour
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Plasma & Fusion (AREA)
- Investigating Or Analysing Materials By The Use Of Chemical Reactions (AREA)
Abstract
The invention discloses a method for establishing a temperature compensation model based on an olfactory visualization technology, which comprises 3 parts of establishing an olfactory sensor reaction device, acquiring data, establishing the temperature compensation model and optimizing the temperature compensation model. The method comprises the following steps of firstly, building a smell sensor reaction device, wherein the smell sensor reaction device comprises a nitrogen cylinder, a constant temperature water bath tank, a smell sensor array, a 3CCD camera, an automatic feeding device, a single chip microcomputer chip and a PC. The automatic input of the sample can be completed through the control of the single chip microcomputer; the temperature in the gas reaction chamber tends to be stable through the constant temperature water bath box; the sensor array images before and after the reaction can be obtained through the 3CCD camera. And correspondingly processing the characteristic image to obtain characteristic data for establishing and optimizing a temperature compensation model. By utilizing the temperature compensation model, data errors caused by temperature in experiments can be greatly reduced, so that the precision is higher; meanwhile, the device is simple to operate, good in sealing performance and wide in application prospect.
Description
Technical Field
The invention relates to the technical field of olfactory visualization detection, in particular to a temperature compensation model establishing device and method based on an olfactory visualization technology, which are suitable for physical parameters which are easily influenced by environmental temperature, such as tea quality, food quality and the like.
Background
The olfactory visualization technology is a novel development and application of the electronic nose technology, and is generally applied to qualitative and quantitative researches in the aspects of tea grade, vinegar age discrimination, meat storage and the like at present. The technology relies on strong interaction between gas molecules and a gas sensor array, such as hydrogen bonds, ionic bonds, pi-pi conjugation and the like, and the enhancement of the acting forces enables the olfactory visualization technology to have better sensitivity and stability in the aspect of detection and stronger anti-interference performance on humidity in the environment. Meanwhile, the technology is not easy to damage samples in the actual operation process, the operation equipment is simple, and the analysis speed is high, so that the technology has wide development space in the aspect of realizing the online detection and control of the computer; on the other hand, in the fields of industry, agriculture and the like, the production quality is also guaranteed, and the method plays an important role in realizing high efficiency of enterprises and rapidness of social economy.
However, in actual operation, the measurement output result is often changed due to the change of the ambient temperature, a certain temperature drift phenomenon exists, the sensitivity of the sensor is correspondingly reduced, and a certain error is generated between the measurement value and the actual value, so that the accuracy of the smell data is reduced. Therefore, the research of a temperature compensation model capable of reducing errors is very critical to the collection of olfactory visual data.
In addition, the temperature compensation method for the olfactory sensor array can be divided into hardware compensation and software compensation in structure, and for the hardware compensation, optimization can be performed in aspects of selection of a sensor array substrate, selection of gas-sensitive materials and the like, but the factors can only realize rough temperature optimization compensation, and influence on the whole olfactory sensor system is very little.
For software compensation, methods such as a table lookup method, a BP neural network method, curve fitting and the like can be adopted, and most of the methods have the function of enabling the model accuracy to be higher. Therefore, the invention also adopts the machine learning method to establish the model to improve the model accuracy.
Disclosure of Invention
Aiming at the problems in the application of the above-mentioned technology, the temperature compensation model is established by taking the temperature as a main variable factor, and the olfactory visualization technology is combined to provide the temperature compensation model correction method which is high in accuracy, strong in stability and good in robustness.
In order to realize the purpose of the above complaint, the invention adopts the following technical scheme:
a method for establishing a temperature compensation model based on an olfactory visualization technology comprises the following steps:
step 1) building a smell visual sensor reaction device;
step 2) collecting smell visual data:
performing a gas-sensitive material screening experiment, weighing 5g of tea leaf samples, placing the tea leaf samples in a culture dish, dripping each gas-sensitive material sample on a C2 reverse phase silicon plate, comparing the color difference change conditions before and after reaction, finally selecting 9 different porphyrin materials to prepare a 3 x 3 gas-sensitive sensor array, and optimizing the reaction time;
performing a blank contrast experiment, namely an experiment without other interference odor, putting the gas-sensitive sensing array into a gas reaction device box, adding no tea sample, heating the gas-sensitive sensing array by using a constant-temperature water bath box to ensure that the air in the reaction box is heated and the constant temperature is constant, respectively acquiring R, G and B characteristic data images before and after reaction by using a scanner, and performing image noise reduction and filtering processing by using software to obtain final color R, G and B differential images; in addition, the center of each dye should be averaged, so as to avoid the nonuniformity in data acquisition, and the data values obtained in the step are all used for establishing a temperature compensation model;
step 3), establishing a temperature compensation model: using the data acquired in the step 2) for establishment and training optimization of a temperature compensation model, using data which is not subjected to interference smell and is influenced by temperature as model input data, and using data which is not subjected to interference smell and is not influenced by temperature as model output data; carrying out optimization training on the BP neural network by adopting a particle swarm PSO optimization algorithm to establish a temperature compensation model, and modifying the threshold value and the weight of the BP neural network according to the actual output and the node error of the sample until the output target error reaches a preset minimum value; adjusting the updating speeds c1 and c2, the iteration times and the population scale of the PSO optimization algorithm according to actual sample data, and optimizing the model again to ensure that the model precision is higher; then, in order to train the performance and the actual utility of the optimization model, the tea odor is used as an interference odor, the data influenced by the interference odor and the temperature is collected and used as model input data, the data not influenced by the temperature and influenced by the interference odor is used as output data, the two groups of data are applied to the model, the model is trained and optimized, and the optimal model parameters are sought.
Further, the olfactory sensation visual sensor reaction device comprises a nitrogen gas bottle (1), a gas two-way valve (2), a constant-temperature water bath box (3), a conical funnel (4), a round metal sheet (5), a relay switch (6), a single chip microcomputer chip (7), a key switch (8), a 3CCD camera (9), an olfactory sensation sensor array (10), a gas reaction device box (11), a gas two-way valve (12), a door handle (13) and a PC (14);
the gas reaction device box (11) is arranged in the constant-temperature water bath box (3), the gas two-way valve (12) is directly arranged outside the gas reaction device box (11), the nitrogen gas bottle (1) is connected into the gas reaction device box (11) through the gas two-way valve (2), and the bottom of the conical funnel (4) directly extends into the gas reaction device box (11); the PC (14) is connected with the 3CCD camera (9) in the gas reaction device box (11) through a data line, and the 3CCD camera (9) is right opposite to the olfactory sensor array (10) in the gas reaction device box (11); a single chip microcomputer chip (7) is connected to the port of the PC machine, one end of the single chip microcomputer chip (7) is connected with a key switch (8), and the other end of the single chip microcomputer chip is connected with a round metal sheet (5) at the neck of the conical funnel (4) through a relay switch (6).
Further, a door handle (13) is also provided on the gas reaction device case (11).
Further, the specific operation steps of the olfactory visual sensor reaction device comprise:
1) Opening a gas two-way valve (2) to enable nitrogen in a nitrogen bottle (1) to flow into a gas reaction device box (11), wherein the nitrogen has the function of enabling gas in the reaction device box to flow and removing other complex gases, and the other complex gases are discharged from a gas reaction device chamber by opening the gas two-way valve (12), and the step is required to be carried out every time the experiment sample is replaced, so that the experiment is more standardized;
2) One end of a singlechip chip (7) is connected with a relay switch (6), one end of the singlechip chip is connected with a key switch (8), the other end of the singlechip chip is connected with a PC (14), the relay switch (6) is closed by pressing the key switch (8) so that the singlechip chip (7) obtains a corresponding instruction, and the circular metal sheet rotates clockwise by 90 degrees when the relay switch (6) is closed, so that a sample in the conical funnel (4) flows into a gas reaction device box along a pipeline, so that the steps can not only achieve the effect that the factors such as stable smell and temperature in a reaction chamber are not damaged by adding the sample, but also can mainly achieve an automatic implementation effect, and are convenient for industrial acquisition in the future;
3) Scanning an olfactory sensor array (10) in a reaction device box by a 3CCD camera (9) to obtain an image before reaction, then scanning the olfactory sensor array by a constant temperature water bath box (3) to enable the gas in the reaction device box to be in a constant temperature state, scanning by the camera again to obtain a sensor array image after reaction influenced by temperature, and inputting the obtained images before and after reaction into a PC (14) for next image processing; meanwhile, regarding the used sample and the reacted sensor array, it is possible to take out waste and replace the sensor by opening the reaction apparatus using a door handle (13).
Further, the specific process of the step 2) is as follows:
step 2.1), firstly, preliminarily screening a plurality of porphyrin materials in a laboratory, weighing 8mg of each porphyrin material, dissolving the porphyrin material in 4ml of dichloromethane, placing the dichloromethane into an ultrasonic oscillator for oscillation for half an hour, taking out the materials, sequentially halating the materials on a C2 reversed phase silica gel plate, and selecting the porphyrin material with better halation effect for next screening;
step 2.2), continuously performing halation on the selected porphyrin material dots on a silica gel plate, reacting with a tea sample, observing the response values of R, G and B components of the materials after reaction, and finally selecting 9 different porphyrin materials which are respectively: tetraphenylporphyrin, tetraphenylporphyrin magnesium (III) chloride, tetra-p-methoxyphenylporphyrin iron (III) chloride, tetraphenylporphyrin copper (II), tetramethoxyphenylporphyrin cobalt (II), tetraphenylporphyrin zinc, tetraporphyrin tetramethyl, meso-tetraphenylporphyrin;
step 2.3), the reaction time optimization comprises the following steps: respectively reacting the prepared 3-by-3 sensor array with tea samples of different grades for 10min,12min,14min, 18min,20min,22min,24min,26min,28min,30min and 32min, and continuously collecting image data of the sensors and the tea odor samples at different times; comprehensively considering the image data of 3 grades, it can be seen that the color change of the image tends to be stable from the 24 th min, so that the optimal reaction time of the sensor array and the tea sample can be determined to be 24min;
performing median filtering to remove noise on the obtained characteristic images before and after reaction in the steps 2.2) and 2.3) through a computer image processing program, then segmenting the sensor image from a background plate by adopting threshold segmentation, then using morphology processing to extract useful image components by using digital morphology as a tool, finally taking out the center of each dye and averaging, and obtaining the final data value for establishing and optimizing a model by using principal component analysis processing;
step 2.4), collecting tea sample smell data, wherein the data obtained in the step is used for verifying the accuracy and precision of the temperature compensation model; collecting odor data of 3 tea leaves with different grades at normal temperature, performing 54 parallel contrasts on the tea leaves with each grade, wherein the data of 3 grades and 162 groups of data are obtained; then, collecting odor data of 3 tea leaves with different grades at another temperature, repeating the above operation, and also forming 162 groups of data; and respectively taking the obtained data as the output and the input of the model so as to train the optimization model.
Because of adopting the above input scheme, the invention has the following advantages:
(1) The detection device established by the invention is simple to operate and convenient to use, and can effectively detect the volatile gas generated in the experiment;
(2) The invention designs a part for automatically inputting the experimental sample, can avoid the condition that the experimental environment which is simulated in the early stage is damaged because the sample needs to be added, has strong practicability, is automatically implemented, and has good industrial application prospect;
(3) By adopting a software compensation method, temperature compensation is realized on the algorithm level, the recognition rate and the operability are higher, the problem of measurement precision is solved, the economic cost brought by hardware compensation is reduced, and the accurate temperature compensation of the olfactory sensor is realized;
(4) The temperature compensation model is simple to operate, the optimal recognition rate can be found only by modifying parameters in the algorithm, and the recognition performance of the olfactory visual detection device is improved.
Drawings
FIG. 1 is a diagram of an array of gas sensors fabricated;
FIG. 2 is a schematic structural diagram of a detection device of an olfactory visual sensor designed by the invention;
fig. 3 is a flow chart for establishing a temperature compensation method based on an olfactory visualization technology.
Detailed Description
The invention is described in detail below with reference to the figures and examples.
In the following detailed description of the embodiments, specific terms and techniques are used so as to have the same meaning as would be understood by one of ordinary skill in the art in the light of the accompanying drawings, which are described in the specification and drawings for purposes of explanation.
As shown in figure 1, 27 porphyrin materials in a laboratory are primarily screened, 8mg of each porphyrin material is weighed and dissolved in 4ml of dichloromethane, the dichloromethane is placed into an ultrasonic oscillator to oscillate for half an hour, a sample application capillary is taken out to absorb the same dose every time and is sequentially subjected to halation on a C2 reversed phase silica gel plate, and the porphyrin material with better halation effect is selected for next screening;
continuously shading the selected porphyrin material points on a silica gel plate, reacting with a tea sample, observing the response values of R, G and B components of the materials after reaction, and finally selecting 9 different porphyrin materials which are respectively: tetraphenylporphyrin, tetraphenylporphyrin magnesium (III) chloride, tetra-p-methoxyphenylporphyrin iron (III) chloride, tetraphenylporphyrin copper (II), tetramethoxyphenylporphyrin cobalt (II), tetraphenylporphyrin zinc, tetraporphyrin tetramethyl, meso-tetraphenylporphyrin;
fig. 2 is a structural diagram of the olfactory sensation visual sensor detection device, which can be divided into a hardware part and a software part. The hardware part consists of a nitrogen cylinder (1), a constant temperature water bath box (3), a singlechip chip (7), a gas reaction device box (11) and a PC (personal computer) machine (14). Gas reaction device case (11) have been put in constant temperature water bath (3), nitrogen cylinder (1) and PC (14) are being connected respectively at the both ends of gas reaction device case (11), inside at gas reaction device case (11), install 3CCD camera (9) and place sense of smell sensor array (10), 3CCD camera (9) are facing to sense of smell sensor array (10), PC (14) are being connected to the other end, and singlechip chip (7) are still being connected to PC (14) outside, the rotation of circular sheetmetal (5) is being controlled in singlechip chip (7). The software part consists of a 3CCD camera (9) driving program, a single chip microcomputer control driving program of a PC (14), an image processing program and a mode recognition system.
The specific use mode of the device is as follows: opening the gas two-way valve (2) to make the nitrogen in the nitrogen bottle (1) flow into the gas reaction device box (11), the nitrogen plays a role of making the gas in the reaction device box flow and removing the rest of the complex gas, the rest of the complex gas is removed from the gas reaction device chamber by opening the gas two-way valve (12), and the two-way valves (2), (12) can be closed after completing the step. Then, scanning the olfactory sensor array (10) by a 3CCD camera (9) to obtain an image before reaction, and automatically putting a sample into the device after the scanning is finished: one end of the singlechip chip (7) is connected with the relay switch (6), the other end is connected with the key switch (8), and the other end is connected with the PC (14). By pressing the key switch (8), the single chip microcomputer chip (7) is enabled to obtain a corresponding instruction to close the relay switch (6), when the relay switch (6) is closed, the circular metal sheet (5) can rotate clockwise by 90 degrees, then a sample in the conical funnel (4) can flow into the gas reaction device box along the pipeline, and when the key switch (8) is closed, the circular metal sheet (5) can rotate anticlockwise by 90 degrees to close the channel. And as for whether data influenced by temperature needs to be acquired, the gas in the reaction device box is heated and tends to be in a constant temperature state through the constant temperature water bath box (3), and the image of the sensor array after reaction influenced by temperature can be obtained by scanning the camera again. And inputting the obtained images before and after the reaction into a PC (personal computer) for further image processing. Meanwhile, regarding the used sample and the reacted sensor array, it is possible to remove the waste and replace the sensor array by opening the reaction apparatus using a door handle (13).
Firstly, the step of disturbing the smell is needed for optimizing the model training in the later period, so that the smell of the tea leaves is selected as the disturbing smell, namely the tea leaves are used as the experimental sample.
The obtained characteristic images before and after reaction are subjected to median filtering to remove noise through a computer image processing program, then a sensor image is segmented from a background plate by adopting threshold segmentation, then morphological processing is used, digital morphology is used as a tool to extract useful image components, finally the center of each dye is taken out and averaged, and principal component analysis processing is used for obtaining the final data value for establishing and optimizing a model.
Fig. 3 is a flow chart for establishing a temperature compensation method based on an olfactory visualization technology. The model establishment needs to collect odor data of 4 different conditions, which are respectively as follows: the method comprises the following steps that (1) data (2) which are free of interfering odor and are not influenced by temperature, data (3) which are free of interfering odor and are influenced by temperature, data (4) which are free of interfering odor and are not influenced by temperature, data (1) and data (2) are used for building a temperature compensation model, data (3) and data (4) are used for training and optimizing the model, and finally the temperature compensation model capable of well improving the olfactory odor detection accuracy of the sensor is obtained.
While the present invention has been particularly shown and described with reference to the accompanying drawings, it will be understood by those skilled in the art that various changes and modifications in detail, and methods of operation thereof, may be made without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (4)
1. A method for establishing a temperature compensation model based on an olfactory visualization technology is characterized by comprising the following steps:
step 1) building a smell visual sensor reaction device;
step 2) collecting olfactory visual data:
performing a gas-sensitive material screening experiment, weighing 5g of tea leaf samples, placing the tea leaf samples in a culture dish, dripping each gas-sensitive material on a C2 reverse phase silicon plate, comparing the color difference change conditions before and after reaction, finally selecting 9 different porphyrin materials to prepare a 3 x 3 gas-sensitive sensor array, and optimizing the reaction time;
performing a blank comparison experiment, namely an experiment without other interfering odors, putting the gas-sensitive sensing array into a gas reaction device box, adding no tea sample, heating the gas-sensitive sensing array by using a constant-temperature water bath box to ensure that the air in the reaction box is heated and constant temperature is constant, respectively acquiring R, G and B characteristic data images before and after reaction by using a scanner, and performing image noise reduction and filtering processing by using software to obtain final color R, G and B difference images; in addition, the center of each dye should be averaged, so as to avoid the nonuniformity in data acquisition, and the data values obtained in the step are all used for establishing a temperature compensation model;
the specific process of the step 2) is as follows:
step 2.1), firstly, preliminarily screening a plurality of porphyrin materials in a laboratory, weighing 8mg of each porphyrin material, dissolving the porphyrin material in 4ml of dichloromethane, placing the dichloromethane into an ultrasonic oscillator for oscillation for half an hour, taking out the materials, sequentially halating the materials on a C2 reversed phase silica gel plate, and selecting the porphyrin material with better halation effect for next screening;
step 2.2), continuously shading the selected porphyrin material points on a silica gel plate and reacting with a tea sample, observing the response values of R, G and B components of the materials after reaction, and finally selecting 9 different porphyrin materials which are respectively: tetraphenylporphyrin, tetraphenylporphyrin magnesium (III) chloride, tetra-p-methoxyphenylporphyrin iron (III) chloride, tetraphenylporphyrin copper (II), tetramethoxyphenylporphyrin cobalt (II), tetraphenylporphyrin zinc, tetraporphyrin tetramethyl, meso-tetraphenylporphyrin;
step 2.3), the reaction time optimization comprises the following steps: respectively reacting the prepared 3-by-3 sensor array with tea samples of different grades for 10min,12min,14min, 18min,20min,22min,24min,26min,28min,30min and 32min, and continuously collecting image data of the sensors and the tea odor samples at different times; comprehensively considering the image data of 3 grades, the color change of the image is seen to tend to be stable from the 24 th min, so that the optimal reaction time of the sensor array and the tea sample is determined to be 24min;
step 2.2) and step 2.3), the obtained characteristic images before and after reaction are subjected to median filtering to remove noise through a computer image processing program, then a sensor image is segmented from a background plate by adopting threshold segmentation, then morphology processing is used, useful image components are extracted by taking digital morphology as a tool, finally the center of each dye is taken out and an average value is taken, and the final data value is obtained by utilizing principal component analysis processing and used for establishing and optimizing a model;
step 2.4), collecting tea sample odor data, wherein the data obtained in the step is used for verifying the accuracy and precision of the temperature compensation model; collecting odor data of 3 tea leaves with different grades at normal temperature, performing 54 parallel contrasts on the tea leaves with each grade, wherein the data of 3 grades and 162 groups of data are obtained; then, collecting smell data of 3 tea leaves with different grades at another temperature, repeating the operation, and also forming 162 groups of data; respectively taking the obtained data as the output and the input of the model so as to train the optimization model;
step 3), establishing a temperature compensation model: using the data acquired in the step 2) for establishing and training optimization of a temperature compensation model, taking the data which is not subjected to interference smell and is influenced by temperature as model input data, and taking the data which is not subjected to interference smell and is not influenced by temperature as model output data; carrying out optimization training on the BP neural network by adopting a particle swarm PSO optimization algorithm to establish a temperature compensation model;
the specific process of the step 3) is as follows: modifying the threshold and the weight of the BP neural network according to the actual output of the sample and the node error until the output target error reaches a preset minimum value; adjusting the updating speeds c1 and c2, the iteration times and the population scale of the PSO optimization algorithm according to actual sample data, and optimizing the model again to ensure that the model precision is higher; then, in order to train the performance and the actual utility of the optimization model, the tea odor is used as an interference odor, the data influenced by the interference odor and the temperature is collected and used as model input data, the data not influenced by the temperature and influenced by the interference odor is used as output data, the two groups of data are applied to the model, the model is trained and optimized, and the optimal model parameters are sought.
2. The method for establishing the temperature compensation model based on the olfactory sensation visualization technology as claimed in claim 1, wherein the olfactory sensation visualization sensor reaction device comprises a nitrogen cylinder (1), a first gas two-way valve (2), a constant temperature water bath box (3), a conical funnel (4), a round metal sheet (5), a relay switch (6), a single chip microcomputer chip (7), a key switch (8), a 3CCD camera (9), an olfactory sensation sensor array (10), a gas reaction device box (11), a second gas two-way valve (12), a door handle (13) and a PC (14);
placing a gas reaction device box (11) in a constant-temperature water bath box (3), directly installing a second gas two-way valve (12) outside the gas reaction device box (11), connecting a nitrogen gas bottle (1) into the gas reaction device box (11) through a first gas two-way valve (2), and directly enabling the bottom of a conical funnel (4) to go deep into the gas reaction device box (11); the PC (14) is connected with the 3CCD camera (9) in the gas reaction device box (11) through a data line, and the 3CCD camera (9) is right opposite to the olfactory sensor array (10) in the gas reaction device box (11); a single chip microcomputer chip (7) is connected to the port of the PC machine, one end of the single chip microcomputer chip (7) is connected with a key switch (8), and the other end of the single chip microcomputer chip is connected with a round metal sheet (5) at the neck of the conical funnel (4) through a relay switch (6).
3. The method for establishing the temperature compensation model based on the olfactory visualization technology as claimed in claim 2, wherein a door handle (13) is further provided on the gas reaction device box (11).
4. The method for establishing the temperature compensation model based on the olfactory visualization technology as claimed in claim 2, wherein the specific operation steps of the olfactory visualization sensor reaction device include:
1) Opening a first gas two-way valve (2) to enable nitrogen in a nitrogen bottle (1) to flow into a gas reaction device box (11), wherein the nitrogen has the functions of enabling gas in the reaction device box to flow and removing other complex gases, the other complex gases are discharged from a gas reaction device chamber by opening a second gas two-way valve (12), and the step is required to be carried out every time the experiment sample is replaced, so that the experiment is more standardized;
2) One end of the singlechip chip (7) is connected with the relay switch (6), one end of the singlechip chip is connected with the key switch (8), the other end of the singlechip chip is connected with the PC (14), the relay switch (6) is closed by pressing the key switch (8) so that the singlechip chip (7) obtains a corresponding instruction, the circular metal sheet rotates clockwise by 90 degrees when the relay switch (6) is closed, and then a sample in the conical funnel (4) flows into the gas reaction device box along a pipeline;
3) Scanning an olfactory sensor array (10) in a reaction device box by a 3CCD camera (9) to obtain an image before reaction, then scanning the olfactory sensor array by a constant temperature water bath box (3) to enable the gas in the reaction device box to be in a constant temperature state, scanning by the camera again to obtain a sensor array image after reaction influenced by temperature, and inputting the obtained images before and after reaction into a PC (14) for next image processing; meanwhile, regarding the used sample and the reacted sensor array, the waste is taken out and the sensor is replaced by opening the reaction apparatus using a door handle (13).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110817323.XA CN113484312B (en) | 2021-07-20 | 2021-07-20 | Method for establishing temperature compensation model based on olfactory visualization technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110817323.XA CN113484312B (en) | 2021-07-20 | 2021-07-20 | Method for establishing temperature compensation model based on olfactory visualization technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113484312A CN113484312A (en) | 2021-10-08 |
| CN113484312B true CN113484312B (en) | 2023-04-18 |
Family
ID=77942319
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110817323.XA Active CN113484312B (en) | 2021-07-20 | 2021-07-20 | Method for establishing temperature compensation model based on olfactory visualization technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113484312B (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101936912A (en) * | 2010-08-25 | 2011-01-05 | 江苏大学 | Method and device for detecting freshness of fish based on olfaction visualization |
| CN103499646A (en) * | 2013-07-30 | 2014-01-08 | 中国标准化研究院 | Honey characteristic fragrance analysis and honey fragrance system simulation method |
| CN108181480A (en) * | 2017-12-29 | 2018-06-19 | 东北电力大学 | A kind of identification device and method of paraffin odor grade |
| CN109406500A (en) * | 2018-09-30 | 2019-03-01 | 江苏大学 | A kind of sausage rapid classification method based on olfaction visualization array |
| CN109447104A (en) * | 2018-09-10 | 2019-03-08 | 江苏大学 | Food smell mobile phone imaging detection method based on olfaction visualization technology |
| CN110702815A (en) * | 2019-10-29 | 2020-01-17 | 江苏大学 | Olfactory sensing detection method for flour storage quality |
-
2021
- 2021-07-20 CN CN202110817323.XA patent/CN113484312B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101936912A (en) * | 2010-08-25 | 2011-01-05 | 江苏大学 | Method and device for detecting freshness of fish based on olfaction visualization |
| CN103499646A (en) * | 2013-07-30 | 2014-01-08 | 中国标准化研究院 | Honey characteristic fragrance analysis and honey fragrance system simulation method |
| CN108181480A (en) * | 2017-12-29 | 2018-06-19 | 东北电力大学 | A kind of identification device and method of paraffin odor grade |
| CN109447104A (en) * | 2018-09-10 | 2019-03-08 | 江苏大学 | Food smell mobile phone imaging detection method based on olfaction visualization technology |
| CN109406500A (en) * | 2018-09-30 | 2019-03-01 | 江苏大学 | A kind of sausage rapid classification method based on olfaction visualization array |
| CN110702815A (en) * | 2019-10-29 | 2020-01-17 | 江苏大学 | Olfactory sensing detection method for flour storage quality |
Non-Patent Citations (5)
| Title |
|---|
| 《Qualitative and quantitative analysis of chlorpyrifos residues in tea by surface-enhanced Raman spectroscopy (SERS) combined with chemometric models》;Jiaji Zhu等;《LWT - Food Science and Technology》;20190726;全文 * |
| 《基于PSO-BP神经网络的湿度传感器温度补偿》;行鸿彦等;《传感技术学报》;20150630;第28卷(第6期);全文 * |
| 《基于嗅觉和光谱技术融合的面粉脂肪酸值定量检测》;江辉等;《农业机械学报》;20210228;第52卷(第2期);全文 * |
| 《引入粒子群方差条件的PSO气体传感器温度补偿》;唐朝国;《电子元器件与信息技术》;20200131;第4卷(第1期);全文 * |
| Evaluating aroma quality of black tea by an olfactory visualization system:Selection of feature sensor using particle swarm optimization;Hui Jiang等;《Food Research International》;20191231;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113484312A (en) | 2021-10-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109977780A (en) | A kind of detection and recognition methods of the diatom based on deep learning algorithm | |
| CN109919934A (en) | A kind of liquid crystal display panel defect inspection method based on the study of multi-source domain depth migration | |
| CN111751376B (en) | Rice nitrogen nutrition estimation method based on canopy image feature derivation | |
| CN105241821A (en) | Slightly decayed fruit quick identification method based on smell spectral imaging technology | |
| CN109447104A (en) | Food smell mobile phone imaging detection method based on olfaction visualization technology | |
| CN110702815A (en) | Olfactory sensing detection method for flour storage quality | |
| Tagliabue et al. | Authigenic mineral phases as a driver of the upper-ocean iron cycle | |
| CN113484312B (en) | Method for establishing temperature compensation model based on olfactory visualization technology | |
| CN111783616A (en) | Data-driven self-learning-based nondestructive testing method | |
| CN112651173B (en) | Agricultural product quality nondestructive testing method based on cross-domain spectral information and generalizable system | |
| CN108694375A (en) | A kind of image conversion white wine recognition methods can be used for polyelectron nose platform | |
| CN105740884B (en) | Hyperspectral Image Classification method based on singular value decomposition and neighborhood space information | |
| Zhu et al. | Development of a novel noninvasive quantitative method to monitor Siraitia grosvenorii cell growth and browning degree using an integrated computer‐aided vision technology and machine learning | |
| Zhu et al. | A prediction method for intervals of trace ions concentration in zinc sulfate solution based on UV-vis spectroscopy | |
| CN115598164A (en) | Machine learning integrated soil heavy metal concentration detection method and system | |
| CN114577791A (en) | Detection method for quality grade of tea | |
| CN102392068A (en) | Method for online in-situ monitoring of solid-state fermentation fungus biomass | |
| CN103048275A (en) | Adaptive soft-instrument device and adaptive soft-instrument construction method based on near infrared spectrum | |
| Danev et al. | Application possibilities of open-source software for microbiological analyses | |
| CN107122496B (en) | Near infrared spectrum substance content indexing method, device and medium based on category analysis method | |
| CN116974201B (en) | Buoyancy separation-based garbage incinerator slag recycling control method and system | |
| CN113671116B (en) | Method for identifying maturity of oil shale in cracking process by using electronic nose | |
| CN104297191B (en) | Construction method of model for detecting guanosine yield by using guanosine fermentation liquid ultraviolet fingerprint spectra | |
| CN116793991B (en) | Glutamic acid concentration measurement method based on near infrared spectrum and mixing loss | |
| Valcárcel et al. | A hierarchical approach to analytical chemistry |
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 |