CN108287183A - A method of reducing cross sensitivity of the semiconductor hydrogen gas sensor to carbon monoxide - Google Patents
A method of reducing cross sensitivity of the semiconductor hydrogen gas sensor to carbon monoxide Download PDFInfo
- Publication number
- CN108287183A CN108287183A CN201711462857.5A CN201711462857A CN108287183A CN 108287183 A CN108287183 A CN 108287183A CN 201711462857 A CN201711462857 A CN 201711462857A CN 108287183 A CN108287183 A CN 108287183A
- Authority
- CN
- China
- Prior art keywords
- hydrogen gas
- gas sensor
- carbon monoxide
- bpnn
- training
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N27/00—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means
- G01N27/02—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating impedance
- G01N27/04—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating impedance by investigating resistance
- G01N27/12—Investigating or analysing materials by the use of electric, electrochemical, or magnetic means by investigating impedance by investigating resistance of a solid body in dependence upon absorption of a fluid; of a solid body in dependence upon reaction with a fluid, for detecting components in the fluid
Landscapes
- Chemical & Material Sciences (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Electrochemistry (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analyzing Materials By The Use Of Fluid Adsorption Or Reactions (AREA)
Abstract
The present invention relates to a kind of reduction semiconductor hydrogen gas sensors to the method for the cross sensitivity of carbon monoxide, includes the following steps：Sensor circuit is established, two-dimensional calibrations experiment is carried out, data is collected in computer；Neural network sample file is made, the training that the sample data that calibration experiment obtains is used as 1/2 2/3 in sum to neural network forms training sample file, forms network structure and weights；Again with remaining 1/2 1/3 sample data to forming test samples file；After sample file is normalized, BPNN is created using Matlab softwares；Training sample file is substituted into, network parameter is set, the BPNN that training has created exports BPNN model structure parameters；Checking by substitution sample file, trained BPNN calculates the output results of test samples to use.The present invention can conveniently reduce cross sensitivity of the semiconductor hydrogen gas sensor to carbon monoxide.
Description
Technical field
The present invention relates to technical field of sensor measurement, more particularly to a kind of reduction semiconductor hydrogen gas sensor pair oxygen
Change the method for the cross sensitivity of carbon.
Background technology
Semiconductor hydrogen gas sensor is a kind of using resistor-type Sensitive Apparatus made of metal oxide, residing for sensor
There are when hydrogen in environment, the conductivity of sensor increases with the increase of Hydrogen in Air concentration.Automotive field, military neck
Domain, chemical field etc. all be unable to do without hydrogen application, semiconductor hydrogen gas sensor also have in the hydrogen measurement in these fields compared with
To be widely applied.Although semiconductor hydrogen gas sensor is good, simple in structure, cheap with stability, is easy to compound spy
Point, but the poor selectivity of this sensor, are vulnerable to the influence of other gases such as carbon monoxide.Therefore, there is an urgent need for a kind of simple
It is convenient to reduce method of the semiconductor hydrogen gas sensor to the cross sensitivity of carbon monoxide.
Invention content
A kind of friendship technical problem to be solved by the invention is to provide reduction semiconductor hydrogen gas sensor to carbon monoxide
Sensitive method is pitched, cross sensitivity of the semiconductor hydrogen gas sensor to carbon monoxide can be conveniently reduced.
The technical solution adopted by the present invention to solve the technical problems is：A kind of reduction semiconductor hydrogen gas sensor pair is provided
The method of the cross sensitivity of carbon monoxide, includes the following steps：
(1) sensor circuit is established, two-dimensional calibrations experiment is carried out, selectes multiple and different carbonomonoxide concentration states to being mended
The hydrogen gas sensor repaid carries out calibration experiment, and data are collected in computer；
(2) neural network sample file is made, the sample data that calibration experiment obtains is used as the 1/2-2/3 in sum
The training of neural network forms training sample file, forms network structure and weights；Again with the sample data of remaining 1/2-1/3
To forming test samples file；
(3) after sample file being normalized, BPNN is created using Matlab softwares；
(4) training sample file is substituted into, network parameter, the BPNN that training has created, output BPNN model structures ginseng are set
Number；
(5) checking by substitution sample file, trained BPNN calculates the output results of test samples to use.
The hydrogen gas sensor and carbon monoxide transducer more biographies used at the same time in sensor circuit in the step (1)
Sensor integration technology, wherein hydrogen gas sensor measures aim parameter, and carbon monoxide transducer measures noisy to hydrogen gas sensor
Carbonomonoxide concentration.
The line number phase of the input layer number of the BPNN models obtained in the step (4) and training input sample file
Together, automatic to obtain, the number of hidden nodes 6, output layer number of nodes is 1.
Advantageous effect
Due to the adoption of the above technical solution, compared with prior art, the present invention having the following advantages that and actively imitating
Fruit：The present invention uses multi-sensor fusion technology, reduces the interference that carbon monoxide detects density of hydrogen；Eliminate original simultaneously
The hardware configuration first detected reduces cost to simplify structure.The present invention is simple and effective, and method is clear, convenient and efficient,
It can be applied to using semiconductor hydrogen gas sensor made of different metal oxides, have the effect of relatively good.
Description of the drawings
Fig. 1 is the flow chart of the present invention；
Fig. 2 is the circuit block diagram of sensor circuit；
Fig. 3 is verification result figure of the hydrogen actual concentrations on forecast sample before improving；
Fig. 4 is to reduce verification result figure of the hydrogen actual concentrations on forecast sample after carbon monoxide interference；
Fig. 5 is the structure chart of the BPNN models used in present embodiment.
Specific implementation mode
Present invention will be further explained below with reference to specific examples.It should be understood that these embodiments are merely to illustrate the present invention
Rather than it limits the scope of the invention.In addition, it should also be understood that, after reading the content taught by the present invention, people in the art
Member can make various changes or modifications the present invention, and such equivalent forms equally fall within the application the appended claims and limited
Range.
Embodiments of the present invention are related to a kind of side reducing semiconductor hydrogen gas sensor to the cross sensitivity of carbon monoxide
Method, as shown in Figure 1, including the following steps：Sensor circuit is established, two-dimensional calibrations experiment is carried out, selectes multiple and different oxidations
Concentration of carbon state carries out calibration experiment to the hydrogen gas sensor compensated, and data are collected in computer；Make neural network
Sample file, the training that the sample data that calibration experiment obtains is used as the 1/2-2/3 in sum to neural network form training
Sample file forms network structure and weights；Again with the sample data of remaining 1/2-1/3 to forming test samples file；It is right
After sample file is normalized, BPNN is created using Matlab softwares；Training sample file is substituted into, setting network is joined
Number, the BPNN that training has created export BPNN model structure parameters；Checking by substitution sample file, trained BPNN is counted use
Calculate the output result of test samples.
In present embodiment, hydrogen gas sensor and carbon monoxide transducer multi-sensor fusion technology used at the same time, hydrogen
Gas sensor measures aim parameter, and carbon monoxide transducer is measured to the noisy carbonomonoxide concentration of hydrogen gas sensor, passes through two
The measurement data that calibration experiment obtains hydrogen gas sensor and carbon monoxide transducer is tieed up, data acquired is deposited into computer,
Using BP neural network algorithm, BP neural network structure and neural network weight are determined using Matlab softwares, will calculated
The model obtained in machine is deposited into multi-sensor measurement system microprocessor, semiconductor hydrogen gas sensor and carbon monoxide sensing
Input value of the device measured value as model reduces by half to the density of hydrogen being reduced after carbon monoxide interference to reach
Purpose of the conductor hydrogen gas sensor to the cross sensitivity of carbon monoxide.
The BPNN model structures used in present embodiment are as shown in figure 5, input layer number and training input sample text
The line number of part is identical, automatic to obtain, the number of hidden nodes 6, and output layer number of nodes is 1.
It is further illustrated the present invention below by a specific embodiment.
Reduce cross sensitivity of the QM-H1 type semiconductor hydrogen-sensitive elements to carbon monoxide：It initially sets up density of hydrogen and measures electricity
Road, carbonomonoxide concentration have a certain impact to QM-H1 type hydrogen gas sensors, using hydrogen gas sensor as master reference, by one
Carbon sensor is aoxidized as aiding sensors, acquires the measurement data of two kinds of sensors.Measuring circuit block diagram, as shown in Fig. 2, main
Sensor and aiding sensors are handled after receiving data into signal processing circuit, then it is AD converted it is laggard in a subtle way
Processor.The data measured are deposited into Matlab programs, reject accidental error point using statistical method, remaining data is made
Data are handled for BP neural network, as the input value of BP neural network after remaining data is normalized, are carried out
Matlab is emulated, and by the emulation of multiplicating property, is compared and is obtained more excellent neural network structure, weighed using BP neural network
The directly determining method of value, obtains neural network weight, to obtain sensing system output valve.
Fig. 3 is verification result of the hydrogen actual concentrations on forecast sample before improving, and Fig. 4 is after reducing carbon monoxide interference
Verification result of the hydrogen actual concentrations on forecast sample.By comparing as it can be seen that can effectively be dropped after method using the present invention
Cross sensitivity of the low semiconductor hydrogen gas sensor to carbon monoxide.
Claims (3)
1. a kind of reducing method of the semiconductor hydrogen gas sensor to the cross sensitivity of carbon monoxide, which is characterized in that including following
Step：
(1) it establishes sensor circuit, carries out two-dimensional calibrations experiment, select multiple and different carbonomonoxide concentration states to being compensated
Hydrogen gas sensor carries out calibration experiment, and data are collected in computer；
(2) neural network sample file is made, the sample data that calibration experiment obtains is used as nerve to the 1/2-2/3 in sum
The training of network forms training sample file, forms network structure and weights；Again with the sample data of remaining 1/2-1/3 to shape
At test samples file；
(3) after sample file being normalized, BPNN is created using Matlab softwares；
(4) training sample file is substituted into, network parameter is set, the BPNN that training has created exports BPNN model structure parameters；
(5) checking by substitution sample file, trained BPNN calculates the output results of test samples to use.
2. according to claim 1 reduce method of the semiconductor hydrogen gas sensor to the cross sensitivity of carbon monoxide, spy
Sign is, the hydrogen gas sensor in sensor circuit and carbon monoxide transducer in the step (1) are used at the same time to be passed more
Sensor integration technology, wherein hydrogen gas sensor measures aim parameter, and carbon monoxide transducer measures noisy to hydrogen gas sensor
Carbonomonoxide concentration.
3. according to claim 1 reduce method of the semiconductor hydrogen gas sensor to the cross sensitivity of carbon monoxide, spy
Sign is that the input layer number of the BPNN models obtained in the step (4) is identical as the training line number of input sample file,
Automatic to obtain, the number of hidden nodes 6, output layer number of nodes is 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711462857.5A CN108287183A (en) | 2017-12-28 | 2017-12-28 | A method of reducing cross sensitivity of the semiconductor hydrogen gas sensor to carbon monoxide |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711462857.5A CN108287183A (en) | 2017-12-28 | 2017-12-28 | A method of reducing cross sensitivity of the semiconductor hydrogen gas sensor to carbon monoxide |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108287183A true CN108287183A (en) | 2018-07-17 |
Family
ID=62832624
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711462857.5A Pending CN108287183A (en) | 2017-12-28 | 2017-12-28 | A method of reducing cross sensitivity of the semiconductor hydrogen gas sensor to carbon monoxide |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108287183A (en) |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101226162A (en) * | 2008-02-18 | 2008-07-23 | 重庆大学 | Intelligent method for inhibiting gas-sensitive sensor decussation sensitivity |
| CN103940453A (en) * | 2014-04-15 | 2014-07-23 | 东华大学 | Method for improving sensor measuring precision |
| CN106841325A (en) * | 2017-01-18 | 2017-06-13 | 西安交通大学 | One kind is based on semiconductor gas sensor array detection exhaled gas device |
-
2017
- 2017-12-28 CN CN201711462857.5A patent/CN108287183A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101226162A (en) * | 2008-02-18 | 2008-07-23 | 重庆大学 | Intelligent method for inhibiting gas-sensitive sensor decussation sensitivity |
| CN103940453A (en) * | 2014-04-15 | 2014-07-23 | 东华大学 | Method for improving sensor measuring precision |
| CN106841325A (en) * | 2017-01-18 | 2017-06-13 | 西安交通大学 | One kind is based on semiconductor gas sensor array detection exhaled gas device |
Non-Patent Citations (3)
| Title |
|---|
| 刘君华 主编: "《智能传感器系统》", 31 May 2010, 西安电子科技大学出版社 * |
| 马戎 等: "基于传感器阵列与神经网络的气体检测系统", 《传感技术学报》 * |
| 黄小燕 等: "电子鼻在气体检测中的应用研究", 《传感器与微系统》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107016236B (en) | Power grid false data injection attack detection method based on nonlinear measurement equation | |
| CN104483598B (en) | A kind of Cable fault examination analysis method | |
| Tian et al. | A novel pattern mismatch based interference elimination technique in E-nose | |
| CN112198209B (en) | Formaldehyde detection method and formaldehyde detection device | |
| CN106706852A (en) | Calibration method and calibration system of gas concentration sensor | |
| CN105911106B (en) | The insulating gas condition detecting system of GIS device | |
| CN105466927A (en) | Identification method, correction method and alarm method for turbidimetry abnormal reaction curve | |
| CN109030566B (en) | Laboratory gas leakage diagnosis device and method | |
| CN1381721A (en) | Portable intelligent electronic nose and its preparing process | |
| CN107036980A (en) | A kind of method and refrigerator for detecting refrigerator food freshness | |
| CN106680592A (en) | Method and system for detecting loss of Cable with arbitrary lengths | |
| CN104303017B (en) | Deconvolution method for emissions measurement | |
| CN205038202U (en) | Hazardous gas real -time detection device | |
| CN112198208B (en) | Formaldehyde detection method and formaldehyde detection device | |
| CN104458833B (en) | A kind of humidity sensor test data analytic method based on Time Domain Reflectometry principle | |
| Marshall et al. | Checking normality for parametric tests | |
| CN111561968A (en) | Sensor-based environmental parameter detection method and device and data processing equipment | |
| CN103940453B (en) | A kind of raising sensor accuracy class method | |
| CN108287183A (en) | A method of reducing cross sensitivity of the semiconductor hydrogen gas sensor to carbon monoxide | |
| Joo et al. | Pattern recognition of gas sensor array using characteristics of impedance | |
| CN104067113B (en) | The conductivity type contact test system of low conductivity | |
| Llobet et al. | Electronic nose simulation tool centred on PSpice | |
| Rodrigues et al. | Tristimulus analysis for sensors set with either positive or negative sensitivities—determination of the relative concentration of an analyte in a binary mixture | |
| CN205719182U (en) | Detection device | |
| Burgués et al. | Using net analyte signal to estimate the limit of detection in temperature-modulated MOX sensors |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180717 |
|
| RJ01 | Rejection of invention patent application after publication |