CN105352907A - Infrared gas sensor based on radial basis network temperature compensation and detection method - Google Patents
Infrared gas sensor based on radial basis network temperature compensation and detection method Download PDFInfo
- Publication number
- CN105352907A CN105352907A CN201510847284.2A CN201510847284A CN105352907A CN 105352907 A CN105352907 A CN 105352907A CN 201510847284 A CN201510847284 A CN 201510847284A CN 105352907 A CN105352907 A CN 105352907A
- Authority
- CN
- China
- Prior art keywords
- radial basis
- basis function
- neural network
- infrared
- gas sensor
- 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
Abstract
The invention provides an infrared gas sensor based on radial basis network temperature compensation and a detection method, that are, a gas concentration detection method based on radial basis function neural network temperature compensation and an infrared gas sensor based on radial basis temperature compensation. The infrared gas sensor can be made to eliminate a measurement error generated due to a detected environment temperature change, and compared with an existing empirical formula method, the temperature compensation process is easier, more convenient and wider in applicability due to the fact that a radial basis network has the advantages that the training speed is higher and structural self-adaption determination and output are irrelevant to an initial weight value; compared with an existing temperature control method, due to the fact that only a chip embedded with a radial basis function neural network algorithm needs to be additionally arranged, increasement of the size of the infrared gas sensor is avoided, and minimization and low cost of the sensor are better facilitated.
Description
Technical field
The invention belongs to infrared gas sensor technical field, be specifically related to a kind of infrared gas sensor based on radial basis function network temperature compensation and detection method, particularly relating to infrared gas sensor detection technique, gas concentration detection method based on radial basis function neural network temperature compensation, is a kind of infrared gas sensor that can carry out temperature compensation.
Background technology
Infrared gas sensor have selectivity good, reliable and stable, be swift in response, the not easily plurality of advantages such as poisoning, long service life, being widely used in numerous places such as chemical industry, coal, metallurgy, electric power, environmental monitoring, is the important tool guaranteeing normal production, support personnel's safety.
Common optical gas detection technique has NDIR (Non-Dispersive Infrared) detection technique, tunable laser spectral technique, acousto-optic spectral technique etc., wherein,, stable performance the simplest based on wide range infrared light supply, non-dispersive infrared gas sensor structure, to be swift in response, be applicable to light and handy equipment, there is huge commercial value.Its ultimate principle make use of the selective spectral absorption approach of gas molecule, totally different due to gas molecule structure, cause and make it have different energy levels, therefore also different to the degree of absorption of infrared radiation at different frequency place, and lambert-Bill's absorption law is followed in the absorption of gas to infrared radiation.The research and development such as Barcelona, ESP university J.Fonollosa detect the application of installation of ethylene gas concentration in the storing process of fruit, and the minimum detectable resolution of instrument is 30ppm.The on-dispersive methane gas sensor of the research and development such as the S.Aleksandrov of St. Petersburg, Russia physical technique research institute, adopt the single light source bifocal path structure of LED and photodiode composition, its precision is well beyond industrial detection standard.Harbin Institute of Technology Sun Hui etc. devises the detection of gas with multiple constituents instrument based on non-dispersive infrared absorption principle, utilizes light source multiplexing structure, can detect sulphuric dioxide, nitrogen monoxide and oxygen simultaneously.
But when carrying out gas concentration and detecting, the change of testing environment temperature is the key factor affecting infrared gas sensor measuring accuracy.Under different environment temperatures, temperature sense electric current can be produced between binary channels pyroelectric detector two electrodes of infrared gas sensor, be mixed in measuring-signal, thus cause the generation of temperature drift.In addition, from equation of gas state PV=nRT, when molal quantity one timing of gas, when pressure P is constant, along with the rising of temperature T, gas volume V will increase.Contrary, along with the reduction of temperature T, gas volume V will reduce.Chamber volume due to infrared gas sensor is fixing, and the rising of temperature will cause the minimizing of gas molecule molal quantity in air chamber, thus reduces the concentration of gas, reduces the absorption of gas to infrared radiation, causes the generation of measuring error.
At present, the method eliminating testing environment influence of temperature change mainly contains two kinds.One is empirical formula method, namely least square fitting is adopted to carry out iteration, determine the related coefficient of experimental formula, thus the error that compensates is brought, this method coefficient of performance after data acquisition is demarcated and is carried out temperature compensation, but calculated amount is comparatively large, and the use of experimental formula has certain limitation; Two is temperature control methods, namely temperature control module is adopted to make testing environment temperature keep mobile equilibrium, thus avoid the measuring error brought because of temperature variation, this method utilizes hardware circuit to make infrared gas sensor be in the testing environment of relative constant temperature, but temperature control module add the manufacturing cost adding infrared gas sensor, and be unfavorable for miniaturization.
Summary of the invention
Object of the present invention provides a kind of infrared gas sensor based on radial basis function network temperature compensation and detection method, namely based on gas concentration detection method and a kind of infrared gas sensor based on radial basis temperature compensation of radial basis function neural network temperature compensation.Not only make infrared gas sensor can eliminate because of testing environment temperature variation produce measuring error, and compared to existing empirical formula method, because radial basis function network has, training speed is very fast, structure adaptive is determined, exports and feature that initial weight has nothing to do, makes that the process of temperature compensation is easier and applicability is wider; Compared to existing temperature control method, because the chip being embedded with radial basis function neural network algorithm only need be added, avoid the increase of infrared gas sensor volume, be more conducive to miniaturization and the low cost of sensor.
In order to overcome deficiency of the prior art, the invention provides a kind of based on the infrared gas sensor of radial basis function network temperature compensation and the solution of detection method, specific as follows:
A kind of infrared gas sensor based on radial basis function network temperature compensation, comprise infrared gas sensor, temperature sensor, embedded in the microprocessor system of radial basis function neural network algorithm, described infrared sensor adopts single light source bifocal path structure, described infrared sensor comprises air chamber, described air chamber top is provided with air intake opening 1 and gas outlet 2, air chamber inwall is provided with gold-plated reflectance coating 3, plenum interior two ends are respectively arranged with infrared wide spectrum light source 4 and binary channels pyroelectric detector 9, and infrared wide spectrum light source 4 is positioned at the focus place of the oval snoot 5 being arranged on plenum interior, in addition, binary channels pyroelectric detector 9 and electrical modulation wide range infrared light supply 4 are positioned on same reference field, bottom in air chamber is provided with temperature sensor 8, described infrared gas sensor, temperature sensor is connected with the microprocessor system 13 that embedded in radial basis function neural network algorithm, the described microprocessor system 13 that embedded in radial basis function neural network algorithm is also connected with display unit 15.
Described infrared wide spectrum light source 4 is the infrared wide spectrum light source of electrical modulation.
Gas concentration detection method based on radial basis function neural network temperature compensation mainly comprises the following steps:
(1) the binary channels pyroelectric detector Measurement channel of infrared gas sensor and the output voltage of reference channel and the output voltage of temperature sensor are normalized;
(2) data obtained after normalized in step (1) are sent into the input vector X=(x of radial basis function neural network input layer
1, x
2, x
3), wherein, x
1for the data of Measurement channel output voltage after normalized of binary channels pyroelectric detector; x
2for the data of reference channel output voltage after normalized of binary channels pyroelectric detector; x
3for the data of output voltage after normalized of temperature sensor;
(3) excitation function of Gaussian function as radial basis function neural network in the present invention is chosen, namely
wherein, || dist|| is the Euclidean distance between input vector and weight vector, is obtained by the product of the row vector of input vector and weighting matrix; δ is the variance of Gaussian function; || x
p-c
i|| be Euclidean Norm, x
pbe p input amendment, c
ifor the center of radial basis function neural network hidden layer node;
(4) being input as by step (3) known radial basis function neural network hidden layer
i, j ∈ [1,3], wherein, w1
jifor the weights that each neuron of hidden layer is connected with input layer; b
ifor threshold value; And the output of radial basis function neural network hidden layer is
wherein, || w1
i-X|| is the distance between weight vector and threshold vector;
(5) by the output of the known radial basis function neural network output layer of step (4) be
Wherein, w2
ifor the weights that each neuron of hidden layer is connected with output layer;
(6) train from 0 neuron, by checking that output error makes network automatically increase neuron, continuous circuit training, input vector corresponding to the maximum error that network is produced is as weight w 1, produce a new hidden layer neuron, then check the error of new network, repeat this process until reach error requirements or maximum hidden layer neuron number;
(7) choose N group sample data, carry out the data prediction of radial basis function neural network, establishment, training according to step (1) to step (6), obtain best weight value w and threshold value b
i, thus determine this network model;
(8) when actual measurement gas concentration, binary channels pyroelectric detector Measurement channel and the output voltage of reference channel and the output voltage of temperature sensor are normalized in the fixed network model of rear feeding step (7) by the infrared gas sensor based on radial basis function network temperature compensation, obtain the gas concentration information to be measured through temperature compensation by the prediction of radial basis function Function Neural Network.
Advantage of the present invention is: compared with prior art, overcome the deficiency of existing infrared gas sensor in temperature compensation, compensate for empirical formula method demarcates complicated, poor for applicability and temperature control method is unfavorable for the shortcoming of miniaturization, by the binary channels pyroelectric detector Measurement channel of infrared gas sensor and the output voltage of reference channel and the output voltage of temperature sensor after normalized, send into the input layer of radial basis function neural network, through the establishment of network, training, after prediction, the gas concentration information through temperature compensation is obtained by output layer.Make that the process of temperature compensation is easier, applicability is wider, be beneficial to the advantages such as miniaturization.
Accompanying drawing explanation
Fig. 1 is the structural representation of infrared gas sensor;
Fig. 2 is the structural representation of radial basis function neural network;
Fig. 3 is the infrared gas sensor workflow schematic diagram based on radial basis function network temperature compensation;
Wherein, 1 is air intake opening; 2 is gas outlet; 3 is gold-plated reflectance coating; 4 is electrical modulation wide range infrared light supply; 5 is oval snoot; 6 is optical path; 7 is reference light paths; 8 is temperature sensor; 9 is binary channels pyroelectric detector; I is the Measurement channel of binary channels pyroelectric detector; II is the reference channel of binary channels pyroelectric detector; 10 is the output voltage of binary channels pyroelectric detector Measurement channel; 11 is the output voltage of binary channels pyroelectric detector reference channel; 12 is the output voltage of temperature sensor; 13 is microprocessor system; 14 is radial basis function neural network algorithm; 15 is display unit.
Embodiment
Below in conjunction with drawings and Examples, summary of the invention is described further:
With reference to shown in Fig. 1, Fig. 2, Fig. 3, based on the infrared gas sensor of radial basis function network temperature compensation, comprise infrared gas sensor, temperature sensor, embedded in the microprocessor system of radial basis function neural network algorithm, described infrared sensor adopts single light source bifocal path structure, described infrared sensor comprises air chamber, described air chamber top is provided with air intake opening 1 and gas outlet 2, and mixed gas to be measured enters air chamber from air intake opening 1, discharges after whole air chamber from gas outlet 2.Air chamber inwall is provided with gold-plated reflectance coating 3, in order to strengthen the reflection of infrared radiation.Plenum interior two ends are respectively arranged with infrared wide spectrum light source 4 and binary channels pyroelectric detector 9, and infrared wide spectrum light source 4 is positioned at the focus place of the oval snoot 5 being arranged on plenum interior, in order to strengthen the light-gathering of infrared radiation.In addition, binary channels pyroelectric detector 9 and electrical modulation wide range infrared light supply 4 are positioned on same reference field, and the bottom in air chamber is provided with temperature sensor 8, in order to obtain the temperature value of testing environment; Described infrared gas sensor, temperature sensor are connected with the microprocessor system 13 that embedded in radial basis function neural network algorithm, and the described microprocessor system 13 that embedded in radial basis function neural network algorithm is also connected with display unit 15.
Described infrared wide spectrum light source 4 is the infrared wide spectrum light source of electrical modulation.
The infrared radiation that electrical modulation wide range infrared light supply 4 sends, after mixed gas to be measured absorbs, arrives the Measurement channel I of binary channels pyroelectric detector and the reference channel II of binary channels pyroelectric detector respectively.Object gas does not produce absorption to infrared radiation in reference light paths 7, in optical path 6, only absorb the infrared radiation of corresponding spectral coverage.Accordingly, background gas does not all produce infrared radiation or only there is very little absorption in reference light paths 7 and optical path 6.The infrared intensity information recorded by reference channel II can be derived from the original value of measurement environment, and the difference of the measured value of reference channel II and Measurement channel I, namely be the uptake of object gas to corresponding spectral coverage infrared radiation, this differential measuring method eliminates the interference because electrical modulation wide range infrared light supply fluctuation problem brings to a certain extent, improves sensitivity and the reliability of infrared gas sensor.
By the binary channels pyroelectric detector Measurement channel of infrared gas sensor and the output voltage of reference channel and the output voltage of temperature sensor after normalized, send into the input layer of radial basis function neural network, after the establishment of network, training, prediction, obtain the gas concentration information to be measured through temperature compensation by output layer.
By the output voltage 10v of the binary channels pyroelectric detector Measurement channel of infrared gas sensor
1, binary channels pyroelectric detector reference channel output voltage 11v
2, the output voltage 12v of temperature sensor
3be designated as one group of input data, corresponding gas concentration c to be measured is designated as one group and exports data.
Gas concentration detection method based on radial basis function neural network temperature compensation mainly comprises the following steps:
(1) N (N >=20) is organized input amendment data and output sample data feeding microprocessor system 13, microprocessor system 13 is embedded with radial basis function neural network algorithm 14.First the output voltage 10 of binary channels pyroelectric detector Measurement channel in sample data and the output voltage 11 of binary channels pyroelectric detector reference channel and the output voltage 12 of temperature sensor are normalized by microprocessor system 13, then the data after normalized are sent into the input layer of radial basis function neural network, N is positive integer; Be normalized by the binary channels pyroelectric detector Measurement channel of infrared gas sensor and the output voltage of reference channel and the output voltage of temperature sensor;
(2) now, the input vector of radial basis function neural network input layer is X=(x
1, x
2, x
3), the data obtained after normalized in step (1) are sent into the input vector X=(x of radial basis function neural network input layer
1, x
2, x
3), wherein, x
1for the data of Measurement channel output voltage after normalized of binary channels pyroelectric detector; x
2for the data of reference channel output voltage after normalized of binary channels pyroelectric detector; x
3for the data of output voltage after normalized of temperature sensor;
(3) excitation function of Gaussian function as radial basis function neural network in the present invention is chosen, namely
wherein, || dist|| is the Euclidean distance between input vector and weight vector, is obtained by the product of the row vector of input vector and weighting matrix; δ is the variance of Gaussian function; || x
p-c
i|| be Euclidean Norm, x
pbe p input amendment, c
ifor the center of radial basis function neural network hidden layer node;
(4) being input as by step (3) known radial basis function neural network hidden layer
i, j ∈ [1,3], wherein, w1
jifor the weights that each neuron of hidden layer is connected with input layer; b
ifor threshold value; And the output of radial basis function neural network hidden layer is
wherein, || w1
i-X|| is the distance between weight vector and threshold vector;
(5) by the output of the known radial basis function neural network output layer of step (4) be
Wherein, w2
ifor the weights that each neuron of hidden layer is connected with output layer;
(6) the hidden layer neuron number choosing radial basis function neural network is 3, train from 0 neuron, by checking that output error makes network automatically increase neuron, continuous circuit training, input vector corresponding to the maximum error that network is produced is as weight w 1, produce a new hidden layer neuron, then check the error of new network, repeat this process until reach error requirements or maximum hidden layer neuron number;
(7) choose N group sample data, carry out the data prediction of radial basis function neural network, establishment, training according to step (1) to step (6), obtain best weight value w and threshold value b
i, thus determine this network model;
(8) when actual measurement gas concentration, binary channels pyroelectric detector Measurement channel and the output voltage of reference channel and the output voltage of temperature sensor are normalized in the fixed network model of rear feeding step (7) by the infrared gas sensor based on radial basis function network temperature compensation, obtain the gas concentration information to be measured through temperature compensation by the prediction of radial basis function Function Neural Network.
Based in the infrared gas sensor actual measurement of radial basis temperature compensation, infrared gas sensor and temperature sensor can export one group of measurement data, these group data are normalized by microprocessor system 13, as the input vector of radial basis function neural network input layer, after the prediction processing of radial basis function neural network algorithm 14, output layer just obtains the gas concentration information after temperature compensation.This gas concentration information carries out renormalization process through microprocessor system 13, is finally exported by display unit 15.
In the empirical formula method used at present, first obtain the output model of infrared gas sensor
wherein, u
1for the output voltage of Measurement channel; u
2for the output voltage of reference channel; C is target gas levels; a
0for the output voltage u of Measurement channel when c equals zero
10with the output voltage u of reference channel
20ratio, namely
a
sfor constant relevant with gas molar absorption coefficient with air chamber length when c equals full scale, namely
then this output model is carried out to the correction of temperature compensation, comprise zero temperature compensation correction and range temperature compensation correction.In zero temperature compensation correction, adopt experimental formula
wherein, k
0for zero temperature compensation coefficient; a
tfor the constant at zero point after temperature compensation; T
tfor testing environment temperature; T
0for demarcating temperature.In range temperature compensation correction, adopt experimental formula
wherein, k
sfor range temperature compensation coefficient.
Finally obtaining gas concentration detection model by above-mentioned output model and temperature compensation experimental formula is
and temperature compensation coefficient wherein needs to carry out iteration according to sample data and determines.
In the present invention, by the output voltage u of Measurement channel
1, reference channel output voltage u
2, testing environment temperature T
tas the input vector of radial basis function neural network input layer, through the training of many group sample datas, obtain final network model, determine the output voltage u of Measurement channel
1, reference channel output voltage u
2, testing environment temperature T
tand the mapping relations between target gas levels c.This invention simplifies the process of temperature compensation, the demarcation of each temperature compensation coefficient need not be carried out, and only carrying out zero temperature compensation and range temperature compensation compared to empirical formula method, the present invention has carried out temperature compensation in whole sensing range, has wider applicability.
The above, it is only preferred embodiment of the present invention, not any pro forma restriction is done to the present invention, although the present invention discloses as above with preferred embodiment, but and be not used to limit the present invention, any those skilled in the art, do not departing within the scope of technical solution of the present invention, make a little change when the technology contents of above-mentioned announcement can be utilized or be modified to the Equivalent embodiments of equivalent variations, in every case be do not depart from technical solution of the present invention content, according to technical spirit of the present invention, within the spirit and principles in the present invention, to any simple amendment that above embodiment is done, equivalent replacement and improvement etc., within the protection domain all still belonging to technical solution of the present invention.
Claims (3)
1. the infrared gas sensor based on radial basis function network temperature compensation, it is characterized in that comprising infrared gas sensor, temperature sensor, embedded in the microprocessor system of radial basis function neural network algorithm, described infrared sensor adopts single light source bifocal path structure, described infrared sensor comprises air chamber, described air chamber top is provided with air intake opening and gas outlet, air chamber inwall is provided with gold-plated reflectance coating, plenum interior two ends are respectively arranged with infrared wide spectrum light source and binary channels pyroelectric detector, and infrared wide spectrum light source is positioned at the focus place of the oval snoot being arranged on plenum interior, in addition, binary channels pyroelectric detector and electrical modulation wide range infrared light supply are positioned on same reference field, bottom in air chamber is provided with temperature sensor, described infrared gas sensor, temperature sensor is connected with the microprocessor system that embedded in radial basis function neural network algorithm, the described microprocessor system that embedded in radial basis function neural network algorithm is also connected with display unit.
2. the infrared gas sensor based on radial basis function network temperature compensation according to claim 1, is characterized in that described infrared wide spectrum light source is the infrared wide spectrum light source of electrical modulation.
3. the detection method of the infrared gas sensor based on radial basis function network temperature compensation according to claim 2, is characterized in that, comprise the following steps:
(1) the binary channels pyroelectric detector Measurement channel of infrared gas sensor and the output voltage of reference channel and the output voltage of temperature sensor are normalized;
(2) data obtained after normalized in step (1) are sent into the input vector X=(x of radial basis function neural network input layer
1, x
2, x
3), wherein, x
1for the data of Measurement channel output voltage after normalized of binary channels pyroelectric detector; x
2for the data of reference channel output voltage after normalized of binary channels pyroelectric detector; x
3for the data of output voltage after normalized of temperature sensor;
(3) excitation function of Gaussian function as radial basis function neural network in the present invention is chosen, namely
wherein, || dist|| is the Euclidean distance between input vector and weight vector, is obtained by the product of the row vector of input vector and weighting matrix; δ is the variance of Gaussian function; || x
p-c
i|| be Euclidean Norm, x
pbe p input amendment, c
ifor the center of radial basis function neural network hidden layer node;
(4) being input as by step (3) known radial basis function neural network hidden layer
i, j ∈ [1,3], wherein, w1
jifor the weights that each neuron of hidden layer is connected with input layer; b
ifor threshold value; And the output of radial basis function neural network hidden layer is
wherein, || w1
i-X|| is the distance between weight vector and threshold vector;
(5) by the output of the known radial basis function neural network output layer of step (4) be
Wherein, w2
ifor the weights that each neuron of hidden layer is connected with output layer;
(6) train from 0 neuron, by checking that output error makes network automatically increase neuron, continuous circuit training, input vector corresponding to the maximum error that network is produced is as weight w 1, produce a new hidden layer neuron, then check the error of new network, repeat this process until reach error requirements or maximum hidden layer neuron number;
(7) choose N group sample data, carry out the data prediction of radial basis function neural network, establishment, training according to step (1) to step (6), obtain best weight value w and threshold value b
i, thus determine this network model;
(8) when actual measurement gas concentration, binary channels pyroelectric detector Measurement channel and the output voltage of reference channel and the output voltage of temperature sensor are normalized in the fixed network model of rear feeding step (7) by the infrared gas sensor based on radial basis function network temperature compensation, obtain the gas concentration information to be measured through temperature compensation by the prediction of radial basis function Function Neural Network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510847284.2A CN105352907A (en) | 2015-11-27 | 2015-11-27 | Infrared gas sensor based on radial basis network temperature compensation and detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510847284.2A CN105352907A (en) | 2015-11-27 | 2015-11-27 | Infrared gas sensor based on radial basis network temperature compensation and detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105352907A true CN105352907A (en) | 2016-02-24 |
Family
ID=55328913
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510847284.2A Pending CN105352907A (en) | 2015-11-27 | 2015-11-27 | Infrared gas sensor based on radial basis network temperature compensation and detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105352907A (en) |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106153567A (en) * | 2016-06-22 | 2016-11-23 | 南京信息工程大学 | Based on the pressure compensated infrared gas sensor of BP neutral net and detection method |
| CN106990769A (en) * | 2017-06-08 | 2017-07-28 | 辽宁省农业科学院 | A kind of greenhouse comprehensive test instrument and method |
| CN107478552A (en) * | 2016-06-07 | 2017-12-15 | 宁波方太厨具有限公司 | Oil smoke concentration sensor and its oil smoke concentration detection apparatus and detection method |
| CN107490556A (en) * | 2017-04-19 | 2017-12-19 | 安徽华脉科技发展有限公司 | A kind of infrared combustable gas concentration detecting system |
| CN108416423A (en) * | 2017-02-10 | 2018-08-17 | 三星电子株式会社 | Automatic threshold for neural network trimming and retraining |
| CN109709058A (en) * | 2018-12-20 | 2019-05-03 | 南京信息工程大学 | Infrared gas sensor and humidity compensation method based on AGA-Elman algorithm |
| CN110009874A (en) * | 2019-04-12 | 2019-07-12 | 泰州职业技术学院 | A kind of flammable gas alarm, combustible gas alarm method and system |
| CN110023721A (en) * | 2016-11-29 | 2019-07-16 | 森西欧有限公司 | Integrated circuit sensor encapsulation and its manufacturing method |
| CN110057773A (en) * | 2019-05-05 | 2019-07-26 | 南京信息工程大学 | A kind of infrared gas sensor system and the temperature-compensation method based on AACA-Elman algorithm |
| CN110618102A (en) * | 2019-09-25 | 2019-12-27 | 成都太阳神鸟智能控制有限公司 | Gas detection method and device based on dispersion spectrum analysis and artificial intelligence |
| CN111948273A (en) * | 2020-08-21 | 2020-11-17 | 山东工商学院 | Gas sensor temperature compensation model establishment method and device and gas concentration measurement method and device |
| CN112683836A (en) * | 2021-01-12 | 2021-04-20 | 杭州麦乐克科技股份有限公司 | Calibration method and system of carbon dioxide sensor based on BP neural network |
| CN112881246A (en) * | 2021-01-26 | 2021-06-01 | 苏州胤煌精密仪器科技有限公司 | Image method particle size analyzer capable of continuously changing magnification |
| CN113888841A (en) * | 2021-12-08 | 2022-01-04 | 成都千嘉科技有限公司 | Gas alarm system |
| CN117288711A (en) * | 2023-11-24 | 2023-12-26 | 四方光电股份有限公司 | Leakage monitoring device and monitoring method for power battery |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE20203759U1 (en) * | 2002-03-08 | 2002-07-18 | Steinel Gmbh & Co Kg | Device for gas measurement |
| JP2009139135A (en) * | 2007-12-04 | 2009-06-25 | Dkk Toa Corp | Infrared absorption-type gas analyzer |
| CN101567120A (en) * | 2009-04-27 | 2009-10-28 | 中北大学 | Portable infrared gas alarm apparatus |
| CN103017991B (en) * | 2012-11-26 | 2015-05-06 | 中国科学技术大学 | Suction type SF6 gas leakage monitoring method |
-
2015
- 2015-11-27 CN CN201510847284.2A patent/CN105352907A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE20203759U1 (en) * | 2002-03-08 | 2002-07-18 | Steinel Gmbh & Co Kg | Device for gas measurement |
| JP2009139135A (en) * | 2007-12-04 | 2009-06-25 | Dkk Toa Corp | Infrared absorption-type gas analyzer |
| CN101567120A (en) * | 2009-04-27 | 2009-10-28 | 中北大学 | Portable infrared gas alarm apparatus |
| CN103017991B (en) * | 2012-11-26 | 2015-05-06 | 中国科学技术大学 | Suction type SF6 gas leakage monitoring method |
Non-Patent Citations (8)
| Title |
|---|
| 于洋编著: "《在线分析仪器》", 31 August 2015, 清华大学出版社 * |
| 何平等: "基于RBF网络的智能气敏传感器温度补偿", 《仪表技术与传感器》 * |
| 史健芳等: "用BPF神经网络改善传感器输出特性", 《微计算机信息》 * |
| 孙继平等编著: "《煤矿安全生产综合监控》", 31 July 2008, 中国矿业大学出版社 * |
| 张晋疆等: "温度补偿型双通道甲烷测试仪的设计", 《计算机测量与控制》 * |
| 张良均等编: "《神经网络实用教程》", 29 February 2008, 机械工业出版社 * |
| 杨应平等编著: "《光电技术》", 31 August 2014, 机械工业出版社 * |
| 谭秋林著: "《红外光学气体传感器及检测系统》", 31 December 2013, 机械工业出版社 * |
Cited By (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107478552B (en) * | 2016-06-07 | 2023-09-15 | 宁波方太厨具有限公司 | Oil smoke concentration sensor and oil smoke concentration detection device and detection method thereof |
| CN107478552A (en) * | 2016-06-07 | 2017-12-15 | 宁波方太厨具有限公司 | Oil smoke concentration sensor and its oil smoke concentration detection apparatus and detection method |
| CN106153567B (en) * | 2016-06-22 | 2019-07-23 | 南京信息工程大学 | Based on the pressure compensated infrared gas sensor of BP neural network and detection method |
| CN106153567A (en) * | 2016-06-22 | 2016-11-23 | 南京信息工程大学 | Based on the pressure compensated infrared gas sensor of BP neutral net and detection method |
| CN110023721A (en) * | 2016-11-29 | 2019-07-16 | 森西欧有限公司 | Integrated circuit sensor encapsulation and its manufacturing method |
| CN108416423A (en) * | 2017-02-10 | 2018-08-17 | 三星电子株式会社 | Automatic threshold for neural network trimming and retraining |
| CN108416423B (en) * | 2017-02-10 | 2024-01-12 | 三星电子株式会社 | Automatic threshold for neural network pruning and retraining |
| CN107490556A (en) * | 2017-04-19 | 2017-12-19 | 安徽华脉科技发展有限公司 | A kind of infrared combustable gas concentration detecting system |
| CN106990769A (en) * | 2017-06-08 | 2017-07-28 | 辽宁省农业科学院 | A kind of greenhouse comprehensive test instrument and method |
| CN109709058A (en) * | 2018-12-20 | 2019-05-03 | 南京信息工程大学 | Infrared gas sensor and humidity compensation method based on AGA-Elman algorithm |
| CN110009874A (en) * | 2019-04-12 | 2019-07-12 | 泰州职业技术学院 | A kind of flammable gas alarm, combustible gas alarm method and system |
| CN110057773A (en) * | 2019-05-05 | 2019-07-26 | 南京信息工程大学 | A kind of infrared gas sensor system and the temperature-compensation method based on AACA-Elman algorithm |
| CN110618102A (en) * | 2019-09-25 | 2019-12-27 | 成都太阳神鸟智能控制有限公司 | Gas detection method and device based on dispersion spectrum analysis and artificial intelligence |
| CN111948273B (en) * | 2020-08-21 | 2023-08-08 | 山东工商学院 | Method and device for establishing temperature compensation model of gas sensor and measuring gas concentration |
| CN111948273A (en) * | 2020-08-21 | 2020-11-17 | 山东工商学院 | Gas sensor temperature compensation model establishment method and device and gas concentration measurement method and device |
| CN112683836B (en) * | 2021-01-12 | 2022-11-01 | 杭州麦乐克科技股份有限公司 | Calibration method and system of carbon dioxide sensor based on BP neural network |
| CN112683836A (en) * | 2021-01-12 | 2021-04-20 | 杭州麦乐克科技股份有限公司 | Calibration method and system of carbon dioxide sensor based on BP neural network |
| CN112881246A (en) * | 2021-01-26 | 2021-06-01 | 苏州胤煌精密仪器科技有限公司 | Image method particle size analyzer capable of continuously changing magnification |
| CN113888841A (en) * | 2021-12-08 | 2022-01-04 | 成都千嘉科技有限公司 | Gas alarm system |
| CN113888841B (en) * | 2021-12-08 | 2022-03-11 | 成都千嘉科技股份有限公司 | Gas alarm system |
| CN117288711A (en) * | 2023-11-24 | 2023-12-26 | 四方光电股份有限公司 | Leakage monitoring device and monitoring method for power battery |
| CN117288711B (en) * | 2023-11-24 | 2024-03-19 | 四方光电股份有限公司 | Leakage monitoring device and monitoring method for power battery |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105352907A (en) | Infrared gas sensor based on radial basis network temperature compensation and detection method | |
| CN104364635B (en) | The collision broadening verified using real-time or near real-time in spectroanalysis instrument is compensated | |
| Qiu | A simple optical model to estimate suspended particulate matter in Yellow River Estuary | |
| CN204924934U (en) | Multicomponent gas is detection device simultaneously based on two kinds of quantum cascade laser spectrum | |
| CN205808925U (en) | A kind of near-infrared crude oil water content detection device | |
| CN103328954A (en) | Temperature calibration methods and apparatus for optical absorption gas sensors, and optical absorption gas sensors thereby calibrated | |
| CN104833645B (en) | Gas sensor with adjustable range, sensing system and sensing method | |
| CN202974860U (en) | High-precision infrared gas detection module | |
| CN101435773A (en) | Gas monitoring method and apparatus based on quasi continuous diode laser modulated spectrum | |
| CN103698298A (en) | Device for measuring gas concentration based on short cavity enhanced correlation spectroscopy, and method for measuring gas concentration by adopting device | |
| CN104020114A (en) | Method for analyzing trace concentration of ammonia gas | |
| Shen et al. | Methane near-infrared laser remote detection under non-cooperative target condition based on harmonic waveform recognition | |
| CN104730021A (en) | Method for calibrating non-dispersive infrared gas sensor | |
| CN105319178A (en) | Real-time detection system of C0 and CO2 concentration in motor vehicle tail gas and control method of real-time detection system | |
| CN104062265A (en) | Detection device and detection method for multi-component gas in transformer oil based on spectrum analysis | |
| CN102507507B (en) | Device and method for detecting concentration of gas to be detected through temperature correction | |
| CN102928390A (en) | On-line detection method and device for chlorophyll concentration in water body based on two detectors | |
| Vita et al. | Development of a portable active long-path differential optical absorption spectroscopy system for volcanic gas measurements | |
| CN111521581B (en) | Method and device for judging components of carbon monoxide and methane and detecting concentration of components of carbon monoxide and methane and application of method and device | |
| CN104267019A (en) | Gas concentration calibrating and detecting method for laser Raman gas analyzer | |
| CN102944546A (en) | Drilling fluid analysis device and method | |
| CN101949825B (en) | Leaf water near infrared non-destructive testing device and method in light open environment | |
| CN101943658A (en) | Method for continuously monitoring water quality by combining principal component analysis theory and spectrum analysis technology | |
| CN111220570A (en) | Infrared multi-gas detection system and gas detection method | |
| CN104091864A (en) | Multi-wavelength near-infrared LED manufacturing method and gas detection method based on PbSe quantum dots |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160224 |