CN201382934Y - Gas detection sensor device - Google Patents
Gas detection sensor device Download PDFInfo
- Publication number
- CN201382934Y CN201382934Y CN200920107385U CN200920107385U CN201382934Y CN 201382934 Y CN201382934 Y CN 201382934Y CN 200920107385 U CN200920107385 U CN 200920107385U CN 200920107385 U CN200920107385 U CN 200920107385U CN 201382934 Y CN201382934 Y CN 201382934Y
- Authority
- CN
- China
- Prior art keywords
- gas
- unit
- information
- module
- pattern recognition
- 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.)
- Expired - Lifetime
Links
Images
Landscapes
- Investigating Or Analyzing Materials By The Use Of Fluid Adsorption Or Reactions (AREA)
Abstract
The utility model provides a gas detection sensor device, which comprises a gas sensing sensor array, a signal pretreatment module and a pattern recognition module, wherein the gas sensing sensor array is used to collect gas information, the signal pretreatment module treats the information in advance, and the pattern recognition module is electrically connected with the signal pretreatment module. The pattern recognition module comprises a first unit, a second unit, a third unit, a forth unit and a decision unit which are orderly connected, wherein the first unit builds an information database for detected gas, the second unit treats gas information, the third unit extracts gas information characteristics, and the forth unit classifies the gas information characteristics.
Description
Technical field
The utility model relates to a kind of sensing device, particularly relates to a kind of sensing device that is used for detection of gas.
Background technology
The burning that can be used as fire alarm that burning produces discharges gas and mainly contains CO, CO2, NOx, CH4, H2, H2O, amine (NH2) etc., can mainly contain as the gas sensor of surveying inflammable gas or fire burning release gas at present: the semiconductor gas bulk detector, infrared absorption type gas sensor, electrochemical sensor etc. mainly are temperature-sensitive, sense cigarette, sense flame detector etc.But these recognition modes are difficult to find reliably can not detect Alcohol Flame as smoke detector by incipient fire, and heat detector is difficult for detecting inflammable burning things which may cause a fire disaster.Owing to, cost an arm and a leg, so the fire gases Detection Techniques are in the condition that is unable to reach always to the reliable detection technology comparison temperature rise of specific gas and the detection technique complexity of flue dust.
The utility model content
At the defective of prior art, the purpose of this utility model is to propose a kind of detection of gas sensing device, can accurately survey the kind and the concentration of tested gas.
To achieve these goals, the utility model proposes a kind of detection of gas sensing device, it is characterized in that, comprise: the gas sensor array that is used to gather gas information, above-mentioned information is carried out pretreated signal pre-processing module, the pattern recognition module that is electrically connected with described signal pre-processing module;
Described pattern recognition module comprises successively and to connect: set up detected gas information database first module, handle gas information Unit second, extract Unit the 3rd of gas information characteristics, with the Unit the 4th and the identifying unit of above-mentioned gas information characteristics classification.
Preferred as technique scheme, described gas sensor display is the resistance-type gas sensor array according to gas with various sputter gas with various sensitive membrane on each sensor base.
Preferred as technique scheme is characterized in that also comprise: described identification module is connected with PC by data collecting card.
The utility model provides a kind of Intelligent Hybrid detection of gas sensing device, and whether this sensing device can not only be measured measured object is inflammable gas, and can distinguish dangerous gas concentration or non-hazardous gas concentration easily.
Below in conjunction with accompanying drawing, embodiment of the present utility model is described in further detail, for the person of ordinary skill in the field, to the detailed description of the present utility model, above-mentioned and other purposes of the present utility model, feature and advantage will be apparent.
Description of drawings
Fig. 1 is the composition frame chart of a kind of detection of gas sensing device of the present utility model;
Fig. 2 is a flow chart of data processing synoptic diagram in the utility model;
Fig. 3 is the structural drawing of a responsive unit of gas sensor of the present utility model;
Fig. 4 is a sensor array data fusion process synoptic diagram of the present utility model;
Fig. 5 is the BP neural network structure of fire gases identification method based of the present utility model.
Embodiment
Below in conjunction with accompanying drawing the utility model is described further.
Be illustrated in figure 1 as the composition frame chart of a kind of detection of gas sensing device of the utility model one preferred embodiment, comprise: the gas sensor array 1 that is used to gather gas information, above-mentioned information is carried out pretreated signal pre-processing module 2, the pattern recognition module 3 that is electrically connected with described signal pre-processing module; Described pattern recognition module comprises successively and to connect: set up detected gas information database first module 4, handle gas information second unit 5, extract the 3rd unit 6 of gas information characteristics, with the 4th unit 7 and the identifying unit 8 of above-mentioned gas information characteristics classification.
As shown in Figure 2, the multidimensional response space that above-mentioned gas sensor display 1 constitutes forms response modes to tested gas, 2 pairs of above-mentioned response modes of signal pre-processing module carry out pre-service then, 3 pairs of pretreated response modes of final mode identification module are handled and are analyzed, thus the kind and the concentration of identification gas.
Gas sensor array 1 is by a plurality of gas sensor forming arrays, and adopts integrated technique to be made, and this array volume is little, low in energy consumption, is convenient to the concentrated collection and the processing of signal.Gas sensor can become electric signal with detected gaseous species and with concentration dependent information translation.Gas sensor array is except the response of each gas sensor, in the hyperspace that all gas sensor is formed, form response modes, the kind of this response modes and tested gas, concentration etc. exist corresponding relation, and this just gas sensor array can debate the key point of knowledge to multiple gases.
A responsive unit of gas sensor as shown in Figure 3 comprises gas sensitization film 21, substrate 22, interdigital electrode 23, heating resistor 24.The sensitive membrane resistance value is linear with the gas concentration that band is analyzed.According to this characteristic, can learn the concentration of adsorbed gas from the variation of resistance.When heating resistor 24 is heated to steady state (SS), on gas to be detected contact gas sensitization film 21 surfaces and when being adsorbed, the molecule that is adsorbed at first freely spreads in object surfaces, lose kinergety, the part molecule is evaporated, the residual molecule of another part produce thermal decomposition and chemisorption nearby.When the work function of gas sensitization film 21 dissociation energy greater than absorbing molecules, absorbing molecules will discharge electronics, form positive ion absorption.Gas with tendency of positive ion absorption has H2, CO, hydrocarbon and alcohols.Owing to is constant on the airborne oxygen content, during resistance also fix.If when gas concentration changed, corresponding variation will take place in its resistance.According to this characteristic, can learn the concentration of adsorbed gas from the variation of resistance.
The response modes of 3 pairs of gas sensor array 1 of signal pre-processing module carries out pre-service: for example adopt wavelet transform (DWT) mode.
Be admitted to pattern recognition module 3, the kind of the gas that is detected to identify and concentration etc. through pretreated response modes information.
As shown in Figure 4, from the curve of each sensor and gas response, extract several features at first respectively, and determine with resolution index whether the characteristic parameter that is extracted is optimum.Secondly characteristic is merged, because characteristic is not The more the better, this is because the complicacy of sorter increases very large soon along with the dimension of model space changes, and is easy to generate shot array.From the performance reduction dimension of sorter also is useful, the design of recognition system generally is based on limited learning sample training classifier, the recognition performance of sorter at dimension more after a little while, the adding of new feature will improve the performance of sorter, but after having arrived to a certain degree, add the degeneration that more eigenwert will cause performance.Feature extraction exists great difference for different problems on its method.The principle of feature extraction is based on the problem of being asked, and should make the maximum information of feature statement of extraction.Extract the requirement that 8 features just can well reach our discriminator in the curve of definite at last gas sensor array and gas response.
Adopt artificial neural network technology in the pattern recognition module 3.By designing three layers of backpropagation (BP) artificial nerve network model, use the BP algorithm of adaptive learning speed, designed fire gases identification method based BP network is trained.Various types of fire all just produce a certain amount of CO2, CO and HCl gas in early days.The amount of CO2 is fixed in the air, and do not contain CO and HCl gas in the air, thus can select CO2, CO and HCl as fire sign property gas, by the fire sign property gas that monitors is carried out identification, carry out " fire " and " non-fire " and differentiate, realize early fire detection.
In Fig. 5, the output layer of BP network has 2 neurons, for the classification of unstructuredness problem, and the functional objective that on behalf of neural network, output neuron must can realize.Therefore, adopt " fire ", " non-fire " state as output neuron.The output valve of output neuron also is the continuous variable of normalization between [0,1].Our used neural network hidden layer number is i, and the neuron number of hidden layer is j.3 * j root line is arranged between input layer and the hidden layer, and the weights between input layer i and the hidden neuron j are W1ij; J * 2 line is arranged between hidden layer and the output layer, and the weights between hidden neuron j and the output layer neuron are W2ij.
In another embodiment of the present utility model, the gas sensor display is the resistance-type gas sensor array according to gas with various sputter gas with various sensitive membrane on each sensor base.
In another embodiment of the present utility model, identification module is connected with PC by data collecting card, sends information datas such as the kind of the gas discerned and concentration thereof to PC, and PC can further be analyzed data.
Certainly; the utility model also can have other embodiment; under the situation of spirit that does not deviate from the utility model and essence; the person of ordinary skill in the field works as can make various corresponding changes according to the utility model, but these corresponding changes all should belong to the protection domain of the utility model claim.
Claims (3)
1. a detection of gas sensing device is characterized in that, comprising: be used to gather the gas sensor array of gas information, above-mentioned information is carried out pretreated signal pre-processing module, the pattern recognition module that is electrically connected with described signal pre-processing module;
Described pattern recognition module comprises successively and to connect: set up detected gas information database first module, handle gas information Unit second, extract Unit the 3rd of gas information characteristics, with the Unit the 4th and the identifying unit of above-mentioned gas information characteristics classification.
2. a kind of detection of gas sensing device according to claim 1 is characterized in that, described gas sensor display is the resistance-type gas sensor array according to gas with various sputter gas with various sensitive membrane on each sensor base.
3. a kind of detection of gas sensing device according to claim 1 is characterized in that also comprise: described identification module is connected with PC by data collecting card.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200920107385U CN201382934Y (en) | 2009-04-15 | 2009-04-15 | Gas detection sensor device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200920107385U CN201382934Y (en) | 2009-04-15 | 2009-04-15 | Gas detection sensor device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN201382934Y true CN201382934Y (en) | 2010-01-13 |
Family
ID=41526406
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200920107385U Expired - Lifetime CN201382934Y (en) | 2009-04-15 | 2009-04-15 | Gas detection sensor device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN201382934Y (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108088876A (en) * | 2017-12-29 | 2018-05-29 | 扬州大学 | It is a kind of can thermometric air-sensitive film test device and its application method |
| CN108831121A (en) * | 2018-05-24 | 2018-11-16 | 歌尔股份有限公司 | The method for early warning and device of mine safety production |
| CN108956700A (en) * | 2018-09-18 | 2018-12-07 | 西安理工大学 | For detecting SF6Substitute the submicron film sensor of gas decomposition product |
-
2009
- 2009-04-15 CN CN200920107385U patent/CN201382934Y/en not_active Expired - Lifetime
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108088876A (en) * | 2017-12-29 | 2018-05-29 | 扬州大学 | It is a kind of can thermometric air-sensitive film test device and its application method |
| CN108831121A (en) * | 2018-05-24 | 2018-11-16 | 歌尔股份有限公司 | The method for early warning and device of mine safety production |
| CN108956700A (en) * | 2018-09-18 | 2018-12-07 | 西安理工大学 | For detecting SF6Substitute the submicron film sensor of gas decomposition product |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103617684B (en) | Interference-type optical fiber circumference vibrating intruding recognizer | |
| KR101852074B1 (en) | Electronic Nose System and Method for Gas Classification | |
| CN201408467Y (en) | Intelligent fire alarm system | |
| CN110319982B (en) | Buried gas pipeline leakage judgment method based on machine learning | |
| Ryan et al. | Polymer–carbon black composite sensors in an electronic nose for air-quality monitoring | |
| CN104318688B (en) | A kind of multisensor fire alarm method based on data fusion | |
| Wang et al. | A Sauter mean diameter sensor for fire smoke detection | |
| Rose-Pehrsson et al. | Multi-criteria fire detection systems using a probabilistic neural network | |
| Dong et al. | The gas leak detection based on a wireless monitoring system | |
| CN109165670A (en) | A kind of TS-RBF fuzzy neural network robust fusion algorithm applied to infra red flame identification | |
| Omatu et al. | E-nose system by using neural networks | |
| CN111613037B (en) | Method for reducing composite smoke sense false alarm based on intelligent algorithm | |
| CN110135466B (en) | Pollutant emission standard exceeding vehicle judgment method and system | |
| CN105975991B (en) | A kind of improved extreme learning machine fire kind recognition methods | |
| CN201382934Y (en) | Gas detection sensor device | |
| Chen et al. | Multi-sourced sensing and support vector machine classification for effective detection of fire hazard in early stage | |
| Zhang et al. | Research on electronic nose system based on continuous wide spectral gas sensing | |
| Fujinaka et al. | Intelligent electronic nose systems for fire detection systems based on neural networks | |
| Chen et al. | A mixed gas composition identification method based on sample augmentation | |
| Omatu et al. | Intelligent electronic nose system independent on odor concentration | |
| Wang et al. | A Research of Neural Network Optimization Technology for Apple Freshness Recognition Based on Gas Sensor Array | |
| Yu et al. | Power transformers fault diagnosis based on a meta-learning approach to kernel extreme learning machine with opposition-based learning sparrow search algorithm | |
| Di Lecce et al. | Discriminating gaseous emission patterns in low-cost sensor setups | |
| Ren et al. | Recognition of oil content for oil shale by electronic nose based on transient and steady-state fusion feature extraction method | |
| Wang et al. | Portable vehicular electronic nose system for detection of automobile exhaust |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Assignee: Hefei Jiadong Technology Co., Ltd. Assignor: Unioriental Optics Co., Ltd. Contract record no.: 2012340000173 Denomination of utility model: Gas detection sensor device Granted publication date: 20100113 License type: Exclusive License Record date: 20120528 |
|
| ASS | Succession or assignment of patent right |
Owner name: HEFEI UNIORIENTAL OPTICS CO., LTD. Free format text: FORMER OWNER: UNIORIENTAL OPTICS CO., LTD. Effective date: 20121224 |
|
| C41 | Transfer of patent application or patent right or utility model | ||
| COR | Change of bibliographic data |
Free format text: CORRECT: ADDRESS; FROM: 100027 CHAOYANG, BEIJING TO: 231202 HEFEI, ANHUI PROVINCE |
|
| TR01 | Transfer of patent right |
Effective date of registration: 20121224 Address after: 231202 Anhui city of Hefei Province Economic and Technological Development Zone Pearl Plaza Building 3 Patentee after: Hefei Jiadong Technology Co., Ltd. Address before: 100027 Beijing city Chaoyang District East Third Ring Road No. 2, room B908, Tianyuan port c Patentee before: Unioriental Optics Co., Ltd. |
|
| CX01 | Expiry of patent term |
Granted publication date: 20100113 |
|
| CX01 | Expiry of patent term |