CN103838210B - Remote wireless monitoring system and method for toxic gas in emergency site - Google Patents

Abstract

The application provides a long-range wireless monitoring system of scene toxic gas of emergency, includes: the system comprises at least one toxic gas wireless detection node, at least one information aggregation transfer node and at least one portable intelligent handheld terminal; the toxic gas wireless detection node is used for detecting first toxic gas information of an emergency scene; the information aggregation transfer node is used for aggregating the second toxic gas information obtained by each toxic gas wireless detection node and sending the second toxic gas information to the portable intelligent handheld terminal; the portable intelligent handheld terminal comprises a third microcontroller module, and the third microcontroller module is used for fusing the second toxic gas information according to the category of the toxic gas to obtain third toxic gas information. Therefore, the problems that toxic gas is quickly detected and corresponding effective countermeasures are taken in the emergency scene are solved.

Description

Accident scene toxic gas remote wireless monitoring system and method

Technical field

The application relates to Detection Technology and Automatic Equipment field, particularly relates to accident scene toxic gas long-range Wireless monitor system and method.

Background technology

In colliery, oil, chemical industry, fire-fighting, nuclear energy, the field such as military affairs, due to operational error, adverse circumstances, ageing equipment Etc. reason, it is susceptible to the poisonous or leakage of fuel gas.Once occur fuel gas to reveal, easily cause large area fire also Cause an explosion accident, and the extreme influence that health is usually caused by toxic gas leakage event, such as: historic India Bhopal toxic gas leakage case can be rated as disaster, and domestication factory occurs that the leakage of the gas such as chlorine, hydrogen sulfide causes casualties Event does not occupies the minority.

In recent years, people give more and more to pay close attention to for sudden toxic gas leakage, simultaneously terrified main in world wide Justice is rampant, and poison (viral) is likely to chemistry (biochemical) weapon used as terrorist, so the grinding of detection of poison gas technology It is imperative with exploitation to study carefully.Reliable fashion is being explored the most as possible to ensure country's people's property life by many countries of the world today Safety, the threat that farthest mankind are caused by reduction accident scene or terrorist incident.

For there is the measurement of the toxic gas in above-mentioned condition, analytical chemistry method, spectrum is currently mainly used to divide The methods such as analysis method, portable gas chromatography-mass spectrograph.Wherein, analytical chemistry method typically gathers sample gas in measure field, so After take back laboratory and carry out chemical analysis and obtain result, the method is limited by condition, it is impossible to carry out toxic gas quickly Detection；Spectroscopic analysis methods is it is generally required to special spectrogrph, apparatus expensive, operation are complicated, not Portable belt, and sampling analysis Speed is slow, it is impossible to realize on-line monitoring；Portable gas chromatography-mass spectrometric method detection toxic gas is the longest, and detection Scope is little.

For toxic gas leakage or the scene by biological attack, find when using above-mentioned existing method, phase The kind of toxic gas at accident scene, concentration cannot quickly be detected by pass personnel, simultaneously also cannot be for tool The situation at the accident scene of body takes effective counter-measure, makes the property of country and enterprise by huge loss, group Many life securities are also by significant damage.

Summary of the invention

This application provides a kind of accident scene toxic gas remote wireless monitoring system and method, to solve prominent Send out love scene and be quickly tested with poisonous gas the problem taking corresponding effective counter-measure.

In order to solve the problems referred to above, this application discloses one accident scene toxic gas long-distance wireless monitoring system System, including: at least one toxic gas radio detection node, at least one converging information transit node with at least one is portable Intelligent hand-held terminal；Wherein, described toxic gas radio detection node, for detecting the first toxic gas that accident is on-the-spot Information, adopts and is sent by described toxic gas radio detection node in various manners to the on-the-spot first toxic gas information of accident Region；

Described toxic gas radio detection node includes poisonous gas sensor array module and the first micro controller module；

Described toxic gas sensor array module is for detection the first toxic gas information, including one-parameter sensor array Row module and multi-parameter sensor array module；

Described first micro controller module is poisonous for analyze the detection of described toxic gas sensor array module first Gas information；Described first micro controller module includes cluster module and artificial neural network identification module；

Described cluster module is for entering the first toxic gas information that described toxic gas sensor array module detects Row cluster, obtains cluster result；

Described artificial neural network identification module is used for, and utilizes the toxic gas identification mould built based on artificial neural network Cluster result described in type identification, obtains the second toxic gas information；

Converging information transit node, for converging the second toxic gas information of each toxic gas radio detection node Poly-, and send to described portable intelligent handheld terminal；

Described portable intelligent handheld terminal includes the 3rd gentle bulk diffusion computing module of micro controller module, the described 3rd Micro controller module, for by described second toxic gas information, is merged by toxic gas generic, and obtaining the 3rd has Poisonous gas information；Utilize effusiometer to calculate module and quickly draw speed and the scope of on-the-spot first toxic gas diffusion of information.

Preferably, described toxic gas radio detection node also include poisonous gas adsorption module, the first locating module and First wireless communication module；Described toxic gas adsorption module is for being adsorbed with the first of poisonous gas sensor array module detection Toxic gas information；Described wireless communication module is for sending the first toxic gas information of the detection of sensor array module To converging information transit node；Described first micro controller module and the first locating module are connected；Described first locating module is used In the location information obtaining toxic gas radio detection node；Described toxic gas radio detection node also includes and toxic gas Sensor array module, toxic gas adsorption module, the first micro controller module, described first micro controller module be connected the One locating module and the first power module being connected with the first wireless communication module；Described first power module is for there being poison Modules in body radio detection node provides power supply；

Further, described system also includes: off-line analysis module, for described first toxic gas information carried out from Line analysis, obtains off-line analysis result；

Further, also include: optimize module, be used for utilizing described off-line analysis modified result artificial neural network identification The output result of module.

Preferably, described cluster module includes:

First cluster module, is used for using Fuzzy c-means clustering algorithm KFCM that the first toxic gas information is passed through letter Number carries out fuzzy C-means clustering FCM again and obtains initial value after being mapped to high-dimensional feature space, recycle particle group optimizing PSO algorithm pair Described initial value is optimized iterative computation to obtain optimal cluster result, and its cluster majorized function used includes:

$f\left(x_i\right)=\frac{1}{J_{\mathrm{\α}}(U,V)+1}=\frac{1}{2\underset{i=1}{\overset{C}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\μ}}_{\mathrm{ij}}^m[1-K(x_i,v_j)]+1}---\left(1\right)$

Wherein, C is cluster number, and N is sample number, U=[μ_ij]_C×NIt is Fuzzy C Matrix dividing, μ_ijFor sample x_iCorrespond to What jth clustered is subordinate to angle value, V=[v_j] it is the set of C cluster centre composition, m is to affect subordinated-degree matrix obfuscation journey The index weight of degree, J_α(U, V) is the cost function of FCM, K (x_i,v_j) it is gaussian kernel function, wherein x_iFor m dimensional vector, v_jWith x_i Being the vector with dimension, the more new formula of cluster centre and degree of membership is as follows:

$v_j=\frac{\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\μ}}_{\mathrm{ij}}^{m}K(x_i,v_j)x_i}{\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\μ}}_{\mathrm{ij}}^{m}K(x_i,v_j)}---\left(2\right)$

${\mathrm{\μ}}_{\mathrm{ij}}=\frac{{(1-K(x_i,v_j))}^{-1/(m-1)}}{\underset{j=1}{\overset{C}{\mathrm{\Σ}}}{(1-K(x_i,v_j))}^{-1/(m-1)}}---\left(3\right)$

In each iteration, each particle updates position and the speed of self by the individual extreme value of tracking and global extremum, Its computing formula is as follows:

V_iα(t+1)=ω V_iα(t)+c₁r₁[P_iα(t)-X_iα(t)]+c₂r₂[P_fα(t)-X_iα(t)] (4)

X_iα(t+1)=X_iα(t)+V_iα(t+1) (5)

In formula, V_iαT () is the speed of particle previous moment, V_iα(t+1) it is the current speed of particle, X_iαT () is particle Current location, X_iα(t+1) be particle produce new position, P_iαT () is individual extreme value, P_fαT () is global extremum, c₁、c₂For learning Practise the factor (being the positive integer more than zero), r₁、r₂For the random number between [0,1], i=1,2 ..., N, ω are inertial factor.

Preferably, described portable intelligent handheld terminal also includes:

Waking up control module up, be suitable to the distribution situation according to described location information, portable intelligent handheld terminal wakes up up can The standby toxic gas radio detection node in monitoring region;

Described portable intelligent handheld terminal also includes the 3rd wireless communication module, the 3rd micro controller module, the 3rd confession Electricity module and display input module；Described 3rd micro controller module is for controlling the 3rd wireless communication module and input module； Described 3rd wireless communication module is for receiving the second toxic gas information that described converging information transit node transmits, and passes through 3rd micro controller module carries out Data Fusion to the second toxic gas information, by poisonous for the 3rd after Data Fusion Gas information sends to described display input module；Described display input module is used for the duty of displaying scene node and has Close data；Described 3rd supply module and the 3rd wireless communication module, the 3rd micro controller module and display input module are connected； Described 3rd supply module provides power supply for the modules in portable intelligent handheld terminal.

Preferably, described converging information transit node includes intra-node Temperature Humidity Sensor module, the second microcontroller Module and the second wireless communication module；Described second wireless communication module is for receiving what toxic gas radio detection node obtained Second toxic gas information；Described intra-node Temperature Humidity Sensor module is integrated on described converging information transit node, uses Temperature and humidity in detecting described converging information transit node；Described second wireless communication module for by location information, Temperature and humidity information sends to portable intelligent handheld terminal；Described converging information transit node also includes second micro-with described The second locating module that controller module is connected；Described second locating module is for obtaining determining of described converging information transit node Position information；Described converging information transit node also include with intra-node Temperature Humidity Sensor module, the second wireless communication module, Second micro controller module, described second micro controller module be connected the second locating module and with described second radio communication mold The second supply module that block is connected；Described second supply module carries for the modules in described converging information transit node Power supply source.

Preferably, described artificial neural network identification module includes: N number of artificial neural network identification submodule, described people Artificial neural networks identification submodule is for being identified the cluster result of corresponding classification.

Preferably, also include:

Fusion Module, for described 3rd toxic gas information being merged, realizes fusion treatment by below equation:

$\left\{\begin{array}{c}\stackrel{\‾}{x}=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{w}_i{z}_i\\ \underset{i=1}{\overset{n}{\mathrm{\Σ}}}{w}_i=1(0\≤w_i\≤1)\end{array}\right.---\left(6\right)$

Wherein, x is true value, z_iFor the measured value of cluster result, w_iFor the weighter factor of cluster result,For data fusion After data；

Wherein, for ensureing the zero deflection estimated, then, its error variance expression formula is

${\mathrm{\σ}}^2=E\left[{(x-\stackrel{\‾}{x})}^2\right]=E\left[\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{w}_i^2{(x-\stackrel{\‾}{x})}^2\right]=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{w}_i^2{\mathrm{\σ}}_i^2---\left(7\right)$

Wherein, weighter factor w_i, $w_i=\frac{{\mathrm{\σ}}_i^{-2}}{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\σ}}_i^{-2}}---\left(8\right)$

(7) and (8) substitute into (6) acquisition data fusion result is

$\stackrel{\‾}{x}=\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{z}_i{\mathrm{\σ}}_i^{-2}}{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\σ}}_i^{-2}}---\left(9\right)$

In order to solve the problems referred to above, this application discloses a kind of toxic gas long-distance wireless monitoring side, accident scene Method, including:

Toxic gas radio detection node is sent to accident on-the-spot;

Utilize the toxic gas sensor array module in toxic gas radio detection node to gather accident scene to have First toxic gas information of poisonous gas radio detection node；

The cluster module in toxic gas radio detection node is utilized the first toxic gas information collected to be gathered Class, obtains cluster result；Use the artificial neural network identification module in toxic gas radio detection node, utilize based on manually The cluster result obtained is identified by the toxic gas identification model that neutral net builds, and obtains the second toxic gas letter Breath；

Utilize converging information transit node described second toxic gas information to be converged, and send to portable intelligent Handheld terminal；

Utilize the 3rd micro controller module in portable intelligent handheld terminal by the second toxic gas information, by there being poison Body generic merges, and obtains the 3rd toxic gas information, and on-the-spot first has to utilize effusiometer calculation module to determine The speed of poisonous gas diffusion of information and scope.

Preferably, described method, also include using Fuzzy c-means clustering algorithm KFCM the first toxic gas information to be led to Crossing Function Mapping and obtain initial value to carrying out FCM cluster after high-dimensional feature space again, recycling particle group optimizing PSO algorithm is to institute Stating initial value and be optimized iterative computation to obtain optimal cluster result, what it used cluster majorized function includes:

$f\left(x_i\right)=\frac{1}{J_{\mathrm{\α}}(U,V)+1}=\frac{1}{2\underset{i=1}{\overset{C}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\μ}}_{\mathrm{ij}}^m[1-K(x_i,v_j)]+1}---\left(1\right)$

Wherein, C is cluster number, and N is sample number, U=[μ_ij]_C×NIt is Fuzzy C Matrix dividing, μ_ijFor sample x_iCorrespond to What jth clustered is subordinate to angle value, V=[v_j] it is the set of C cluster centre composition, m is to affect subordinated-degree matrix obfuscation journey The index weight of degree, J_α(U, V) is the cost function of FCM, K (x_i,v_j) it is gaussian kernel function, wherein x_iFor m dimensional vector, v_jWith x_i Being the vector with dimension, the more new formula of cluster centre and degree of membership is as follows:

$v_j=\frac{\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\μ}}_{\mathrm{ij}}^{m}K(x_i,v_j)x_i}{\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\μ}}_{\mathrm{ij}}^{m}K(x_i,v_j)}---\left(2\right)$

${\mathrm{\μ}}_{\mathrm{ij}}=\frac{{(1-K(x_i,v_j))}^{-1/(m-1)}}{\underset{j=1}{\overset{C}{\mathrm{\Σ}}}{(1-K(x_i,v_j))}^{-1/(m-1)}}---\left(3\right)$

In each iteration, each particle updates position and the speed of self by the individual extreme value of tracking and global extremum, Its computing formula is as follows:

V_iα(t+1)=ω V_iα(t)+c₁r₁[P_iα(t)-X_iα(t)]+c₂r₂[P_fα(t)-X_iα(t)] (⁴)

X_iα(t+1)=X_iα(t)+V_iα(t+1) (5)

In formula, V_iαT () is the speed of particle previous moment, V_iα(t+1) it is the current speed of particle, X_iαT () is particle Current location, X_iα(t+1) be particle produce new position, P_iαT () is individual extreme value, P_fαT () is global extremum, c₁、c₂For learning Practise the factor (being the positive integer more than zero), r₁、r₂For the random number between [0,1], i=1,2 ..., N, ω are inertial factor.

Preferably, also include having utilize the toxic gas identification model built based on artificial neural network to obtain second Poisonous gas information carries out data fusion, realizes Data Fusion by below equation:

$\left\{\begin{array}{c}\stackrel{\‾}{x}=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{w}_i{z}_i\\ \underset{i=1}{\overset{n}{\mathrm{\Σ}}}{w}_i=1(0\≤w_i\≤1)\end{array}\right.---\left(6\right)$

Wherein, x is true value, z_iFor the measured value of cluster result, w_iFor the weighter factor of cluster result, after x is data fusion Data；

Wherein, for ensureing the zero deflection estimated, then, its error variance expression formula is:

${\mathrm{\σ}}^2=E\left[{(x-\stackrel{\‾}{x})}^2\right]=E\left[\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{w}_i^2{(x-\stackrel{\‾}{x})}^2\right]=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{w}_i^2{\mathrm{\σ}}_i^2---\left(7\right)$

Wherein, weighter factor w_i, $w_i=\frac{{\mathrm{\σ}}_i^{-2}}{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\σ}}_i^{-2}}---\left(8\right)$

(7) and (8) substitute into (6) acquisition data fusion result is:

$\stackrel{\‾}{x}=\frac{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{z}_i{\mathrm{\σ}}_i^{-2}}{\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{\mathrm{\σ}}_i^{-2}}---\left(9\right)$

Compared with prior art, the application includes advantages below:

First, toxic gas radio detection node can be sent by the application by emitter or aircraft or robot On-the-spot, by the one-parameter sensor array module in toxic gas radio detection node and multi-parameter sensor to accident Array module quickly detects the first toxic gas information that accident is on-the-spot.

Secondly, the application is used in combination by cluster module and artificial neural network identification module, can effectively determine The type of the second toxic gas and concentration, and by experiment and on-the-spot application Data correction Network Recognition module, to improve constantly The accuracy of detection of on-the-spot toxic gas.

Again, can to monitor accident scene toxic gas in real time wireless for the portable intelligent handheld terminal in the application Detect the duty of node and wake up the standby toxic gas radio detection node that can monitor region up, and having provided poison Body in time with the diffusion concentration field of spatial variations.

Accompanying drawing explanation

Fig. 1 is toxic gas remote wireless monitoring system structure chart in accident scene described in the embodiment of the present application；

Fig. 1-A shows the on-the-spot described toxic gas radio detection node section knot of accident described in the embodiment of the present application Structure schematic diagram；

Fig. 1-B shows the on-the-spot described portable intelligent handheld terminal part-structure of accident described in the embodiment of the present application Schematic diagram；

Fig. 2 is the transmission means signal of toxic gas radio detection node in accident scene described in the embodiment of the present application Figure；

Fig. 3 is the structure chart of accident scene toxic gas radio detection node described in the embodiment of the present application；

Fig. 4 is the structure chart that accident field data described in the embodiment of the present application converges transit node；

Fig. 5 is the structure chart of accident field portable intelligent hand-held terminal described in the embodiment of the present application；

Fig. 6 is the online test method of accident scene toxic gas described in the embodiment of the present application；

Fig. 7 is accident scene toxic gas detection and recognition principle described in the embodiment of the present application；

Fig. 8 is that described in the embodiment of the present application, accident scene toxic gas remote wireless monitoring system (is breaking down Time) information transmission mode.

Detailed description of the invention

Understandable for enabling the above-mentioned purpose of the application, feature and advantage to become apparent from, real with concrete below in conjunction with the accompanying drawings The application is described in further detail by mode of executing.

With reference to Fig. 1, it is shown that toxic gas remote wireless monitoring system knot in accident scene described in the embodiment of the present application Composition.Wherein, including trochanter at least one toxic gas radio detection node (A101-A10n), at least one converging information Point (A201-A20m) and at least one portable intelligent handheld terminal (A301-A30p).Wherein, the wireless inspection of described toxic gas Survey node, for detecting the first toxic gas information that accident is on-the-spot, adopt in various manners by wireless for described toxic gas Detection node sends the first toxic gas information area to accident scene；Converging information transit node, for being respectively arranged with The second toxic gas information that poisonous gas radio detection node obtains converges, and sends to described portable intelligent hand-held end End.

First toxic gas information includes common are poisonous gas and chemical and biological weapons discharge containing lifetime killer as thin Bacterium, virus, the gas of toxin.Wherein common are poisonous gas kind more, such as: carbon monoxide, nitric oxide, hydrogen sulfide, two The gases such as sulfur oxide, chlorine, ammonia, fluohydric acid gas.

The on-the-spot described toxic gas radio detection node of accident described in the embodiment of the present application is shown with reference to Fig. 1-A Separation structure schematic diagram.Wherein, toxic gas radio detection node includes poisonous gas sensor array module (101) and first micro- Controller module (102)；Described toxic gas sensor array module (101), for detection the first toxic gas information, can be wrapped Include one-parameter sensor array module and multi-parameter sensor array module；In embodiments of the present invention, sensor array module Including chemical sensor array module and biosensor array module.

Described first micro controller module (102) is used for analyzing the detection of described toxic gas sensor array module (101) The first toxic gas information；Described first micro controller module (102) includes that cluster module (103) and artificial neural network are known Other module (104)；Described cluster module (103) is used for first described toxic gas sensor array module (101) detected Toxic gas information clusters, and obtains cluster result；Described artificial neural network identification module (104) is used for, utilize based on Cluster result described in the toxic gas identification Model Identification that artificial neural network builds, obtains the second toxic gas information.

One-parameter sensor array module includes one-parameter biosensor array module and one-parameter chemical sensor battle array Row module, wherein, one-parameter biosensor array module is for detecting the virus discharged by chemical and biological weapons or antibacterial etc.； One-parameter chemical sensor array module is used for detecting the toxic gas such as common carbon monoxide or nitric oxide.Multi-parameter sensor Array module includes multiparameter biosensor array module and multiparameter chemical sensor array module；Wherein, multiparameter Learn sensor array module to mix with nitric oxide production mixed gas or other gas etc. for the carbon monoxide detecting mixing Gas；Multiparameter biosensor array module is for detecting virus or antibacterial or the syphilis etc. of mixing.

Fig. 1-B shows the on-the-spot described portable intelligent handheld terminal part-structure of accident described in the embodiment of the present application Schematic diagram.Wherein, described portable intelligent handheld terminal includes the 3rd micro controller module 105, described 3rd microcontroller mould Block 105, for by described second toxic gas information, is merged by toxic gas generic, obtains the 3rd toxic gas letter Breath.

Effusiometer calculates module (504), for speed and the model of the on-the-spot first toxic gas diffusion of information of quick obtaining Enclose.As when accident scene occurs toxic gas leakage, do not required nothing more than the kind quickly detecting the first toxic gas information And concentration, in addition it is also necessary to the speed of the on-the-spot first toxic gas diffusion of information of quick obtaining and scope.According to Fick diffusion differential side Journey, has

$\frac{\∂C}{\∂t}+u_x\frac{\∂C}{\∂x}+u_y+\frac{\∂C}{\∂y}+u\frac{\∂C}{\∂z}=D_x\frac{{\∂}^{2}C}{{\∂y}^2}+D_y\frac{{\∂}^{2}C}{{\∂y}^2}+D_z\frac{{\∂}^{2}C}{{\∂z}^2}---\left(10\right)$

In formula, C is gaseous mass concentration, mg/m³；u_x,u_y,u_zFor x, the mean wind speed m/s on y, z direction；D_x,D_y,D_z For x, the diffusion coefficient on y, z direction, m²/s。

If only considering the situation of plane wind, then u_z=0, if wind angle is α, wind speed is f, then

u_x=fcos α, u_y=fsin α (¹¹)

For the ease of solving, if diffusion coefficient is constant, i.e. D_x=D_y=D_z=D, then diffusion equation can be reduced to

$\frac{\∂C}{\∂t}+f\cos \mathrm{\α}\frac{\∂C}{\∂x}+f\sin \mathrm{\α}\frac{\∂C}{\∂y}=D[\frac{{\∂}^{2}C}{{\∂x}^2}+\frac{{\∂}^{2}C}{{\∂y}^2}+\frac{{\∂}^{2}C}{{\∂z}^2}]---\left(12\right)$

Being provided with a height of H of storage tank of poisonous gas, radius is R, initial concentration C=C in tank₀, concentration C=0 in tank.Initial condition For

$C=\left\{\begin{array}{cc}{C}_0,& \sqrt{x^2+y^2}\≤R,\left|z\right|\≤H,t=0\\ 0,& \sqrt{x^2+y^2}\≥R,t=0\\ 0,& \sqrt{x^2+y^2}=\±\∞\\ 0,& z=\∞\end{array}\right.---\left(13\right)$

Boundary condition is

$\frac{\∂C}{\∂z}=0,z=0---\left(14\right)$

By initial condition and Boundary Condition for Solving diffusion equation (10), on-the-spot first toxic gas information concentration can be obtained Process and on-the-spot first toxic gas diffusion of information radius process over time over time, in order to take as early as possible Effective counter-measure.

Said system is mainly for the detection of the first toxic gas information at remote accident scene, for low coverage From accident scene can use toxic gas radio detection node (A101-A10n), directly the most hand-held with portable intelligent Terminal communicates.

The application is used in combination by cluster module and artificial neural network identification module, can effectively determine that second has The type of poisonous gas and concentration.

Preferably, accident is on-the-spot is all not detect node in advance, once accident occurs, can not be true Determine in the case of which kind of toxic gas accident scene be, the emitter in the application or aircraft or machine to be used Described toxic gas radio detection node is sent the toxic gas region to accident scene by the modes such as people, can quickly detect Go out the first toxic gas information that accident is on-the-spot.Show described in the embodiment of the present application there be accident scene with reference to Fig. 2 The transmission means schematic diagram of poisonous gas radio detection node.Wherein, toxic gas radio detection node can provide n radio detection Node.Road between the first toxic gas information area (201) that toxic gas radio detection node to accident is on-the-spot When road is the most smooth, robot (202) can be used to carry the mode of poisonous gas radio detection node, by toxic gas Radio detection node is sent to the on-the-spot first toxic gas information area (201) of accident.

Between the first toxic gas information area (201) that toxic gas radio detection node to accident is on-the-spot During road unevenness, aircraft (203) can be used to carry the mode of poisonous gas radio detection node, toxic gas is wireless Detection node is sent to the on-the-spot first toxic gas information area (201) of accident.

Between the first toxic gas information area (201) that toxic gas radio detection node to accident is on-the-spot When environment is more severe, such as strong wind, typhoon or contrary wind, road bumpiness inequality situation, emitter (204) can be used to launch poisonous The mode of gas wireless detection node, is sent to the on-the-spot first toxic gas letter of accident by toxic gas radio detection node Breath region (201).

Three of the above mode is that toxic gas radio detection node sends the first toxic gas letter to accident scene The optimum way in breath region, those skilled in the art can adopt in other ways according to concrete applied environment, and the application is to tool The mode that body uses is without restriction.

With reference to Fig. 3, it is shown that the structure of accident scene toxic gas radio detection node described in the embodiment of the present application Figure.Wherein, by including poisonous gas chemical sensor array module (301) and toxic gas biosensor array module (302) the toxic gas sensor array module of the present invention, toxic gas chemical sensor array module (301) and poisonous are constituted Gaseous bio sensor array module (302) all can by one-parameter sensor array module and or multi-parameter sensor array Module composition, the first toxic gas information that its detection obtains sends to the first micro controller module (102).Toxic gas is wireless Detection node also includes poisonous gas adsorption module (303), the first locating module (305) and the first wireless communication module (304)；

Also include poisonous gas adsorption module (303) to be used for adsorbing site of the accident toxic gas radio detection node institute in place Put the toxic gas of adnexa.It can be appreciated that also will be acquired by the original poison in the scene of the accident, it is sealed in toxic gas In adsorption module, carry out off-line analysis in case follow-up at laboratory；

Wherein, the first wireless communication module (304) has for the first micro controller module (102) is analyzed second obtained Poisonous gas information sends to converging information transit node；

Wherein the first micro controller module (102) is connected with the first locating module (305)；Described first locating module (305) for obtaining the location information of toxic gas radio detection node；

Described toxic gas radio detection node also includes the first supply module (306), for the wireless inspection of toxic gas Survey the modules in node and power supply is provided.

Described first micro controller module (102) can be also used for controlling the suction of toxic gas adsorption module (303) in Fig. 3 Attached state.

Further, described system also includes: off-line analysis module, for described first toxic gas information carried out from Line analysis, obtains off-line analysis result；

Will the scene of the accident toxic gas radio detection node reclaim after, by having of each toxic gas radio detection node Toxic gas in poisonous gas adsorption module (303) utilizes (the most large-scale poison analysis of off-line analysis module in the lab Instrument etc.) carry out accurate special analysis, obtain the dense of accurate poison analysis result, such as poison type and this poison type Degree.

Further, also include: optimize module, be used for utilizing described off-line analysis modified result artificial neural network identification The output result of module.

It is appreciated that, at laboratory, each toxic gas radio detection node of recovery is carried out off-line with the equipment of specialty Analyze the result obtained, be likely more accurately, then the application (can also include that poison type is with dense according to off-line analysis result Degree), compare with the second toxic gas information (including poison type and concentration) of toxic gas radio detection node itself, If error exceedes threshold value, then illustrate that the toxic gas identification Model Parameter built based on artificial neural network exists deviation, The application then utilizes off-line analysis result to be modified above-mentioned parameter.

See Fig. 4 and show that accident field data described in the embodiment of the present application converges the structure chart of transit node.Its In, converging information transit node include intra-node Temperature Humidity Sensor module (403), the second micro controller module (402) and Second wireless communication module (401)；Described second wireless communication module (401) is used for receiving toxic gas radio detection node and obtains The the second toxic gas information arrived；Described intra-node Temperature Humidity Sensor module (403) is integrated in described converging information transfer On node, for detecting the temperature and humidity in described converging information transit node, and send to the second micro controller module；Institute State the second locating module (404) that converging information transit node also includes being connected with described second micro controller module (402)；Institute State the second locating module (404) for obtaining location information and the relevant information of described converging information transit node；Described second Wireless communication module (401) is for receiving the location information of accident scene toxic gas radio detection node, temperature and wet Degree information, and by location information, temperature and humidity information by (certainly can also believing after the process of the second micro controller module Breath preserves), location information, temperature and humidity information after processing send to portable intelligent handheld terminal；Described letter Breath converges transit node and also includes the second supply module (405), for the modules in described converging information transit node Power supply is provided.

See Fig. 5 and show the structure chart of accident field portable intelligent hand-held terminal described in the embodiment of the present application. Wherein, described portable intelligent handheld terminal also include the 3rd wireless communication module (501), the 3rd micro controller module (105), Display input module (502), data fusion module (503), effusiometer calculate module (504), wake up up control module (505) and 3rd supply module (506)；Described 3rd micro controller module (105) is for controlling the 3rd wireless communication module (501), display Input module (502), bulk diffusion computing module (504), wake up control module (505), data fusion module (503) up；Described Three wireless communication modules (501) are for receiving the second toxic gas information of described converging information transit node detection, by described Second toxic gas information sends extremely described data fusion module (503) block by the 3rd microcontroller mould (105) and carries out data Fusion treatment.The 3rd toxic gas information after data fusion module (503) Data Fusion sends to described display input Module (502)；Described effusiometer calculates speed and the scope that module (504) spreads for quick detection toxic gas, and will Testing result is sent to the 3rd micro controller module (105), and described toxic gas is spread by the 3rd micro controller module (105) Speed and scope are sent to show that input module (502) shows；Described display input module (502) is saved for displaying scene The duty of point and the various detection data of toxic gas；Described 3rd supply module (506) is for portable intelligent hands Hold the modules in terminal and power supply is provided.

Described portable intelligent handheld terminal also includes: wake up control module (505) up, is suitable to according to described location information Distribution situation, portable intelligent handheld terminal wakes up the standby toxic gas radio detection node that can monitor region up, i.e. by the Three micro controller modules (105) and the 3rd wireless communication module (501) wake up the wireless inspection of standby toxic gas that can monitor region up Survey node.

Throwing in toxic gas radio detection node when, there may come a time when to there will be the feelings that detection Node distribution is uneven Condition, at this point it is possible to according to the distribution situation of location information, in the region that can monitor, by portable intelligent handheld terminal Wake module wake up the standby toxic gas radio detection node in the region that can monitor up, make toxic gas radio detection node It is evenly distributed in the region that can monitor.When certain toxic gas radio detection one malfunctions in accident scene, It is poisonous that portable intelligent handheld terminal can also utilize the first locating module in toxic gas radio detection node to obtain first The location information of gas wireless detection node, according to the distribution situation of location information, in the region that can monitor, adjusting first has The delivering path of poisonous gas information.Such as: send out when monitoring a certain toxic gas radio detection node at portable handheld terminal During raw fault, can on-line tuning originally by the path of this node-node transmission information, the first toxic gas information that will be original is not Reach this node, and change and reach the non-faulting node (i.e. toxic gas radio detection node) nearest away from this malfunctioning node, thus Guarantee that the detection information of whole toxic gas reliably reaches converging information transit node.

In like manner, when the respective nodes in converging information transit node or portable intelligent handheld terminal breaks down, all Method during toxic gas radio detection one malfunctions is used to carry out information transmission.I.e. converging information transit node or portable Formula intelligent hand-held terminal selects the non-faulting node nearest away from malfunctioning node to carry out information transmission.

Described portable intelligent handheld terminal also has the duty of observable Site Detection node；Change detection information Network delivering path；Manually wake up redundant detection node up；Find work on the spot detection node and the function of malfunctioning node.Can also transport With Data fusion technique, improve the certainty of measurement of accident scene toxic gas and analyze the diffusion tendency of toxic gas.

Described cluster module includes: the first cluster module, is used for using Fuzzy c-means clustering algorithm KFCM to have first Poisonous gas information obtains initial value by carrying out FCM cluster after Function Mapping to high-dimensional feature space again, and recycling population is excellent Changing PSO algorithm to be optimized to obtain optimal cluster result to described initial value, its cluster majorized function used includes:

$f\left(x_i\right)=\frac{1}{J_a(U,V)+1}=\frac{1}{2\underset{i=1}{\overset{C}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\μ}}_{\mathrm{ij}}^m[1-K(x_i,v_j)]+1}---\left(1\right)$

Wherein, C is cluster number, and N is sample number, U=[μ_ij]_C×NIt is Fuzzy C Matrix dividing, μ_ijFor sample x_iCorrespond to What jth clustered is subordinate to angle value, V=[v_j] it is the set of C cluster centre composition, m is to affect subordinated-degree matrix obfuscation journey The index weight of degree, J_α(U, V) is the cost function of FCM, K (x_i,v_j) it is gaussian kernel function, wherein x_iFor m dimensional vector, v_jWith x_i It it is the vector with dimension.The more new formula of cluster centre and degree of membership is as follows:

$v_j=\frac{\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\μ}}_{\mathrm{ij}}^{m}K(x_i,v_j)x_i}{\underset{i=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\μ}}_{\mathrm{ij}}^{m}K(x_i,v_j)}---\left(2\right)$

${\mathrm{\μ}}_{\mathrm{ij}}=\frac{{(1-K(x_i,v_j))}^{-1/(m-1)}}{\underset{j=1}{\overset{C}{\mathrm{\Σ}}}{(1-K(x_i,v_j))}^{-1/(m-1)}}---\left(3\right)$

PSO algorithm is fitness based on population Yu particle, and each particle has position and two features of speed.Often In secondary iteration, each particle is by following the tracks of individual extreme value and the position of global extremum renewal self and speed, and its computing formula is such as Under:

V_iα(t+1)=ω V_iα(t)+c₁r₁[P_iα(t)-X_iα(t)]+c₂r₂[P_fα(t)-X_iα(t)] (4)

X_iα(t+1)=X_iα(t)+V_iα(t+1) (5)

In formula, V_iαT () is the speed of particle previous moment, V_iα(t+1) it is the current speed of particle, X_iαT () is particle Current location, X_iα(t+1) be particle produce new position, P_iαT () is individual extreme value, P_fαT () is global extremum, c₁、c₂For learning Practise the factor (being the positive integer more than zero), r₁、r₂For the random number between [0,1], i=1,2 ..., N, ω are inertial factor.

Cluster is a kind of important data analysis technique, and by cluster, people are capable of identify that intensive and sparse region, Thus found that the distribution pattern of the overall situation, and the mutual relation between data attribute.Fuzzy C-Means Cluster Algorithm is that k average is gathered The popularizing form of class algorithm, neither can guarantee that the optimal solution finding problem, is likely to converge to local extremum.Therefore, The initial cluster center of application particle group optimizing (PSO) algorithm optimization Fuzzy c-means clustering algorithm (KFCM) realizes data and gathers Class, to realize higher cluster accuracy.

The is exported according to the toxic gas sensor array module in the toxic gas radio detection node of accident scene The numerical value of one toxic gas information, determines the number C(i.e. C kind mixed gas of the first toxic gas information cluster).N is for determine Time period in, by toxic gas sensor array module sampling output the first toxic gas information value sum (i.e. N is total Sample number).First toxic gas information is entered by the computing formula utilizing fuzzy C-means clustering (FCM) and core fuzzy C-means clustering (KFCM) Row cluster, concrete fuzzy C-means clustering (FCM) and the computing formula of core fuzzy C-means clustering (KFCM) can use meter of the prior art Calculating formula, this is not any limitation as by the application.

Further, in the first cluster module, KFCM algorithm steps based on PSO can briefly be described below:

1. number C(C determining the first toxic gas information cluster can be that toxic gas sensor array module is detectable The number of poison type), set allowable error ε, and set iterations t=1, set total sample number N, set Inertia Weight ω, Studying factors c1 and c2 and index weight, (particle can be understood as toxic gas and senses to calculate particle by above parameter Device array module transmission each detected value) speed and position.The formula of the concrete speed calculating particle and position can use Computing formula of the prior art, this is not any limitation as by the application.

2. population v is initialized₁,v₂..., v_j,…,v_C, (detect for toxic gas sensor array module from sample set Each poison information) in appoint take C vector to initialize v_jIf, sample space X=[x1, x2 ..., x_N], calculate gaussian kernel function K (x_i,v_j)；

Wherein, v_jFor cluster centre.

3. calculate subordinated-degree matrix U for each sample, then calculate cluster majorized function

$f\left(x_i\right)=\frac{1}{J_{\mathrm{\α}}(U,V)+1}=\frac{1}{2\underset{i=1}{\overset{C}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\μ}}_{\mathrm{ij}}^m[1-K(x_i,v_j)]+1}---\left(1\right)$ According still further to formula (4) formula (5) revise particle rapidity and Position, according to the cluster individual extreme value of majorized function value amendment and global extremum, in order to produce particle of future generation；

4. judge: whether current iteration number of times reaches maximum times set in advance, if reached, stopping iteration, i.e. obtaining Obtain the optimal solution of cluster centre；Otherwise forward step to 3.；

5. update the degree of membership calculating particle colony, update the cluster centre calculating colony；Carry out more by formula (3), (2) Newly.

6. calculating difference E of adjacent generations subordinated-degree matrix, if E < ε, then iterative computation stops；Otherwise forward to 5..

Described artificial neural network identification module includes: N number of artificial neural network identification submodule, described ANN Network identification submodule is for being identified the cluster result of corresponding classification.

Build artificial neural network submodule by historical data and experimental data, and it is trained, then could It is applied to scene.Therefore, it is integrated in the number of sensors on toxic gas radio detection node (to include: one-parameter chemical sensitisation Device+one-parameter biosensor+multiparameter chemical sensor+multiparameter biosensor) and relevant experimental data determine people Artificial neural networks submodule quantity.

Further, described system also includes: Fusion Module, for described 3rd toxic gas information is merged, Fusion treatment is realized by below equation:

$\left\{\begin{array}{c}\stackrel{\&OverBar;}{x}=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{w}_i{z}_i\\ \underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{w}_i=1(0\&le;w_i\&le;1)\end{array}\right.---\left(6\right)$

Wherein, x is true value, z_iFor the measured value of cluster result, w_iFor the weighter factor of cluster result, after x is data fusion Data；

Wherein, for ensureing the zero deflection estimated, then, its error variance expression formula is

${\mathrm{\&sigma;}}^2=E\left[{(x-\stackrel{\&OverBar;}{x})}^2\right]=E\left[\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{w}_i^2{(x-\stackrel{\&OverBar;}{x})}^2\right]=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{w}_i^2{\mathrm{\&sigma;}}_i^2---\left(7\right)$

Wherein, weighter factor w_i,

$w_i=\frac{{\mathrm{\&sigma;}}_i^{-2}}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_i^{-2}}---\left(8\right)$

Formula (7) and formula (8) are substituted into formula (6) acquisition data fusion result is

$\stackrel{\&OverBar;}{x}=\frac{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{z}_i{\mathrm{\&sigma;}}_i^{-2}}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_i^{-2}}---\left(9\right)$

Described portable intelligent handheld terminal also includes:

For said system embodiment, present invention also offers a kind of accident scene toxic gas long-distance wireless monitoring Method, shows the online test method of accident scene toxic gas described in the embodiment of the present application with reference to Fig. 6, including: will Toxic gas radio detection node sends to accident on-the-spot;The toxic gas in toxic gas radio detection node is utilized to pass Sensor array module gathers the first toxic gas information of accident scene toxic gas radio detection node.Toxic gas passes Sensor array includes one-parameter sensor array module and multi-parameter sensor array module.One-parameter sensor array module bag Include one-parameter biosensor array module and one-parameter chemical sensor array module；Multi-parameter sensor array module includes Multiparameter biosensor array module and multiparameter chemical sensor array module.Utilize in toxic gas radio detection node Cluster module the first toxic gas information collected is clustered, obtain cluster result；As shown in Figure 6, x₁-x_nRepresent The output of one-parameter chemical sensor array module (601), x₁₁-x_mmIt is expressed as multiparameter chemical sensor array module (602) Output, y₁-y_nIt is expressed as the output of one-parameter biosensor array module (603), y₁₁-y_nnIt is expressed as multiparameter biology to pass The output of sensor array module (604).The output data of each sensor array module all use Fuzzy c-means clustering algorithm KFCM by the first toxic gas information by Function Mapping at the beginning of carrying out again after FCM is clustered after high-dimensional feature space Initial value, the purpose of cluster is to be combined by identical gas, the initial value after the cluster obtained.Tie for reaching optimal cluster Really, recycling particle group optimizing PSO algorithm is optimized iterative computation to described initial value, its cluster majorized function bag used Include:

$f\left(x_i\right)=\frac{1}{J_{\mathrm{\&alpha;}}(U,V)+1}=\frac{1}{2\underset{i=1}{\overset{C}{\mathrm{\&Sigma;}}}\underset{j=1}{\overset{N}{\mathrm{\&Sigma;}}}{\mathrm{\&mu;}}_{\mathrm{ij}}^m[1-K(x_i,v_j)]+1}---\left(1\right)$

Wherein, C is cluster number, and N is sample number, U=[μ_ij]_C×NIt is Fuzzy C Matrix dividing, μ_ijFor sample x_iCorrespond to What jth clustered is subordinate to angle value, V=[v_j] it is the set of C cluster centre composition, m is to affect subordinated-degree matrix obfuscation journey The index weight of degree, J_α(U, V) is the cost function of FCM, K (x_i,v_j) it is gaussian kernel function, wherein x_iFor m dimensional vector, v_jWith x_i It it is the vector with dimension.The more new formula of cluster centre and degree of membership is as follows:

$v_j=\frac{\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{\mathrm{\&mu;}}_{\mathrm{ij}}^{m}K(x_i,v_j)x_i}{\underset{i=1}{\overset{N}{\mathrm{\&Sigma;}}}{\mathrm{\&mu;}}_{\mathrm{ij}}^{m}K(x_i,v_j)}---\left(2\right)$

${\mathrm{\&mu;}}_{\mathrm{ij}}=\frac{{(1-K(x_i,v_j))}^{-1/(m-1)}}{\underset{j=1}{\overset{C}{\mathrm{\&Sigma;}}}{(1-K(x_i,v_j))}^{-1/(m-1)}}---\left(3\right)$

PSO algorithm is fitness based on population Yu particle, and each particle has position and two features of speed.Often In secondary iteration, each particle is by following the tracks of individual extreme value and the position of global extremum renewal self and speed, and its computing formula is such as Under:

V_iα(t+1)=ω V_iα(t)+c₁r₁[P_iα(t)-X_iα(t)]+c₂r₂[P_fα(t)-X_iα(t)] (4)

X_iα(t+1)=X_iα(t)+V_iα(t+1) (5)

In formula, V_iαT () is the speed of particle previous moment, V_iα(t+1) it is the current speed of particle, X_iαT () is particle Current location, X_iα(t+1) be particle produce new position, P_iαT () is individual extreme value, P_fαT () is global extremum, c₁、c₂For learning Practise the factor (being the positive integer more than zero), r₁、r₂For the random number between [0,1], i=1,2 ..., N, ω are inertial factor.

Use the artificial neural network identification module in toxic gas radio detection node, utilize based on artificial neural network The initial value of the cluster result obtained is identified by the toxic gas identification model built, and obtains the second toxic gas letter Breath；Utilize converging information transit node described second toxic gas information to be converged, and send to portable intelligent hand-held Terminal；Utilize the 3rd micro controller module in portable intelligent handheld terminal by the second toxic gas information, by toxic gas Generic merges, and obtains the 3rd toxic gas information, so that it is determined that the type of the 3rd final toxic gas information And concentration.

Preferably, described method also includes: by the toxic gas adsorption module in toxic gas radio detection node (606) the first toxic gas information that absorption accident is on-the-spot, will get described first toxic gas information and use special Analytical tool (607) carries out off-line analysis, obtains off-line analysis result；Utilize described off-line analysis result to revise artificial neuron The recognition result of Network Recognition module.

Preferably, Parameter Map 7 shows that accident scene toxic gas detection described in the embodiment of the present application is former with identification Manage concrete application example.When the generation of accident scene is single or is mixed with poisonous gas leakage, by biographies various in the application Defeated mode, is sent to accident on-the-spot by n toxic gas radio detection node.The poison on-the-spot due to accident may It is pure gas, it is also possible to mixed gas (i.e. in figure 701), causes the radio detection node being distributed in zones of different to be passed back Detection data will have certain difference, the most relatively accurately detect and identify in mixed gas the type of each gas and Concentration, has bigger difficulty.To this end, use particle group optimizing (PSO) algorithm and Fuzzy c-means clustering algorithm (KFCM) phase In conjunction with method, i.e. PSO+KFCM algorithm.It is then based on PSO+KFCM algorithm detection data are effectively classified (702), Data class 1-data class m(703 after cluster and optimization).Again by m artificial neural network, i.e. ANN1-ANNm(704) Gas 1-gas m after data class 1-data class m is identified one by one, after being processed.Finally give and determine gas 1-gas The type of body m and concentration.Further, after each accident treatment completes, each toxic gas detection node there is poison The toxic gas of body adsorption module absorption carries out off-line analysis, according to this toxic gas detection module of analysis result correction, uses In revising weights that the toxic gas identification model built based on artificial neural network knows (i.e. in figure 705, based on off-line analysis number According to, revise neural network weight), to realize improving constantly the purpose of certainty of measurement.

When sending to certain toxic gas radio detection one malfunctions that accident is on-the-spot, portable intelligent hands Holding terminal can utilize the first locating module in toxic gas radio detection node to obtain the first toxic gas radio detection joint The location information of point, according to the distribution situation of location information, in the region that can monitor, adjusts the biography of the first toxic gas information Defeated approach.When being accident scene toxic gas detection node section transmission fault described in the embodiment of the present application with reference to Fig. 8 Detection method.Portable handheld terminal can detect that the state of malfunctioning node, and it is on-the-spot (801) to adjust accident easily In radio detection node and the information transmission path of converging information transit node that converges in transit node set (802).

Assume that toxic gas radio detection node (2) sent to accident on-the-spot (801) breaks down, have poison The first toxic gas information that body radio detection node (1) detects is sent to the non-faulting node nearest away from malfunctioning node i.e. to be had Poisonous gas radio detection node (3), then sent to converging information transit node (1) by toxic gas radio detection node (3).

Assume, when the converging information transit node (1) in converging information transit node set (802) breaks down, have poison The second toxic gas information that body radio detection node (2) detection obtains is transferred to from fault transit node converging information transfer In non-faulting transit node converging information transit node (2) that node (1) is nearest, then converging information transit node (2) reaches just Take formula handheld terminal (2)；

Assume when the portable handheld terminal (1) in portable handheld terminal set (803) breaks down, in converging information The non-faulting portable hand-held that the second toxic gas information that trochanterion (1) converges is transferred to from fault transit node is nearest is whole End i.e. portable handheld terminal (2).

Described method, also includes the toxic gas identification model built based on artificial neural network will be utilized to obtain the Two toxic gas information carry out data fusion, realize Data Fusion by below equation:

$\left\{\begin{array}{c}\stackrel{\&OverBar;}{x}=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{w}_i{z}_i\\ \underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{w}_i=1(0\&le;w_i\&le;1)\end{array}\right.---\left(6\right)$

Wherein, x is true value, z_iFor the measured value of cluster result, w_iFor the weighter factor of cluster result, after x is data fusion Data；

Previously according to the historical data of each toxic gas radio detection node, calculate n toxic gas in measurement system The variance of radio detection node is respectivelyIf the true value that certain measurement moment need to be estimated is x, it is respectively arranged with The poisonous gas radio detection node measured value in this moment is respectively z₁, z₂..., z_n, each measurement result is independent, respectively each other The weighter factor of toxic gas radio detection node is respectively w₁, w₂..., w_n, then after fusion treatmentValue and weighter factor should This meets formula (4).After time accident analysis is complete, will add when the data of each toxic gas radio detection node in time accident Enter described historical data, re-evaluate the variance of correspondence.

Wherein, for ensureing the zero deflection estimated, then, its error variance expression formula is:

${\mathrm{\&sigma;}}^2=E\left[{(x-\stackrel{\&OverBar;}{x})}^2\right]=E\left[\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{w}_i^2{(x-\stackrel{\&OverBar;}{x})}^2\right]=\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{w}_i^2{\mathrm{\&sigma;}}_i^2---\left(7\right)$

Wherein, weighter factor w_i, $w_i=\frac{{\mathrm{\&sigma;}}_i^{-2}}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_i^{-2}}---\left(8\right)$

Formula (7) and formula (8) are substituted into formula (6) acquisition data fusion result is:

$\stackrel{\&OverBar;}{x}=\frac{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{z}_i{\mathrm{\&sigma;}}_i^{-2}}{\underset{i=1}{\overset{n}{\mathrm{\&Sigma;}}}{\mathrm{\&sigma;}}_i^{-2}}---\left(9\right)$

By being distributed in the toxic gas sensor array module in the toxic gas radio detection node of accident scene, Detection is the toxic gas around this toxic gas radio detection node, completes basic process and identification, including pre-place Reason, cluster analysis, pattern recognition, to draw kind and the concentration of toxic gas that respective node detected.Then can be the most portable Formula intelligent hand-held terminal, maintenance data integration technology, final obtain associating inferred results (the i.e. kind of scene toxic gas and dense Degree).

Toxic gas radio detection node can be sent to prominent by the application by emitter or aircraft or robot Send out love scene, by the one-parameter sensor array module in toxic gas radio detection node and multi-parameter sensor array Module quickly detects the first toxic gas information that accident is on-the-spot.Secondly, by cluster module and people in the application Artificial neural networks identification module is used in combination, and can effectively determine type and the concentration of the second toxic gas.Again, in the application Portable intelligent handheld terminal can monitor in real time accident scene toxic gas radio detection node duty and Wake up the standby toxic gas radio detection node that can monitor region up.The main innovation point of the application includes: (1) detecting system Innovation: i.e. fully take into account the difficult point of toxic gas Site Detection, carry out overall improvement and framework from system, i.e. can pass through Various ways throws in the toxic gas detection node of modes such as () robot input, aircraft input, emitter inputs, is used for counting According to the converging information transit node of transfer with may be located remotely from the handheld terminal of poison field monitor, with in unknown emergency case Under, can relatively accurate measurement poison type and concentration under the safety ensureing personnel；(2) detection algorithm integrated innovation: this The bright environmental consideration on-the-spot based on toxic gas, in order to detect toxic gas accurately, it is proposed that the first detection number to poison According to carrying out KFCM based on PSO cluster, toxic gas identification model based on artificial neuron model construction is then used to know Not, finally handheld terminal carry out all data fusion this particularly combine, obtained pinpoint accuracy toxic gas know Other result.And available described off-line analysis result revises the recognition result of artificial neural network identification module, to enter one Step improves the accuracy of detection of accident scene toxic gas long distance wireless；(3) the intelligent principle innovation of handheld terminal: the present invention Handheld terminal can quickly determine the on-the-spot toxic gas diffusion scope of poison and speed, all toxic gas detection nodes can be received Data, and it is carried out fusion treatment.And can make according to the transmission path of on-the-spot practical situation on-line tuning detection information Whole scheme overall architecture is simply effective, facilitates technical staff to make follow-up decision.

Above to a kind of accident scene toxic gas remote wireless monitoring system provided herein and method, enter Having gone and be discussed in detail, principle and the embodiment of the application are set forth by specific case used herein, above enforcement The explanation of example is only intended to help and understands the present processes and core concept thereof；General technology people simultaneously for this area Member, according to the thought of the application, the most all will change, in sum, and this explanation Book content should not be construed as the restriction to the application.

Claims (9)

1. an accident scene toxic gas remote wireless monitoring system, it is characterised in that including: at least one has poison Body radio detection node, at least one converging information transit node and at least one portable intelligent handheld terminal；Wherein, described Toxic gas radio detection node, for detecting the first toxic gas information that accident is on-the-spot, adopts in various manners by institute State toxic gas radio detection node to send to the on-the-spot first toxic gas information area of accident；Described various mode bag Include: robot is thrown in, aircraft is thrown in, emitter is thrown in；

Described toxic gas radio detection node includes poisonous gas sensor array module and the first micro controller module；

Described toxic gas sensor array module is for detection the first toxic gas information, including one-parameter sensor array mould Block and multi-parameter sensor array module；

Described first micro controller module is for analyzing the first toxic gas of described toxic gas sensor array module detection Information；Described first micro controller module includes cluster module and artificial neural network identification module；

Described cluster module is for gathering the first toxic gas information that described toxic gas sensor array module detects Class, obtains cluster result；Described cluster includes that KFCM based on PSO clusters；

Described artificial neural network identification module is used for, and utilizes the toxic gas identification model built based on artificial neural network to know The most described cluster result, obtains the second toxic gas information；

Converging information transit node, for the second toxic gas information of each toxic gas radio detection node is converged, And send to described portable intelligent handheld terminal；

Described portable intelligent handheld terminal includes the 3rd gentle bulk diffusion computing module of micro controller module, described 3rd micro-control Device module processed, for by described second toxic gas information, is merged by toxic gas generic, and obtaining the 3rd has poison Body information；Utilize effusiometer to calculate module and quickly draw speed and the scope of on-the-spot first toxic gas diffusion of information；

Described portable intelligent handheld terminal also includes: wake up control module up；The described control module that wakes up up is for having described in basis The distribution situation of the location information of poisonous gas radio detection node, portable intelligent handheld terminal wakes up up and can monitor the standby of region Toxic gas radio detection node；

Described portable intelligent handheld terminal is additionally operable in the region that can monitor, and adjusts the transmission way of the first toxic gas information Footpath.

System the most according to claim 1, it is characterised in that described toxic gas radio detection node also includes poison Body adsorption module, the first locating module and the first wireless communication module；Described toxic gas adsorption module is used for being adsorbed with poison First toxic gas information of body sensor array module detection；Described wireless communication module is for by sensor array module First toxic gas information of detection sends to converging information transit node；Described first micro controller module and the first location mould Block is connected；Described first locating module is for obtaining the location information of toxic gas radio detection node；Described toxic gas without Line detection node also includes and toxic gas sensor array module, toxic gas adsorption module, the first micro controller module, institute State the first locating module that the first micro controller module is connected and the first power module being connected with the first wireless communication module；Institute State the first power module and provide power supply for the modules in toxic gas radio detection node；

Further, described system also includes: off-line analysis module, divides for described first toxic gas information is carried out off-line Analysis, obtains off-line analysis result；

Further, also include: optimize module, be used for utilizing described off-line analysis modified result artificial neural network identification module Output result.

System the most according to claim 1, it is characterised in that described cluster module includes:

First cluster module, is used for using Fuzzy c-means clustering algorithm KFCM the first toxic gas information to be reflected by function Carrying out fuzzy C-means clustering FCM after being mapped to high-dimensional feature space again and obtain initial value, recycling particle group optimizing PSO algorithm is to described Initial value is optimized iterative computation to obtain optimal cluster result, and its cluster majorized function used includes:

$f\left(x_i\right)=\frac{1}{J_{\mathrm{\&alpha;}}(U,V)+1}=\frac{1}{2\underset{i=1}{\overset{C}{\&Sigma;}}\underset{j=1}{\overset{N}{\&Sigma;}}{\mathrm{\&mu;}}_{ij}^m\&lsqb;1-K(x_i,v_j)\&rsqb;+1}---\left(1\right)$

Wherein, C is cluster number, and N is sample number, U=[μ_ij]_C×NIt is Fuzzy C Matrix dividing, μ_ijFor sample x_iCorresponding to jth Cluster is subordinate to angle value, V=[v_j] it is the set of C cluster centre composition, m is the finger affecting subordinated-degree matrix obfuscation degree Number weight, J_α(U, V) is the cost function of FCM, K (x_i,v_j) it is gaussian kernel function, wherein x_iFor m dimensional vector, v_jWith x_iIt is with dimension Vector, the more new formula of cluster centre and degree of membership is as follows:

$v_j=\frac{\underset{i=1}{\overset{N}{\&Sigma;}}{\mathrm{\&mu;}}_{ij}^{m}K(x_i,v_j)x_i}{\underset{i=1}{\overset{N}{\&Sigma;}}{\mathrm{\&mu;}}_{ij}^{m}K(x_i,v_j)}---\left(2\right)$

${\mathrm{\&mu;}}_{ij}=\frac{{(1-K(x_i,v_j\left)\right)}^{-1/(m-1)}}{\underset{j=1}{\overset{C}{\&Sigma;}}{(1-K(x_i,v_j\left)\right)}^{-1/(m-1)}}---\left(3\right)$

PSO algorithm is fitness based on population Yu particle, and each particle has position and two features of speed；The most repeatedly Dai Zhong, each particle is by following the tracks of individual extreme value and the position of global extremum renewal self and speed, and its computing formula is as follows:

V_iα(t+1)=ω V_iα(t)+c₁r₁[P_iα(t)-X_iα(t)]+c₂r₂[P_fα(t)-X_iα(t)] (4)

X_iα(t+1)=X_iα(t)+V_iα(t+1) (5)

In formula, V_iαT () is the speed of particle previous moment, V_iα(t+1) it is the current speed of particle, X_iαT () is the current of particle Position, X_iα(t+1) be particle produce new position, P_iαT () is individual extreme value, P_fαT () is global extremum, c₁、c₂For study because of Son (being the positive integer more than zero), r₁、r₂For the random number between [0,1], i=1,2 ..., N, ω are inertial factor.

System the most according to claim 1, it is characterised in that described portable intelligent handheld terminal also includes:

Waking up control module up, be suitable to the distribution situation according to described location information, portable intelligent handheld terminal wakes up up and can monitor The standby toxic gas radio detection node in region；

Described portable intelligent handheld terminal also includes the 3rd wireless communication module, the 3rd micro controller module, the 3rd power supply mould Block and display input module；Described 3rd micro controller module is for controlling the 3rd wireless communication module and input module；Described 3rd wireless communication module is for receiving the second toxic gas information that described converging information transit node transmits, and passes through the 3rd Micro controller module carries out Data Fusion to the second toxic gas information, by the 3rd toxic gas after Data Fusion Information sends to described display input module；Described display input module is used for the duty of displaying scene node and relevant number According to；Described 3rd supply module and the 3rd wireless communication module, the 3rd micro controller module and display input module are connected；Described 3rd supply module provides power supply for the modules in portable intelligent handheld terminal.

System the most according to claim 1, it is characterised in that described converging information transit node includes that intra-node is warm and humid Degree sensor assembly, the second micro controller module and the second wireless communication module；Described second wireless communication module is used for receiving The second toxic gas information that toxic gas radio detection node obtains；Described intra-node Temperature Humidity Sensor module is integrated in On described converging information transit node, for detecting the temperature and humidity in described converging information transit node；Described second nothing Line communication module is for sending location information, temperature and humidity information to portable intelligent handheld terminal；Described converging information Transit node also includes the second locating module being connected with described second micro controller module；Described second locating module is used for obtaining Take the location information of described converging information transit node；Described converging information transit node also includes passing with intra-node humiture The second location that sensor module, the second wireless communication module, the second micro controller module, described second micro controller module are connected Module and the second supply module being connected with described second wireless communication module；Described second supply module is for described information Converge the modules in transit node and power supply is provided.

System the most according to claim 1, it is characterised in that described artificial neural network identification module includes: N number of manually Neural network recognization submodule, described artificial neural network identification submodule is for knowing the cluster result of corresponding classification Not.

System the most according to claim 1, it is characterised in that also include:

Fusion Module, for described 3rd toxic gas information being merged, realizes fusion treatment by below equation:

$\{\begin{array}{c}\stackrel{\&OverBar;}{x}=\underset{i=1}{\overset{n}{\&Sigma;}}{w}_i{z}_i\\ \begin{array}{cc}\underset{i=1}{\overset{n}{\&Sigma;}}{w}_i=1& (0\&le;w_i\&le;1)\end{array}\end{array}---\left(6\right)$

Wherein, x is true value, z_iFor the measured value of cluster result, w_iFor the weighter factor of cluster result, x is the number after data fusion According to；

Wherein, for ensureing the zero deflection estimated, thenIts error variance expression formula is

${\mathrm{\&sigma;}}^2=E\&lsqb;{(x-\stackrel{\&OverBar;}{x})}^2\&rsqb;=E\&lsqb;\underset{i=1}{\overset{n}{\&Sigma;}}{w}_i^2{(x-\stackrel{\&OverBar;}{x})}^2\&rsqb;=\underset{i=1}{\overset{n}{\&Sigma;}}{w}_i^2{\mathrm{\&sigma;}}_i^2---\left(7\right)$

Wherein, weighter factor w_i,

(7) and (8) substitute into (6) acquisition data fusion result is

$\stackrel{\&OverBar;}{x}=\frac{\underset{i=1}{\overset{n}{\&Sigma;}}{z}_i{\mathrm{\&sigma;}}_i^{-2}}{\underset{i=1}{\overset{n}{\&Sigma;}}{\mathrm{\&sigma;}}_i^{-2}}---\left(9\right).$

8. an accident scene toxic gas long-range wireless monitoring method, it is characterised in that including:

Toxic gas radio detection node is sent to accident on-the-spot；Described transmission includes that robot is thrown in, aircraft is thrown Put, emitter is thrown in；

Utilize the toxic gas sensor array module in toxic gas radio detection node to gather accident scene and have poison First toxic gas information of body radio detection node；

The cluster module in toxic gas radio detection node is utilized the first toxic gas information collected to be clustered, To cluster result；Use the artificial neural network identification module in toxic gas radio detection node, utilize based on artificial neuron The cluster result obtained is identified by the toxic gas identification model of network struction, and obtains the second toxic gas information；Institute State cluster and include that KFCM based on PSO clusters；

Utilize converging information transit node described second toxic gas information to be converged, and send to portable intelligent hand-held Terminal；

Utilize the 3rd micro controller module in portable intelligent handheld terminal by the second toxic gas information, by toxic gas institute Belong to classification to merge, obtain the 3rd toxic gas information, and on-the-spot first has poison to utilize effusiometer calculation module to determine The speed of body diffusion of information and scope；

Described portable intelligent handheld terminal also includes: wake up control module up；The described control module that wakes up up is for having described in basis The distribution situation of the location information of poisonous gas radio detection node, portable intelligent handheld terminal wakes up up and can monitor the standby of region Toxic gas radio detection node；

Described portable intelligent handheld terminal is additionally operable in the region that can monitor, and adjusts the transmission way of the first toxic gas information Footpath.

Method the most according to claim 8, it is characterised in that also include utilizing based on having that artificial neural network builds The second toxic gas information that poisonous gas identification model obtains carries out data fusion, is realized at data fusion by below equation Reason:

$\{\begin{array}{c}\stackrel{\&OverBar;}{x}=\underset{i=1}{\overset{n}{\&Sigma;}}{w}_i{z}_i\\ \begin{array}{cc}\underset{i=1}{\overset{n}{\&Sigma;}}{w}_i=1& (0\&le;w_i\&le;1)\end{array}\end{array}---\left(6\right)$

Wherein, x is true value, z_iFor the measured value of cluster result, w_iFor the weighter factor of cluster result,For the number after data fusion According to；

Wherein, for ensureing the zero deflection estimated, thenIts error variance expression formula is:

${\mathrm{\&sigma;}}^2=E\&lsqb;{(x-\stackrel{\&OverBar;}{x})}^2\&rsqb;=E\&lsqb;\underset{i=1}{\overset{n}{\&Sigma;}}{w}_i^2{(x-\stackrel{\&OverBar;}{x})}^2\&rsqb;=\underset{i=1}{\overset{n}{\&Sigma;}}{w}_i^2{\mathrm{\&sigma;}}_i^2---\left(7\right)$

Wherein, weighter factor w_i,

(7) and (8) substitute into (6) acquisition data fusion result is:

$\stackrel{\&OverBar;}{x}=\frac{\underset{i=1}{\overset{n}{\&Sigma;}}{z}_i{\mathrm{\&sigma;}}_i^{-2}}{\underset{i=1}{\overset{n}{\&Sigma;}}{\mathrm{\&sigma;}}_i^{-2}}---\left(9\right).$