CN103630588A - Rapid response method of electrochemical gas sensor - Google Patents

Abstract

The invention provides a rapid response method of an electrochemical gas sensor. The method concretely comprises the following steps: quickly collecting a plurality of gas density values in the gas density transfer transition process to build a discrete grey model (DGM) (1,1), then training a gaussian process regression model by virtue of a grey prediction result and the collected actual gas density values, and finally, acquiring the gas density values in advance based on the trained gaussian process regression models. By the rapid response method, the gas density value can be quickly and accurately obtained, and the response time of the electrochemical gas sensor is greatly shortened.

Description

A kind of fast response method of galvanochemistry firedamp sensor

Technical field

The present invention relates to firedamp sensor technical field, be specifically related to a kind of fast response method of galvanochemistry firedamp sensor.

Background technology

Firedamp sensor is used in the airborne methane content of progress of coal mining continuous monitoring, i.e. gas density.Existing firedamp sensor mainly contains the types such as optical profile type and electric chemical formula (carrier catalyst element formula).At present in China's coal-mine industry, the most extensively and a large amount of what use is electric chemical formula methane transducer.

It is detecting element that electric chemical formula methane transducer adopts carrier catalyst element.During work, the metal sponge that the methane in test environment has seen through buffer action with diffusion way enters sensor probe air chamber and sensitive element generation chemical reaction, thereby produces the electric signal corresponding with methane concentration.This structure of sensor and principle have determined that it has considerable transfer delay.

The dynamic response characteristic of firedamp sensor is to approach gradually measured concentration value according to exponential law.When the methane concentration of gas changes, sensor can not indicate the actual value of methane concentration immediately.The process that needs a gradual change.Until transient process could obtain the actual value of concentration after finishing.And the signal processing systems such as single-chip microcomputer also need certain signal to process and signal transmission time, so firedamp sensor exists certain hysteresis delay time, has operating lag phenomenon.The time delay of all kinds of firedamp sensors (as optical profile type, infrared type, carrier catalyst element formula) is not etc.The time delay of the electric chemical formula methane transducer that China the most generally adopts is at present greatly about 10～30 seconds.

Along with the development of observation and control technology, the performance of sensor has been proposed to more and more higher requirement, sensor must have good dynamic perfromance and quick susceptibility.Coal production field very high to security requirement, need obtain on-the-spot gas density quickly and accurately by firedamp sensor, require firedamp sensor dynamic perfromance high, and time delay is the smaller the better, can respond fast the variation of methane concentration.In prior art, mostly be by improving sensitive element itself to improve the response speed of firedamp sensor, and the inherent characteristic of sensitive material itself makes response speed room for promotion less.

Based on aforementioned analysis, the present invention proposes a kind of fast response method of firedamp sensor, and to reduce firedamp sensor time delay, this case produces thus.

Summary of the invention

Object of the present invention, is to provide a kind of fast response method of galvanochemistry firedamp sensor, and it can shorten the time delay of firedamp sensor, improves the dynamic response characteristic of firedamp sensor, to obtain fast and accurately tested gas concentration.

In order to reach above-mentioned purpose, solution of the present invention is:

A kind of fast response method of galvanochemistry firedamp sensor, comprises the steps:

(1) with galvanochemistry firedamp sensor former time delay of Δ t's for the sampling period, gather m gas concentration T _i, 0<m<<n wherein, m, n are positive integer, i=1,2 ..., m;

(2) m the gas concentration T that utilizes step (1) to gather _iset up discrete grey model DGM (1,1);

(3) utilize the discrete grey model DGM that builds of institute (1,1) in step (2) to carry out forward prediction, obtain gray prediction value

wherein ${\tilde{T}}_1={\mathrm{<figure-callout\; id="1"\; label="T"\; filenames="HDA0000438822900000011.png"\; state="\{\{state\}\}">T</figure-callout>}}_1,0<m<<n;$

(4) the gray prediction value to be obtained in step (3)

for input, actual gas concentration T _ifor output, training Gaussian process regression model;

(5) utilize the gray prediction value of obtaining in step (3)

and step (4) the Gaussian process regression model of training, calculate gas concentration C _nbe gas concentration to be asked, firedamp sensor is existed

time in concentration while having obtained Δ t, be also to have reduced the firedamp sensor response time

wherein 0<k<m<n and k are integer.

A kind of fast response method of galvanochemistry firedamp sensor of the present invention, the concrete grammar of described step (2) is:

First, order $Y=\left[\begin{array}{c}{T}^{\left(2\right)}\left(2\right)\\ T^{\left(2\right)\left(3\right)}\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ T^{\left(2\right)}\left(m\right)\end{array}\right],B=\left[\begin{array}{cc}{T}^{\left(2\right)}\left(1\right)& 1\\ T^{\left(2\right)}\left(2\right)& 1\\ \&CenterDot;& \&CenterDot;\\ \&CenterDot;& \&CenterDot;\\ \&CenterDot;& \&CenterDot;\\ T^{\left(2\right)}(m-1)& 1\end{array}\right],$ Ask for $\mathrm{\&beta;}=\left[\begin{array}{c}{\mathrm{\&beta;}}_1\\ {\mathrm{\&beta;}}_2\end{array}\right]{(B^{\&prime;}\&CenterDot;B)}^{-1}\&CenterDot;B^{\&prime;}\&CenterDot;Y,$ Wherein $T^{\left(2\right)}\left(i\right)=\underset{j=1}{\overset{i}{\mathrm{\&Sigma;}}}{T}^{\left(1\right)}\left(j\right),T^{\left(1\right)}\left(i\right)=\underset{j=1}{\overset{i}{\mathrm{\&Sigma;}}}T\left(j\right),i=\mathrm{1,2},\&CenterDot;\&CenterDot;\&CenterDot;,m,$ B ' is the transposed matrix of B;

Then, setting up discrete grey model DGM (1,1) is: $\left\{\begin{array}{c}\tilde{T}\left(i\right)={{\mathrm{\&beta;}}_1}^{i-2}[({\mathrm{\&beta;}}_1-1)T\left(1\right)+{\mathrm{\&beta;}}_2],i=\mathrm{2,3},\&CenterDot;\&CenterDot;\&CenterDot;,m\\ \tilde{T}\left(i\right)=T\left(1\right),i=1\end{array}\right\}.$

A kind of fast response method of galvanochemistry firedamp sensor of the present invention, in described step (3), the concrete steps of Gaussian process regression model training are:

(3.1) the covariance function K (x that structure Gaussian process returns ⁱ, x ^j) be a square index covariance function K _sE(x ⁱ, x ^j), Rational Quadratic covariance function K _rQwith Matern covariance function K _mthree's quadratic sum, that is:

$K_{\mathrm{SE}}(x^i,x^j)={\mathrm{\&sigma;}}_f^2\&CenterDot;e^{-\frac{1}{2}{(x^i-x^j)}^{\&prime;}\&CenterDot;M(x^i-x^j)}+{\mathrm{\&sigma;}}_n^2{\mathrm{\&delta;}}^{\mathrm{ij}}$

$K_{\mathrm{RQ}}(x^i,x^j)={\mathrm{\&sigma;}}_f^2\&CenterDot;{[1+\frac{{(x^i-x^j)}^{\&prime;}\&CenterDot;M(x^i-x^j)}{2\mathrm{\&alpha;}}]}^{-\mathrm{\&alpha;}}$

$K_M(x^i,x^j)={\mathrm{\&sigma;}}_f^2\&CenterDot;[1+\sqrt{3M}(x^i-x^j)\&CenterDot;e^{\sqrt{3M}(x^i-x^j)}]$

K(x ⁱ,x ^j)=[K _SE(x ⁱ,xj)] ²+[K _RQ(x ⁱ,x ^j)] ²+[K _M(x ⁱ,x ^j)] ²

Wherein, M=diag (l ^-2) be the symmetric matrix of super parameter, l is the super parameter of relevance measure,

σ _n, δ ^ij, α is undetermined parameter, (x ⁱ-x ^j) ^'for (x ⁱ-x ^j) transposed matrix, x ⁱ, x ^jfor input sample, i ≠ j;

(3.2) the gray prediction value of obtaining from step (3)

in choose successively k value for input, take actual gas density as output, form (m-k) individual input and output training sample pair, that is: input sample is

output sample is T _i+k, i=1,2 ..., m-k;

(3.3) utilize (m-k) the individual input and output training sample pair in step (3.2), training obtains the model that Gaussian process regression model is the single output of k input, and wherein 0<k<m and k are integer.

A kind of fast response method of galvanochemistry firedamp sensor of the present invention, described step (5) obtain gas concentration C to be asked _nconcrete grammar be: with the gray prediction value of being obtained in step (3)

the input of the Gaussian process regression model of training for step (4), the output of Gaussian process regression model is gas concentration C to be asked _n.

A kind of fast response method of a kind of galvanochemistry firedamp sensor of the present invention, its specific implementation process is calculated realization by write software program in the data processing chips such as single-chip microcomputer, and without changing the component of firedamp sensor own and structure thereof.

Adopt after such scheme, the present invention transmits the some gas concentration T in transient process by Quick Acquisition gas density ₁, T ₂..., T _mits T.T. used of sampling is far smaller than settling time, thereby in transient process, can obtain final required gas density actual value, and just obtain required gas density actual value when reaching real stable state, can effectively solve the latency issue of firedamp sensor.

Accompanying drawing explanation

Fig. 1 is the process flow diagram of a kind of fast response method of galvanochemistry firedamp sensor.

Embodiment

Below with reference to accompanying drawing, technical scheme of the present invention is elaborated.

The invention provides a kind of fast response method of galvanochemistry firedamp sensor, its general thought is to transmit the some gas concentrations in transient process by Quick Acquisition gas density, employing is based on DGM-GPR(Discrete Grey Model-Gaussian Processes Regression) prediction algorithm obtain in advance gas density, and then when gas transmission transient process does not reach stable state, gas concentration when obtaining stable state in advance, greatly degree has reduced the response time of firedamp sensor.

Take gray theory as basic gray prediction, on processing small sample, poor information, uncertain problem, there is unique advantage.Discrete grey model DGM (1,1), by traditional gray model GM (1,1) precision, has improved the stability of prediction to a certain extent.Gaussian process (Gaussian Processes, GP) as a kind of new machine learning techniques, it is a kind of method based on kernel function, first GP sets up the priori function of model with the form of probability distribution, then under Bayesian frame, realization is the conversion to posteriority function by priori function, and can calculate " the super parameter " of kernel function.Gaussian process regression algorithm has good adaptability to processing the challenges such as small sample, non-linear, high dimension, and reduces the computing time of export target amount predicted value, can be for the quick calculating of firedamp sensor concentration value.

Below, in connection with Fig. 1, step of the present invention is elaborated.A kind of fast response method of galvanochemistry firedamp sensor of the present invention, comprises the following steps:

(1) with galvanochemistry firedamp sensor former time delay of Δ t's for the sampling period, gather m gas concentration T _i, 0<m<<n wherein, m, n are positive integer, i=1,2 ..., m;

(2) m the gas concentration T that utilizes step (1) to gather _iset up discrete grey model DGM (1,1).

First, order $Y=\left[\begin{array}{c}{T}^{\left(2\right)}\left(2\right)\\ T^{\left(2\right)\left(3\right)}\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ T^{\left(2\right)}\left(m\right)\end{array}\right],B=\left[\begin{array}{cc}{T}^{\left(2\right)}\left(1\right)& 1\\ T^{\left(2\right)}\left(2\right)& 1\\ \&CenterDot;& \&CenterDot;\\ \&CenterDot;& \&CenterDot;\\ \&CenterDot;& \&CenterDot;\\ T^{\left(2\right)}(m-1)& 1\end{array}\right],$ Ask for $\mathrm{\&beta;}=\left[\begin{array}{c}{\mathrm{\&beta;}}_1\\ {\mathrm{\&beta;}}_2\end{array}\right]{(B^{\&prime;}\&CenterDot;B)}^{-1}\&CenterDot;B^{\&prime;}\&CenterDot;Y,$ Wherein $T^{\left(2\right)}\left(i\right)=\underset{j=1}{\overset{i}{\mathrm{\&Sigma;}}}{T}^{\left(1\right)}\left(j\right),T^{\left(1\right)}\left(i\right)=\underset{j=1}{\overset{i}{\mathrm{\&Sigma;}}}T\left(j\right),i=\mathrm{1,2},\&CenterDot;\&CenterDot;\&CenterDot;,m,$ B ' is the transposed matrix of B;

Then, setting up discrete grey model DGM (1,1) is: $\left\{\begin{array}{c}\tilde{T}\left(i\right)={{\mathrm{\&beta;}}_1}^{i-2}[({\mathrm{\&beta;}}_1-1)T\left(1\right)+{\mathrm{\&beta;}}_2],i=\mathrm{2,3},\&CenterDot;\&CenterDot;\&CenterDot;,m\\ \tilde{T}\left(i\right)=T\left(1\right),i=1\end{array}\right\};$

Finally, according to built discrete grey model and gas concentration T (1), can ask for each value of gray prediction constantly.

(3) utilize the discrete grey model DGM that builds (1,1) in step (2), carry out forward prediction, obtain gray prediction value

wherein ${\tilde{T}}_1={\mathrm{<figure-callout\; id="1"\; label="T"\; filenames="HDA0000438822900000011.png"\; state="\{\{state\}\}">T</figure-callout>}}_1,0<m<<n.$

(4) the gray prediction value to be obtained in step (3)

for input, actual gas concentration T _ifor output, training Gaussian process regression model, concrete steps are:

(4.1) the covariance function K (x that structure Gaussian process returns ⁱ, x ^j) be a square index covariance function K _sE(x ⁱ, x ^j), Rational Quadratic covariance function K _rQwith Matern covariance function KM three's quadratic sum, that is:

$K_{\mathrm{SE}}(x^i,x^j)={\mathrm{\&sigma;}}_f^2\&CenterDot;e^{-\frac{1}{2}{(x^i-x^j)}^{\&prime;}\&CenterDot;M(x^i-x^j)}+{\mathrm{\&sigma;}}_n^2{\mathrm{\&delta;}}^{\mathrm{ij}}$

$K_{\mathrm{RQ}}(x^i,x^j)={\mathrm{\&sigma;}}_f^2\&CenterDot;{[1+\frac{{(x^i-x^j)}^{\&prime;}\&CenterDot;M(x^i-x^j)}{2\mathrm{\&alpha;}}]}^{-\mathrm{\&alpha;}}$

$K_M(x^i,x^j)={\mathrm{\&sigma;}}_f^2\&CenterDot;[1+\sqrt{3M}(x^i-x^j)\&CenterDot;e^{\sqrt{3M}(x^i-x^j)}]$

K(x ⁱ,x ^j)=[K _SE(x ⁱ,x ^j)] ²+[K _RQ(x ⁱ,x ^j)] ²+[K _M(x ⁱ,x ^j)] ²

Wherein, M=diag (l ^-2) be the symmetric matrix of super parameter, l is the super parameter of relevance measure,

, σ _n, δ ^ij, α is undetermined parameter, (x ⁱ-x ^j) ^￠for (x ⁱ-x ^j) transposed matrix, x ⁱ, x ^jfor input sample, i ≠ j;

Each parameter (l,

, σ _n, δ ^ij, α) value initialization is random value, adopts method of conjugate gradient, by the maximization to the log-likelihood function of training sample (formula (1)), obtains optimized parameter.Method of conjugate gradient is prior art, does not repeat them here.

$L=-\frac{1}{2}{y}^{\&prime;}{K}^{-1}y--\frac{1}{2}\lg \left|K\right|-\frac{N}{2}\lg \left(2\mathrm{\&pi;}\right)$ Formula (1)

In formula (1), y is Gaussian process regression model output data variable, and K is the covariance function setting, and N is number of training, the transposition that y ' is y.

(4.2) the gray prediction value of obtaining from step (3) in choose successively k value for input, take actual gas density as output, form (m-k) individual input and output training sample pair, that is: input sample is

output sample is T _i+k, i=1,2 ..., m-k, utilizes this (m-k) individual input and output training sample pair, and training obtains the model that Gaussian process regression model is the single output of k input, and wherein 0<k<m and k are integer.

(5) the gray prediction value of obtaining from step (3)

in choose

as the input of the Gaussian process regression model of training for step (4), the output of Gaussian process regression model is gas concentration C to be asked _n, firedamp sensor is existed

time in obtained Δ _ttime concentration, be also to have reduced the firedamp sensor response time

wherein 0<k<m<n and k are integer.

A kind of fast response method of galvanochemistry firedamp sensor of the present invention, its specific implementation process is calculated realization by write software program in the data processing chips such as single-chip microcomputer, and without changing the component of firedamp sensor own and structure thereof.

Above embodiment only, for explanation technological thought of the present invention, can not limit protection scope of the present invention with this, every technological thought proposing according to the present invention, and any change of doing on technical scheme basis, within all falling into protection domain of the present invention.

Claims (5)

1. a kind of fast response method of galvanochemistry firedamp sensor, is characterized in that comprising the following steps:

(1) with galvanochemistry firedamp sensor former time delay of Δ t's

for the sampling period, gather m gas concentration T _i, 0<m<<n wherein, m, n are positive integer, i=1,2 ..., m;

(2) m the gas concentration T that utilizes step (1) to gather _iset up discrete grey model DGM (1,1);

(3) utilize the discrete grey model DGM that builds of institute (1,1) in step (2) to carry out forward prediction, obtain gray prediction value ,

wherein

0<m<<n;

(4) the gray prediction value to be obtained in step (3) for input, actual gas concentration T _ifor output, training Gaussian process regression model GPR (Gaussian Processes Regression);

(5) utilize the gray prediction value of obtaining in step (3)

and step (4) the Gaussian process regression model of training, calculate gas concentration C _nbe gas concentration to be asked, firedamp sensor is existed

time in concentration while having obtained Δ t, be also to have reduced the firedamp sensor response time

wherein 0<k<m<n and k are integer.

2. a kind of fast response method of galvanochemistry firedamp sensor as claimed in claim 1, is characterized in that, the concrete grammar of described step (2) is:

First, order $Y=\left[\begin{array}{c}{T}^{\left(2\right)}\left(2\right)\\ T^{\left(2\right)\left(3\right)}\\ \&CenterDot;\\ \&CenterDot;\\ \&CenterDot;\\ T^{\left(2\right)}\left(m\right)\end{array}\right],B=\left[\begin{array}{cc}{T}^{\left(2\right)}\left(1\right)& 1\\ T^{\left(2\right)}\left(2\right)& 1\\ \&CenterDot;& \&CenterDot;\\ \&CenterDot;& \&CenterDot;\\ \&CenterDot;& \&CenterDot;\\ T^{\left(2\right)}(m-1)& 1\end{array}\right],$ Ask for $\mathrm{\&beta;}=\left[\begin{array}{c}{\mathrm{\&beta;}}_1\\ {\mathrm{\&beta;}}_2\end{array}\right]{(B^{\&prime;}\&CenterDot;B)}^{-1}\&CenterDot;B^{\&prime;}\&CenterDot;Y,$ Wherein $T^{\left(2\right)}\left(i\right)=\underset{j=1}{\overset{i}{\mathrm{\&Sigma;}}}{T}^{\left(1\right)}\left(j\right),T^{\left(1\right)}\left(i\right)=\underset{j=1}{\overset{i}{\mathrm{\&Sigma;}}}T\left(j\right),i=\mathrm{1,2},\&CenterDot;\&CenterDot;\&CenterDot;,m,$ B ' is the transposed matrix of B;

Then, setting up discrete grey model DGM (1,1) is: $\left\{\begin{array}{c}\tilde{T}\left(i\right)={{\mathrm{\&beta;}}_1}^{i-2}[({\mathrm{\&beta;}}_1-1)T\left(1\right)+{\mathrm{\&beta;}}_2],i=\mathrm{2,3},\&CenterDot;\&CenterDot;\&CenterDot;,m\\ \tilde{T}\left(i\right)=T\left(1\right),i=1\end{array}\right\}.$

3. a kind of fast response method of galvanochemistry firedamp sensor as claimed in claim 1, is characterized in that, in described step (3), the concrete steps of Gaussian process regression model training are:

(3.1) the covariance function K (x that structure Gaussian process returns ⁱ, x ^j) be a square index covariance function K _sE(x ⁱ, x ^j), Rational Quadratic covariance function K _rQwith Matern covariance function K _mthree's quadratic sum, that is:

$K_{\mathrm{SE}}(x^i,x^j)={\mathrm{\&sigma;}}_f^2\&CenterDot;e^{-\frac{1}{2}{(x^i-x^j)}^{\&prime;}\&CenterDot;M(x^i-x^j)}+{\mathrm{\&sigma;}}_n^2{\mathrm{\&delta;}}^{\mathrm{ij}}$

$K_{\mathrm{RQ}}(x^i,x^j)={\mathrm{\&sigma;}}_f^2\&CenterDot;{[1+\frac{{(x^i-x^j)}^{\&prime;}\&CenterDot;M(x^i-x^j)}{2\mathrm{\&alpha;}}]}^{-\mathrm{\&alpha;}}$

$K_M(x^i,x^j)={\mathrm{\&sigma;}}_f^2\&CenterDot;[1+\sqrt{3M}(x^i-x^j)\&CenterDot;e^{\sqrt{3M}(x^i-x^j)}]$

K(x ⁱ,x ^j)=[K _SE(x ⁱ,x ^j)] ²+[K _RQ(x ⁱ,x ^j)] ²+[K _M(x ⁱ,x ^j)] ²

Wherein, M=diag (l ^-2) be the symmetric matrix of super parameter, l is the super parameter of relevance measure,

σ _n, δ ^ij, α is undetermined parameter, (x ⁱ-x ^j) ^'for (x ⁱ-x ^j) transposed matrix, x ⁱ, x ^jfor input sample, i ≠ _j;

(3.2) the gray prediction value of obtaining from step (3)

in choose successively k value for input, take actual gas density as output, form (m-k) individual input and output training sample pair, that is:

Input sample is:

Output sample is: T _i+k,

(3.3) utilize the individual input and output training sample pair of step (3.2) constructed (m-k), training obtains the model that Gaussian process regression model is the single output of k input, and wherein 0<k<m and k are integer.

4. a kind of fast response method of galvanochemistry firedamp sensor as claimed in claim 1, is characterized in that, the concrete steps of described step (5) are:

With the gray prediction value of being obtained in step described in claim 1 (3)

the input of the Gaussian process regression model of training for step (4), the output of Gaussian process regression model is gas concentration C to be asked _n, firedamp sensor is existed

time in concentration while having obtained Δ t, be also to have reduced the firedamp sensor response time $\frac{n-m}{n}\&CenterDot;\mathrm{\&Delta;t}.$

5. a kind of fast response method of galvanochemistry firedamp sensor as claimed in claim 1, it is characterized in that, its specific implementation process is calculated realization by write software program in the data processing chips such as single-chip microcomputer, and without changing the component of firedamp sensor own and structure thereof.

Images