CN101701850A - Method for detecting temperature and blackness of flame - Google Patents

Abstract

The invention provides a method for detecting the temperature and the blackness of flame, belonging to an analysis method of a flame spectrum, and solving the problem that the existing method can not predict the accuracy of counting results. The method comprises the steps of: (1) measuring: measuring the spectrum output intensity I (lambda) of output flame; (2) modifying: modifying the I (lambda) to obtain spectrum radiation intensity Id (lambda); (3) counting: selecting any two wavelengths with distance of delta lambda from the Id (lambda), and counting to obtain the distribution curve of flame temperature T lambda and flame darkness epsilon lambda along with wavelength lambda; (4) judging: judging whether the flame spectrum radiation intensity belongs to a grey wave band according to the distribution curve of the flame darkness epsilon lambda along with the wavelength lambda, wherein if the flame spectrum radiation intensity belongs to grey wave band, entering into step (5), otherwise giving up the detection; and (5) outputting: outputting the flame temperature and the flame darkness of the flame spectrum radiation belonging to the grey wave band. The method has no interference to the environment of a measuring object, counts to obtain the temperature and the darkness of the flame when the flame is accordant with the grey wave band, and can guarantee the counting accuracy and effectiveness.

Description

The detection method of a kind of flame temperature and blackness

Technical field

The invention belongs to the flame spectrometric analysis method, be specifically related to the detection method of a kind of flame temperature and blackness.

Background technology

In heat radiation was analyzed, grey body was meant the object of spectral absorption and Wavelength-independent.When Fire Radiation is analyzed, a lot of scholars see flame as grey body for convenience, but such grey body hypothesis has no basis, and the grey body of object supposes generally also be to set up, not necessarily set up in other wavelength coverage in certain wavelength coverage.Blindly analyze if can not determine object to meet the wavelength coverage of grey body hypothesis, may produce bigger error, so the wavelength coverage that flame is met the grey body hypothesis judges it is necessary as grey body.

With the fiber spectrometer of ccd array as the spectral measurement element, can obtain the spectral radiance information of internal combustion flame in a big way, realize detection and diagnosis by analysis, for realizing providing condition to the judgement of the wavelength coverage of flame grey body hypothesis to flame to spectrum.

Compare with the image pick-up type flame detector system, fiber spectrometer is simple in structure, cost is low, and data processing amount is also much smaller; And compare with ordinary optical formula flame detector system, the Fire Radiation information that it obtained is much more, more and more is subjected to the attention of Chinese scholars at present.Abroad, Romero C, scholars such as Li X have carried out the spectroscopy measurements electric furnace based on fiber spectrometer, the research that glass is stored the temperature of stove, see Spectrometer-based combustion monitoring for flame stoichiometryand temperature control[J] .Appl Them Eng 2005,25 (5-6): 659-676, equally under being the prerequisite of ash, spectral radiance adopted duochrome method to calculate temperature in this article, but do not calculate the distribution of blackness, do not analyze the foundation that adopts the ash hypothesis yet.

Domestic scholars Cai Xiao Shu has carried out the method research of spectroscopy measurements pulverized coal flame temperature and blackness, sees Cai Xiaoshu, Luo Wude, " research of spectroscopy measurements pulverized coal flame temperature and blackness ", Engineering Thermophysics journal, 2000,21 (6): 779-782; The pulverized coal flame spectrum of this method pairing approximation gray-body radiation adopts the method for least square curve fitting to obtain the medial temperature and the blackness of flame.Domestic scholars all clean the carried out temperature of spectroscopy measurements hydrocarbon gas flame and the method research of blackness, see Zhou Jie, " adopt optimized Algorithm analytic combustion Fire Radiation spectrum to ask for flame temperature ", 2000,15 (3): 223-225, this method supposes that equally flame is gray-body radiation, adopts the Levenberg-Marquardt optimization algorithm curve of spectrum to be analyzed the temperature that obtains flame.Whether said method not only requires spectral radiance to be necessary for ash, and does not provide the distribution of blackness with wavelength, support for ash provides theoretical for spectral radiance in certain wavelength coverage.

Summary of the invention

The present invention proposes the detection method of a kind of flame temperature and blackness, solve existing detection method and do not judging that whether the measurand spectral radiance meets the simplification computation model that has adopted under the situation of ash based on the ash hypothesis, can't expect the problem of result of calculation accuracy.

The detection method of a kind of flame temperature of the present invention and blackness, step is:

(1) measuring process: utilize the curve of spectrum of fiber spectrometer acquisition flame, and the spectrum output intensity I (λ) of the interior flame different wave length of output optical fibre spectrometer measurement scope;

(2) revise step: according to fiber spectrometer correction relational expression f (λ) under the different wave length, the flame spectrum output intensity I (λ) that obtains is revised, revise back spectral radiance I _d(λ) be:

I _d(λ)＝I(λ)/f(λ)；

(3) calculation procedure: to spectral radiance I _d(λ), choose any two wavelength X, the λ+Δ λ that are spaced apart Δ λ, calculate flame temperature T _λDistribution curve with wavelength X:

$T_{\mathrm{\λ}}=-C_2(\frac{1}{\mathrm{\λ}}-\frac{1}{\mathrm{\λ}+\mathrm{\Δ\λ}})/\ln \left(\frac{I_d\left(\mathrm{\λ}\right)}{I_d(\mathrm{\λ}+\mathrm{\Δ\λ})}\frac{{\mathrm{\λ}}^5}{\mathrm{\λ}+\mathrm{\Δ\λ}}\right)$

Planck constant C in the formula ₂=1.4388E-2; I _d(λ), I _d(λ+Δ λ) is respectively the spectral radiance under wavelength X, the λ+Δ λ, wherein 30nm＜Δ λ＜150nm;

Calculate flame blackness ε _λDistribution curve with wavelength X:

ε _λ＝I _d(λ)/I _b(λ，T _ave)，

In the formula, I _d(λ) be spectral radiance under the wavelength X, I _b(λ, T _Ave) be wavelength X and temperature T _AveUnder the black matrix spectral radiance, medial temperature T _AveBe described flame temperature T _λIn the distribution curve with wavelength X, temperature T in the corresponding whole wavelength coverage _λMean value;

(4) determination step: according to flame blackness ε _λWith the distribution curve of wavelength X, judge in the fiber spectrometer measurement range, whether to exist flame blackness not with the wave band of wavelength variations, to be that then this wavelength band flame spectrum radiation belongs to ash, to carry out step (5); Otherwise withdraw from detection;

(5) output step: at flame temperature T _λε _λDistribution curve and flame blackness ε with wavelength X _λOn the distribution curve with wavelength X, the flame spectrum radiation is belonged to the pairing flame temperature T of wavelength band and the flame blackness ε output of ash.

The detection method of described flame temperature and blackness is characterized in that:

In the described correction step, fiber spectrometer correction relational expression f (λ) demarcates under the blackbody radiation light source by fiber spectrometer and obtains under the described different wave length, comprises following substep:

(2.1) software that utilizes fiber spectrometer to carry obtains the curve of spectrum of blackbody furnace under a certain temperature, and the spectrum output intensity I (λ) of output different wave length;

(2.2) obtain the blackbody radiation intensity I of the different wave length under the relevant temperature according to Wien formula _b(λ);

(2.3) obtain spectrometer correction relational expression f (λ)=I (λ)/I under the different wave length _b(λ).

The detection method of described flame temperature and blackness in the described calculation procedure, is calculated flame blackness ε _λDuring with the distribution curve of wavelength X, described medial temperature T _AveBe described flame temperature T _λIn the distribution curve with wavelength X, temperature fluctuation less than 100K and wavelength coverage greater than temperature T in the wavelength coverage of 50nm _λMean value.Temperature fluctuation in this wavelength coverage is less, and the wavelength coverage of calculating is bigger, and its mean value more can be represented its temperature levels, and the error of calculating is littler.

If can not find temperature fluctuation less than 100K and wavelength coverage wavelength coverage, then medial temperature T greater than 50nm _AveBe described flame temperature T _λIn the distribution curve with wavelength X, temperature T in the corresponding whole wavelength coverage _λMean value.

The present invention is based on analysis to flame spectrum, the Fire Radiation characteristic is judged and obtained meeting temperature and blackness in the grey body hypothesis wavelength coverage, it is a kind of contactless measurement, environment to measuring object does not disturb, method with respect to other spectral measurement flame temperature, provide flame and met the wavelength coverage that grey body is supposed, in this wavelength coverage, calculated spectrum corresponding temperature and blackness, can guarantee accuracy and validity that temperature and blackness calculate like this.

Description of drawings

Fig. 1 (a) is that blackbody furnace output intensity and the blackbody radiation intensity of fiber spectrometer under five different temperatures distributes;

Fig. 1 (b) is the demarcation correction factor curve of fiber spectrometer under five different temperatures;

Fiber spectrometer blackbody furnace output spectrum intensity and correction spectral intensity distributed when Fig. 2 was 1230 ℃;

Temperature when Fig. 3 is four different wave lengths intervals distributes with wavelength change;

Fig. 4 is a FB(flow block) of the present invention;

Fig. 5 is a fiber spectrometer spectra collection system schematic;

Fig. 6 is the output spectrum intensity of solidified gasoline flame and revises the spectral intensity distribution;

Fig. 7 (a) is that the temperature of solidified gasoline flame spectrum distributes with wavelength change;

Fig. 7 (b) is that the blackness of solidified gasoline flame spectrum distributes with wavelength change;

Fig. 8 is the output spectrum intensity of red phosphorus flame and revises the spectral intensity distribution;

Fig. 9 (a) is that the temperature of red phosphorus flame spectrum distributes with wavelength change;

Fig. 9 (b) is that the blackness of red phosphorus flame spectrum distributes with wavelength change;

Figure 10 is the spectra collection location arrangements synoptic diagram of coal-burning boiler in the commerical test of the present invention;

Figure 11 is that the correction spectral intensity of eight different acquisition points of coal-burning boiler distributes;

Figure 12 (a) is that the temperature of flame spectrum at #2 collection point place is with wavelength change;

Figure 12 (b) is that the blackness of flame spectrum at #2 collection point place is with wavelength change.

Embodiment

The present invention is further described below in conjunction with drawings and Examples.

(1) model of fiber spectrometer is AvaSpec-2048-USB2 among the present invention, provides every technical parameter of fiber spectrometer in the table 1.

Table 1

Fiber spectrometer adopts blackbody furnace to demarcate for the standard radiant heat source, obtains spectrometer spectrum output intensity I (λ) and spectral radiance I _bCalibration coefficient f (λ) (λ):

f(λ)＝I(λ)/I _b(λ)

Employed blackbody furnace is the horizontal blackbody furnace of WJL-11 that Kunming calm and peaceful instrument and meter company limited makes in the demarcation, and its temperature range is 800 ℃～3000 ℃, and its temperature-controlled precision is 3～4 ℃ in the time of 1000 ℃.

Among Fig. 1 (a) be fiber spectrometer at temperature 1560K between the 1640K degree, distribute every the blackbody furnace spectrum output intensity of 20K and corresponding blackbody radiation intensity.Fig. 1 (b) has provided the fiber spectrometer calibration coefficient curve under the different temperatures in this temperature range, and as can be seen from the figure 5 curves are basic coincidences, illustrate that the calibration coefficient f (λ) of this fiber spectrometer is temperature independent.

(2) to adopt double-colored temperature checking method be to calculate curve of spectrum temperature T under a series of dual wavelengths of Δ λ in the wavelength interval to revised spectral radiance in the present invention _λDistribute with wavelength change, wherein the selection of the wavelength interval Δ λ of dual wavelength can exert an influence to result of calculation, and according to the principle of duochrome method thermometric, two wavelength apart can not be too far away to guarantee that two radiances under the wavelength equate, if but wavelength can bring the very big error of calculation to calculating at a distance of too near.Be that 1230 curves of spectrum of gathering when spending are the span that example is analyzed the wavelength interval of dual wavelength with fiber spectrometer in the blackbody furnace temperature below.

Fig. 2 has provided spectrum output intensity curve and the revised radiation intensity curve of process that fiber spectrometer is gathered when the blackbody furnace temperature is 1230 ℃.In the 1000nm wavelength coverage, choosing Δ λ respectively is a series of wavelength combinations of 10nm, 30nm, 40nm, 60nm, is calculated as follows temperature T at 500nm _λ:

$T_{\mathrm{\λ}}=-C_2(\frac{1}{\mathrm{\λ}}-\frac{1}{\mathrm{\λ}+\mathrm{\Δ\λ}})/\ln \left(\frac{I_d\left(\mathrm{\λ}\right)}{I_d(\mathrm{\λ}+\mathrm{\Δ\λ})}\frac{{\mathrm{\λ}}^5}{\mathrm{\λ}+\mathrm{\Δ\λ}}\right)$

Planck constant C in the formula ₂=1.4388E-2; I _d(λ), I _d(λ+Δ λ) is respectively the spectral radiance under wavelength X, the λ+Δ λ.

Temperature T _λDistribute as shown in Figure 3 with wavelength change.

For to temperature T _λDistribution situation with wavelength is analyzed, and introduces relative mean square deviation σ _TCome the fluctuation situation of analysis temperature distribution curve:

${\mathrm{\σ}}_T=\sqrt{\frac{1}{n-1}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{(T_i-T_{\mathrm{ave}})}^2}/T_{\mathrm{ave}}$

In the formula, T _iBe the arbitrary temp in the temperature distribution history wavelength coverage, n is the number of temperature in the temperature distribution history wavelength coverage, T _AveMean value for all temperature.σ _TMore little, illustrate that then temperature fluctuation is more little, can be used as the characteristic temperature of the curve of spectrum with the mean value of temperature.By different temperatures among Fig. 3 is analyzed as can be seen, when being 10nm in the wavelength interval, the temperature of calculating is very big with the fluctuation of wavelength change curve, and relative mean square deviation reaches 3.98%; When being 30nm, 40nm and 60nm in the wavelength interval, it is little much that temperature curve changes fluctuation, relative mean square deviation is all less, and particularly obvious precision is higher when 40nm and 60nm, and relative mean square deviation is all less than 1%, and both temperature computation results change are very little, medial temperature is identical, and after greater than 30nm, choosing of the result of calculation of temperature and wavelength interval is irrelevant in the wavelength interval in this explanation, and consider that the duochrome method medium wavelength at interval can not be too big, the wavelength interval should be less than 150nm.So the wavelength interval of dual wavelength should all adopt wavelength interval 40nm to come accounting temperature at 30nm in the 150nm scope in the embodiments of the invention in duochrome method thermometric process.

Fig. 4 provides FB(flow block) of the present invention.

Fig. 5 is the synoptic diagram of fiber spectrometer spectra collection system, native system is made up of portable computer, spectrometer, USB connecting line and optical fiber, portable computer is connected by the USB connecting line with spectrometer, and optical fiber is connected on the spectrometer, and optical fiber end can have also and can not have collimation lens.

Below at the solidified gasoline red phosphorus flame of laboratory environment, method of the present invention is described.

1. measure and revise step: Fig. 6 and Fig. 8 and provided the solidified gasoline flame that fiber spectrometer collects and the output spectrum intensity distributions of red phosphorus flame respectively, shown in the black solid-line curve among the figure; Calibration coefficient according to resulting spectrometer is revised the spectrum output intensity that collects, and obtains revising spectral intensity and distributes, shown in black dashed curve among the figure.As can be seen from the figure, less in the spectral radiance intensity of wavelength during less than 500nm, be unfavorable for analysis to spectrum; And it is lower in the signal to noise ratio (S/N ratio) of the curve of spectrum of wavelength during greater than 900nm, this be since this moment near the upper limit of measurement range of spectrometer, the signal sensitivity of spectrum reduces, so chosen wavelength range distributes to the spectral radiance between 900nm at 500nm and analyzes.

2. calculation procedure: wavelength coverage is distributed to the spectral radiance of two kinds of flames between 900nm at 500nm, select wherein two wavelength X of arbitrary interval 40nm ₁, λ ₂, obtain the correction spectral intensity S of their correspondences by fiber spectrometer ₁, S ₂, calculate temperature T _λDistribute with wavelength change, shown in Fig. 7 (a) and 9 (a).

Temperature Distribution to Fig. 7 (a) is analyzed discovery, and the temperature of solidified gasoline flame is less to the temperature fluctuation between 800nm at 550nm, and relative mean square deviation only is 1.07%, with the medial temperature in this wavelength coverage as characteristic temperature T _AveBring formula ε into _λ=I _d(λ)/I _b(λ, T _Ave) in, then can obtain blackness ε _λDistribute with wavelength change, shown in Fig. 7 (b).In the formula, I _d(λ) be spectral radiance under the wavelength X, I _b(λ, T _Ave) be wavelength X and temperature T _AveUnder the black matrix spectral radiance.

Equally for the Temperature Distribution among Fig. 9 (a), chosen wavelength range 500nm to the mean value of the temperature between 550nm as characteristic temperature T _Ave, calculate blackness ε _λDistribute with wavelength change, shown in Fig. 9 (b).

Determination step: distributing with wavelength change according to the flame blackness that obtains, whether ash is judged for spectral radiance to flame, if flame blackness is constant substantially with wavelength in certain wavelength coverage, think that then this medium is the ash medium in this wavelength coverage, if flame blackness changes with wavelength variations in certain wavelength coverage, think that then this medium is non-ash medium in this wavelength coverage.The same relative mean square deviation σ that introduces blackness _εAnalyze blackness with the fluctuation situation that wavelength change distributes, be shown below:

${\mathrm{\σ}}_{\mathrm{\ϵ}}=\sqrt{\frac{1}{m-1}\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{({\mathrm{\ϵ}}_i-{\mathrm{\ϵ}}_{\mathrm{ave}})}^2}/{\mathrm{\ϵ}}_{\mathrm{ave}}$

Solidified gasoline flame blackness distribution situation analysis among Fig. 7 (b) is found, very little at 500nm to the variation that the blackness in the 900nm scope distributes, relative mean square deviation only is 0.85%, can think that blackness is constant with wavelength change in this wavelength coverage, that is to say that solidified gasoline flame can be regarded grey body medium as in this wavelength coverage.And the red phosphorus flame blackness distribution situation analysis among Fig. 9 (b) is found, the variation of blackness is bigger in the measurement range of spectrometer, there be not the flame blackness wavelength coverage constant, illustrate that red phosphorus flame can not regard grey body medium as in 200 to 1000nm scopes with wavelength variations.

3. output step: in flame can be regarded as the wavelength coverage of grey body medium, adopt duochrome method to calculate temperature T under the different wave length _λWith blackness ε _λ, also export as spectral radiance distribution corresponding temperature T and blackness ε with mean value.For example for solidified gasoline flame, 500nm is exactly the temperature and the blackness of curve of spectrum correspondence to the mean value 1731.76K of the temperature between 900nm and the mean value 0.02 of blackness, and can not adopt temperature and the blackness that calculates flame spectrum curve correspondence based on the method for ash hypothesis for red phosphorus flame.

As shown in figure 10, on a 300MW unit coal-burning boiler, carry out type approval test of the present invention.This boiler is to manufacture and design by the technology of Harbin Boiler Plant introduction U.S. combustion enginnering company and in conjunction with China's coal characteristics, adopt the corner tangential firing mode, in the burner hearth of boiler on four jiaos, arrange two-layer totally 8 burners, represent with the thick rectangle frame of black respectively, the position of spectra collection be exactly 12.97 meters A places of absolute altitude and 16.08 meters B places of absolute altitude four angles of boiler on, the order of following one deck collection is collection point 1,2,3,4, and the order of last layer collection is collection point 5,6,7,8.

The spectral radiance that Figure 11 has provided through 8 collection points after the calibration coefficient finishing distributes, and represents each collection point with different marks respectively; Select wavelength to analyze to the spectral radiance data between 1000nm at 500nm.Below just be distributed as that example is calculated the pairing temperature of the curve of spectrum and blackness distributes with the spectral radiance of collection point 2.Two wavelength of choosing wavelength interval 40nm arbitrarily calculate temperature with wavelength change, the result is shown in Figure 12 (a), as can be seen from the figure, very little at 600nm to the temperature variation in the 900nm wavelength coverage, relative mean square deviation has only 1.62%, with the characteristic temperature of medial temperature 1361.8K as spectral distribution, calculate blackness at wavelength 600nm to the distribution between the 1000nm scope, shown in Figure 12 (b).As can be seen from the figure, the variation of blackness in this wavelength coverage is very little, and relative mean square deviation has only 2.49%, and this explanation pulverized coal flame in this wavelength coverage can be regarded grey body medium as.

For the accuracy of verifying that the inventive method is calculated, analytical calculation is carried out in the spectral radiance distribution of 8 collection points obtained corresponding medial temperature, the relative mean square deviation of temperature, average blackness, the relative mean square deviation of blackness, and adopt the method in the aforementioned documents " research of spectroscopy measurements pulverized coal flame temperature and blackness " to calculate corresponding characteristic temperature and feature blackness, the result is as shown in table 2, label 1～8 is represented 8 collection points respectively in the table, and the temperature of the present invention and two kinds of methods of documents and the relative error of blackness are respectively in table 2 the 7th, the 9th row provides.

Table 2

As can be seen from Table 2, the temperature that the method is calculated and the relative mean square deviation of blackness are all less, except #7 since spectral intensity curve signal to noise ratio (S/N ratio) lower, the error that causes result of calculation is than outside big, the maximum relative mean square deviation of temperature only is 4.32%, the maximum relative mean square deviation of blackness only is 3.15%, and the relative error of this method and documents method result calculated is also less, except the result of #7, the maximum relative error of temperature is 0.48%, the maximum relative error of blackness is 3.33%.

The result of above embodiment shows, the inventive method not only can satisfy the wavelength coverage of ash hypothesis by the analysis of flame spectrum radiation intensity distribution being judged Fire Radiation, and the temperature and the blackness of the flame spectrum correspondence that in this wavelength coverage, calculates, the result is more accurate and effective.

Claims (3)

1. the detection method of flame temperature and blackness, step is:

(1) measuring process: utilize the curve of spectrum of fiber spectrometer acquisition flame, and the spectrum output intensity I (λ) of the interior flame different wave length of output optical fibre spectrometer measurement scope;

(2) revise step: according to fiber spectrometer correction relational expression f (λ) under the different wave length, the flame spectrum output intensity I (λ) that obtains is revised, revise back spectral radiance I _d(λ) be:

I _d(λ)＝I(λ)/f(λ)；

(3) calculation procedure: to spectral radiance I _d(λ), choose any two wavelength X, the λ+Δ λ that are spaced apart Δ λ, calculate flame temperature T _λDistribution curve with wavelength X:

$T_{\mathrm{\λ}}=-C_2(\frac{1}{\mathrm{\λ}}-\frac{1}{\mathrm{\λ}+\mathrm{\Δ\λ}})/\ln \left(\frac{I_d\left(\mathrm{\λ}\right)}{I_d(\mathrm{\λ}+\mathrm{\Δ\λ})}\frac{{\mathrm{\λ}}^5}{\mathrm{\λ}+\mathrm{\Δ\λ}}\right)$

Planck constant C in the formula ₂=1.4388E-2; I _d(λ), I _d(λ+Δ λ) is respectively the spectral radiance under wavelength X, the λ+Δ λ, wherein 30nm＜Δ λ＜150nm;

Calculate flame blackness ε _λDistribution curve with wavelength X:

ε _λ＝I _d(λ)/I _b(λ，T _ave)，

In the formula, I _d(λ) be spectral radiance under the wavelength X, I _b(λ, T _Ave) be wavelength X and temperature T _AveUnder the black matrix spectral radiance, medial temperature T _AveBe described flame temperature T _λIn the distribution curve with wavelength X, temperature T in the corresponding whole wavelength coverage _λMean value;

(4) determination step: according to flame blackness ε _λWith the distribution curve of wavelength X, judge in the fiber spectrometer measurement range, whether to exist flame blackness not with the wave band of wavelength variations, to be that then this wavelength band flame spectrum radiation belongs to ash, to carry out step (5); Otherwise withdraw from detection;

(5) output step: at flame temperature T _λε _λDistribution curve and flame blackness ε with wavelength X _λOn the distribution curve with wavelength X, the flame spectrum radiation is belonged to the pairing flame temperature T of wavelength band and the flame blackness ε output of ash.

2. the detection method of flame temperature as claimed in claim 1 and blackness is characterized in that:

In the described correction step, fiber spectrometer correction relational expression f (λ) demarcates under the blackbody radiation light source by fiber spectrometer and obtains under the described different wave length, comprises following substep:

(2.1) software that utilizes fiber spectrometer to carry obtains the curve of spectrum of blackbody furnace under a certain temperature, and the spectrum output intensity I (λ) of output different wave length;

(2.2) obtain the blackbody radiation intensity I of the different wave length under the relevant temperature according to Wien formula _b(λ);

(2.3) obtain spectrometer correction relational expression f (λ)=I (λ)/I under the different wave length _b(λ).

3. the detection method of flame temperature as claimed in claim 1 or 2 and blackness is characterized in that:

In the described calculation procedure, calculate flame blackness ε _λDuring with the distribution curve of wavelength X, described medial temperature T _AveBe described flame temperature T _λIn the distribution curve with wavelength X, temperature fluctuation less than 100K and wavelength coverage greater than temperature T in the wavelength coverage of 50nm _λMean value.

Images