Spectral gas parameter measurement method and device based on signal power spectrum analysis

Abstract

The present invention relates to a spectral gas parameter measurement method and device based on signal power spectrum analysis, the method comprising: obtaining modulated light; dividing the modulated light into three beams through an optical fiber beam splitter, one beam passing through a measuring cell to obtain a transmitted light intensity signal #imgabs0# containing gas concentration information; one beam obtaining a background light intensity signal I0(t); and another beam obtaining an etalon signal I containing a time-frequency relationship. υ (t); I υ (t) Combined with the initial set concentration value X, the spectral absorbance signal α(υ) is converted; I0(t) is combined with α(υ), and the simulated transmitted light intensity signal #imgabs1# is obtained according to the Beer-Lambert law. The power spectral density analysis algorithm is used for analysis #imgabs2# The power ratio R of the second harmonic to the first harmonic in the frequency domain is obtained. S , and then get the representation of X and the corresponding R S The relationship between S ‑X database; use the power spectrum density analysis algorithm to analyze #imgabs3# and get the power ratio R of the second harmonic to the first harmonic in the frequency domain M , R M Substitute R S ‑X database, interpolation to obtain the measured value of gas concentration X M . The sensitivity and accuracy of gas parameter measurement are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of gas detection, and in particular to a spectral gas parameter measurement method and device based on signal power spectrum analysis. Background Art

[0002] In the prior art, gas detection related technologies are mainly divided into two categories according to the measurement principle: spectroscopy and non-spectroscopy. Non-spectroscopy mainly includes chemical analysis and electrical gas detection. Chemical analysis can effectively separate mixed gases, but the response time is slow. Electrical gas detection has high sensitivity and wide detection range, but it needs to contact the measurement environment and has poor ability to identify gas types. Spectroscopy measures the relevant parameters of the target gas spectrum and inverts the gas parameter information. It can detect many types of gases and has the advantages of non-contact measurement and fast response time. Spectroscopy mainly includes Fourier transform infrared absorption spectroscopy (FTIR) and tunable semiconductor laser absorption spectroscopy (TDLAS). FTIR technology is mainly based on the principle of Michelson interferometer. The infrared light source is collimated by a collimating lens to emit parallel light, which is absorbed by the gas to be measured and received by the telescope system. Then it is converged to the detector through the interferometer to obtain the interference signal of the gas to be measured. After Fourier transform, the absorption spectrum information of the gas at different concentrations can be obtained, so as to calculate the concentration of the gas. However, FTIR equipment is relatively large, the response speed is relatively slow, and the price is relatively expensive. TDLAS technology is a spectral measurement method based on the narrow linewidth characteristics of semiconductor DFB lasers. It can realize the simultaneous measurement of multiple components and multiple parameters of mixed gases. It has strong versatility and high measurement resolution. The concentration of trace gases can be measured by selecting the appropriate characteristic absorption spectrum of the gas to be measured. Among them, the measurement technology of wavelength modulation spectroscopy (WMS) based on TDLAS has the advantages of non-invasiveness and non-interference to the measurement object, fast response time, strong anti-electromagnetic interference ability, small space occupation and easy installation of measurement instrument and equipment system, and has been widely used to monitor the temperature and component concentration of combustion fields. In order to accurately obtain the target spectrum without calibration, many scholars have tried 1f normalization, multi-parameter harmonic fitting based on Fourier transform, absorption line correction, center wavelength drift correction and other processing methods. Among them, the traditional calibration-free 2f / 1f method is the most commonly used due to its advantages such as easy operation, no need for complex analysis models, and applicability to various measurement environments.

[0003] However, in the detection of gas parameters in complex environments, due to the influence of interference factors such as turbulence and strong vibration in the environment, the signal measured by the traditional WMS method will be distorted. The secondary signal extracted by the commonly used harmonic extraction method has large errors or even exceeds the specified range when inverting gas parameter related information. When there are mechanical vibrations, beam jitter, and window contamination in the measurement environment, the measured normalized second harmonic signal is easily contaminated by noise, resulting in reduced measurement sensitivity and accuracy. In order to eliminate noise and enhance detection sensitivity at the same time, data averaging is often used in calibration-free absorption spectroscopy. However, analysis shows that environmental noise not only causes harmonic fluctuations, but also causes harmonic pseudo-enhancement. As a result, averaging the measurement data can only improve the measurement accuracy, but it is powerless against the systematic errors caused by noise. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a spectral gas parameter measurement method and device based on signal power spectrum analysis, aiming to enhance the detection sensitivity and gas parameter detection accuracy of the gas parameter measurement system based on calibration-free wavelength modulation, while improving the noise resistance of the measurement system.

[0005] The technical solution adopted by the present invention is as follows:

[0006] On the one hand, the present application provides a spectral gas parameter measurement method based on signal power spectrum analysis, comprising the following steps:

[0007] (1) The function generator will sweep the frequency f s The signal superposition modulation frequency is f m The signal is input into the laser controller, and the laser controller tunes the output wavelength and light intensity of the DFB laser;

[0008] (2) The modulated light obtained in step (1) is divided into three light signals by a fiber optic beam splitter. One light signal is emitted by the laser collimator transmitting end I, passes through the measuring cell and is received by the photodetector I to obtain a transmitted light intensity signal I containing gas concentration information. t M (t); one beam is emitted from the laser collimator emitting end II and directly received by the photodetector II to obtain the background light intensity signal I0(t) without absorption information; the other beam is emitted from the laser collimator emitting end III and received by the photodetector III after passing through the optical etalon to obtain the etalon signal I containing the time-frequency relationship υ (t);

[0009] (3) The etalon signal I υ (t) converting the wavelength-based spectral absorbance signal α(υ) by combining the HITRAN database, the initial set concentration value X and other known parameters;

[0010] (4) Combining the background light intensity signal I0(t) with the spectral absorbance signal α(υ), and obtaining a simulated transmission light intensity signal containing gas-related information according to the Beer-Lambert law

[0011] (5) The simulated transmission light intensity signal The power spectrum density analysis algorithm is used to analyze and obtain The power ratio R of the second harmonic to the first harmonic in the frequency domain S , so that the R corresponding to different initial set concentration values ​​X can be obtained S , establish the representation of X and R S The simulation of the relationship between S -X database;

[0012] (6) The transmitted light intensity signal I obtained by step (2) t M (t) The power spectrum density analysis algorithm is also used for analysis, and I t M (t) The power ratio R of the second harmonic to the first harmonic in the frequency domain M , R M Substitute the R obtained in step (5) into S -X database, obtain the measured value X of gas concentration through interpolation M .

[0013] Further technical solutions are:

[0014] The power ratio R of the second harmonic to the first harmonic in the frequency domain S Calculated by the following formula:

[0015]

[0016] Where P 2f , P 1f are the second harmonic and the first harmonic respectively, a f is the amplitude of the transmitted light intensity after Fourier transformation, f m is the modulation frequency, f c1 、f c2 are the cutoff frequencies of the first harmonic and the second harmonic respectively; λ1 and λ2 are the average values ​​of the power spectral density of the transmitted light intensity in the non-absorption frequency range near the first harmonic and the second harmonic respectively.

[0017] λ1 can be calculated by the following formula:

[0018]

[0019] λ2 can be calculated by the following formula:

[0020]

[0021] On the other hand, the present application provides a spectral gas parameter measurement device based on signal power spectrum analysis, comprising a function generator, a laser controller, a DFB laser, a fiber beam splitter, three laser collimator transmitting ends, a measurement cell containing a gas to be measured, an optical etalon, three photodetectors, a data acquisition card and a computer;

[0022] The function generator inputs the scanning signal superimposed with the modulation signal into the laser controller through its internal channel;

[0023] The laser controller simultaneously tunes the output wavelength and light intensity of the DFB laser;

[0024] The laser emitted by the DFB laser is divided into three light signals by the two optical fiber beam splitters. One light signal is emitted by the laser collimator emitting end I, passes through the measuring cell, is received by the photodetector I and converted into an electrical signal, and a transmission light intensity signal I containing gas concentration information is obtained. t M (t);

[0025] One beam is emitted from the laser collimator emitting end II and directly received by the photodetector II to obtain the background light intensity signal I0(t) without absorption information;

[0026] The other beam is emitted by the laser collimator emission end III, passes through the optical etalon 7 and is received by the photodetector III to obtain the etalon signal I containing the time-frequency relationship. υ (t);

[0027] All three signals are collected by the data acquisition card and input into the computer for real-time analysis or stored for other subsequent signal analysis and processing;

[0028] It also includes a signal analysis and processing module, which is used to:

[0029] The etalon signal I υ (t) converting the wavelength-based spectral absorbance signal α(υ) by combining the HITRAN database, the initial set concentration value X and other known parameters;

[0030] The background light intensity signal I0(t) is combined with the spectral absorbance signal α(υ) to obtain a simulated transmission light intensity signal containing gas-related information according to the Beer-Lambert law.

[0031] The simulated transmitted light intensity signal The power spectrum density analysis algorithm is used to analyze and obtain The power ratio R of the second harmonic to the first harmonic in the frequency domain S , establish the R corresponding to different initial set concentration values ​​X S , and get the simulated R S -X database;

[0032] The transmitted light intensity signal The power spectrum density analysis algorithm is also used for analysis, and we get The power ratio R of the second harmonic to the first harmonic in the frequency domain M , R M Substitute the R S -X database, obtain the measured value X of gas concentration through interpolation M .

[0033] The beneficial effects of the present invention are as follows:

[0034] Compared with the traditional calibration-free wavelength modulation spectroscopy technology based on first and second harmonic signals, the method proposed in the present invention only needs to obtain the power ratio of the second frequency and the fundamental frequency of the measurement signal through the power spectrum analysis algorithm in the frequency domain, and can calculate the concentration of the gas by interpolation in the pre-simulated database, thereby improving the sensitivity and accuracy of gas parameter measurement; at the same time, due to the sparse characteristics of the modulated signal power spectrum, the present invention can eliminate the influence of noise on the measurement results to a certain extent, and expand the application scope of the spectral absorption method; the method of the present invention has the characteristics of good applicability and wide application scenarios, so the method of the present invention can be tried for the diagnostic measurement of flame temperature and component concentration in complex environments such as aerospace engine combustion chambers.

[0035] Other features and advantages of the present invention will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a schematic diagram of the process of the present invention.

[0037] Figure 2 The figures are the measurement results obtained by continuously measuring methane with a concentration of 1.10% using the traditional method and the method of the present invention.

[0038] Figure 3 The figure shows the analysis results of Allan variance versus integration time of the measurement results using the traditional method and the method of the present invention.

[0039] Figure 4 It is a schematic diagram of the structure of the device of the present invention. DETAILED DESCRIPTION

[0040] The specific implementation of the present invention is described below with reference to the accompanying drawings.

[0041] The combustion chamber is an important component of an aircraft engine. It releases the chemical energy in the fuel through combustion, converts the chemical energy into thermal energy, produces high-temperature, high-pressure combustion gas, and improves the engine's working capacity. When designing a combustion chamber, it is necessary to ensure combustion characteristics such as good combustion stability, high combustion efficiency, and low emission pollution. The combustion products of the aircraft engine combustion chamber mainly include residual air (O2 and N2), carbon oxides (CO and CO2), nitrogen oxides (NO and NO2), unburned hydrocarbon compounds (TCH) and other gases and solid fine particles. Characteristic gases are important indicative gases that reflect combustion characteristics. Its presence indicates that the fuel is not completely burned and the combustion efficiency needs to be improved. The detection of characteristic gas parameters provides important data for optimizing combustion efficiency, obtaining combustion chamber design, and selecting fuel equivalence ratio. In this regard, the present application provides a spectral gas parameter measurement method based on signal power spectrum analysis, see Figure 1 , including the following steps:

[0042] (1) The function generator will sweep the frequency f s The signal superposition modulation frequency is f m The signal is input into the laser controller, and the laser controller tunes the output wavelength and light intensity of the DFB laser;

[0043] (2) The modulated light obtained in step (1) is divided into three light signals by a fiber optic beam splitter. One light signal is emitted by the laser collimator transmitting end I, passes through the measuring cell and is received by the photodetector I to obtain a transmitted light intensity signal I containing gas concentration information. t M (t); one beam is emitted from the laser collimator emitting end II and directly received by the photodetector II to obtain the background light intensity signal I0(t) without absorption information; the other beam is emitted from the laser collimator emitting end III and received by the photodetector III after passing through the optical etalon to obtain the etalon signal I containing the time-frequency relationship υ (t);

[0044] (3) The standard signal I υ (t) is subjected to peak fitting processing to obtain the wave number signal v(t) that varies with time. Then, combined with the information in the HITRAN database, the initial set concentration value X and other known parameters, the wave number-based spectral absorbance signal α(υ) is obtained by the following conversion:

[0045]

[0046] Where, P, T, L, S(T), and v0 are respectively the ambient pressure, temperature, measurement area length, line intensity, and spectral line center wave number; T is temperature; is the absorption spectrum line function, Δv C is the collision broadening, Δv D is Doppler broadening.

[0047] (4) The background light intensity signal I0(t) is combined with the spectral absorbance signal α(υ) to obtain a simulated transmission light intensity signal containing gas-related information according to the Beer-Lambert law as follows:

[0048]

[0049] (5) The simulated transmission light intensity signal The power spectral density analysis algorithm (PSD) is used for analysis and the The power ratio R of the second harmonic to the first harmonic in the frequency domain S , so as to obtain the R corresponding to different initial set concentration values ​​X S , through fitting, we can get the form of R S =kX+b represents X and R S The simulation of the relationship between S -X database;

[0050] The power ratio R of the second harmonic to the first harmonic in the frequency domain S Calculated by the following formula:

[0051]

[0052] Where P 2f , P 1f are the second harmonic and the first harmonic respectively, a f and b f are the transmitted light intensity and the amplitude of the noise after Fourier transformation, respectively. m is the modulation frequency, f c1 and f c2 are the cutoff frequencies of the first and second harmonics, respectively;

[0053] The noise term in the equation can be obtained in the frequency range that is irrelevant to absorption in the power spectrum density of the transmitted light intensity. Therefore, the above equation can be transformed into the following equation:

[0054]

[0055] Where λ1 and λ2 are the average values ​​of the power spectral density of the transmitted light intensity in the non-absorption frequency range near the first harmonic and the second harmonic, respectively.

[0056] λ1 can be calculated by the following formula:

[0057]

[0058] λ2 can be calculated by the following formula:

[0059]

[0060] (6) The transmitted light intensity signal measured in step (2) The power spectrum density analysis algorithm is also used for analysis, and we get The power ratio R of the second harmonic to the first harmonic in the frequency domain M , R M Substitute the R obtained in step (5) as the ordinate S =kX+b represents X and R S The database of the relationship between them is used to obtain the measured value X of the abscissa, i.e. the gas concentration, by interpolation. M .

[0061] like Figure 2 As shown, the measurement results obtained by continuously measuring methane with a concentration of 1.10% using the traditional 2f / 1f method and the method of this embodiment.

[0062] The measurement conditions are the same, all under the influence of noise in the actual measurement environment. Figure 2 It can be seen that the final measurement values ​​of the two measurement methods have certain fluctuations, but it can be clearly seen that the measurement results of the PSD analysis method proposed in this embodiment have smaller errors and smaller fluctuations than those of the traditional method, which increases the detection sensitivity of the gas parameter measurement system and the detection accuracy of the gas parameters.

[0063] like Figure 3 As shown, the analysis results of the Allan variance of the measurement results using the traditional method and the method of this embodiment versus the integration time. It can be seen that within the integration time range of 1s to 60s, the performance of the PSD analysis method of this embodiment is always better than the traditional 2f / 1f method. It can be seen from the Allan variance diagram that when the integration time is 1s, the CH4 gas detection limits of the traditional 2f / 1f method and the PSD analysis method of this embodiment are 31.34ppmv and 11.85ppmv, respectively; when the integration time is 60s, the CH4 gas detection limits of the traditional 2f / 1f method and the PSD analysis method of this embodiment are reduced to 8.20ppmv and 6.79ppmv, respectively. That is, the method proposed in this application has a lower detection limit.

[0064] The present application also provides a spectral gas parameter measurement device based on signal power spectrum analysis, see Figure 4, including a function generator 1, a laser controller 2, a DFB laser 3, a fiber beam splitter 4, three laser collimator emitting ends 5, a measuring cell 6 containing a gas to be measured, an optical etalon 7, three photoelectric detectors 8, a data acquisition card 9 and a computer 10;

[0065] The function generator 1 inputs the scanning signal superimposed with the modulation signal into the laser controller 2 through its internal channel;

[0066] The laser controller 2 simultaneously tunes the output wavelength and light intensity of the DFB laser 3;

[0067] The laser emitted by the DFB laser 3 is divided into three light signals by two optical fiber beam splitters 4. One beam is emitted by the laser collimator emitting end I, passes through the measuring cell, is received by the photodetector I and converted into an electrical signal, and a transmitted light intensity signal I containing gas concentration information is obtained. t M (t);

[0068] One beam is emitted from the laser collimator emitting end II and directly received by the photodetector II to obtain the background light intensity signal I0(t) without absorption information;

[0069] The other beam is emitted by the laser collimator emission end III, passes through the optical etalon 7 and is received by the photodetector III to obtain the etalon signal I containing the time-frequency relationship. υ (t);

[0070] The three signals are collected by the data acquisition card 9 and input into the computer 10 for real-time analysis or storage for other subsequent signal analysis and processing;

[0071] It also includes a signal analysis and processing module, which is used to:

[0072] The etalon signal I υ (t) converting the wavelength-based spectral absorbance signal α(υ) by combining the HITRAN database, the initial set concentration value X and other known parameters;

[0073] The background light intensity signal I0(t) is combined with the spectral absorbance signal α(υ) to obtain the simulated transmission light intensity signal containing gas-related information according to the Beer-Lambert law.

[0074] For the simulated transmitted light intensity signal The power spectrum density analysis algorithm is used to analyze and obtain The power ratio R of the second harmonic to the first harmonic in the frequency domain S , establish the R corresponding to different initial set concentration values ​​X S , and get the simulated RS -X database;

[0075] Transmitted light intensity signal The power spectrum density analysis algorithm is also used for analysis, and we get The power ratio R of the second harmonic to the first harmonic in the frequency domain M , R M Substitute R S -X database, obtain the measured value X of gas concentration through interpolation M .

[0076] The DFB laser 3, i.e., the distributed feedback laser, can continuously emit stable laser light according to set conditions, and its characteristic output wavelength needs to be determined according to the gas to be measured.

[0077] Those skilled in the art can understand that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention is described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions recorded in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A spectral gas parameter measurement method based on signal power spectrum analysis, characterized in that: The following steps are involved: (1) The function generator will sweep the frequency f s The signal superposition modulation frequency is f m The signal is input into the laser controller, and the laser controller tunes the output wavelength and light intensity of the DFB laser; (2) The modulated light obtained in step (1) is divided into three light signals by a fiber optic beam splitter. One light signal is emitted by the laser collimator transmitting end I, passes through the measuring cell and is received by the photodetector I to obtain a transmitted light intensity signal I containing gas concentration information. t M (t); one beam is emitted from the laser collimator emitting end II and directly received by the photodetector II to obtain the background light intensity signal I0(t) without absorption information; the other beam is emitted from the laser collimator emitting end III and received by the photodetector III after passing through the optical etalon to obtain the etalon signal I containing the time-frequency relationship υ (t); (3) The etalon signal I υ (t) converting the wavelength-based spectral absorbance signal α(υ) by combining the HITRAN database, the initial set concentration value X and other known parameters; (4) Combining the background light intensity signal I0(t) with the spectral absorbance signal α(υ), and obtaining a simulated transmission light intensity signal containing gas-related information according to the Beer-Lambert law (5) The simulated transmission light intensity signal The power spectrum density analysis algorithm is used to analyze and obtain The power ratio R of the second harmonic to the first harmonic in the frequency domain S , so that the R corresponding to different initial set concentration values ​​X can be obtained S , establish the relationship between X and R S The simulation of the relationship between S -X database; (6) The transmitted light intensity signal I obtained by step (2) t M (t) The power spectrum density analysis algorithm is also used for analysis, and I t M (t) The power ratio R of the second harmonic to the first harmonic in the frequency domain M , R M Substitute the R obtained in step (5) into S -X database, obtain the measured value X of gas concentration through interpolation M ; The power ratio R of the second harmonic to the first harmonic in the frequency domain S Calculated by the following formula: Where P 2f , P 1f are the second harmonic and the first harmonic respectively, a f is the amplitude of the transmitted light intensity after Fourier transformation, f m is the modulation frequency, f c1 、f c2 are the cutoff frequencies of the first harmonic and the second harmonic respectively; λ1 and λ2 are the average values ​​of the power spectrum density of the transmitted light intensity in the non-absorption frequency range near the first harmonic and the second harmonic respectively; λ1 is calculated by the following formula: λ2 is calculated by the following formula:

2. A device for implementing the spectral gas parameter measurement method based on signal power spectrum analysis according to claim 1, characterized in that: It includes a function generator, a laser controller, a DFB laser, a fiber beam splitter, three laser collimator emitting ends, a measuring cell containing the gas to be measured, an optical etalon, three photoelectric detectors, a data acquisition card and a computer; The function generator inputs the scanning signal superimposed with the modulation signal into the laser controller through its internal channel; The laser controller simultaneously tunes the output wavelength and light intensity of the DFB laser; The laser emitted by the DFB laser is divided into three light signals by the two optical fiber beam splitters. One light signal is emitted by the laser collimator emitting end I, passes through the measuring cell, is received by the photodetector I and converted into an electrical signal, and a transmission light intensity signal I containing gas concentration information is obtained. t M (t); One beam is emitted from the laser collimator emitting end II and directly received by the photodetector II to obtain the background light intensity signal I0(t) without absorption information; The other beam is emitted by the laser collimator emission end III, passes through the optical etalon 7 and is received by the photodetector III to obtain the etalon signal I containing the time-frequency relationship. υ (t); All three signals are collected by the data acquisition card and input into the computer for real-time analysis or stored for other subsequent signal analysis and processing; It also includes a signal analysis and processing module, which is used to: The etalon signal I υ (t) converting the wavelength-based spectral absorbance signal α(υ) by combining the HITRAN database, the initial set concentration value X and other known parameters; The background light intensity signal I0(t) is combined with the spectral absorbance signal α(υ) to obtain a simulated transmission light intensity signal containing gas-related information according to the Beer-Lambert law. The simulated transmitted light intensity signal The power spectrum density analysis algorithm is used to analyze and obtain The power ratio R of the second harmonic to the first harmonic in the frequency domain S , establish the R corresponding to different initial set concentration values ​​X S , and get the simulated R S -X database; The transmitted light intensity signal I t M (t) The power spectrum density analysis algorithm is also used for analysis, and I t M (t) The power ratio R of the second harmonic to the first harmonic in the frequency domain M , R M Substitute the R S -X database, obtain the measured value X of gas concentration through interpolation M .

Images