1. A portable greenhouse gas detection system based on a spectrum reconstruction technology is composed of an infrared light source (1), an input optical fiber (2), an absorption air chamber (3), an air pump (4), a sensor (5), an output optical fiber (6), an optical fiber collimator (7), a focusing lens (8), a photoelectric detector (9), a transimpedance amplifier (10), an analog-digital conversion circuit (11), a bias circuit (12) and a signal processing system (13); in the system, infrared light emitted by an infrared light source (1) enters an absorption air chamber (3) through an input optical fiber (2), gas to be detected is introduced into the absorption air chamber (3) at a certain speed and air pressure through an air pump (4), a sensor (5) is used for detecting the temperature and the pressure of the gas, the light emitted from the absorption air chamber (3) irradiates a single photoelectric detector (9) through an output optical fiber (6), an optical fiber collimator (7) and a focusing lens (8), the photocurrent generated by the photoelectric detector (9) is converted into a voltage signal through a transimpedance amplifier (10), an analog-digital conversion circuit (11) converts the voltage signal into a digital signal and inputs the digital signal into a signal processing system (13), a bias circuit (12) provides fixed bias voltage for the signal processing system (13) so as to enable the signal processing system to work normally, and the bias voltage input into the photoelectric detector (9) by the bias circuit (12) is controlled by the signal processing system (13), the spectral response function of the photoelectric detector (9) is tested in advance and stored in a signal processing system (13); the signal processing system (13) takes the signals obtained from the analog-digital conversion circuit (11) and the spectral response function as input, carries out calculation processing through a reconstruction algorithm to reconstruct spectral information, then the signal processing system (13) compares the spectral information with a standard gas absorption spectrum database to detect gas components, and calculates the concentration of the gas components through a Lambert-beer formula.

2. The portable greenhouse gas detection system based on spectral reconstruction technology of claim 1, wherein: the infrared light source (1) is used for providing broad spectrum light with the wavelength of an infrared region, the wavelength range of the infrared light comprises the absorption spectrum wavelength range of the gas to be detected, the infrared light source emits infrared light with fixed light intensity, namely the light intensity corresponding to each wavelength in the infrared region range is fixed, the air pump (4) is used for inputting the gas to be detected into the absorption air chamber (3) and controlling the speed and the air pressure of the input gas, the sensor (5) is used for detecting the temperature and the pressure of a gas sample, so that the influence brought by the temperature and pressure change is considered in the analysis result, and the gas to be detected is kept at constant temperature and pressure as much as possible.

3. The portable greenhouse gas detection system based on spectral reconstruction technology of claim 1, wherein: the photoelectric detector (9) is used for detecting the light incident from the focusing lens (8), the photoelectric detector (9) generates a corresponding electric signal under the irradiation of the incident light, and the type of the photoelectric detector (9) can be any one of a PN junction type photoelectric diode, a PIN junction type photoelectric diode and an avalanche type photoelectric diode.

4. The portable greenhouse gas detection system based on spectral reconstruction technology of claim 1, wherein: the function of the bias circuit (12) is to provide a fixed bias voltage for the signal processing system (13), the signal processing system (13) controls the bias voltage input to the photoelectric detector (9) by the bias circuit (12), the spectral response of the detector is different under the condition of different bias voltages, the spectrum with high resolution can be obtained by combining different bias voltages, and the function of the bias circuit (12) is to convert the input voltage (direct current or alternating current) into the direct current voltage required by the work of the photoelectric detector (9).

5. The portable greenhouse gas detection system based on spectral reconstruction technology of claim 1, wherein: the output of the photoelectric detector (9) is connected to a circuit formed by a transimpedance amplifier (10) and an analog-digital conversion circuit (11), the transimpedance amplifier (10) is used for converting weak photocurrent generated by the photoelectric detector (9) into voltage, and the analog-digital conversion circuit (11) is used for converting an analog signal generated by the transimpedance amplifier (10) into a digital signal.

6. The portable greenhouse gas detection system based on spectral reconstruction technology of claim 1, wherein: the spectral response function information of the photoelectric detector (9) is stored in a signal processing system (13), the signal processing system (13) stores the information that the photocurrent generated by the detector under a certain bias voltage changes along with the change of the wavelength, the signal processing system (13) normalizes the input photocurrent value, and the parameters of the spectral response function comprise the bias voltage, the normalized current value and the wavelength of the detector.

7. The portable greenhouse gas detection system based on spectral reconstruction technology of claim 1, wherein: the method for measuring the spectral response function comprises the following steps that the bias voltage of a photoelectric detector (9) is fixed firstly, the light intensity of incident light emitted by an adjustable light source is the same, monochromatic light with fixed wavelength is emitted by the light source to irradiate the photoelectric detector (9), the system stores photocurrent information obtained by testing into a signal processing system (13), then the wavelength of the incident light is gradually changed within the detectable wavelength range of the system according to a certain step length, the photocurrent information generated by the detector is sequentially and continuously stored into the signal processing system (13), then the bias voltage of the detector is changed, the steps are repeatedly carried out to measure the photocurrent under different wavelength conditions, and the bias voltage is changed according to a certain step length.

8. The portable greenhouse gas detection system based on spectral reconstruction technology of claim 1, wherein: the input of the signal processing system (13) is a bias circuit (12) and data output from an analog-digital conversion circuit (11), the content stored by the signal processing system (13) comprises a spectral response function under each bias voltage, a standard gas absorption spectrum database and a gas absorption coefficient alpha (lambda) of each greenhouse gas, one of the functions of the signal processing system (13) is that the signal processing system (13) associates a plurality of groups of data obtained from the analog-digital conversion circuit (11) with the spectral response function, calculates and processes the input through a reconstruction algorithm so as to reconstruct spectral information, and compares the absorption peak area of the spectrum with the standard gas absorption spectrum database so as to judge the gas component and obtain the concentration of the gas through a Lambert-beer equation.

9. The portable greenhouse gas detection system based on spectral reconstruction technology of claim 1, wherein: the micro spectrometer based on the spectrum reconstruction technology is composed of a photoelectric detector (9), a bias circuit (12), a signal processing system (13), a trans-impedance amplifier (10) and an analog-digital conversion circuit (11), wherein an unknown quantity required to be solved by the signal processing system (13) is F (lambda), the F (lambda) represents spectral information of incident light, parameters are light intensity and wavelength, and a known quantity R is_j(λ) indicates that the photodetector (9) is biased at V_jSpectral response function of time, F (lambda) R_j(λ) represents a bias voltage V_jAnd photocurrent I generated by the detector under the irradiation of incident light with wavelength lambda_jThe detectable wavelength range of the system is from λ_minTo lambda_maxThe incident light source contains light with different wavelengths and each monochromatic light with different wavelength in the detection range can generate photocurrent in the detector, so that the detector is biased at V_jPhotocurrent I generated during the time_jFor F (lambda) R in the wavelength range_j(lambda), i.e. the photocurrent generated by the detector is superimposed by light of each wavelength, and assuming that there are m bias voltages to be adjusted by the photodetector (9), m equations representing the photocurrent of the detector are listed as a set of equations, the variables of which include I_j、R_j(λ) and an unknown quantity F (λ); the signal processing system (13) obtains the photocurrent I through testing_jAnd a spectral response function R_jAnd (lambda) forming a linear equation system by the m equations, and finally solving F (lambda) by calculation processing to reconstruct the spectral information of the incident light.

(II) background of the invention

The traditional gas analysis and detection system generally adopts a Fourier Transform Infrared (FTIR) spectrometer, the Fourier transform infrared spectrometer utilizes an interferometer in the Fourier transform infrared spectrometer to generate interference information and obtains spectral information through Fourier transform, the near infrared Fourier transform spectroscopy has the advantages of high signal-to-noise ratio, high sensitivity and the like, but the whole system is large and high in cost, is lack of portability, and is not suitable for detecting gas on site in real time.

The reconfigured spectrometer solves the problem of limiting the miniaturization of the spectrometer due to factors such as optical elements and optical path length, the technology can compensate the performance reduction caused by miniaturization, and the spectrometer can improve the performance and reduce the size by utilizing the technical progress of hardware and new algorithm processing.

In order to realize a portable gas detection system with high performance, t.scharf et al, which applies an interferometer using a silicon micromachined layered grating to a Fourier transform infrared spectrometer of a Micro Electro Mechanical System (MEMS) technology, designs a system for multiple gas detection analysis, and improves the miniaturization of the system (t.scharf, et al, "Miniaturized Fourier transform spectrometer for gas detection in the MIR region", IEEE, 2010); shaoda Zhang et Al devised a gas sensing system BASED on a linear variable optical filter and a MEMS BASED INFRARED thermopile detector array using p and n type polysilicon or Al metal as thermocouple material, the optical filter being mounted as a narrow band filter array over the thermopile detector array to form a micro mid-INFRARED spectrometer (Shaoda Zhang, et Al, "A COMPACT MEMS-BASED INDIRED SPECTROMETERFORM-GAS MEASUREMENT", IEEE, 2019); alaa Fathy et al designs a gas sensing system of FTIR spectrometer based ON MEMS technology, integrates two separate MEMS interferometers ON the same chip in parallel by deep reactive ion etching technology, and improves the system resolution by increasing the scanning optical path difference OPD (Alaa Fathy, et al, "ON-CHIP PARALLEL ARCHITECTURE MEMS SPECTROMETERS ENBLINGHIGH SPECTRAL RESOLUTION NFOR ENVIRONMENT GAS ANALYSIS" IEEE, 2019); jarkko Antila et al developed a compact mid-infrared spectrometer for gas sensing using a micro-opto-electro-mechanical systems (MOEMS) based surface micro-mechanically tunable Fabry-Perot interferometer that used a single pixel detector for high quality spectral measurements, avoiding the expensive linear array detector (Jarkko Antila, et al, "Miniaturized MOEMS spectrometer technology for gas sensing" IEEE, 2013); seawav et al 2020 discloses a sheet ZnO/graphene single sphere micro/nano structure gas sensor and a manufacturing method thereof (Chinese patent: CN202010028965.7), wherein the interference spectrum of the sensor generates large drift due to the change of the concentration of nitrogen dioxide gas by means of the prepared sheet ZnO/graphene sensitive material and a strong interference optical fiber structure, the drift amount can measure the corresponding gas concentration, and the signal of the gas sensor is transmitted to a spectrometer through an optical fiber circulator.

The above invention has the following drawbacks and disadvantages: 1. the Fourier transform infrared spectrometer structure has the advantages that the miniaturization, the volume and the price are high due to the limitation of various optical components, optical path length and other factors, and the Fourier transform infrared spectrometer adopting the Micro Electro Mechanical System (MEMS) technology is reduced in size to a certain extent but has lower resolution; 2. the two Michelson interferometers are integrated on the chip in parallel, so that the resolution of the spectrograph is increased, the requirement on chip processing is high, the system size is increased, and the miniaturization is limited; 3. the spectrometer adopts a tunable optical filter, and an optical element needs to be manufactured independently, so that the manufacturing difficulty and cost of the spectrometer are increased, the absorption efficiency and sensitivity of the detector are reduced, and the size of the system is increased; 4. the Fabry-Perot interferometer with the tunable surface micro-machinery is adopted, although the system size can be reduced by adopting the single-pixel detector, the interferometer still adopts the structures of optical parts and the like, so that the size is larger, and the surface micro-machining technology has high requirements on the chip machining process, high manufacturing cost and complex design; 5. the high-sensitivity optical fiber gas sensor is manufactured by fusing the prepared flaky ZnO/graphene sensitive material and the strong-interference optical fiber structure, the method has high material preparation requirement and high manufacturing cost, and the traditional infrared spectrometer is large in size and high in price and is not suitable for field portable real-time gas detection.

In order to solve the problems, the invention discloses a portable greenhouse gas detection system based on a spectrum reconfiguration type spectrometer, the system is provided with a micro spectrometer based on a spectrum reconfiguration technology, a large number of different spectrum response conditions can be obtained by combining a single infrared detector with different bias voltages, and a spectrum response function with high sampling resolution is obtained, so that the spectrometer has the advantages of high resolution and high integration, and the system can realize the portable gas detection system with simple structure and high performance by detecting gas by an infrared absorption spectroscopy.

Incident light striking the photodetector generates electron-hole pairs, shorter wavelength light is absorbed near the surface, while light with longer wavelength penetrates deeper into the semiconductor material until it is absorbed, the rate G of surface electron-hole pair generation at the PN junction depends on the incident light flux M, the wavelength λ of the incident light, the light absorption coefficient α of the material, and the distance y from the silicon surface, as shown in equation (1):

G(y)＝Mα(λ)exp(-α(λ)y) (1)

due to the generation of electron-hole pairs, light irradiation to the PN junction generates a photocurrent, and the photocurrent caused by the light irradiation is shown in formula (2):

I_L＝-qAG(y)×(Ln+Xd+Lp) (2)

wherein A is the junction area of the PN junction, Xd is the barrier width, Lp and Ln are the hole and electron diffusion lengths respectively, and the photo-generated current of the photodetector depends on the incident light and the structural property of the PN junction; under the condition of no radiation effect, the characteristics of the photoelectric detector are the same as those of a common photoelectric diode, and the current equation of the external bias voltage is shown as the formula (3):

wherein I₀In order to achieve the reverse saturation current, KT is the thermal equivalent and the value at room temperature is 26mev, the photodetector is normally biased at an external negative voltage, the reverse bias voltage cannot exceed the breakdown voltage, and it can be seen from the above equations (2) and (3) that the current generated by the light radiation applied to the photodiode when the external bias voltage is V is as shown in equation (4):

the spectral responsivity refers to the response capability of the photodetector to monochromatic light, the spectral response function represents the change of photocurrent generated under a certain bias voltage along with the change of wavelength, and a typical spectral response curve when the bias voltage is V can be obtained through formula (4).

Potential barrier width X of PN junction_dThe formula is shown in (5) in relation to the external bias voltage V of the semiconductor,

wherein N is₀To reduce the concentration,. epsilon_sIs dielectric constant, V_biFor the built-in potential, it can be known from formulas (4) and (5) that the external bias voltage affects the photocurrent generated by the photodetector and changes the photocurrent expressed by formulas (3) and (2), the external bias voltage mainly changes the photocurrent generated by illumination, as the absolute value of the external negative voltage increases, the current generated by the photodetector correspondingly increases, that is, the spectral responsivity increases, and the monotonic relation between the spectral responsivity and the bias voltage is not simply in direct proportion, that is, the spectral response of the photodetector under different bias voltages has specificity, so the detector can have different spectral responses by adopting different bias voltages.

Under the irradiation of incident light with a certain fixed wavelength, the spectral response function of the detector under a certain fixed bias voltage is obtained by measuring the photocurrent generated by the photoelectric detector under the fixed bias voltage, the photocurrent needs to be normalized (the photocurrent value is compared with the light intensity value of the incident light), and the external bias voltage value of the detector is V_jThe spectral response of R_j(λ)。

F (lambda) represents the spectral information of the incident light, the parameters are light intensity and wavelength, the incident light F (lambda) irradiates the photoelectric detector, and the external bias voltage of the photoelectric detector is V_jPhotocurrent I obtained by time measurement_jCan be expressed by equation (6):

f (lambda) represents the light intensity of incident light with the wavelength of lambda, under the working condition of being lower than the breakdown voltage, the photocurrent formed by the detector is in direct proportion to the light intensity, the normalized current value of the spectral response function represents the proportionality coefficient of the photocurrent and the light intensity, so F (lambda) R_j(λ) represents a bias voltage V_jAnd a photocurrent generated by the detector under the irradiation of incident light having a wavelength of λ, the spectrometer being capable ofThe wavelength range of detection is from λ_minTo lambda_maxSo that the detector is biased at V_jPhotocurrent I generated in the case of (1)_jIs to F (lambda) R in the whole wavelength range_j(λ), equation (6) can also be written in discrete form, as expressed in equation (7):

I＝R×F (7)

wherein R represents the spectral response function matrix of the detector and I is the photocurrent data matrix obtained by the test, so that the unknown target spectrum F can be reconstructed by solving equation (7).

The infrared spectrum is a molecular spectrum, the molecular structure of a substance has uniqueness, so the molecular structure of the substance can be determined by detecting the spectral data of a sample, the unknown sample is qualitatively and quantitatively analyzed by analyzing the spectral absorbance, the absorption peak area and the shape of the unknown sample, the theoretical basis of the infrared spectrum quantitative analysis method is the Lambert-beer law, greenhouse gases mainly comprise CO2, CH4, N2O and water vapor, and the absorption peak areas of the gas components are all in the infrared spectrum area.

When light irradiates gas molecules, if the spectrum intersects with the absorption peak spectral line of the gas molecules, a part of the light is consumed, resulting in the reduction of the total light intensity, and the specific explanation of the Lambert-beer law is that when infrared light takes the light intensity as P₀The light intensity of the emerging light P upon incidence into an unknown gas sample can be expressed as equation (8):

P(t)＝P₀(t)×exp[-α(λ)LC] (8)

where C is the gas concentration, α is the absorption coefficient of the gas, L is the optical path length, and P is₀The intensity of incident light and the intensity of emergent light are respectively calculated, a broadband infrared light source with a proper detection waveband is selected through researching a molecular characteristic absorption peak of the gas to be detected, the gas to be detected and a transmission light path complete absorption, and the concentration of the gas to be detected is calculated by utilizing the Lambert-beer law.

After a system structure is designed, the same gas is filled into the gas chamber, the absorption coefficient alpha and the optical path L of the gas are determined to be unchanged, the gas concentration can be reflected by detecting the light intensity difference between input and output, and the expression of the gas concentration C can be expressed as (9):

the system obtains the absorption spectrum of the gas to be measured, gas components are obtained by comparing the absorption spectrum with a standard gas absorption spectrum database, the light intensity difference obtained at two ends of the gas chamber is obtained by testing, and the concentration of the gas components is obtained by a Lambert-beer formula.

Disclosure of the invention

The invention aims to provide a portable greenhouse gas detection system based on a spectrum reconstruction technology, which can realize a low-cost high-performance portable gas detection analysis system and belongs to the technical field of gas measurement.

The portable greenhouse gas detection system based on the spectrum reconstruction technology is composed of an infrared light source (1), an input optical fiber (2), an absorption air chamber (3), an air pump (4), a sensor (5), an output optical fiber (6), an optical fiber collimator (7), a focusing lens (8), a photoelectric detector (9), a transimpedance amplifier (10), an analog-digital conversion circuit (11), a bias circuit (12) and a signal processing system (13); in the system, infrared light emitted by an infrared light source (1) enters an absorption air chamber (3) through an input optical fiber (2), gas to be detected is introduced into the absorption air chamber (3) at a certain speed and air pressure through an air pump (4), a sensor (5) is used for detecting the temperature and the pressure of the gas, the light emitted from the absorption air chamber (3) irradiates a single photoelectric detector (9) through an output optical fiber (6), an optical fiber collimator (7) and a focusing lens (8), the photocurrent generated by the photoelectric detector (9) is converted into a voltage signal through a transimpedance amplifier (10), an analog-digital conversion circuit (11) converts the voltage signal into a digital signal and inputs the digital signal into a signal processing system (13), a bias circuit (12) provides fixed bias voltage for the signal processing system (13) so as to enable the signal processing system to work normally, and the bias voltage input into the photoelectric detector (9) by the bias circuit (12) is controlled by the signal processing system (13), the spectral response function of the photoelectric detector (9) is tested in advance and stored in a signal processing system (13); the signal processing system (13) takes the signals obtained from the analog-digital conversion circuit (11) and the spectral response function as input, carries out calculation processing through a reconstruction algorithm to reconstruct spectral information, then the signal processing system (13) compares the spectral information with a standard gas absorption spectrum database to detect gas components, and calculates the concentration of the gas components through a Lambert-beer formula.

The infrared light source (1) is used for providing broad spectrum light with the wavelength of an infrared region, the wavelength range of the infrared light comprises the absorption spectrum wavelength range of a gas to be detected, the infrared light source emits infrared light with fixed light intensity, namely the light intensity corresponding to each wavelength in the infrared region is fixed, the air pump (4) is used for inputting the gas to be detected into the absorption air chamber (3) and controlling the speed and the air pressure of the input gas, and the sensor (5) is used for detecting the temperature and the pressure of a gas sample so as to consider the influence caused by the temperature and pressure change during the analysis result and keep the gas to be detected at the constant temperature and pressure as much as possible.

The photodetector (9) is used for detecting the light incident from the focusing lens (8), the photodetector (9) generates a corresponding electric signal under the irradiation of the incident light, and the type of the photodetector (9) can be any one of a PN junction type photodiode, a PIN junction type photodiode and an avalanche type photodiode.

The function of the bias circuit (12) is to provide a fixed bias voltage for the signal processing system (13), the bias voltage input to the photoelectric detector (9) by the bias circuit (12) is controlled by the signal processing system (13), the spectral response of the detector is different under the condition of different bias voltages, the spectrum with high resolution can be obtained by combining different bias voltages, and the function of the bias circuit (12) is to convert the input voltage (direct current or alternating current) into the direct current voltage required by the work of the photoelectric detector (9).

The output of the photoelectric detector (9) is connected to a circuit formed by a transimpedance amplifier (10) and an analog-digital conversion circuit (11), the transimpedance amplifier (10) is used for converting weak photocurrent generated by the photoelectric detector (9) into voltage, and the analog-digital conversion circuit (11) is used for converting an analog signal generated by the transimpedance amplifier (10) into a digital signal.

The spectral response function information of the photoelectric detector (9) is stored in a signal processing system (13), the signal processing system (13) stores the information that the photocurrent generated by the detector under a certain bias voltage changes along with the change of the wavelength, the signal processing system (13) normalizes the input photocurrent value, and the parameters of the spectral response function comprise the bias voltage, the normalized current value and the wavelength of the detector.

The method for measuring the spectral response function comprises the following steps that the bias voltage of a photoelectric detector (9) is fixed firstly, the light intensity of incident light emitted by an adjustable light source is the same, monochromatic light with fixed wavelength is emitted by the light source to irradiate the photoelectric detector (9), the system stores photocurrent information obtained by testing into a signal processing system (13), then the wavelength of the incident light is gradually changed within the detectable wavelength range of the system according to a certain step length, the photocurrent information generated by the detector is sequentially and continuously stored into the signal processing system (13), then the bias voltage of the detector is changed, the steps are repeatedly carried out to measure the photocurrent under different wavelength conditions, and the bias voltage is changed according to a certain step length.

The input of the signal processing system (13) is a bias circuit (12) and the data output from the analog-digital conversion circuit (11), the content stored by the signal processing system (13) includes a spectral response function under each bias, a standard gas absorption spectrum database and a gas absorption coefficient alpha (lambda) of each greenhouse gas, one of the functions of the signal processing system (13) is to reconstruct spectral information by connecting a plurality of sets of data obtained from the analog-digital conversion circuit (11) with the spectral response function and performing calculation processing on the input through a reconstruction algorithm, and further, the signal processing system (13) compares the absorption peak area of the spectrum with the standard gas absorption spectrum database to judge the gas composition and find the concentration of the gas through a Lambert-beer equation.

The micro spectrometer based on the spectrum reconstruction technology is composed of a photoelectric detector (9), a bias circuit (12), a signal processing system (13), a trans-impedance amplifier (10) and an analog-digital conversion circuit (11), wherein an unknown quantity required to be solved by the signal processing system (13) is F (lambda), the F (lambda) represents spectral information of incident light and parameters are light intensity and wavelength, and a known quantity R is_j(λ) indicates that the photodetector (9) is biased at V_jSpectral response function of time, F (lambda) R_j(λ) represents a bias voltage V_jAnd photocurrent I generated by the detector under the irradiation of incident light with wavelength lambda_jThe detectable wavelength range of the system is from λ_minTo lambda_maxThe incident light source contains light with different wavelengths and each monochromatic light with different wavelength in the detection range can generate photocurrent in the detector, so that the detector is biased at V_jPhotocurrent I generated during the time_jFor F (lambda) R in the wavelength range_j(lambda), i.e. the photocurrent generated by the detector is superimposed by light of each wavelength, and assuming that there are m bias voltages to be adjusted by the photodetector (9), m equations representing the photocurrent of the detector are listed as a set of equations, the variables of which include I_j、R_j(λ) and an unknown quantity F (λ); the signal processing system (13) obtains the photocurrent I through testing_jAnd a spectral response function R_jAnd (lambda) forming a linear equation system by the m equations, and finally solving F (lambda) by calculation processing to reconstruct the spectral information of the incident light.

Compared with the prior art, the invention has the advantages that: the invention adopts the micro spectrometer based on the spectrum reconstruction technology, avoids the problems that the traditional spectrometer (such as a Fourier transform spectrometer) is limited by factors such as optical components, optical path length and the like to be miniaturized and bring performance reduction, can obtain a large number of different spectrum response conditions by combining different bias voltages through a single detector so as to obtain a spectrum response function with high sampling resolution, and can obtain a large number of effective data to carry out spectrum reconstruction by adjusting the bias voltage conditions, so the spectrum reconstruction type spectrometer has the advantages of high resolution and high integration level, and the system detects greenhouse gases through an infrared absorption spectrum method, and can realize a portable greenhouse gas detection system with simple structure and high performance.

(IV) description of the drawings

Fig. 1 is a schematic diagram of a portable greenhouse gas detection system based on a spectrum reconstruction technology, and the system is composed of an infrared light source (1), an input optical fiber (2), an absorption air chamber (3), an air pump (4), a sensor (5), an output optical fiber (6), an optical fiber collimator (7), a focusing lens (8), a photoelectric detector (9), a transimpedance amplifier (10), an analog-to-digital conversion circuit (11), a bias circuit (12) and a signal processing system (13).

FIG. 2 is a diagram of the detection of CO in greenhouse gases based on a spectral reconstruction technique₂The system comprises an infrared LED (IRL715) (1), an input optical fiber (2), an absorption air chamber (3), an air pump (4), a sensor (5), an output optical fiber (6), an optical fiber collimator (7), a focusing lens (8), an indium gallium arsenide (InGaAs) PIN junction type photodiode (9), a transimpedance amplifier (10), an analog-digital conversion circuit (11), a bias circuit (12) and a signal processing system (13).

(V) detailed description of the preferred embodiments

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be described below in conjunction with the drawings in the embodiments of the present invention.

FIG. 2 shows a method for detecting CO in greenhouse gases based on spectral reconstruction₂The system comprises an infrared LED (IRL715) (1), an input optical fiber (2), an absorption air chamber (3), an air pump (4), a sensor (5), an output optical fiber (6), an optical fiber collimator (7), a focusing lens (8), an indium gallium arsenide (InGaAs) PIN junction type photodiode (9), a transimpedance amplifier (10), an analog-digital conversion circuit (11), a bias circuit (12) and a signal processing system (13); this example shows a single infrared detector combined with 9 biases to achieve a reconstructed spectrum, followed by detection of greenhouse gases by Lambert-beer's law, including CO in greenhouse gases₂Identification of gas composition and CO₂And (5) detecting the gas concentration.

CO₂The gas has absorption bands with different absorption intensities in near infrared and middle and far infrared bands, the strong absorption bands are distributed at the wavelengths of 2.7 microns, 4.26 microns and 11.4 microns, and the absorption line at the central wavelength of 4.26 microns is selected as the detection basis, because the absorption and attenuation of the band are the strongest; to ensure the light source is at CO₂The infrared absorption peak wavelength range has stronger light intensity, the selected infrared light source is an infrared LED with the model of IRL715, the wavelength range of the infrared light source is wide, and the LED is suitable for CO₂(absorption peak wavelength range 4.15 to 4.4 μm) measurement of gas.

The photoelectric detector mainly has parameters such as quantum efficiency, spectral response, response time and the like, the parameters can influence the specific performance indexes of the detector, all the parameters are comprehensively considered when the detector is selected, the system uses an indium gallium arsenide (InGaAs) PIN junction type photoelectric diode (9) as the photoelectric detector, the detector is the PIN junction type photoelectric diode, the infrared light wave band can be detected, and the system has the characteristics of low dark current and high sensitivity.

The light intensity P of the light emitted by the infrared LED (IRL715) (1)₀(lambda) is fixed, infrared light emitted by an infrared LED (IRL715) (1) enters an absorption air chamber (3) through an input optical fiber (2), gas to be detected is introduced into the absorption air chamber (3) through an air pump (4), light emitted from the absorption air chamber (3) irradiates to a single InGaAs (InGaAs) PIN junction type photodiode (9) through an output optical fiber (6), an optical fiber collimator (7) and a focusing lens (8), photocurrent generated by a detector is converted into a voltage signal through a trans-impedance amplifier (10), an analog-digital conversion circuit (11) converts the voltage signal into a digital signal and inputs the digital signal to a signal processing system (13), a bias circuit (12) provides fixed bias voltage for the signal processing system (13), the trans-impedance amplifier (10) and the analog-digital conversion circuit (11), and the bias voltage provided by the bias circuit (12) for the InGaAs (InGaAs) PIN junction type photodiode (9) is controlled by the signal processing system (13), spectral response function R of detector under different bias voltages_jThe (lambda) is obtained by testing in advance and stored in a signal processing system (13), the signal processing system (13) takes the signal obtained from the analog-digital conversion circuit (11) and the spectral response function as input, and the spectral information is reconstructed by performing calculation processing on the input through a reconstruction algorithm.

The steps for reconstructing the absorption spectrum of a gas will be described first, where F (λ) represents the spectral information of the incident light and the parameters are light intensity and wavelength, R_j(λ) denotes the detector at bias voltage V_jSpectral response function of time, at bias voltage V_jPhotocurrent I generated by time detector_jIt can be considered that light of each wavelength in the incident light is superimposed on the photocurrent generated by the detector, so I_jCan be expressed as F (lambda) R over a range of wavelengths_j(λ) integration.

Spectral response functionR_j(λ) the information expressed is that the detector has a wavelength of light λ and a bias voltage V_jTime-normalized current value, at bias voltage of V_jIn the case of (A) R_j(lambda) the value of the wavelength collected therein is lambda₁、λ₂、...、λ_kAnd corresponding normalized current valuesWherein λ_min≤λ₁＜λ₂＜…＜λ_k≤λ_maxThe number of samples taken by the signal processing system (13) at each bias is k.

The bias voltage set in this example is V₁、V₂、V₃、V₄、V₅、V₆、V₇、V₈、V₉The system adjusts the bias voltage of an indium gallium arsenide (InGaAs) PIN junction type photodiode (9) through a bias circuit (12) to obtain a digital signal I which represents photocurrent information and is output from an analog-digital conversion circuit (11)_j(where j 1, 2.., 9) which are input to a signal processing system (13), an indium gallium arsenide (InGaAs) PIN junction photodiode (9) is biased at V_jThe spectral response function of time is R_j(λ), a calculation processing formula obtained by integrating the data is as follows:

F(λ₁)R₁(λ₁)+F(λ₂)R₁(λ₂)+…+F(λ_k)R₁(λ_k)＝I₁

F(λ₁)R₂(λ₁)+F(λ₂)R₂(λ₂)+…+F(λ_k)R₂(λ_k)＝I₂

F(λ₁)R₃(λ₁)+F(λ₂)R₃(λ₂)+…+F(λ_k)R₃(λ_k)＝I₃

F(λ₁)R₄(λ₁)+F(λ₂)R₄(λ₂)+…+F(λ_k)R₄(λ_k)＝I₄

F(λ₁)R₅(λ₁)+F(λ₂)R₅(λ₂)+…+F(λ_k)R₅(λ_k)＝I₅

F(λ₁)R₆(λ₁)+F(λ₂)R₆(λ₂)+…+F(λ_k)R₆(λ_k)＝I₆

F(λ₁)R₇(λ₁)+F(λ₂)R₇(λ₂)+…+F(λ_k)R₇(λ_k)＝I₇

F(λ₁)R₈(λ₁)+F(λ₂)R₈(λ₂)+…+F(λ_k)R₈(λ_k)＝I₈

F(λ₁)R₉(λ₁)+F(λ₂)R₉(λ₂)+…+F(λ_k)R₉(λ_k)＝I₉

I_jit can be understood that light with various wavelengths superposes the photocurrent signal generated by the detector, and when the sampling value k of the spectral response function is large, F (lambda) R can be used_jThe accumulation of (λ) is expressed in integral form as follows:

the above calculation processing formula can be expressed by a matrix, and the expression is as follows:

I＝R×F

where R is a 9 × K matrix representing the spectral response function, I is a 9 × 1 matrix representing the photocurrent information, and F is a K × 1 matrix representing the light source spectral information, where I and R are known quantity matrices and F is an unknown quantity.

There are two cases where the solution to the system of R × F linear equations needs to satisfy that the rank of the R matrix is equal to the rank of the augmented matrix (R, I), there are two cases where there are unique solutions or infinite solutions, and in order for the equation to have a unique solution needs to satisfy that the rank of the R matrix is equal to the rank of the augmented matrix (R, I) and the rank is equal to the number of samples K.

By making the system of linear equations have a unique solution through reasonable arrangement, the solution of the F matrix can be expressed as:

F＝R^-1×I

R^-1is an inverse matrix of R, which exists if the determinant of the matrix is not zero, and the row vectors of the R matrix are not linearly related due to the difference of spectral response functions between different detectors, i.e. the determinant of the R matrix is not zero, so R exists.

The linear system of equations is uniquely solved by selecting a suitable value of k, the signal processing system (13) takes the photocurrent signal and the spectral response function as inputs, knowing the R matrix and the I matrix, and solves the inverse matrix R^-1The inverse matrix is multiplied by the matrix I to obtain an unknown matrix F, and a linear equation set consisting of 9 equations is obtained through calculation processing to reconstruct the absorption spectrum F (lambda) of the gas.

The absorption coefficient of the gas corresponds to the wavelength of the absorption peak position, and the intensity of the gas absorption is directly proportional to the gas concentration, so the wavelength corresponding to the absorption peak and the magnitude of the peak attenuation need to be determined in the corresponding calculation of the gas detection.

Next, the steps of gas detection and the analysis process are described, first, the whole of the absorption gas chamber (3) is charged with N₂Due to N₂No absorption peak exists in the near infrared band, at this time, the spectrum obtained by testing that no greenhouse gas (3) is introduced into the absorption gas chamber is considered as the original spectrum of the infrared light source, and the light intensity P of the infrared light source can be obtained through the original spectrum₀(λ)。

After the absorption gas chamber (3) is filled with greenhouse gas to be detected, an absorption spectrum is measured, whether an absorption peak exists in the wavelength range of 4.15-4.4 mu m in the absorption spectrum is detected, if the absorption peak exists in the wavelength range, the existence of CO in the greenhouse gas can be judged₂And (3) components.

In confirming the presence of CO in the gas₂Under the condition of components, the light intensity P (t) corresponding to each wavelength of light emitted from the absorption gas chamber (3) can be obtained after the absorption spectrum of the gas is obtained, the absorption spectrum is processed in a signal processing system (13), and the signal processing system (13) comprises a standard gas absorption spectrum database and gas absorption coefficients alpha (lambda), CO of various greenhouse gases₂The gas absorption coefficient alpha (lambda) of the gas component is known in advance, and is fixed by the structure of the absorption gas chamber (3), so the optical path length L is fixed, and the light intensity P of the infrared light source is fixed₀Since (. lamda.) is also fixed, CO can be obtained by Lambert-beer's law₂The concentration of the gas, C, the expression of lambert-beer's law is as follows:

this example demonstrates that the system reconstructs the absorption spectra of the gases by a single detector combined with 9 biases, followed by infrared spectral absorption and combined with the lambert-beer law for CO in greenhouse gases₂Detection of CO in greenhouse gases₂Identification of gas composition, in addition to CO₂The detection of the gas concentration shows the feasibility of the system, and the system can obtain high-resolution spectral information by increasing the sampling rate k of the spectral response function and combining more bias settings, thereby improving the gas detection performance of the system.