Multi-component gas detection system and method based on VMD adaptive frequency division

Abstract

The invention belongs to the field of gas detection and measurement, and discloses a multi-component gas detection system and method based on vmd self-adaptive frequency division, the system innovatively combines a frequency division multiplexing technology and a variational mode decomposition (VMD) algorithm, and realizes high-precision detection of multi-component gas. A digital signal carrying multi-component gas concentration information is obtained through a data acquisition card, a complex signal is subjected to self-adaptive decomposition by using a VMD algorithm, and a characteristic absorption signal of the multi-component gas is effectively separated. According to the method, the modal parameters of the VMD are iteratively optimized by adopting a k optimization algorithm, the accurate distinguishing of overlapped absorption spectral lines is realized by increasing the modal density, the analysis capability of a dense spectrum is improved, and the interference of interference noise and electronic noise is remarkably reduced. Second harmonic signals are extracted through software demodulation, and gas concentration information is inverted according to the second harmonic signals. Experiments show that compared with a traditional frequency division multiplexing scheme, the method has the advantages that the detection precision is improved, and the method is particularly suitable for synchronous detection of multi-component trace gas under a complex background.

Description

Technical Field

[0001] The present invention belongs to, but is not limited to, the technical field of gas detection and measurement, and particularly relates to a multi-component gas detection system and method based on VMD adaptive frequency division. Background Art

[0002] Tunable diode laser absorption spectroscopy (TDLAS) technology is a spectral detection technology with high resolution, strong selectivity, fast and convenient operation, accuracy and reliability, and can achieve online non-contact detection. It is based on the theory of molecular absorption spectroscopy. When a laser beam passes through a target gas, the measured substance selectively absorbs photons, causing attenuation of light energy, that is, the transmitted light intensity weakens. The transmitted light intensity after the initial light intensity fully interacts with the gas to be measured satisfies the Beer-Lambert law, and the concentration of the gas to be measured is calculated based on the attenuation of light energy before and after.

[0003] In TDLAS multi-component gas detection technology, time-division multiplexing and frequency-division multiplexing technologies are often used to meet specific transmission and processing requirements. The time-division multiplexing technology uses lasers of multiple wavelengths to enter the gas chamber for absorption in different time periods, and the absorption signals are demodulated by a lock-in amplifier to obtain the concentration information of multi-component gases. The structure is simple, but the real-time performance is poor. The frequency-division multiplexing technology combines different frequency modulation and demodulation technologies to achieve signal separation, but while providing high time resolution, it often comes with an increase in the cost of detection equipment, such as the number of photodetectors and lock-in amplifiers.

[0004] In these measurements, due to the weak signals, a long optical path cell is often used to ensure sufficient absorption between the light and the gas to be measured. However, the light reflected back and forth by the mirror and the diffraction effect easily cause the light to overlap in space and generate interference noise. These interference fringes are different from electronic noise. They have sufficient amplitude and periodic oscillation to obscure weak signals, and the frequency distribution is similar to that of the absorption signal frequency. The existing technology generally uses a lock-in amplifier for signal demodulation. In the simultaneous detection of multi-component gases, multiple lock-in amplifiers need to work together, which not only greatly increases the hardware cost and system volume, but also there is a clock synchronization problem between the reference signal and the modulation signal. In addition, the parameters of the digital filter used for filtering are fixed, with poor adaptability and a wide frequency transition bandwidth, and it cannot retain the complete absorption signal while filtering out the noise components.

[0005] In view of the above analysis, the technical problems urgently to be solved in the existing technology are as follows:

[0006] The existing multi-component gas detection system has poor integration, and the filter parameters in the lock-in amplifier are fixed during demodulation, with poor adaptability and a wide cut-off frequency transition bandwidth, and it cannot retain the complete absorption signal while filtering out the interference noise components. Summary of the Invention

[0007] Aiming at the problems existing in the prior art, the present invention provides a multi-component gas detection system and method based on variational mode decomposition (VMD) adaptive frequency division.

[0008] The present invention is implemented as follows. A multi-component gas detection system based on VMD adaptive frequency division, characterized in that the multi-component gas detection system based on VMD adaptive frequency division specifically includes: a signal generation module, two distributed feedback lasers (DFB Lasers) with wavelengths of 1572 nm and 1653 nm respectively, a wavelength division multiplexer, an optical fiber collimator, a gas absorption cell, a gas distribution system, a photodetector, a data acquisition card, and a host computer.

[0009] The multi-channel signal generation module simultaneously outputs two modulation signals, which are respectively composed of a low-frequency sawtooth wave with the same frequency and two different high-frequency sine waves, and are input to the corresponding laser driver module and then connected to the lasers.

[0010] The laser driver is a temperature and current control module. Two laser control modules are used to drive the two lasers to emit laser beams with the central wavelengths being the absorption spectral lines of the gas to be detected.

[0011] The wavelength division multiplexer couples the two beams of light into a laser signal. The laser output through the collimating mirror enters the gas absorption cell and is attenuated due to gas absorption.

[0012] The photodetector converts the optical signal into an electrical signal, which is directly connected to the data acquisition card to obtain a digital signal carrying the concentration information of the dual-component gas simultaneously.

[0013] The host computer is used to perform VMD frequency division, filtering, and software demodulation processing on the digital signal.

[0014] Further, the data acquisition software includes: LabVIEW.

[0015] Further, the host computer processing software includes: MATLAB.

[0016] Another object of the present invention is to provide a multi-component gas detection method based on VMD adaptive frequency division. The method specifically includes:

[0017] S1: Use the LabVIEW software to control the data acquisition card to collect the digital signal output by the detector, and adjust the temperatures and driving currents of the lasers corresponding to the dual-component gas to make the sawtooth wave signal include the complete absorption process of the multi-component gas.

[0018] S2: Select the optimal decomposition layer number k1 of the VMD algorithm for frequency division. Use Fourier transform to perform spectral transformation on the modal components, and observe the frequency distribution and composition of the signal to obtain the value of k1;

[0019] S3: Use the VMD algorithm for frequency division. The signal is decomposed into k1 modal components, and the signals with different frequency components are separated to achieve frequency division operation. According to the modulation frequency, match the high-frequency effective absorption components and modal components of each component gas;

[0020] S4: Multiply the two modal components with the corresponding synchronously generated reference signals respectively to achieve the software demodulation process;

[0021] S5: Observe the signal spectrogram obtained through Fourier transform. Determine the optimal decomposition layer number k2 for the second filtering operation using the VMD algorithm according to the frequency of the interference fringes near the frequency of the high-frequency effective absorption component and the k-value optimization iteration algorithm;

[0022] S6: Filter the digital signal. The two signals obtained in step S4 will both undergo a second VMD decomposition process. When using the VMD algorithm for the second time, generally use a higher decomposition layer number to decompose the signal frequency into k2 components with smaller frequency intervals, so as to separate the interference noise frequency information and the lowest-frequency second harmonic information;

[0023] S7: Directly select the lowest-frequency components of the two signals after processing as the low-frequency DC components to obtain the second harmonic information of the binary gas.

[0024] Furthermore, the VMD frequency division algorithm performs frequency domain decomposition on the collected digital signal carrying binary gas information, and decomposes the original digital signal into k modal components with specific central frequencies and finite bandwidths according to the frequency components.

[0025] Furthermore, the effective absorption component matches the effective absorption component of the corresponding gas according to the magnitude of the sine modulation frequency and the number of layers of the modal component. The higher the sine modulation frequency, the lower the number of layers of the modal component corresponding to the gas effective absorption component.

[0026] Furthermore, for the software demodulation, the software generates a reference signal and multiplies it with the effective absorption component of the corresponding gas. The reference signal is a continuous cosine wave with a frequency twice that of the high-frequency sine modulation frequency and an amplitude of 1.

[0027] Further, for the VMD algorithm filtering, the spectrum is analyzed through Fourier transform to observe the interval between the modulation frequency and the interference noise frequency. An iterative optimization strategy of k-value optimization is adopted. By analyzing the spectrum overlap degree and cross-correlation coefficient of the last two modal components, the optimal decomposition layer number k is obtained, and the appropriate bandwidth is adaptively found to ensure that the multiplied signal can filter out the frequency components of the interference noise while retaining the integrity of the effective information.

[0028] Further, the host computer software includes: MATLAB.

[0029] Combined with the above technical solutions and the solved technical problems, the advantages and positive effects of the technical solutions to be protected by the present invention are as follows:

[0030] First, by adopting the frequency division multiplexing technology, the present invention modulates two signals with different sine modulation frequencies and the same frequency sawtooth wave to drive two lasers to emit laser beams with specific wave numbers respectively. The wavelength division multiplexer combines the laser signals emitted by the two lasers, so that the effective absorption components of the binary gas can be signal-converted by one detector. The host computer replaces the hardware lock-in amplifier: the dual adaptive decomposition mechanism based on the VMD algorithm replaces the filter, effectively avoiding the disadvantages caused by the fixed parameters of the traditional filter and the too wide transition band of the cut-off frequency, and reducing the influence of interference noise on the second harmonic. Using software demodulation is more conducive to the integration of the multi-component gas detection system, reduces noise interference, and improves the detection accuracy of the system. This research also provides theoretical support for the performance enhancement of more component gas sensing systems, and its adaptive frequency division mechanism also opens up a new technical path for the detection scenario of complex gases with high spectral density.

[0031] Second, the technical solution of the present invention fills the domestic and foreign industry technical gaps: aiming at the technical bottleneck that the fixed filter parameters of the lock-in amplifier in the TDLAS-WMS system cannot accurately separate the high spectral density gas signals, an innovative multi-component gas detection method based on VMD adaptive frequency division is proposed. Through dual variational mode decomposition, the precise frequency domain separation and synchronous demodulation of the CO2 and CH4 absorption signals are realized, effectively overcoming the problem of suppressing the system electronic noise and interference noise, and successfully extracting the high-purity second harmonic characteristic signal. Experimental verification shows that this technology not only significantly improves the system integration, but also realizes the high-precision inversion of the concentration. Its unique adaptive frequency division mechanism completely breaks through the frequency domain analysis limitation of the traditional method, provides a revolutionary solution for the synchronous detection of complex gas components in a high spectral density environment, and has great application value in the field of synchronous detection of multi-component gases such as industrial emission monitoring and energy security early warning. Description of the Drawings

[0032] Figure 1It is the flowchart of the multi-component gas detection system based on VMD adaptive frequency division provided by the embodiments of the present invention;

[0033] Figure 2 It is the flowchart of the multi-component gas detection method based on VMD adaptive frequency division provided by the embodiments of the present invention;

[0034] Figure 3 It is the VMD algorithm adaptive frequency division diagram provided by the embodiments of the present invention;

[0035] Figure 4 It is the comparison diagram of the electrical signal and the detector signal spectrum provided by the embodiments of the present invention;

[0036] Figure 5 It is the second harmonic diagram of each of the two-component gases provided by the embodiments of the present invention;

[0037] Figure 6 It is the least squares fitting diagram provided by the embodiments of the present invention. Specific embodiments

[0038] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0039] As Figure 1 shown, the system first generates two composite modulation signals by a multi-channel signal generation module. Each signal is formed by superimposing a low-frequency sawtooth wave (for linear wavelength scanning) and a high-frequency sine wave (for wavelength modulation spectroscopy, WMS), constituting a frequency-domain composite modulation structure. These two signals are respectively input into the driving modules of two DFB (Distributed Feedback) lasers to achieve precise control of the injection current of the lasers. Since the wavelength of the DFB laser is highly sensitive to the driving current, the modulation signal will directly cause a periodic shift in the output optical wavelength, realizing high-resolution spectral scanning and frequency calibration, which is convenient for subsequent identification of multi-component gases.

[0040] The central wavelengths output by the two DFB lasers are different, corresponding to different absorption characteristic lines of the gas to be measured, and respectively cover the absorption bands of multiple target gas molecules. The two lasers are spectrally coupled by a wavelength division multiplexer (WDM) to achieve single-light-path multi-wavelength composite output. This composite beam is converted into a collimated beam by an optical fiber collimator and vertically incident into a multi-pass gas absorption cell (usually using a Herriot or White cell structure), and the absorption sensitivity is improved by increasing the effective optical path length, realizing highly sensitive detection of trace gases, which conforms to the enhanced absorption principle of the Beer-Lambert law.

[0041] In the gas absorption cell, the composite modulated laser interacts with the target mixed gas for multiple long-path interactions, and different gas components selectively absorb the laser intensity at the corresponding wavelengths. The absorbed optical signal is converted by a high-speed photodetector, and the output electrical signal contains the harmonic components of the modulation frequency (mainly the second harmonic signal), and the amplitude of these components is proportional to the gas concentration. The signal is then sampled at high resolution by a data acquisition card (DAQ), retaining key information such as modulation depth and center wavelength displacement, providing raw data support for subsequent multi-group demodulation.

[0042] After the host computer receives the time domain signal from the data acquisition card, it first performs background subtraction and baseline correction, and then uses variational mode decomposition (VMD) to adaptively decompose the composite signal. VMD effectively separates the second harmonic components superimposed on different modulation frequencies by introducing variational constraints on the center frequency of the narrowband sub-mode, thereby achieving decoupling of different gas signals. Combined with the calibrated response function, the system can perform high-precision inversion of the concentration of each component, and realize simultaneous and interference-free detection of multiple gases such as CO, CH4, and NH3, meeting the requirements of high dynamic range and high resolution detection.

[0043] The multi-component gas detection system based on VMD adaptive frequency division provided by the present invention has the following working principle:

[0044] Firstly, the multi-channel signal generation module generates two composite modulation signals. Each signal adopts frequency division multiplexing technology, in which the low-frequency sawtooth wave signal is used for wavelength scanning to cover the absorption spectrum of the gas to be measured, and the high-frequency sine wave signal is used to realize harmonic demodulation of wavelength modulation spectroscopy (WMS). The high-frequency carrier frequencies of the two independent modulation signals are set differently to ensure that the output signals of the two DFB lasers are fully separated in the frequency domain to prevent crosstalk interference during the subsequent VMD frequency division.

[0045] Secondly, the operating temperature and injection current of the DFB laser are controlled by the laser driving module to achieve precise control of the laser center wavelength and wavelength modulation depth, so that the output laser accurately corresponds to the characteristic absorption peak position (1572nm, 1653nm) of the target gas (such as CO, CH4). The light beam merged by the wavelength division multiplexer (WDM) is coupled into a long-range gas absorption cell (such as a Herriot cell or a White cell) through a fiber collimator. The laser fully interacts with the multi-component gas to be measured in the cell, causing the spectral line intensity of the optical signal carrying the gas concentration information to attenuate.

[0046] Then, the attenuated optical signal is converted into a current signal by a high-sensitivity InGaAs photodiode and amplified into a voltage signal by a transimpedance amplifier (TIA), and then input into a data acquisition card (DAQ) for high-speed synchronous sampling (such as using the virtual instrument software platform built by LabVIEW for data acquisition control), so as to realize the simultaneous recording of the mixed spectral signals of multiple gas components.

[0047] Next, the collected mixed signal is preprocessed and adaptively frequency-divided by the host computer software (such as MATLAB). In this process, the variational mode decomposition algorithm is used to adaptively decompose the mixed signal into several intrinsic mode functions (IMFs) with non-overlapping frequencies by optimizing the mode bandwidth and central frequency parameters under variational constraints, effectively separating the independent gas spectral line information corresponding to different laser signals.

[0048] Furthermore, band-pass filtering is performed on each IMF mode signal obtained after VMD adaptive frequency division to suppress the interference of low-frequency scanning sawtooth signals and background noise. Then, using the harmonic detection algorithm of WMS, such as the second harmonic (2f) demodulation technique, the high-sensitivity harmonic characteristic signals carrying gas concentration in each channel signal are extracted through the principle of lock-in amplifier, and the integral area of the absorption spectral line or the harmonic peak intensity of each gas component is calculated.

[0049] Finally, based on the calibrated spectral parameters of known gas components (such as the absorption cross-section data in the HITRAN database), a quantitative relationship between the concentration of the gas to be measured and the measured harmonic signal is established, and the concentrations of each component of the mixed gas are accurately solved through linear or non-linear fitting algorithms, realizing the on-line synchronous detection of fast, highly sensitive, and multi-component gases, and significantly improving the anti-interference ability and detection stability of the system in actual complex environment applications.

[0050] As Figure 1 shown, the embodiment of the present invention provides a multi-component gas detection system based on VMD adaptive frequency division, which specifically includes: a multi-channel signal generation module, two laser drivers, two lasers with wavelengths of 1572 nm and 1653 nm respectively, a wavelength division multiplexer, a collimating mirror, a gas mixing system, a gas absorption cell, a detector, a data acquisition card, and a host computer;

[0051] The multi-channel signal generator module is connected to the laser drive module, and its output composite modulation signal is composed of the synchronous superposition of multiple groups of programmable sine waves and high-precision sawtooth waves. By dynamically optimizing the modulation parameters through a real-time feedback control system, it is ensured that the laser can stably and efficiently output a laser beam with a specified wave number during use;

[0052] The laser driving module has two functions. One is to provide appropriate current and voltage to the laser according to the modulation signal, thereby controlling the output of the laser. The other is that since semiconductor lasers are sensitive to temperature, the laser driver usually includes a temperature control circuit to maintain the laser within the optimal operating temperature range. The two methods control the laser beam to be located on the absorption line simultaneously. Here, the present invention adjusts the voltage range and temperature of the modulation signal so that a sawtooth wave can completely cover the absorption spectral segments of the binary gas;

[0053] Two DFB lasers with wavelengths of 1572 nm and 1653 nm respectively, whose central wavelengths cover the absorption peaks of the target gas, are respectively connected to the corresponding laser drivers; the wavelength division multiplexer outputs with double single-mode tail fibers, combining optical signals with different wavelengths into one beam. Its receiving end is connected to the double lasers through a flange, and the combined beam passes through a collimating mirror at the transmitting end and enters the gas chamber; the detector is placed behind the gas chamber, and when the laser signal exits the gas chamber, it is received by the detector and the optical signal is converted into an electrical signal; the data acquisition card is connected to the detector, and the signal collected by the detector is stored in the upper computer;

[0054] The upper computer processes the detector signal through two software, LabVIEW and MATLAB. Through the cooperation of the LabVIEW program and the acquisition card, its sampling rate and total number of sampling points are set to collect the detector signal with a fixed period, and the signal is imported into MATLAB for processing;

[0055] For the signal processed by MATLAB, the VMD toolbox should be downloaded first to ensure that the vmd function is built in MATLAB and added to the MATLAB path.

[0056] As Figure 2 shown, a multi-component gas detection method based on vmd adaptive frequency division provided by an embodiment of the present invention specifically includes:

[0057] S1: Use the Labview software to control the data acquisition card to collect the digital signal output by the detector, and adjust the temperature and driving current of the lasers corresponding to the binary gas so that the sawtooth wave signal includes the complete absorption process of the multi-component gas.

[0058] S2: Select the optimal decomposition layer number k1 for VMD algorithm frequency division, perform spectral transformation on the modal components using Fourier transform, and observe the frequency distribution and composition of the signal to obtain the k1 value.

[0059] S3: Use the VMD algorithm for frequency division, the signal is decomposed into k1 modal components, the signals with different frequency components are separated to achieve frequency division operation, and the high-frequency effective absorption components and modal components of each component gas are matched according to the modulation frequency size.

[0060] S4: Multiply the two modal components with the corresponding synchronously generated reference signals respectively to implement the software demodulation process.

[0061] S5: Observe the signal spectrogram obtained through Fourier transform. Determine the optimal decomposition layer number k2 for the second use of the VMD algorithm for filtering operation according to the frequency of the interference fringes near the frequency of the high-frequency effective absorption component and the k-value optimization iteration algorithm.

[0062] S6: Filter the digital signal. Both of the two signals obtained in step S4 will undergo a second VMD decomposition process. When using the VMD algorithm for the second time, generally a higher decomposition layer number is adopted to decompose the signal frequency into K2 components with smaller frequency intervals, so as to separate the interference noise frequency information and the lowest-frequency second harmonic information.

[0063] S7: Directly select the lowest-frequency components after processing the two signals as the low-frequency DC components to obtain the second harmonic information of the two-component gas.

[0064] In the embodiment of the present invention, by adopting the variational mode decomposition (VMD) algorithm, the mixed multi-component gas signal collected by the data acquisition card can be adaptively analyzed in the frequency domain. This algorithm decomposes the composite signal into k intrinsic mode functions (IMFs) with specific central frequencies and finite bandwidths, realizing the effective separation of the characteristic absorption spectral lines of each component. The specific implementation process is as follows: First, perform a fast Fourier transform (FFT) spectral analysis on the original signal, and determine the initial mode number k of the VMD algorithm by observing the spectral energy distribution characteristics and the number of peaks. On this basis, the system generates orthogonal reference signals matching the absorption lines of the target gas through the host computer. Mix the reference signals with the high-frequency absorption components of the target gas obtained by VMD decomposition digitally to realize digital phase-locked demodulation based on software. To eliminate the interference noise in adjacent frequency bands introduced during the modulation process, the system innovatively adopts an iterative optimization algorithm for k-value optimization: by analyzing the spectral overlap degree and cross-correlation coefficient of the last two modal components, dynamically determine the optimal decomposition layer number K. Finally, select the DC component of the lowest-frequency mode as the effective signal output. The innovation of this method lies in realizing the collaborative optimization of signal demodulation and noise suppression through a dual adaptive decomposition mechanism.

[0065] The VMD frequency division algorithm performs frequency domain decomposition on the collected digital signal carrying two-component gas information, and obtains k modal components with specific central frequencies and finite bandwidths from the original digital signal according to the frequency components.

[0066] The effective absorption component and the modal component present a strictly descending distribution characteristic in the frequency domain, that is, each IMF is arranged in order from high to low according to its center frequency. Therefore, the effective absorption component of the corresponding gas can be matched according to the magnitude of the sine modulation frequency and the number of layers of the modal component. The higher the sine modulation frequency, the lower the number of layers of the modal component corresponding to the effective absorption component of the gas;

[0067] For the software demodulation, the software generates a reference signal and multiplies it with the effective absorption component of the corresponding gas. The reference signal is a continuous cosine wave with a frequency twice that of the high-frequency sine modulation frequency and an amplitude of 1.

[0068] For the VMD algorithm filtering, the spectrum is analyzed through Fourier transform to observe the interval between the modulation frequency and the interference noise frequency. An iterative optimization strategy of k-value optimization is adopted: by analyzing the spectrum overlap degree and cross-correlation coefficient of the last two modal components, the optimal decomposition layer number k is obtained, and the appropriate bandwidth is adaptively found to ensure that the multiplied signal can filter out the frequency components of the interference noise while retaining the integrity of the effective information;

[0069] The upper computer software includes: MATLAB.

[0070] Next, in combination with an embodiment, the gases to be measured are CH4 and CO2 respectively. In this embodiment, a dual adaptive decomposition mechanism based on VMD is provided, which can avoid the disadvantages of the above traditional filters when detecting binary gas using frequency division multiplexing technology.

[0071] The multi-component gas detection system device provided by the embodiment of the present invention includes:

[0072] A multi-channel signal generation module, two laser drivers, two lasers with wavelengths of 1572nm and 1653nm respectively, a wavelength division multiplexer, a collimating mirror, a gas mixing system, a gas absorption cell, a detector, a data acquisition card, and an upper computer.

[0073] The connection sequence is as follows: The multi-channel signal generation module is connected to two laser drivers, and then respectively connected to the corresponding lasers. The tail-end optical fibers of the two lasers are connected to the wavelength division multiplexer through flanges. The collimating mirror is placed at the light input end of the gas absorption cell. The light of the wavelength division multiplexer is incident into the gas absorption cell through the collimating mirror. After absorption, the detector placed at the rear end of the gas absorption cell performs signal conversion. The detector is connected to the data acquisition card, and the data acquisition card is connected to the upper computer.

[0074] Step 1, the multi-channel signal generation module outputs a modulation signal. The modulation signal for driving the laser with a center wavelength of 1572nm is a low-frequency scanning sawtooth wave of 1Hz and a high-frequency modulation sine wave of 2kHz. The modulation signal for driving the laser with a center wavelength of 1653nm is a low-frequency scanning sawtooth wave of 1Hz and a high-frequency modulation sine wave of 3kHz.

[0075] Step 2: Connect and check according to the above connection sequence, set up the optical path of the system, adjust the gas chamber, and introduce the gas to be measured, so that the laser signals with central wavelengths of 1572 nm and 1653 nm emitted by the two lasers enter the light inlet of the gas chamber, exit from the light outlet of the gas chamber, and are received by the detector. Then, the acquisition card receives the electrical signal of the detector and transmits it to the upper computer.

[0076] Step 3: Modify the temperature parameter so that the sawtooth wave can include the effective absorption segments of the two gases, and modify the sine wave amplitude parameter and the reference signal phase so that the peak value of the second harmonic demodulated by the upper computer through software is the largest.

[0077] Step 4: Determine the concentration of the introduced gas according to the spectral absorption coefficient. Introduce CO2 with concentrations of 7500 ppm, 15000 ppm, 22500 ppm, 30000 ppm, 37500 ppm, and 45000 ppm and CH4 with concentrations of 250 ppm, 500 ppm, 750 ppm, 1000 ppm, 1250 ppm, and 1500 ppm into the gas chamber respectively. Collect the detector signals of ten cycles for each concentration for the dual adaptive decomposition mechanism based on VMD and software demodulation, and continuously adjust the decomposition layer number to make the second harmonic waveform smooth.

[0078] Step 5: Input the two-component gas to be measured with unknown concentration, and substitute the corresponding second harmonic peak value into the fitting function to calculate the concentration.

[0079] Figure 3 This is the illustration of the adaptive frequency division process implemented based on the variational mode decomposition (VMD) algorithm in this embodiment. The upper part of the figure shows the time-domain waveforms of the original signal and the three intrinsic mode functions (IMF 1 - IMF 3) obtained after decomposition, and the residual signal is shown below. It can be clearly observed from the decomposed IMF images that the signal has achieved effective frequency separation among different IMFs, and the signal characteristics of different components are clearly revealed.

[0080] Figure 4 This is the illustration of the spectrum comparison between the electrical signal without interference noise generated by direct digital synthesis (DDS) and the actual detection signal collected by the photodetector. The figure clearly shows the spectral amplitudes and frequency structures of the DDS modulation signal and the actual detector output signal. Through the local enlarged inset (Zoom-in), the matching situation of the detection signal at the characteristic frequency position with the DDS signal can be observed, verifying the consistency and effectiveness of the actual photoelectric detection signal and the modulation carrier frequency.

[0081] Figure 5The waveforms of the second harmonic signals of the binary gas (CO2 and CH4) measured under modulation of different laser wavelengths (1572 nm and 1653 nm) are given. The figure clearly shows the respective second harmonic characteristic spectral lines of CO2 and CH4. The peak shapes and positions of the curves reflect the characteristic information of the gas absorption spectral lines, corresponding to the concentration and absorption intensity characteristics of the gas components respectively.

[0082] Figure 6 It is a linear fitting diagram between the concentrations of the binary gas and the amplitudes of the second harmonic signals in the system embodiment. The figure gives the relationship curves of the second harmonic peaks of CO2 and CH4 gases changing with the concentration, and uses the least squares method to fit, obtaining the fitting straight line equation and the correlation coefficient (R 2 ). The results show that there is a significant linear relationship between the concentrations of the two gas components and the amplitudes of the second harmonic signals, verifying the quantitative analysis ability of the system.

[0083] The adaptive frequency division detection method based on VMD has extensive and important application values in the fields of industry, environmental protection, medical treatment, scientific research, etc. Its core lies in realizing the synchronous, fast and accurate monitoring of various gas components in complex environments.

[0084] It should be noted that the embodiments of the present invention can be implemented by hardware, software or a combination of software and hardware. The hardware part can be realized by using special logic; the software part can be stored in a memory and executed by an appropriate instruction execution system, such as a microprocessor or special designed hardware. Those of ordinary skill in the art can understand that the above-mentioned devices and methods can be implemented using computer-executable instructions and / or included in the processor control code, for example, such code is provided on a carrier medium such as a disk, CD or DVD-ROM, a programmable memory such as a read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The devices and their modules of the present invention can be realized by the hardware circuits of programmable hardware devices such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, etc., or programmable logic devices such as field programmable gate arrays, and can also be realized by software executed by various types of processors, or can be realized by a combination of the above hardware circuits and software such as firmware.

[0085] As mentioned above, it is only the specific implementation manners of the present invention, but the protection scope of the present invention is not limited thereto. Any modification, equivalent replacement, improvement, etc. made by those skilled in the art within the technical scope disclosed by the present invention and within the spirit and principle of the present invention should be covered by the protection scope of the present invention.

Claims

1. A multi-component gas detection system based on VMD adaptive frequency division, characterized in that, Including: A multi-channel signal generation module, two distributed feedback (DFB) lasers, a laser driving module, a wavelength division multiplexer, an optical fiber collimator, a multi-reflection gas absorption cell, a gas distribution system, a photodetector, a data acquisition card, and an upper computer processing module; The two composite modulation signals output by the multi-channel signal generation module are respectively formed by superimposing a low-frequency sawtooth wave and high-frequency sine waves of different frequencies, and act on the two DFB lasers via the laser driving module; The lasers output by the two DFB lasers are coupled via the wavelength division multiplexer and enter the multi-reflection gas absorption cell through the optical fiber collimator; The output end of the photodetector is connected to the input end of the data acquisition card, and the output end of the data acquisition card is connected to the input end of the upper computer processing module.

2. The multi-component gas detection system based on VMD adaptive frequency division as described in claim 1, wherein The multi-channel signal generation module includes at least two independent modulation channels, and the high-frequency sine wave frequencies of each channel are different from each other.

3. The multi-component gas detection system based on VMD adaptive frequency division according to claim 1, wherein The laser driving module includes a temperature control unit and a current control unit.

4. The multi-component gas detection system based on VMD adaptive frequency division according to claim 1, characterized in that The gas absorption cell is of a multi-reflection structure, and an optical mirror group with a multi-reflection optical path is provided in the cell body.

5. The multi-component gas detection system based on VMD adaptive frequency division according to claim 1, wherein The upper computer processing module includes a variational mode decomposition (VMD) algorithm module and a wavelength modulation spectroscopy harmonic demodulation algorithm module.

6. A multi-component gas detection method based on VMD adaptive frequency division, characterized in that, This method specifically includes: S1: Control the data acquisition card through Labview software to collect the digital signal output by the detector, and adjust the temperatures and driving currents of the lasers corresponding to the two-component gas so that the sawtooth wave signal includes the complete absorption process of the multi-component gas; S2: Select the optimal decomposition layer number k1 for frequency division by the VMD algorithm, perform spectral transformation on the modal components using Fourier transform, and observe the frequency distribution and composition of the signal to obtain the k1 value; S3: Use the VMD algorithm for frequency division. The signal is decomposed into k1 modal components, the signals with different frequency components are separated to achieve frequency division operation, and the high-frequency effective absorption components and modal components of each component gas are matched according to the modulation frequency; S4: Multiply the two modal components with the corresponding synchronously generated reference signals respectively to achieve the software demodulation process; S5: Observe the signal spectrogram obtained through Fourier transform, and determine the optimal decomposition layer number k2 for the second filtering operation using the VMD algorithm according to the frequency of the interference fringes near the frequency of the high-frequency effective absorption component and the k-value optimization iteration algorithm; S6: Perform filtering processing on the digital signal. The two signals obtained in step S4 will both be subjected to a second VMD decomposition process. When using the VMD algorithm for the second time, generally a higher decomposition layer number is adopted to decompose the signal frequency into K2 components with smaller frequency intervals, so as to separate the interference noise frequency information and the lowest-frequency second harmonic information; S7: Directly select the lowest-frequency components after processing the two signals as the low-frequency DC components to obtain the second harmonic information of the two-component gas.

7. The multi-component gas detection method based on VMD adaptive frequency division according to claim 6, characterized in that, The VMD frequency division algorithm performs frequency domain decomposition on the collected digital signal carrying two-component gas information, and the original digital signal is decomposed into k modal components with specific central frequencies and limited bandwidths according to the frequency components.

8. The multi-component gas detection method based on VMD adaptive frequency division according to claim 6, wherein, The effective absorption component matches the effective absorption component of the corresponding gas according to the magnitude of the sine modulation frequency and the number of layers of the modal component. The higher the sine modulation frequency, the lower the number of layers of the modal component corresponding to the gas effective absorption component.

9. The multi-component gas detection method based on VMD adaptive frequency division according to claim 6, wherein, For the software demodulation, the software generates a reference signal and multiplies it with the effective absorption component of the corresponding gas. The reference signal is a continuous cosine wave with a frequency twice that of the high-frequency sine modulation frequency and an amplitude of 1.

10. The multi-component gas detection method based on VMD adaptive frequency division according to claim 6, characterized in that, For the VMD algorithm filtering, the spectrum is analyzed through Fourier transform to observe the interval between the modulation frequency and the interference noise frequency. An iterative optimization strategy of k-value optimization is adopted. By analyzing the spectrum overlap degree and cross-correlation coefficient of the last two modal components, the optimal decomposition layer number k is obtained, and the appropriate bandwidth is adaptively found to ensure that the multiplied signal can filter out the frequency components of the interference noise while retaining the integrity of the effective information.

Images