Trace water content detection method based on tunable laser absorption spectrum

Abstract

The invention discloses a trace water content detection method based on a tunable laser absorption spectrum, and relates to the technical field of gas moisture detection.The method comprises the steps that a laser generating device is adopted to generate an optical signal, the optical signal is irradiated into an integrated small gas chamber storing to-be-detected gas, and a first harmonic signal and a second harmonic signal of the optical signal passing through the to-be-detected gas are collected; performing light energy compensation by using the first harmonic signal and the second harmonic signal to obtain an inversion parameter; and performing water vapor concentration nonlinear fitting inversion by using the inversion parameters to obtain the moisture content of the gas to be detected. The extracted first harmonic signal and the extracted second harmonic signal are utilized, and a stable signal amplitude ratio is adopted as a subsequent processing signal. The signal amplitude ratio makes the detection signal no longer depend on the absolute value of the light intensity, but is determined by the relative relationship between the two harmonic components. The influence of light intensity fluctuation is compensated, periodic noise such as optical interference fringes is inhibited to a certain degree, and then rapid and high-precision detection of the gas water content is achieved.

Description

Technical Field

[0001] This application relates to the field of gas moisture detection technology, specifically to a method for detecting trace water content based on tunable laser absorption spectroscopy. Background Technology

[0002] Moisture content (humidity) in gases is a crucial parameter in industrial processes, environmental monitoring, scientific research, and daily life. In fields such as natural gas processing, semiconductor manufacturing, power equipment (e.g., SF6 circuit breakers), pharmaceuticals, aerospace, and HVAC, accurate and reliable monitoring of gas humidity directly impacts process efficiency, product quality, equipment safety, and energy consumption control.

[0003] Early humidity measurements relied primarily on mechanical methods such as wet-bulb and dry-bulb hygrometers. While simple in structure, these methods suffered from low accuracy, slow response, and the inability to automate or perform online measurements. To meet the demands of modern industry for high-precision, real-time online measurement, a series of electronic humidity sensing technologies have been developed, including cold mirror, polymer capacitive, alumina, and laser spectroscopy. However, while cold mirror methods offer high accuracy, they suffer from high cost, difficulty in cleaning mirror contamination, and poor portability. Polymer capacitive methods are small and inexpensive, but their accuracy is limited and their anti-interference capabilities are poor. Alumina methods are simple in principle and highly feasible, but similarly, their accuracy is limited and their response is slow. Common laser spectroscopy methods have poor vibration resistance and a narrow measurement range, making them suitable only for high water content applications. They also have a limited background gas, making them unsuitable for detecting alkanes and corrosive gases.

[0004] Therefore, how to quickly and accurately detect the water content in gases is a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0005] In order to solve the above-mentioned technical problems, this application proposes the following technical solution: In a first aspect, embodiments of this application provide a method for detecting trace water content based on tunable laser absorption spectroscopy, comprising: A laser generator produces a light signal that is irradiated into an integrated miniature gas chamber, which stores the gas to be tested. Collect the first and second harmonic signals of the optical signal after it passes through the gas to be tested; Inversion parameters are obtained by using the first harmonic signal and the second harmonic signal for optical energy compensation. The moisture content of the gas to be measured is obtained by nonlinear fitting and inversion of water vapor concentration using the inversion parameters.

[0006] In one possible implementation, the use of a laser generator to generate an optical signal that irradiates the integrated miniature gas chamber includes: The signal generation module generates high-frequency sine wave and low-frequency triangular wave signals; The high-frequency sine wave and low-frequency triangular wave signals are combined into a modulation signal by an adder to drive the laser. After being excited by the modulation signal, the laser generates an optical signal that enters the integrated miniature gas chamber and penetrates the gas to be measured.

[0007] In one possible implementation, the acquisition of the first harmonic signal and the second harmonic signal of the optical signal after passing through the gas to be measured includes: The projected light signal penetrates the gas to be measured using a photodetector; The projected light signal is input into a transimpedance gain amplifier for signal amplification; The amplified projected light signal is sequentially input into a lock-in amplifier and a low-pass filter to demodulate and obtain the first harmonic signal and the second harmonic signal.

[0008] In one possible implementation, obtaining the inversion parameters by using the first harmonic signal and the second harmonic signal for optical energy compensation includes: Determine the signal strength of the projected light signal detected by the photodetector. And the formula is expressed as follows: in: It is the intensity of the incident light. It is the laser modulation depth. Where is the instantaneous frequency of the laser, L is the length of the gas chamber, and c is the gas concentration; According to the above Formula representation and known , , And L obtains the c; The amplitudes of the first harmonic signal and the second harmonic signal are determined according to c. The amplitude ratio of the second harmonic signal to the first harmonic signal is determined as the inversion parameter.

[0009] In one possible implementation, determining the amplitudes of the first and second harmonic signals based on c includes: in: It is the average light intensity. It is the intensity modulation coefficient. It is the phase difference between intensity modulation and frequency modulation. , It is a parameter related to system gain. It is the first harmonic component of the Fourier expansion. It is the second harmonic component.

[0010] In one possible implementation, determining the amplitude ratio of the second harmonic signal to the first harmonic signal as the inversion parameter includes: Wherein: when the concentration of the gas to be measured is very low The term can be ignored, and the above formula can be simplified to: Since other system parameters are constants when the modulation parameter is fixed, the final result is: .

[0011] In one possible implementation, the step of using the inversion parameters to perform nonlinear fitting inversion of water vapor concentration to obtain the moisture content of the gas to be measured includes: The inversion parameters are used as input for spectral signal preprocessing to obtain a discrete signal sequence; The moisture content of the gas to be measured is obtained by nonlinear fitting and inversion of the water vapor concentration using the discrete signal sequence.

[0012] In one possible implementation, the inversion parameters are used as input for spectral signal preprocessing to obtain a discrete signal sequence, including: Let the signal collected in one instance be: ,in: For deterministic signals, For the first Noise from the second measurement; After N averages Output discrete signal sequence ; Regarding the Use a fixed length The window, with the center of the window aligned with the currently pending processing number. For each data point, the arithmetic mean of all data points within the window is used as the output value for that point. ,but: in, The window length determines the smoothness. The window slides through the entire signal sequence sequentially, achieving smooth processing of each data point.

[0013] In one possible implementation, obtaining the moisture content of the gas to be measured by nonlinear fitting and inversion of the water vapor concentration using the discrete signal sequence includes: Establish dew point temperature and Minimum residual model for measured values: in, Forward theoretical model: in: These are the calibration parameters for the instrument. For the intensity of the absorption spectral line, For optical path, It is the value of the spectral line at its center frequency. For other relevant parameters, Dew point temperature The saturated water vapor pressure below Defined as: Enter the initial value of the dew point temperature. and parameter initial values ; Calculate the residual of the k-th iteration: Calculate the corresponding Jacobian matrix: in, for Dimensions ; From the Magnus formula established above, we can obtain: Solve the LM equations: in, , It is the identity matrix. The damping factor, Update parameters: When obtaining the minimum value of E in the minimum residual model corresponding .

[0014] In one possible implementation, when obtaining the minimum value of the minimum residual model E... corresponding ,include: Set iteration parameters ; when Less than When the maximum number of iterations is reached, the inversion iteration terminates, and the final output is given. .

[0015] In this embodiment, the extracted first and second harmonic signals are used, and a stable signal amplitude ratio is employed as the subsequent processing signal. This signal amplitude ratio ensures that the detection signal no longer depends on the absolute value of the light intensity, but rather on the relative relationship between the two harmonic components. This not only compensates for the influence of light intensity fluctuations but also suppresses periodic noise such as optical interference fringes to a certain extent, thereby achieving rapid and high-precision detection of gaseous water content. Attached Figure Description

[0016] Figure 1 A schematic flowchart of a trace water content detection method based on tunable laser absorption spectroscopy provided in this application embodiment; Figure 2 A schematic diagram of an apparatus for implementing a trace water content detection method based on tunable laser absorption spectroscopy, provided in an embodiment of this application; Figure 3 This is a schematic diagram of laser installation provided for an embodiment of this application. Detailed Implementation

[0017] The present solution will now be described in conjunction with the accompanying drawings and specific embodiments.

[0018] See Figure 1 The trace water content detection method based on tunable laser absorption spectroscopy provided in this embodiment includes: S101, a laser generator is used to generate a light signal that is irradiated into an integrated small gas chamber, which stores the gas to be tested.

[0019] See Figure 2 The signal generation module generates high-frequency sine wave and low-frequency triangular wave signals. These signals are combined into a modulation signal by an adder to drive the laser. Once excited by the modulation signal, the laser generates an optical signal that enters the integrated miniature gas chamber and penetrates the gas being measured.

[0020] In this embodiment, regions with strong absorption lines for each target substance are selected from the Hisran database, and absorption simulation software is used to analyze suitable single absorption lines in each region. It is found that when the background gas is a non-alkane gas, the absorption line is relatively abundant at 1370 nm, and when the background gas is an alkane gas, the absorption point is more obvious at 1877 nm. Therefore, in this embodiment, the 1877 nm laser from Vertilis is selected.

[0021] The laser is driven using a dual-channel DDS design: a low-frequency 10Hz triangular wave and a high-frequency 4751Hz sine wave superimposed to drive the laser signal. The generated triangular wave signal is used for current driving, generating a square wave, and triggering subsequent signal acquisition. The generated sine wave is also used as a reference signal, entering the frequency doubling and phase-shifting circuits. Simultaneously, a separate heat insulation structure is designed to avoid the influence of temperature on the semiconductor laser output. Figure 3 The laser mounting base 1 is placed on the cold side of the TEC cooler 2, with its hot side in close contact with the heat sink 3. The laser mounting base 1 is isolated from the outside air by a heat insulation cover 4 made of bakelite with low thermal conductivity. To improve the thermal conductivity of the laser tube shell, the aluminum material of the laser mounting base 1 is replaced with copper to improve thermal conductivity.

[0022] The laser's temperature control module includes a temperature acquisition module and a temperature control module. The temperature acquisition section collects the resistance value of a thermistor using a bridge circuit design. The voltage difference from the thermistor is input to a 16-bit AD acquisition chip with built-in gain. A 10K negative temperature coefficient thermistor is used, and the bridge circuit reference voltage is 2.5V. The AD converter samples the resistance change. Temperature control employs PID control. The execution section mainly consists of a control circuit and a TEC (hotspot controller). The control circuit primarily comprises current direction control and duty cycle control. The current direction is controlled by two optocouplers, and the duty cycle is output by a microcontroller. The TEC selected is model TES1-04903, used for temperature control.

[0023] In this embodiment, the integrated miniature air chamber is based on the Heriot-Ribbent cell, with a designed flow cell optical path of 3m and an overall internal cavity size of [missing information]. To improve the reflectivity of the mirror, a hard dielectric film is applied using tantalum oxide as the dielectric, which improves both reflectivity and corrosion resistance. Liquid sealant is used for sealing to ensure long-term airtightness.

[0024] S102 collects the first and second harmonic signals of the optical signal after passing through the gas to be tested.

[0025] See further Figure 2 When the laser signal passes through the gas to be tested, the intensity of the laser signal changes, and the projected signal is detected by a photodetector. Because the intensity of the projected signal is too low, an amplifier is used to amplify the signal before processing. The amplified projected light signal enters a lock-in amplifier and a low-pass filter, which demodulates the amplified projected light signal into first and second harmonic signals. The first and second harmonic signals are acquired by the signal acquisition module and then sent to the signal processing module.

[0026] S103, using the first harmonic signal and the second harmonic signal to perform optical energy compensation to obtain inversion parameters.

[0027] Traditional tunable diode laser absorption spectroscopy detection methods typically use the peak amplitude of the second harmonic signal as the inversion parameter for gas concentration. While this approach, which relies directly on a single harmonic amplitude, has the advantage of being simple and clear in its theoretical model, its applicability in practical measurement systems faces significant challenges. Because this method bases the detection signal entirely on the absolute amplitude of the second harmonic, which is not only related to the intensity attenuation caused by gas absorption but also closely related to the initial light intensity level throughout the optical transmission path, the system cannot effectively distinguish and compensate for various intensity interference factors that may occur during the measurement process. Specifically, factors such as contaminants gradually accumulating in the optical window over time, minor changes in the collimation state of the optical path due to mechanical vibration or thermal deformation, and fiber optic connector insertion and removal losses can all lead to significant attenuation of transmitted light energy. These light intensity changes caused by non-gas absorption factors are directly superimposed on the harmonic signal, causing systematic deviations in the concentration inversion results.

[0028] To effectively address this issue, a wavelength modulation-based optical energy compensation method is proposed, which uses the ratio of the second harmonic signal to the first harmonic signal as the input parameter for subsequent concentration inversion. Although the first harmonic signal component is also affected by gas absorption, its main component reflects the light energy intensity reference of the detection system. With optimized modulation depth, the amplitude of the first harmonic signal maintains a good linear relationship with the average light intensity incident on the detector. When the light intensity attenuates due to window contamination or optical path misalignment in the optical transmission path, the amplitudes of the second and first harmonics decrease proportionally, while their ratio remains relatively stable. The ratio calculation makes the detection signal no longer dependent on the absolute value of the light intensity, but rather on the relative relationship between the two harmonic components. This not only compensates for the influence of light intensity fluctuations but also suppresses periodic noise such as optical interference fringes to a certain extent, because these noises have different transfer functions on different harmonic components.

[0029] Based on the phase-locked detection principle in a lock-in amplifier, a digital phase-locked detector is designed to demodulate the first and second harmonics (1f and 2f signals) in the transmitted signal. The signal strength of the projected light signal detected by the photodetector is then determined. And the formula is expressed as follows: in: It is the intensity of the incident light. It is the laser modulation depth. is the instantaneous frequency of the laser, L is the length of the gas chamber, and c is the gas concentration.

[0030] After receiving the aforementioned light intensity signal, the photodetector detects the components in the projected light intensity signal that are integer multiples of the reference frequency (i.e., harmonics) through a lock-in amplifier, and outputs their amplitude.

[0031] According to the above Formula representation and known , , And L obtain c. Based on c, determine the amplitudes of the first harmonic signal and the second harmonic signal: in: It is the average light intensity. It is the intensity modulation coefficient. It is the phase difference between intensity modulation and frequency modulation. , It is a parameter related to system gain. It is the first harmonic component of the Fourier expansion. It is the second harmonic component.

[0032] Then 2f / 1f can be expressed as: When there is no absorption or the concentration is very low, the denominator contains The item can be ignored, and system calibration can make it possible. and Given that... Therefore, the above equation can be simplified to: The above equation can be simplified to: Since other system parameters are constants when the modulation parameter is fixed, the final result is: The amplitude ratio of the second harmonic signal to the first harmonic signal is determined as the inversion parameter. This method can be used to eliminate the initial light intensity. The impact of fluctuations improves detection accuracy. S104, The moisture content of the gas to be measured is obtained by nonlinear fitting and inversion of water vapor concentration using the inversion parameters.

[0033] After extracting the second and first harmonics using a lock-in amplifier, their ratio is used as the input for spectral signal preprocessing. The original signal has characteristics such as spikes and fluctuations, while the water vapor absorption spectrum for trace water detection may be completely submerged in noise. Therefore, the signal needs to be preprocessed before inversion calculation in order to better extract characteristic signal parameters and subsequent solutions.

[0034] In this embodiment, time averaging filtering and spectral smoothing filtering algorithms are mainly used. Time averaging filtering is a classic signal processing technique based on statistical theory. Its core principle is to enhance the signal-to-noise ratio by utilizing the different statistical characteristics of deterministic signals and random noise in the time dimension. When the system performs N repeated measurements, the useful signal components with periodicity and phase consistency exhibit coherent superposition characteristics, and their amplitude increases proportionally to the number of averages N. Random noise, as a stationary random process, is uncorrelated between different measurement instances, exhibiting incoherent superposition, and its amplitude increases only proportionally to the number of averages N. The improvement in signal-to-noise ratio (SNR) is directly proportional to this difference. Proportional.

[0035] Let the signal collected in one instance be: ,in: For deterministic signals, For the first The noise from the first measurement. Then, after N averagings... Output discrete signal sequence The noise power is thus reduced to 1 / N of its original value. In the trace gas detection of this system, this technology effectively improves the detection capability of weak absorption signals through the synchronous accumulation and averaging of multi-cycle signals.

[0036] Moving average filtering is a classic digital filtering technique that smooths signals in the time or spatial domain. Its core principle is to replace the current data point with the local average value of consecutive data points in the signal, thereby suppressing random noise and short-term fluctuations while preserving the long-term trend and main characteristics of the signal.

[0037] For a discrete signal sequence Using a fixed length The window, The number of data points is usually odd. The center of the window is aligned with the nth data point to be processed, and the arithmetic mean of all data points within the window is used as the output value for that point. ,but: in, The window length determines the smoothness.

[0038] The window slides sequentially, traversing the entire signal sequence to achieve smoothing of each data point. Window length The choice of window size is crucial for moving average filtering, requiring a balance between noise suppression and signal fidelity: a smaller window provides weaker smoothing but better preserves rapid signal changes and details, suitable for signals that change rapidly or where real-time performance is critical; a larger window provides significant smoothing but may lead to loss of signal details and edge effects, suitable for slowly changing signals or applications requiring extremely high smoothness. In this project, the team adopted an adaptive window strategy: automatically increasing the window length when signal noise is high, and decreasing the window length to maintain response speed when signal quality is good or changes rapidly.

[0039] Two filtering methods are used to process the original signal, effectively suppressing high-frequency random noise with extremely low computational complexity, thereby significantly improving signal quality. This is crucial for subsequent accurate extraction of second harmonic peaks or spectral fitting, effectively reducing inversion errors caused by random noise and ensuring the system maintains rapid response capabilities while ensuring detection accuracy.

[0040] In this embodiment, the dew point temperature is established. and Minimum residual model for measured values: in, Forward theoretical model: in: These are the calibration parameters for the instrument. For the intensity of the absorption spectral line, For optical path, It is the value of the spectral line at its center frequency. For other relevant parameters, Dew point temperature The saturated water vapor pressure below Defined as: Enter the initial value of the dew point temperature. and parameter initial values ; Calculate the residual of the k-th iteration: Calculate the corresponding Jacobian matrix: in, for Dimensions ; From the Magnus formula established above, we can obtain: Solve the LM equations: in, , It is the identity matrix. The damping factor, Update parameters: Set iteration parameters The entire inversion iteration process will terminate upon the fulfillment of any of the following conditions: when Less than When the maximum number of iterations is reached, the inversion iteration terminates, and the final output is given. When the minimum residual model E is obtained... corresponding Moisture content of the reactant gas.

[0041] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, the simultaneous existence of A and B, or the existence of B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects have an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, and c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0042] The above description is merely a specific embodiment of this application. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the protection scope of this application. The protection scope of this application should be determined by the protection scope of the claims.

Claims

1. A method for detecting trace water content based on tunable laser absorption spectroscopy, characterized in that, include: A laser generator produces a light signal that is irradiated into an integrated miniature gas chamber, which stores the gas to be tested. Collect the first and second harmonic signals of the optical signal after it passes through the gas to be tested; Inversion parameters are obtained by using the first harmonic signal and the second harmonic signal for optical energy compensation. The moisture content of the gas to be measured is obtained by nonlinear fitting and inversion of water vapor concentration using the inversion parameters.

2. The method for detecting trace water content based on tunable laser absorption spectroscopy according to claim 1, characterized in that, The method of using a laser generator to generate a light signal and irradiate it into an integrated miniature gas chamber includes: The signal generation module generates high-frequency sine wave and low-frequency triangular wave signals; The high-frequency sine wave and low-frequency triangular wave signals are combined into a modulation signal by an adder to drive the laser. After being excited by the modulation signal, the laser generates an optical signal that enters the integrated miniature gas chamber and penetrates the gas to be measured.

3. The method for detecting trace water content based on tunable laser absorption spectroscopy according to claim 1, characterized in that, The first and second harmonic signals of the optical signal collected after passing through the gas to be tested include: The projected light signal penetrates the gas to be measured using a photodetector; The projected light signal is input into a transimpedance gain amplifier for signal amplification; The amplified projected light signal is sequentially input into a lock-in amplifier and a low-pass filter to demodulate and obtain the first harmonic signal and the second harmonic signal.

4. The method for detecting trace water content based on tunable laser absorption spectroscopy according to claim 1, characterized in that, The method of obtaining inversion parameters by optical energy compensation using the first harmonic signal and the second harmonic signal includes: Determine the signal strength of the projected light signal detected by the photodetector. And the formula is expressed as follows: in: It is the intensity of the incident light. It is the laser modulation depth. Where is the instantaneous frequency of the laser, L is the length of the gas chamber, and c is the gas concentration; According to the above Formula representation and known , , And L obtains the c; The amplitudes of the first harmonic signal and the second harmonic signal are determined according to c. The amplitude ratio of the second harmonic signal to the first harmonic signal is determined as the inversion parameter.

5. The method for detecting trace water content based on tunable laser absorption spectroscopy according to claim 4, characterized in that, Determining the amplitudes of the first and second harmonic signals based on c includes: in: It is the average light intensity. It is the intensity modulation coefficient. It is the phase difference between intensity modulation and frequency modulation. , It is a parameter related to system gain. It is the first harmonic component of the Fourier expansion. It is the second harmonic component.

6. The method for detecting trace water content based on tunable laser absorption spectroscopy according to claim 5, characterized in that, Determining the amplitude ratio of the second harmonic signal to the first harmonic signal as the inversion parameter includes: Wherein: when the concentration of the gas to be measured is very low The term can be ignored, and the above formula can be simplified to: Since other system parameters are constants when the modulation parameter is fixed, the final result is: .

7. The method for detecting trace water content based on tunable laser absorption spectroscopy according to claim 6, characterized in that, The step of obtaining the moisture content of the gas to be measured by performing nonlinear fitting inversion of water vapor concentration using the inversion parameters includes: The inversion parameters are used as input for spectral signal preprocessing to obtain a discrete signal sequence; The moisture content of the gas to be measured is obtained by nonlinear fitting and inversion of the water vapor concentration using the discrete signal sequence.

8. The method for detecting trace water content based on tunable laser absorption spectroscopy according to claim 7, characterized in that, Using the inversion parameters as input, spectral signal preprocessing is performed to obtain a discrete signal sequence, including: Let the signal collected in one instance be: ,in: For deterministic signals, For the first Noise from the second measurement; After N averages Output discrete signal sequence ; Regarding the Use a fixed length The window, with the center of the window aligned with the currently pending processing number. For each data point, the arithmetic mean of all data points within the window is used as the output value for that point. ,but: in, The window length determines the smoothness. The window slides through the entire signal sequence sequentially, achieving smooth processing of each data point.

9. The method for detecting trace water content based on tunable laser absorption spectroscopy according to claim 8, characterized in that, The step of obtaining the moisture content of the gas to be measured by nonlinear fitting and inversion of water vapor concentration using the discrete signal sequence includes: Establish dew point temperature and Minimum residual model for measured values: in, Forward theoretical model: in: These are the calibration parameters for the instrument. For the intensity of the absorption spectral line, For optical path, It is the value of the spectral line at its center frequency. For other relevant parameters, Dew point temperature The saturated water vapor pressure below Defined as: Enter the initial value of the dew point temperature. and parameter initial values ; Calculate the residual of the k-th iteration: Calculate the corresponding Jacobian matrix: in, for Dimensions ; From the Magnus formula established above, we can obtain: Solve the LM equations: in, , It is the identity matrix. The damping factor, Update parameters: When obtaining the minimum value of E in the minimum residual model corresponding .

10. The method for detecting trace water content based on tunable laser absorption spectroscopy according to claim 9, characterized in that, When obtaining the minimum value of the minimum residual model E corresponding ,include: Set iteration parameters ; when Less than When the maximum number of iterations is reached, the inversion iteration terminates, and the final output is given. .

Images