An industrial gas concentration detection system

Abstract

The present invention relates to the technical field of gas detection, and specifically discloses an industrial gas concentration detection system, which is used to solve the problems that the transmissive optical machine adopts a single path and a fixed compensation method, cannot measure multi-component gases simultaneously, environmental interference and multi-laser beam combination lead to signal attenuation, low detection accuracy, high cost and poor stability. The system includes a transmitting end and a receiving end. A laser is provided in the transmitting end, and a red light generator is inside. An off-axis parabolic mirror is provided on one side of the red light generator. A filter is provided above the off-axis parabolic mirror, and a photodetector is provided above the filter. A window one is provided on one side of the off-axis parabolic mirror. A window two is encapsulated in the receiving end, and a full-angle reflector is provided on one side of the window two. The photodetector is sequentially connected to a data inversion analysis module, a dynamic environment compensation module and an environmental sensing module. The present invention realizes multi-gas concentration detection by optimizing optical design and dynamic environment compensation.

Description

Technical Field

[0001] The present invention relates to the technical field of gas detection, and more specifically, the present invention relates to an industrial gas concentration detection system. Background Art

[0002] In the petrochemical, metallurgical, and environmental industries, a heating furnace is a key production equipment. Usually, multiple gas components coexist in the furnace. Among them, CO, N2O, and H2O are the most common gas components. The detection of the concentrations of these three gases plays an important role in production. The opposed-beam optical machine is a conventional detection device for detecting the gas component concentrations in a heating furnace. The traditional structure of the opposed-beam optical machine uses a near-infrared laser as the emission end. After the emitted laser is absorbed by the gas to be measured, a photodetector is placed as the receiving end to receive the optical signal. This method can only obtain the straight-line path from the emission end to the receiving end and can only detect one gas. Electrical wire lines need to be arranged at both the emission end and the receiving end, and the signal is easily interfered. There is also a method of combining the emitted lights of two or more lasers. After the combined emitted laser is absorbed by the gas to be measured, a reflecting mirror is placed to reflect the laser back to the emission end, and then the laser is separated by wavelength using a spectroscope with different wavelengths, and then corresponding photodetectors are sequentially placed for optical signal detection. Although this method obtains twice the straight-line path from the emission end to the receiving end, the use of multiple lasers and photodetectors will increase the use cost and bring trouble to the mechanical structure layout of the transceiver end. The use of multiple spectroscopes will also weaken the optical signal and affect the optical signal detection. The internal environment temperature and pressure of the heating furnace change violently and have strong time-variation. The traditional temperature and pressure fixed compensation method uses the ideal gas state equation. According to the standard data at the temperature and pressure pre-determined in the laboratory, a correction factor is calculated, and then a very simple linear compensation is performed according to the current temperature and pressure during measurement. However, usually a fixed coefficient is adopted, which is difficult to meet the requirements of non-linear changes in a dynamic environment. Summary of the Invention

[0003] In order to overcome the above defects of the prior art, the present invention provides an industrial gas concentration detection system, which realizes high-precision and stable multi-gas concentration detection through optimizing the optical design and dynamic environment compensation.

[0004] In industrial heating furnaces, there are significant differences in the sensitivity of the absorption characteristics of different gases to environmental changes. The absorption line of CO is wider than those of N2O and H2O, and its response to environmental fluctuations is relatively stable. Linear function compensation can be used to smoothly adjust the environmental dynamic compensation effect; N2O is very sensitive to temperature and pressure changes, and its absorption spectrum line is narrow. A function with a faster non-linear response is required to quickly adjust the compensation weight when the environment changes suddenly; the absorption characteristics of H2O are extremely dependent on humidity changes, and its response shows obvious non-linearity. It is necessary to significantly amplify the error caused by humidity changes during feedback to effectively compensate. The setting of different gating functions can effectively reduce the measurement error caused by environmental changes, reduce the dependence of overall industrial gas detection on the performance of the environmental sensing module, reduce the use cost, and reduce the cross-interference of each gas in multi-component gas detection, thereby improving the accuracy of gas concentration inversion.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] An industrial gas concentration detection system includes a transceiver and a reflection end. There is a gas to be measured between the transceiver and the reflection end. A laser is provided in the transceiver. The wavelength range of the emitted laser beam after temperature and current modulation by the laser is 2202.8 - 2205.6 cm -1 (4533.91 - 4539.67 nm). A red light generator is encapsulated inside it. An off-axis paraboloid mirror with a central opening is provided on one side of the laser. A filter is provided above the off-axis paraboloid mirror with a central opening. A photodetector is provided above the filter, which is responsible for converting the processed optical signal into an electrical signal. A window plate I is provided on the side of the off-axis paraboloid mirror away from the laser. A window plate II is encapsulated in the reflection end. A full-angle reflector is provided on one side of the window plate II. The photodetector is connected to a data inversion analysis module. The data inversion analysis module is connected to a dynamic environment compensation module. The dynamic environment compensation module is connected to an environmental sensing module. The environmental sensing module real-time collects the temperature, humidity, and pressure in the heating furnace and transmits them to the dynamic environment compensation module. The dynamic environment compensation module uses the acquired data to construct a normalized vector of environmental variables, non-linearly maps the normalized deviation of environmental parameters using a preset coupling analysis formula to obtain an environmental parameter mapping term, weights and accumulates the historical gas concentration inversion errors within a set time window to obtain an environmental parameter time integral term, and based on the environmental parameter mapping term and the environmental parameter time integral term, uses a preset dynamic environment compensation formula to obtain a dynamic environment compensation value. The preset dynamic environment compensation formula is:

[0007] ;

[0008] In the formula: is the time variable, is the dynamic environment compensation factor, is the environmental parameter mapping term, is the environmental parameter time integral term, is the gating function, and respective appropriate gating functions are used for the detection of the concentrations of CO, N2O, and H2O gases. is the overall scaling factor, which is calibrated offline.

[0009] As a further solution of the present invention, the optical path between the transceiver and the reflector includes:

[0010] The red light indicating light emitted by the red light generator and the infrared light emitted by the laser are combined as the outgoing light beam. First, it passes through the middle opening of the off-axis parabolic mirror and the first window piece and enters the area of the gas to be measured to complete the first absorption; then the laser beam passes through the second window piece and shoots towards the total angle reflector. After being reflected along the original path, it passes through the second window piece again and enters the area of the gas to be measured for the second absorption (to effectively extend the optical path and enhance the absorption signal). The reflected light beam passes through the first window piece and is incident on the parabolic surface of the off-axis parabolic mirror again. Its reflected light passes through the filter and is focused on the light-sensitive surface of the photodetector. After the optical signal collected by the photodetector is photoelectrically converted into an electrical signal, the data inversion analysis module performs detection electrical signal processing and gas concentration inversion calculation, performs dynamic environment correction based on the dynamic environment compensation algorithm, and feeds back to control the laser.

[0011] As a further solution of the present invention, in the dynamic environment compensation module, in the dynamic environment compensation formula, the gating function used for the detection of the CO gas concentration uses a linear combination of the environmental parameter mapping term and the environmental parameter time integral term and limits its output range within [0,1] to smoothly adjust the detection of the CO gas concentration. The formula of the gating function used for the detection of the CO gas concentration in the dynamic environment compensation formula is:

[0012] ;

[0013] In the formula: is the gating function value used for the detection of the CO gas concentration, , are respectively the environmental parameter mapping term and the environmental parameter time integral term during the detection of the CO gas concentration. The environmental parameters involve temperature, humidity, and pressure. , are respectively the weight factors of the environmental parameter mapping term and the environmental parameter time integral term during the detection of the CO gas concentration.

[0014] As a further solution of the present invention, in the dynamic environment compensation module, the gating function used in the dynamic environment compensation formula for N2O gas concentration detection enables the N2O optoelectronic signal to adjust the dynamic environment compensation weight at a set speed when reaching the set critical point, improving the feedback sensitivity during environmental mutations. The gating function formula used in the dynamic environment compensation formula for N2O gas concentration detection is as follows:

[0015] ;

[0016] In the formula: is the gating function value for N2O gas concentration detection, 、 are the environmental parameter mapping term and the environmental parameter time integration term respectively during N2O gas concentration detection. The environmental parameters involve temperature and pressure, 、 are the weight factors of the environmental parameter mapping term and the environmental parameter time integration term respectively during N2O gas concentration detection.

[0017] As a further solution of the present invention, in the dynamic environment compensation module, the gating function used in the dynamic environment compensation formula for H2O gas concentration detection adopts an exponential function to compensate for the error caused by the change in water vapor concentration due to humidity. The gating function formula used in the dynamic environment compensation formula for H2O gas concentration detection is as follows:

[0018] ;

[0019] In the formula: is the gating function value for H2O gas concentration detection, 、 are the environmental parameter mapping term and the environmental parameter time integration term respectively during H2O gas concentration detection. The environmental parameters involve humidity, 、 are the weight factors of the environmental parameter mapping term and the environmental parameter time integration term respectively during H2O gas concentration detection.

[0020] As a further solution of the present invention, in the data inversion analysis module, the process of processing the detection electrical signal, inversely calculating the gas concentration, performing dynamic environment correction based on the dynamic environment compensation algorithm, and feedback controlling the laser includes:

[0021] Step 1, signal acquisition and preprocessing: The photodetector converts the optical signal after two absorptions into an electrical signal, and filters the electrical signal to remove background noise and baseline drift.

[0022] Step 2, Calculation of dynamic environment compensation factor: The environment sensing module collects the temperature, humidity, and pressure inside the heating furnace in real time, constructs a normalized environment state vector, performs a non-linear mapping on the environment parameters using a preset coupling analysis formula to obtain the environment parameter mapping terms. At the same time, within a set time window, an exponentially weighted integral is performed on the historical gas concentration inversion error to obtain the environment parameter time integral term. According to the type of the gas to be measured, the dynamic environment compensation module uses a corresponding gating function to determine the relative weights between the environment parameter mapping terms and the environment parameter time integral term, and calculates the current dynamic environment compensation factor according to the dynamic environment compensation formula;

[0023] Step 3, Signal correction and concentration inversion: Use the dynamic environment compensation factor obtained in Step 2 to correct the electrical signal obtained in Step 1. The corrected electrical signal undergoes an integration process to extract the integrated area of the absorption peak of the target gas. Combining with the Beer-Lambert law and a preset inversion formula, the current gas concentration is calculated;

[0024] Step 4, Closed-loop feedback regulation: Compare the current gas concentration obtained in Step 3 with the standard gas concentration, calculate the concentration error, feedback this concentration error into the dynamic environment compensation factor through the environment parameter time integral term, and at the same time transmit it as a feedback signal to the laser. The laser uses this error to adjust the modulation parameters of the temperature and current to ensure that the output wavelength is maintained within the preset range, optimize the absorption signal, and minimize the data inversion error;

[0025] Step 5, Continuous online adaptive update: The environment sensing module continuously collects data. The dynamic environment compensation module uses the latest environment parameter mapping terms and environment parameter time integral terms to feedback and calculate the latest dynamic environment compensation factor. The corrected electrical signal undergoes data inversion to obtain the gas concentration, and the gas concentration error is fed back into the dynamic environment compensation and laser control. The weight parameters in each gas gating parameter are continuously optimized through an online learning and gradient descent adaptive update algorithm.

[0026] As a further solution of the present invention, in the dynamic environment compensation module, the preset coupling analysis formula considers the non-linear coupling effect between the temperature, pressure, and humidity inside the heating furnace. Based on the coupling analysis formula of the radial basis function network, the normalized environment parameters are mapped into environment parameter mapping terms. The preset coupling analysis formula is:

[0027] ;

[0028] In the formula: is the index of the preset basis function, is the number of preset basis functions, is the weight of each basis function, through offline calibration, is the The central vector of a base function, which is derived from the clustering analysis of historical environmental data, is the normalized environmental parameter vector, , 、 、 are the temperature, pressure, and humidity data collected in real time by the environmental sensing module at time respectively, 、 、 are the temperature reference value, pressure reference value, and humidity reference value for calibrating the standard conditions respectively, 、 、 are the standard deviations of temperature, humidity, and pressure respectively, which are statistically obtained based on historical data, is the inverse matrix of the covariance matrix of the th base function, which is used to quantify the influence of actual deviation on the mapping value.

[0029] As a further solution of the present invention, in the data inversion analysis module, step 4 of the closed-loop feedback adjustment of the temperature and current modulation parameters of the laser includes temperature modulation feedback, current modulation feedback, and joint control. Among them, the temperature modulation feedback uses the PID algorithm to obtain the temperature adjustment amount according to the detected gas concentration error, and transmits the temperature adjustment amount to the laser to adjust the working temperature of the laser chip, so that the output laser wavelength shifts towards the target absorption peak. The current modulation feedback uses the PID control algorithm to obtain the current adjustment amount, and transmits the current adjustment amount to the laser to adjust the baseline current and modulation amplitude of the drive circuit, and adjusts the output power and wavelength accordingly. The joint control is to simultaneously adjust the working temperature of the laser chip, the output power, and wavelength of the drive circuit based on the decoupling algorithm and the online self-learning strategy, and keep the output stable within the set threshold.

[0030] As a further solution of the present invention, the central opening diameter of the off-axis parabolic mirror is set between 3 and 5 mm, and the appropriate focal length and curvature are selected according to the optical path length and the expected spot size. The installation angle accuracy is controlled within ±0.1°. Both window pane 1 and window pane 2 are high-temperature resistant window panes, made of sapphire, with a thickness between 1 and 3 mm. Their surfaces are provided with anti-reflection and anti-heat coatings. The filter is a mid-infrared filter with a central wavelength of 4.535 μm, a bandwidth of 10 nm, a transmittance greater than or equal to 90% in the target wavelength band, and an optical density greater than or equal to 4 in the out-of-band region of 2.2 - 2.4 μm and related interference wavelength bands.

[0031] As a further solution of the present invention, the laser is a mid-infrared quantum cascade laser, and the output power of the red light indicating light accounts for 0.5% - 2% of the output power of the infrared light.

[0032] Technical effects of the industrial gas concentration detection system of the present invention:

[0033] The system proposed by the present invention is simple. When collimating the optical path for alignment, the operation is simple. By using the filter, it can filter out the interference of the CO absorption spectrum line within the set threshold range of 2.3 μm and the N2O and H2O absorption spectrum lines at other positions, improving the signal-to-noise ratio. Using a single optical path, a single laser, and a single photodetector can simultaneously detect the concentrations of three molecules, CO, N2O, and H2O. Utilizing the reflection function of the passive end total angle mirror, the laser detection optical path length is twice that of the ordinary opposed beam optical path, increasing the absorption optical path to improve the detection accuracy of gas concentration; using the dynamic environment compensation algorithm and closed-loop feedback control technology, it realizes the real-time adaptive correction of the laser output error caused by environmental fluctuations such as temperature, pressure, and humidity in the heating furnace, improving the response speed and anti-interference performance under complex working conditions in the heating furnace. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is the system block diagram of the system proposed by the present invention;

[0035] Figure 2 It is the schematic diagram of the system optical path proposed by the present invention;

[0036] Figure 3 It is the direct absorption spectrogram of simulating CO, N2O, and H2O of the present invention;

[0037] Figure 4 It is the second harmonic absorption signal diagram of simultaneously measuring CO, N2O, and H2O of the present invention;

[0038] Figure 5 It is the working flowchart of the data inversion analysis module of the present invention;

[0039] In the figure: 01 - laser, 02 - off-axis paraboloid mirror, 03 - window one, 04 - filter, 05 - photodetector, 06 - window two, 07 - total angle mirror, 101 - transceiver end, 102 - reflection end. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0040] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0041] Embodiment 1

[0042] As Figure 1As shown in the figure, an industrial gas concentration detection system proposed by the present invention includes a transceiver 101 and a reflection end 102. There is a gas to be measured between the transceiver 101 and the reflection end 102. A laser 01 is provided inside the transceiver 101. The wavelength range of the emitted laser beam after temperature and current modulation of the laser 01 is 2202.8 - 2205.6 cm -1 (4533.91 - 4539.67 nm). A red light generator is encapsulated inside it. An off-axis paraboloid mirror 02 with a middle opening is provided on one side of the laser 01. A filter 04 is provided above the off-axis paraboloid mirror 02 with a middle opening. A photodetector 05 is provided above the filter 04, which is responsible for converting the processed optical signal into an electrical signal. A window plate 03 is provided on the side of the off-axis paraboloid mirror 02 away from the laser 01. A window plate 06 is encapsulated inside the reflection end 102. A full-angle reflector 07 is provided on one side of the window plate 06. The photodetector 05 is connected to a data inversion analysis module. The data inversion analysis module is connected to a dynamic environment compensation module. The dynamic environment compensation module is connected to an environment sensing module. The environment sensing module collects the temperature, humidity and pressure in the heating furnace in real time and transmits them to the dynamic environment compensation module. The dynamic environment compensation module constructs a normalized vector of environmental variables using the acquired data, non-linearly maps the normalized deviation of environmental parameters using a preset coupling analysis formula to obtain an environmental parameter mapping term, weights and accumulates the historical gas concentration inversion errors within a set time window to obtain an environmental parameter time integration term. Based on the environmental parameter mapping term and the environmental parameter time integration term, a dynamic environment compensation value is obtained using a preset dynamic environment compensation formula. The preset dynamic environment compensation formula is:

[0043] ;

[0044] In the formula: is the time variable, is the dynamic environment compensation factor, is the environmental parameter mapping term, is the environmental parameter time integration term, is the gating function. Appropriate gating functions are used for the detection of CO, N2O, and H2O gas concentrations respectively (in industrial heating furnaces, the absorption characteristics of different gases have significant differences in sensitivity to environmental changes. By setting different gating functions for CO, N2O, and H2O respectively, the measurement error caused by environmental changes is effectively reduced, the dependence of the system on the performance of the environment sensing module is reduced, thereby reducing the use cost and weakening the cross-interference between gases in multi-component gas detection, and finally significantly improving the gas concentration inversion accuracy), is the overall scaling coefficient, which is calibrated offline.

[0045] It should be noted that the central opening diameter of the off-axis parabolic mirror 02 is set between 3 and 5 mm. The focal length and curvature are selected according to the optical path length and the expected spot size, and the installation angle accuracy is controlled within ±0.1°. Both the first window 03 and the second window 06 are high-temperature resistant windows made of sapphire with a thickness between 1 and 3 mm, and their surfaces are provided with anti-reflection and anti-heat coatings. The central wavelength of the filter 04 is 4.535 μm, the bandwidth is 10 nm, the transmittance in the target band is greater than or equal to 90%, and the optical density in the out-of-band region of 2.2 - 2.4 μm and related interference bands is greater than or equal to 4. In the laser 01, the output power of the red indicating light accounts for 0.5% - 2% of the output power of the infrared light.

[0046] By strictly limiting the central opening diameter of the off-axis parabolic mirror 02 between 3 and 5 mm, and precisely selecting the appropriate focal length and curvature according to the optical path length and the expected spot size, and controlling the installation angle within ±0.1°, it is ensured that the laser beam remains highly collimated and has a low wavefront error after reflection, achieving stable focusing and efficient optical transmission. At the same time, the first window 03 and the second window 06 made of sapphire with a thickness controlled between 1 and 3 mm and anti-reflection and anti-heat coatings applied on their surfaces are used to ensure high light transmittance and excellent thermal stability in a high-temperature environment, avoiding the degradation of optical performance caused by thermal expansion or thermal interference. In addition, the filter 04 is designed with a central wavelength of 4.535 μm and a bandwidth of 10 nm, with a transmittance not less than 90% in the target band, and an optical density reaching or exceeding 4 in interference bands such as 2.2 - 2.4 μm, effectively filtering out stray light and interference signals other than the absorption lines of CO, N2O, and H2O to ensure the purity and high signal-to-noise ratio of the detection spectrum. Furthermore, the red light generator encapsulated in the laser 01 has an output power of the red indicating light accounting for only 0.5% - 2% of the output power of the infrared light, which can provide sufficient visual alignment assistance without interfering with the infrared signal.

[0047] It should be noted that, as Figure 2 shown, the optical path between the transceiver 101 and the reflector 102 includes:

[0048] The red indicating light emitted by the red light generator is combined with the infrared light emitted by the laser 01 to form an outgoing light beam. First, it passes through the middle opening of the off-axis parabolic mirror 02 and the window pane 03, enters the area of the gas to be measured, and completes the first absorption. Subsequently, the laser beam passes through the window pane 06 and then shoots towards the full-angle reflector 07. After being reflected along the original path, it passes through the window pane 06 again and enters the area of the gas to be measured for the second absorption (to effectively extend the optical path and enhance the absorption signal). The reflected light beam passes through the window pane 03 and then is incident on the parabolic surface of the off-axis parabolic mirror 02. Its reflected light passes through the filter 04 and is focused on the photosensitive surface of the photodetector 05. The optical signal collected by the photodetector 05 is converted into an electrical signal through photoelectric conversion, and then the data inversion analysis module performs detection electrical signal processing and gas concentration inversion calculation, conducts dynamic environment correction based on the dynamic environment compensation algorithm, and feedback controls the laser 01.

[0049] The system proposed by the present invention is simple, and the operation of collimating the optical path during optical alignment is easy. By using the filter 04, it can filter out the interference of the CO absorption spectrum line within the set threshold range of 2.3μm and the N2O and H2O absorption spectrum lines at other positions, improving the signal-to-noise ratio. Using a single optical path, a single laser 01, and a single photodetector 05 can simultaneously detect the concentrations of three molecules, namely CO, N2O, and H2O. By utilizing the reflection function of the passive full-angle reflector 07, the length of the laser detection optical path is twice that of the ordinary opposed-beam optical path, enhancing the absorption optical path to improve the detection accuracy of the gas concentration.

[0050] It should be noted that in the dynamic environment compensation module, in the dynamic environment compensation formula, the gating function used for CO gas concentration detection uses a linear combination of the environmental parameter mapping term and the environmental parameter time integration term and limits its output range within [0, 1] to smoothly adjust the CO gas concentration detection. The formula of the gating function used for CO gas concentration detection in the dynamic environment compensation formula is:

[0051] ;

[0052] In the formula: is the gating function value used for CO gas concentration detection, 、 are respectively the environmental parameter mapping term and the environmental parameter time integration term during CO gas concentration detection. The environmental parameters involve temperature, humidity, and pressure. 、 are respectively the weight factors of the environmental parameter mapping term and the environmental parameter time integration term during CO gas concentration detection, which are determined based on offline calibration and historical data statistics.

[0053] Since CO has a relatively wide absorption line and responds relatively smoothly to temperature, pressure, and humidity fluctuations, linear function compensation is used to achieve smooth adjustment. By using the environmental parameter mapping term and the environmental parameter time integral term are linearly combined with weight factors and respectively, and the output is limited within the range of [0, 1] by using the min–max function to form a gating function . This technical solution enables smooth adjustment of CO gas concentration detection in the heating furnace, reduces errors caused by temperature, humidity, and pressure fluctuations, thereby reducing the dependence on the performance of environmental sensors, improving the accuracy of data inversion and the overall stability of the system. At the same time, it reduces cross-interference in multi-component gas detection and ensures reliable detection results.

[0054] It should be noted that in the dynamic environment compensation module, the gating function used in the dynamic environment compensation formula for N2O gas concentration detection enables the N2O optoelectronic signal to adjust the dynamic environment compensation weight at a set speed when reaching the set critical point, improving the feedback sensitivity during environmental mutations. The gating function formula used in the dynamic environment compensation formula for N2O gas concentration detection is:

[0055] ;

[0056] In the formula: is the gating function value for N2O gas concentration detection, , are the environmental parameter mapping term and the environmental parameter time integral term respectively during N2O gas concentration detection. The environmental parameters involve temperature and pressure, , are the weight factors of the environmental parameter mapping term and the environmental parameter time integral term respectively during N2O gas concentration detection.

[0057] The absorption spectral line of N2O is narrow and extremely sensitive to temperature and pressure changes. A function with a rapid non-linear response is required to quickly adjust the compensation weight during environmental mutations. By using function to construct the gating function for N2O gas concentration detection , and introducing the linear combination of the environmental parameter mapping term and the environmental parameter time integral term , where and is a weighting factor, enabling the dynamic environmental compensation of N2O concentration to rapidly adjust the compensation weight according to the change of input signal in a short time when the environment mutates. Especially when dealing with environmental fluctuations such as temperature and pressure changes, it enhances the sensitivity of feedback, improves the response speed and accuracy of the detection system under rapid environmental changes, avoids the feedback delay caused by overcompensation, makes the real-time measurement of N2O gas concentration more stable and accurate, and effectively improves the adaptive ability to environmental changes in high-temperature or severe working conditions, thus enhancing the anti-interference performance of the whole system and the detection accuracy of multi-component gases.

[0058] It should be noted that in the dynamic environmental compensation module, the gating function used in the dynamic environmental compensation formula for H2O gas concentration detection adopts an exponential function to compensate for the error caused by humidity change in water vapor concentration. The formula of the gating function used in the dynamic environmental compensation formula for H2O gas concentration detection is:

[0059] ;

[0060] In the formula: is the gating function value for H2O gas concentration detection, 、 are respectively the environmental parameter mapping term and the environmental parameter time integral term during H2O gas concentration detection, and the environmental parameters involve humidity, 、 are respectively the weighting factors of the environmental parameter mapping term and the environmental parameter time integral term during H2O gas concentration detection.

[0061] The absorption characteristics of H2O extremely depend on humidity change, and its response shows obvious nonlinearity. A function that can significantly amplify the error caused by humidity change during feedback needs to be designed. This technical solution realizes precise compensation for the nonlinear error caused by humidity change in H2O gas concentration detection. The principle is that first, the environmental sensing module is used to collect the humidity data in the heating furnace in real time, and after normalization processing, the environmental parameter mapping term (reflecting the deviation of humidity relative to the reference value and standard deviation) is constructed. At the same time, the exponential weighted integral of the historical concentration inversion error within the set time window is obtained to get the environmental parameter time integral term ; Then, through the preset weighting factors and for 、 Perform non-linear weighted combination, and utilize the fast decay characteristic of the exponential function to significantly amplify the influence of humidity fluctuations on the water vapor absorption signal, enabling the system to rapidly adjust the dynamic environment compensation weight when encountering rapid humidity changes, real-time correct the measurement deviation caused by humidity fluctuations, significantly improve the accuracy of water vapor concentration inversion and the overall stability of the system, reduce the dependence on the high-precision performance of environmental sensors and the cross-interference risk between multi-component gases, thereby providing a strong technical guarantee for the real-time, accurate, and stable on-line detection of H2O gas concentration in industrial heating furnaces, and further enhancing the response speed and anti-interference ability of the system in complex high-temperature dynamic environments.

[0062] It should be noted that, as Figure 5 shown, in the data inversion analysis module, the process of processing the detection electrical signal, inverting and calculating the gas concentration, performing dynamic environment correction based on the dynamic environment compensation algorithm, and feedback controlling the laser 01 includes:

[0063] Step 1, signal acquisition and preprocessing: The photodetector 05 converts the optical signal after two absorptions into an electrical signal, and filters the electrical signal to remove background noise and baseline drift.

[0064] Step 2, calculation of dynamic environment compensation factor: The environment sensing module real-time collects the temperature, humidity, and pressure inside the heating furnace, constructs a normalized environment state vector, performs non-linear mapping on the environmental parameters using a preset coupling analysis formula to obtain the environmental parameter mapping terms. At the same time, within a set time window, perform exponential weighted integration on the historical gas concentration inversion error to obtain the environmental parameter time integration term. According to the different measured gases, the dynamic environment compensation module uses corresponding gating functions to determine the relative weights between the environmental parameter mapping terms and the environmental parameter time integration terms, and calculates the current dynamic environment compensation factor according to the dynamic environment compensation formula.

[0065] Step 3, signal correction and concentration inversion: Use the dynamic environment compensation factor obtained in Step 2 to correct the electrical signal obtained in Step 1. The corrected electrical signal is processed by integration, the integral area of the absorption peak of the target gas is extracted, and the current gas concentration is calculated in combination with the Beer-Lambert law and a preset inversion formula.

[0066] Step 4, closed-loop feedback regulation: Compare the current gas concentration obtained in Step 3 with the standard gas concentration, calculate the concentration error, feedback this concentration error into the dynamic environment compensation factor through the environmental parameter time integration term, and at the same time transmit it as a feedback signal to the laser 01. The laser 01 uses this error to adjust the modulation parameters of temperature and current to ensure that the output wavelength is maintained within the preset range, optimize the absorption signal, and minimize the data inversion error.

[0067] Step 5, continuous online adaptive update: The environmental sensing module continuously collects data. The dynamic environmental compensation module uses the latest environmental parameter mapping term and the environmental parameter time integral term to feedback and calculate the latest dynamic environmental compensation factor. The corrected electrical signal is inversely calculated to obtain the gas concentration. The gas concentration error is fed back into the dynamic environmental compensation and laser 01 control, and the weight parameters in each gas gating parameter are continuously optimized through an online learning and gradient descent adaptive update algorithm.

[0068] This technical solution implements closed-loop feedback regulation in the data inversion analysis module. First, the signal collected by the photodetector 05 is preprocessed, and then the temperature, humidity, and pressure data collected in real time by the environmental sensing module are used to construct a normalized vector. The dynamic environmental compensation factor is calculated through non-linear mapping and the exponentially weighted integral of the historical concentration error, and the original signal is corrected and the accurate gas concentration is inversely calculated. Subsequently, the concentration error is fed back to the laser 01 to automatically adjust the temperature and current modulation parameters, ensuring that the laser output wavelength is always maintained within the preset range, thereby effectively reducing environmental interference and data inversion error, improving the real-time performance, stability, and accuracy of the multi-component gas concentration detection in the heating furnace, enhancing the overall reliability of the system and reducing the maintenance cost.

[0069] It should be noted that in the dynamic environmental compensation module, the preset coupling analysis formula considers the non-linear coupling effect among the temperature, pressure, and humidity in the heating furnace. Based on the coupling analysis formula of the radial basis function network, the normalized environmental parameters are mapped into the environmental parameter mapping term. The preset coupling analysis formula is:

[0070] ;

[0071] In the formula: is the index of the preset basis function, is the number of preset basis functions, is the weight of each basis function, through offline calibration, is the center vector of the th basis function, originating from the clustering analysis of historical environmental data, is the normalized environmental parameter vector, 、 、 are the temperature, pressure, and humidity data collected in real time by the environmental sensing module at time respectively, 、 、 are the temperature reference value, pressure reference value, and humidity reference value for calibrating the standard conditions respectively, 、 、 They are the standard deviations of temperature, humidity, and pressure respectively, statistically obtained from historical data. is the inverse matrix of the covariance matrix of the

[0072] th basis function, used to quantify the impact of actual deviations on the mapping values. By adopting the coupling analysis formula based on the radial basis function network, the temperature, pressure, and humidity data collected in real time in the heating furnace are normalized to form the environmental parameter vector , and the weights of each basis function , center vectors and their inverse covariance matrix pairs obtained through offline calibration are used for non-linear mapping, so as to obtain the environmental parameter mapping term

[0073] This method can accurately capture and quantify the complex non-linear coupling effects among the temperature, pressure, and humidity in the heating furnace, provide an accurate environmental compensation basis for the dynamic environment compensation module, enable the system to adjust the compensation factor in real time when the environmental fluctuations cause laser output errors and absorption signal drifts, effectively reduce the measurement error, improve the accuracy of gas concentration inversion and the stability of the overall system, and at the same time reduce the dependence on the performance of high-precision sensors, thus providing a solid technical support for the long-term stable operation of the industrial gas detection system under high-temperature and dynamic working conditions.

[0074] It should be noted that in the data inversion analysis module, step 4 of the closed-loop feedback to adjust the temperature and current modulation parameters of the laser 01 includes temperature modulation feedback, current modulation feedback, and joint control. Among them, the temperature modulation feedback uses the PID algorithm to obtain the temperature adjustment amount according to the detected gas concentration error, transmits the temperature adjustment amount to the laser 01 to adjust the working temperature of the laser 01 chip, so that the output laser wavelength shifts towards the target absorption peak. The current modulation feedback uses the PID control algorithm to obtain the current adjustment amount, transmits the current adjustment amount to the laser 01 to adjust the baseline current and modulation amplitude of the drive circuit, and adjusts the output power and wavelength accordingly. The joint control is to simultaneously adjust the working temperature of the laser 01 chip and the output power and wavelength of the drive circuit based on the decoupling algorithm and the online self-learning strategy to keep the output stable within the set threshold.

[0074] The environmental dynamic compensation module proposed by the present invention utilizes the dynamic environment compensation algorithm and the closed-loop feedback control technology to achieve real-time adaptive correction of the laser output error caused by environmental fluctuations such as temperature, pressure, and humidity in the heating furnace, and improves the response speed and anti-interference performance under complex working conditions in the heating furnace.

[0075] Embodiment 2

[0076] Different from Example 1, this example is a specific description of the measurement of the concentrations of three molecules, namely CO, N2O, and H2O, in an industrial heating furnace based on the industrial gas concentration detection system proposed in Example 1. The limited temperature for the test is 296, the pressure is one standard atmosphere, and the absorption optical path length is 48.3 meters. The direct absorption spectrum is detected by the industrial gas concentration detection system proposed by the present invention, and the CO concentration in the industrial heating furnace at the current measurement point is obtained as 600 ppd, the N2O concentration is 280 ppd, and H2O is 0.3%. The direct absorption spectra of the three molecules, CO, N2O, and H2O, are as Figure 3 shown, Figure 4 showing the measured signals of the simultaneous measurement of the three molecules, CO, N2O, and H2O. In Figure 4 the three molecules, CO, N2O, and H2O, and their second harmonic absorption signals in the infrared band are sequentially marked. The temperature collected by the environmental sensing module is 300K, the pressure is 101.3 kPa, and the humidity is 65%, while the reference values of the system are 296K, 101.3 kPa, and 50% respectively, and the standard deviations of temperature, pressure, and humidity are set to = 2K, = 1 kPa, and = 5%. Based on this, the normalized environmental vector is calculated as:

[0077] =(2, 0, 3);

[0078] Through the preset radial basis function network coupling analysis formula, using the offline calibrated weight and basis function parameters, the environmental mapping term = 0.85 is calculated. At the same time, after exponentially weighted integration of the historical concentration error within the set time window, the feedback integration term = 0.15 is obtained.

[0079] For the CO concentration detection, a linear gating function is adopted:

[0080] ;

[0081] It is calculated that = 0.64.

[0082] Substitute these data into the dynamic environment compensation formula:

[0083] ;

[0084] (taking the overall scaling factor = 1), and it is calculated that = 0.946. When dynamic compensation was not carried out, the error of the detected CO concentration by the system was about 8%, while after calibration, using this compensation factor to correct the original optical signal, the error of the retrieved CO concentration was reduced to 1% - 2%, significantly improving the measurement accuracy. Similar data processing was also applied to the detection of N2O and H2O. By specifically adjusting the parameters of their respective gating functions, the whole system effectively compensated for the non - linear deviation caused by environmental factors, achieving high precision and robustness in the detection of multi - component gas concentrations.

[0085] In summary, as can be seen from Example 1 and Example 2, an industrial gas concentration detection system proposed by the present invention realizes high - precision, real - time, and stable online detection of the concentrations of multi - component gases such as CO, N2O, and H2O in the high - temperature and dynamically complex environment of an industrial heating furnace by optimizing the optical structure, strictly defining the parameters of key components, and combining intelligent regulation technologies based on dynamic environment compensation and closed - loop feedback control. At the same time, it significantly reduces the dependence of the system on high - precision sensors and the cross - interference between gases, and has high cost - effectiveness and the ability to operate stably for a long time.

[0086] The above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed by this application can easily think of changes or substitutions, which should all be covered within the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claimed rights.

[0087] Finally: The above is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An industrial gas concentration detection system, comprising a transceiver end (101) and a reflector end (102), characterized in that: A laser (01) is arranged in the transceiver end (101), a red light generator is arranged in the laser (01), an off-axis parabolic mirror (02) with a hole in the middle is arranged on one side of the laser (01), a filter (04) is arranged above the off-axis parabolic mirror (02), a photodetector (05) is arranged above the filter (04), a window piece 1 (03) is arranged on the side of the off-axis parabolic mirror (02) away from the laser (01), a window piece 2 (06) is arranged in the reflective end (102), and a full-angle reflector is arranged on one side of the window piece 2 (06) (07), the photoelectric detector (05) is connected to a data inversion analysis module, the data inversion analysis module is connected to a dynamic environment compensation module, the dynamic environment compensation module is connected to an environment sensing module, the dynamic environment compensation module uses the environmental interference data to construct a normalized vector and a normalized deviation of the environmental variable, obtains an environmental parameter mapping item, weightedly accumulates the historical gas concentration inversion error within a set time window, obtains an environmental parameter time integral item, and inputs the two into a dynamic environment compensation formula to dynamically compensate for environmental interference. The dynamic environment compensation formula is: K CF (t)=1+κ{[A(t)] G(t) ·[1+B(t)] 1-G(t) -1}; Where: t is the time variable, K CF (t) is the dynamic environment compensation factor, A(t) is the environment parameter mapping term, B(t) is the environment parameter time integral term, G(t) is the gating function, and the appropriate gating function is used for CO, N2O, and H2O gas concentration detection, and κ is the overall scaling factor; In the dynamic environment compensation module, the gate function used for CO gas concentration detection in the dynamic environment compensation formula uses a linear combination of environmental parameter mapping items and environmental parameter time integral items and limits its output range to [0,1] to smoothly adjust the CO gas concentration detection. The gate function formula used by the dynamic environment compensation formula for CO gas concentration detection is: G CO (t)=min(max(γ CO A CO (t)+δ CO B CO (t),0),1); Where: G CO (t) is the gating function value used for CO gas concentration detection, A CO (t), B CO (t)) are the environmental parameter mapping item and environmental parameter time integral item when detecting CO gas concentration, and the environmental parameters involve temperature, humidity and pressure, γ CO , δ CO They are the weight factors of the environmental parameter mapping item and the environmental parameter time integral item when performing CO gas concentration detection; In the dynamic environment compensation module, the gate function used for N2O gas concentration detection in the dynamic environment compensation formula enables the N2O photoelectric signal to adjust the dynamic environment compensation weight at a set speed when reaching the set critical point, thereby improving the feedback sensitivity when the environment changes suddenly. The gate function formula used for N2O gas concentration detection in the dynamic environment compensation formula is: Where: is the gating function value for N2O gas concentration detection, They are respectively the environmental parameter mapping item and the environmental parameter time integral item when performing N2O gas concentration detection. The environmental parameters involve temperature and pressure. They are the weight factors of the environmental parameter mapping item and the environmental parameter time integral item when performing N2O gas concentration detection; In the dynamic environment compensation module, the gate function used for H2O gas concentration detection in the dynamic environment compensation formula adopts an exponential function to compensate for the error of water vapor concentration caused by humidity changes. The gate function formula used for H2O gas concentration detection in the dynamic environment compensation formula is: Where: is the gating function value for H2O gas concentration detection, They are respectively the environmental parameter mapping item and the environmental parameter time integral item when performing H2O gas concentration detection. The environmental parameters involve humidity. They are respectively the weight factors of the environmental parameter mapping item and the environmental parameter time integral item when detecting the H2O gas concentration.

2. An industrial gas concentration detection system according to claim 1, characterized in that: The optical path between the transceiver end (101) and the reflector end (102) includes: the red light indicator light emitted by the red light generator and the infrared light emitted by the laser (01) are combined as an outgoing light beam, which first passes through the middle opening of the off-axis parabolic mirror (02) and the window piece 1 (03), enters the gas area to be measured, and completes the first absorption; then the laser beam passes through the window piece 2 (06) and is emitted to the full-angle reflector (07), after being reflected along the original path, it passes through the window piece 2 (06) again and enters the gas area to be measured, and is absorbed for the second time; the reflected light beam passes through the window piece 1 (03) and is incident on the parabola of the off-axis parabolic mirror (02); its reflected light passes through the filter (04) and is focused on the light sensitive surface of the photoelectric detector (05); the optical signal collected by the photoelectric detector (05) is converted into an electrical signal through photoelectric conversion, and then the data inversion analysis module performs detection electrical signal processing and gas concentration inversion calculation, performs dynamic environment correction based on the dynamic environment compensation algorithm, and feedback controls the laser (01).

3. An industrial gas concentration detection system according to claim 1, characterized in that: In the data inversion analysis module, the process of performing detection electrical signal processing, gas concentration inversion calculation, dynamic environment correction based on a dynamic environment compensation algorithm, and feedback control of the laser (01) includes: Step 1, signal acquisition and preprocessing: the photodetector (05) converts the light signal after two absorptions into an electrical signal, filters the electrical signal, removes background noise and baseline drift; Step 2, calculation of dynamic environmental compensation factor: the environmental sensing module collects the temperature, humidity and pressure in the heating furnace in real time, constructs a normalized environmental state vector, uses a preset coupling analysis formula to perform nonlinear mapping on the environmental parameters, obtains environmental parameter mapping items, and at the same time, performs exponential weighted integration on the historical gas concentration inversion error within the set time window to obtain the environmental parameter time integral item. According to the different gases being measured, the dynamic environmental compensation module uses the corresponding gating function to determine the relative weights between the environmental parameter mapping items and the environmental parameter time integral items, and calculates the current dynamic environmental compensation factor according to the dynamic environmental compensation formula; Step 3, signal correction and concentration inversion: The electrical signal obtained in step 1 is corrected using the dynamic environment compensation factor obtained in step 2. The corrected electrical signal is integrated to extract the integral area of ​​the target gas absorption peak. The current gas concentration is calculated by combining the Beer-Lambert law and the preset inversion formula. Step 4, closed-loop feedback control: compare the current gas concentration obtained in step 3 with the standard gas concentration, calculate the concentration error, feed back the concentration error to the dynamic environment compensation factor through the environmental parameter time integral term, and transmit it to the laser (01) as a feedback signal. The laser (01) uses this error to adjust the modulation parameters of temperature and current to ensure that the output wavelength is maintained within a preset range, optimize the absorption signal and minimize the data inversion error. Step 5, continuous online adaptive update: the environmental sensing module continuously collects data, the dynamic environmental compensation module uses the latest environmental parameter mapping item and the environmental parameter time integral item feedback to calculate the latest dynamic environmental compensation factor, the corrected electrical signal is inverted to obtain the gas concentration, the gas concentration error is fed back to the dynamic environmental compensation and laser (01) control, and the weight parameters in each gas gating parameter are continuously optimized through the adaptive update algorithm of online learning and gradient descent.

4. The industrial gas concentration detection system according to claim 1, characterized in that: In the dynamic environment compensation module, the nonlinear coupling effect between the temperature, pressure and humidity in the heating furnace is considered based on the preset coupling analysis formula. Based on the coupling analysis formula of the radial basis function network, the normalized environmental parameters are mapped into environmental parameter mapping items. The preset coupling analysis formula is: Where: j is the index of the preset basis function, J is the number of preset basis functions, ω j is the weight of the jth basis function, which is calibrated offline. j is the center vector of the jth basis function, which is derived from the cluster analysis of historical environmental data, x(t) is the normalized environmental parameter vector, T(t), P(t), and H(t) are the temperature, pressure, and humidity data collected by the environmental sensor module in real time at time t, respectively. ref , P ref , H ref are the temperature reference value, pressure reference value, and humidity reference value of the calibration standard conditions, respectively, T , σ P , σ H They are the standard deviations of temperature, humidity, and pressure based on historical data statistics. is the inverse matrix of the covariance matrix of the jth basis function, which is used to quantify the impact of the actual deviation on the mapping value.

5. An industrial gas concentration detection system according to claim 3, characterized in that: In the data inversion analysis module, step 4 closed-loop feedback adjustment of the temperature and current modulation parameters of the laser (01) includes temperature modulation feedback, current modulation feedback and joint control, wherein the temperature modulation feedback uses a PID algorithm to obtain a temperature adjustment amount according to the detected gas concentration error, and transmits the temperature adjustment amount to the laser (01) to adjust the operating temperature of the laser (01) chip so that the output laser wavelength shifts toward the target absorption peak; the current modulation feedback uses a PID control algorithm to obtain a current adjustment amount, and transmits the current adjustment amount to the laser (01) to adjust the baseline current and modulation amplitude of the drive circuit, and adjust the output power and wavelength accordingly; the joint control is based on a decoupling algorithm and an online self-learning strategy to simultaneously adjust the operating temperature of the laser (01) chip and the output power and wavelength of the drive circuit to keep the output stable within a set threshold.

6. An industrial gas concentration detection system according to claim 1, characterized in that: The diameter of the central opening of the off-axis parabolic mirror (02) is set between 3 and 5 mm, and the appropriate focal length and curvature are selected according to the optical path length and the expected spot size. The installation angle accuracy is controlled within ±0.1°. The window piece 1 (03) and the window piece 2 (06) are both high-temperature resistant windows made of sapphire with a thickness between 1 and 3 mm. The surface is provided with anti-reflection and heat-resistant coatings. The filter (04) is a mid-infrared filter with a central wavelength of 4.535 μm and a bandwidth of 10 nm. The projection rate in the target band is greater than or equal to 90%, and the optical density in the 2.2-2.4 μm out-band area and related interference bands is greater than or equal to 4.

7. An industrial gas concentration detection system according to claim 1, characterized in that: Laser (01) is a mid-infrared quantum cascade laser, which emits 2202.8-2205.6cm after modulation. -1 For mid-infrared lasers, the output power of the red indicator light accounts for 0.5% to 2% of the infrared light output power.

Images