A tdlas optical lens contamination detection method and system

Abstract

The application discloses a TDLAS optical lens pollution detection method and system, the method comprises the following steps: collecting an absorption curve signal of a TDLAS gas concentration sensor; according to a gas absorption line type model, deducting a gas absorption part from the absorption curve signal to obtain a signal containing a background baseline of a non-gas absorption area; segmenting the background baseline, calculating a discrete degree function value of the signal value in each segment; and judging whether the optical lens is polluted according to the fluctuation of the discrete degree function value. The application does not rely on the absolute value of the received light intensity to make the pollution judgment, and can effectively avoid the light intensity criterion failure problem caused by the reflection or interference of the pollutants in the remote measurement and weak signal measurement scene, accurately identifies the pollution state of the optical lens, and guarantees the long-term stable operation of the TDLAS gas concentration sensor.

Description

Technical Field

[0001] This application relates to the field of laser detection technology, and in particular to a TDLAS optical lens contamination detection method and system. Background Technology

[0002] During long-term online industrial operation, TDLAS gas concentration sensors are prone to dust accumulation, fogging, and oil buildup on the optical lens, causing the contaminant layer to directly reflect part of the output beam. Traditional transmitter-receiver gas chambers determine the impact of contaminant adhesion on the measurement by measuring the received light intensity; that is, they use a light intensity threshold to determine whether the sensor is working properly, and if it falls below the threshold, they will prompt an alarm for maintenance.

[0003] TDLAS telemetry emits a light beam through an optical lens, which is transmitted through space and diffusely reflected by a reflective surface before being received by the sensor's detector. Because the telemetry signal is weak, the actual light intensity may not decrease after the light reflected directly from the contaminant layer is received by the detector. In fact, if there is some oil film attached, the reflected light from the oil film will interfere with the emitted light, resulting in a stronger light intensity received by the detector.

[0004] Both the transmission-reception method and the telemetry method, in their original approaches, revolved around using the light intensity reduction threshold as a condition for determining contamination.

[0005] However, in scenarios involving weak signals, especially telemetry, contaminants on the transmitting optical lens not only fail to pass the threshold for contamination detection but also create periodic interference fringes on the gas absorption curve. Therefore, the extent of contaminant adhesion cannot be determined, affecting subsequent maintenance and potentially causing sensor failure due to contaminant measurement. Summary of the Invention

[0006] This application provides a TDLAS optical lens contamination detection method and system, aiming to solve the problem that existing methods that rely on light intensity reduction thresholds to determine lens contamination cannot effectively determine the contaminant adhesion status in telemetry and weak signal measurement scenarios because contaminant reflection / interference causes the received light intensity to increase instead of decrease, and contaminants superimpose interference fringes on the gas absorption curve.

[0007] Firstly, a TDLAS optical lens contamination detection method is provided, including:

[0008] Step S1: Acquire the absorption curve signal of the TDLAS gas concentration sensor;

[0009] Step S2: Based on the gas absorption line model, subtract the gas absorption portion from the absorption curve signal to obtain a background baseline signal that includes the non-gas absorption region.

[0010] Step S3: Divide the background baseline into segments and calculate the discreteness function value of the signal value in each segment; determine whether there is contamination in the optical lens based on the fluctuation of the discreteness function value.

[0011] Optionally, in the above scheme, step S3 specifically includes:

[0012] Extract the front-end background baseline and the back-end background baseline from the signal containing the background baseline of the non-gas absorption region, and calculate the piecewise variance of the front-end background baseline and the back-end background baseline respectively;

[0013] Based on the calculated piecewise variance, determine the K value that characterizes the degree of baseline fluctuation;

[0014] The K value is compared with a preset contamination threshold to determine the contamination status of the optical lens.

[0015] In the above scheme, optionally, the K value is the maximum value among the piecewise variances of the front-end background baseline and the back-end background baseline.

[0016] Optionally, in the above scheme, the determination of the contamination status of the optical lens specifically involves: when the K value is greater than the preset contamination determination threshold, determining that the optical lens is contaminated and issuing a maintenance warning message.

[0017] Optionally, in the above scheme, the segmented standard deviations of the front-end background baseline and the back-end background baseline can be calculated separately.

[0018] Based on the calculated segmented standard deviation, the K value, which characterizes the degree of baseline fluctuation, is determined.

[0019] In the above scheme, optionally, the gas absorption line model is a Lorentz line model, and the step of subtracting the gas absorption part from the absorption curve signal specifically involves: constructing an absorption curve model based on the Beer-Lambert law, and using the Lorentz line function to fit and subtract the gas absorption signal in the absorption curve that satisfies a fixed line width and shape.

[0020] Optionally, in the above scheme, the preset pollution judgment threshold is determined based on the actual pollution data corresponding to different pollutant types.

[0021] Optionally, in the above scheme, the absorption curve signal is acquired under the condition of a fixed number of sampling points and uniform wavelength sampling;

[0022] The segmentation of the background baseline specifically involves:

[0023] Extract the background baseline of the non-gas absorption region in the absorption curve, and divide the background baseline into equal-number segments according to the segment boundaries of the absorption curve, so that each segment contains the same number of sampling points.

[0024] Optionally, the above scheme can be applied to TDLAS gas sensing applications with telemetry or weak through-beam signals.

[0025] Secondly, a TDLAS optical lens contamination detection system is provided, comprising:

[0026] The signal acquisition module is used to acquire the absorption curve signal of the TDLAS gas concentration sensor;

[0027] The baseline extraction module is used to subtract the gas absorption portion from the absorption curve signal according to the gas absorption line model to obtain a background baseline signal that includes the non-gas absorption region.

[0028] The calculation and processing module is used to segment the background baseline, calculate the discreteness function value of the signal value in each segment, and determine whether there is contamination in the optical lens based on the fluctuation of the discreteness function value.

[0029] The output module is used to output the pollution determination result.

[0030] Compared with the prior art, this application has at least the following beneficial effects:

[0031] This application, based on further analysis and research into the problems of existing technologies, recognizes that existing methods relying on a light intensity reduction threshold to determine lens contamination have limitations. In telemetry and weak signal measurement scenarios, contaminant reflection / interference can cause the received light intensity to increase instead of decrease, and contaminants can superimpose interference fringes on the gas absorption curve, thus failing to effectively determine contaminant adhesion. This application addresses this issue by subtracting the gas absorption portion from the absorption curve to extract the background baseline, and then calculating the dispersion function value of the background baseline in segments. The fluctuation of the dispersion function value between each segment is used as the contamination criterion, eliminating reliance on the absolute value of the received light intensity. Therefore, it can accurately identify lens contamination even in scenarios where contaminants cause light intensity to increase instead of decrease. Lens contamination detection overcomes the limitations of traditional light intensity threshold methods in weak signal scenarios by transforming the presence of contaminants into local fluctuations in the background baseline. Furthermore, by quantitatively characterizing the fluctuation of the background baseline using discrete indices such as piecewise variance or piecewise standard deviation, it can sensitively capture the local fluctuations caused by interference fringes due to contaminants, enabling objective and quantitative judgment of the optical lens contamination status. This provides a reliable lens contamination detection method for TDLAS gas concentration sensors, especially in remote sensing and weak through-beam signal applications, ensuring the long-term stable operation and measurement effectiveness of the sensor. Attached Figure Description

[0032] Figure 1 A schematic flowchart of a TDLAS optical lens contamination detection method provided in one embodiment of this application;

[0033] Figure 2 An absorption curve and front-end and back-end baseline schematic diagram provided for one embodiment of this application;

[0034] Figure 3 This is an example of an embodiment of the present application showing the absorption curves of the same sensor with different contaminant attachments.

[0035] Figure 4 This is an overall flowchart of a TDLAS optical lens contamination detection method provided in one embodiment of this application. Detailed Implementation

[0036] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0037] In the description of this application, unless otherwise stated, the terms "including", "comprising", "having", etc., also mean "not limited to" (certain units, components, materials, steps, etc.).

[0038] Explanation of technical terms:

[0039] TDLAS, Tunable Diode Laser Absorption Spectroscopy, is a technique that allows for the absorption of light by a tunable laser diode.

[0040] The purpose of this application is to overcome the shortcomings of existing light intensity criteria and provide a TDLAS optical lens contamination detection method based on baseline variance fluctuation to determine the contaminant adhesion and provide maintenance information.

[0041] In one embodiment, reference Figure 1 and Figure 4 A TDLAS optical lens contamination detection method is provided, comprising:

[0042] Step S1: Acquire the absorption curve signal of the TDLAS gas concentration sensor;

[0043] Step S2: Based on the gas absorption line model, subtract the gas absorption portion from the absorption curve signal to obtain a background baseline signal that includes the non-gas absorption region.

[0044] Step S3: Divide the background baseline into segments and calculate the discreteness function value of the signal value in each segment; determine whether there is contamination in the optical lens based on the fluctuation of the discreteness function value.

[0045] This embodiment uses the background baseline of the non-gas absorption region in the TDLAS absorption curve as the analysis object. By detecting the local fluctuations of the background baseline caused by the interference fringes of contaminants, the lens contamination status is determined, thereby overcoming the failure defect of the traditional light intensity threshold method in weak signal scenarios.

[0046] In one embodiment, the absorption curve signal is acquired under the condition of a fixed number of sampling points and uniform wavelength sampling.

[0047] Specifically, the TDLAS gas concentration sensor uses a distributed feedback (DFB) laser. By controlling the injection current or temperature of the laser, the output wavelength of the laser is periodically scanned near the absorption line of the target gas. During the scan, the output wavelength is changed in constant step sizes, and the light intensity signal output by the detector is acquired at each wavelength step, forming an absorption curve within one scan cycle.

[0048] In one specific implementation, N sampling points are acquired in each scanning cycle, with equal wavelength intervals between adjacent sampling points. For example, N can be 1024, and the laser's scanning frequency is 100Hz, meaning 100 complete absorption curves can be obtained per second. The horizontal axis of the absorption curve signal represents the wavelength or the corresponding scanning time, and the vertical axis represents the light intensity signal received by the detector. The absorption curve signal is then sent to the processor for further processing after analog-to-digital conversion.

[0049] As an optional implementation, if the TDLAS sensor uses wavelength modulation spectroscopy, the absorption curve signal can also be a second harmonic signal demodulated by a lock-in amplifier, which also contains gas absorption information and background baseline information, and the method of this application is also applicable.

[0050] In one embodiment, the gas absorption line model is a Lorentz line model, and the step of subtracting the gas absorption portion from the absorption curve signal specifically involves: constructing an absorption curve model based on the Beer-Lambert law, and using the Lorentz line function to fit and subtract the gas absorption signal from the absorption curve that satisfies a fixed linewidth and shape.

[0051] Specifically, in TDLAS technology, the selective absorption of a specific wavelength of laser light by gas molecules satisfies the Beer-Lambert law, as shown in the wavelength expansion formula (1):

[0052] (1);

[0053] In the formula, This represents the transmitted light intensity, which is the actual light intensity value received by the detector at different wavelengths; This represents the incident light intensity, i.e., the initial light intensity emitted by the laser. denoted by , represents the absorption coefficient, indicating the strength of light absorption by a gas at different wavelengths. The value is larger at the gas absorption peak and close to zero in the non-absorption region; L represents the optical path length, which is the distance the laser beam travels in the gas. Wavelength represents the independent variable in the formula.

[0054] At normal pressure, gas absorption follows the Lorentz curve, which has a fixed linewidth and shape, so the absorption curve with this linewidth can be removed.

[0055] At normal pressure and temperature, the gas absorption line shape expressed in terms of wavelength is the Lorentz line shape, and its line shape formula is shown in (2).

[0056] (2);

[0057] The half-width of the expansion can be expressed as ; Represents the Lorentz linear function; Indicates wavelength; This represents the center wavelength, which is a fixed value under normal pressure and temperature.

[0058] Based on the above model, the gas absorption peak portion of the absorption curve is fitted: the Lorentz linear function is used to fit the gas absorption peak, and the position, amplitude, and linewidth parameters of the absorption peak are determined. Then, the fitted gas absorption signal is subtracted from the original absorption curve signal to obtain the background baseline signal after removing the influence of gas absorption.

[0059] In one specific implementation, the absorption curve can be fitted using the least squares method. The fitting model is the convolution of equation (1) and equation (2). The remaining signal after subtraction is the background baseline signal.

[0060] When the measurement environment is under low pressure, the gas absorption curve may conform to a Gaussian curve or a Voith curve. Accordingly, the gas absorption curve model may also be a Gaussian curve model or a Voith curve model, and should also fall within the protection scope of this application.

[0061] The "background baseline" mentioned in this application refers to the signal corresponding to the wavelength region in the absorption curve where there is no gas absorption. (Reference) Figure 2 The absorption curve diagram shown is composed of three parts: the front background baseline before the gas absorption peak, the gas absorption peak part at the center of the absorption peak, and the back background baseline after the gas absorption peak.

[0062] In one embodiment, segmenting the background baseline specifically involves:

[0063] Extract the background baseline of the non-gas absorption region in the absorption curve, and divide the background baseline into equal-number segments according to the segment boundaries of the absorption curve, so that each segment contains the same number of sampling points.

[0064] For example, if the front-end background baseline contains 200 sampling points, it is divided into 10 evenly spaced segments, each containing 20 sampling points; if the back-end background baseline contains 300 sampling points, it is also divided into 10 evenly spaced segments, each containing 30 sampling points. The number of sampling points in each segment is the same to ensure that the statistics of each segment are comparable.

[0065] The absorption curves are classified into two categories: the background baselines at both ends of the absorption curve and the gas absorption curve section. Then, the variance of the background baselines at both ends is calculated.

[0066] The background baseline is the baseline whose output wavelength increases as the laser is driven by current. Because the wavelength-current linearity of the DFB laser is very good, it can be considered as a straight line with a constant slope. At the same time, there is no gas absorption effect. Therefore, the normal uncontaminated background baseline is a linear baseline with variance close to 0 at the front and back ends. When lens contamination produces interference fringes, it causes uneven local fluctuations in the overall baseline, drastic changes in piecewise variance, and a significant increase in the K value.

[0067] Given a linear baseline signal:

[0068] (3);

[0069] Where k is the slope of the baseline and b is the intercept. λ is the wavelength.

[0070] Take any segment of the background baseline, and calculate the mean within that segment:

[0071] (4);

[0072] For a smooth linear baseline, the signal values ​​of each sampling point within any segment are randomly distributed around the mean of that segment, and the fluctuation amplitude is determined only by the system noise.

[0073] In this embodiment, the dispersion function is a variance function. For a certain segment of the background baseline, this segment contains n sampling points with signal values ​​x1, x2, ..., xn. n The variance S of this segment 2 Calculate according to the following formula:

[0074] (5);

[0075] in, It is the arithmetic mean of the signal values ​​of n sampling points within this segment.

[0076] Substitute any segment of the background baseline and the mean into the variance formula:

[0077] (6);

[0078] Simplifying the expression, we get:

[0079] (7);

[0080] Since the absorption curve is acquired under conditions of fixed number of sampling points and uniform sampling, the wavelength interval between adjacent sampling points is equal. Therefore, for any segment:

[0081] (8);

[0082] C is a constant. This represents the wavelength of the i-th sampling point; This represents the arithmetic mean of the wavelengths of the n sampling points within this segment.

[0083] From equations (7) and (8), it can be seen that for a linear baseline under pollution-free conditions, the variance of all pieces... Exactly the same.

[0084] Therefore, the piecewise variance / standard deviation K = 0. When more pollutants adhere, the intensity of the emitted light reflected by the pollutants increases, and the amplitude of the interference fringes increases, so K > 0, and increases significantly with the increase of pollutants.

[0085] In one embodiment, step S3 specifically includes:

[0086] Extract the front-end background baseline and the back-end background baseline from the signal containing the background baseline of the non-gas absorption region, and calculate the piecewise variance of the front-end background baseline and the back-end background baseline respectively;

[0087] Based on the calculated piecewise variance, determine the K value that characterizes the degree of baseline fluctuation;

[0088] The K value is compared with a preset contamination threshold to determine the contamination status of the optical lens.

[0089] In this embodiment, the "fluctuation of the discreteness function value" is quantitatively characterized by calculating the standard deviation K between the variances of each segment.

[0090] Let the variance values ​​of each segment be respectively (where m is the total number of segments), then the standard deviation K between the variances of each segment is calculated according to formula (9):

[0091] (9);

[0092] in, This is the arithmetic mean of the variances of each segment.

[0093] As can be seen from the previous derivation, when the optical lens is in a clean state, the background baseline is a linear smooth baseline, and the variance of each segment is exactly the same, so K=0.

[0094] When contaminants adhere to an optical lens, the contaminant layer reflects part of the output beam. This reflected light interferes with the output light, superimposing periodic interference fringes onto the background baseline. Since the periodicity of the interference fringes is wavelength-dependent, the amplitude and shape of the fringes differ in different segments, leading to variations in the fluctuation of the signal value within each segment, and consequently, the variance of each segment. They are no longer the same; the K value is significantly greater than 0.

[0095] Furthermore, as the amount of pollutants attached increases, the intensity of the emitted light reflected by the pollutants increases, the amplitude of the interference fringes increases, and the difference between the variances of each segment increases. Therefore, the K value increases significantly with the increase of pollutants.

[0096] In one embodiment, determining the contamination status of the optical lens specifically involves: when the K value is greater than the preset contamination determination threshold, determining that the optical lens is contaminated, and issuing a maintenance warning message.

[0097] In one embodiment, the preset pollution determination threshold is determined based on actual pollution data corresponding to different pollutant types.

[0098] Lens contamination can be determined purely by using the threshold value of the maximum K value of the front and back background baselines, providing a basis for judging optical lens contamination under weak signals. Because the reflectivity of the reflected light formed by contaminants does not increase linearly with the contaminant, different threshold values ​​need to be set according to different types of contaminants for sensor contamination early warning judgment.

[0099] In one embodiment, the segmented standard deviations of the front-end background baseline and the back-end background baseline are calculated respectively;

[0100] Based on the calculated segmented standard deviation, the K value, which characterizes the degree of baseline fluctuation, is determined.

[0101] As an alternative implementation, the dispersion function can also be a standard deviation function. Variance and standard deviation are both commonly used statistics for measuring the dispersion of data, and there is a definite mathematical conversion relationship between them. Those skilled in the art can choose to use them according to actual needs, and both can achieve the technical effects of this application.

[0102] In one embodiment, the K value is the maximum value among the piecewise variances of the front-end background baseline and the back-end background baseline.

[0103] In this embodiment, the maximum value of the K value between the front-end background baseline and the back-end background baseline is used as the contamination determination index:

[0104] (10);

[0105] when When the preset contamination threshold is exceeded, the optical lens is determined to be contaminated, and a maintenance warning is triggered.

[0106] Because different types of pollutants (such as dust, water mist, and oil) have different characteristics in reflecting and interfering light, the amplitude and shape of the interference fringes they cause vary. Therefore, different pollution detection thresholds can be set for different types of pollutants. In practical applications, one or more thresholds can be pre-stored based on the specific application environment of the sensor, historical pollution data, and manual calibration experiments to achieve effective detection of different types of pollutants.

[0107] Figure 3 For the same TDLAS methane gas measurement sensor, when pollutants increase, the maximum values ​​of the variances of the gas absorption curves and the corresponding standard deviation K values ​​from signal 1 to signal 5 at the beginning and end are 45.4, 39.3, 14.6, 18.2, and 3.7, respectively. This K value is positively correlated with the amplitude of the interference signal, thus indicating that this method can determine the increase of pollutants.

[0108] In one embodiment, the method is applicable to TDLAS gas sensing applications with telemetry or weak through-beam signals.

[0109] The methods and systems provided in this application are particularly suitable for telemetry-based TDLAS gas sensing applications and weak through-beam TDLAS gas sensing applications.

[0110] In telemetry-based TDLAS applications, after the laser beam is emitted from the sensor's transmitter, it travels through open space to a distant reflective surface (such as the ground, wall, or target gas cloud). After diffuse reflection by the reflective surface, part of the light signal returns to the sensor's detector. Due to the long-distance, two-way transmission of the light signal, the signal at the receiving end is extremely weak; the light intensity received by the detector is often only one ten-thousandth or even less than the emitted light intensity. Under these conditions, even if the transmitting optical lens is covered with an extremely thin layer of contaminants, the direct reflected light from the contaminant layer is not negligible compared to the weak telemetry echo signal, and the interference effect between the contaminant reflected light and the emitted light is significant, rendering the traditional light intensity threshold method completely ineffective. This application determines contamination by analyzing the dispersion fluctuations of the background baseline, without relying on the absolute value of the received light intensity, and is therefore particularly suitable for telemetry-based TDLAS sensing scenarios.

[0111] Similarly, this application is also applicable to weak through-beam signal TDLAS sensing scenarios where the transmit and receive signals are separated.

[0112] In summary, the embodiments of this application achieve at least the following beneficial effects:

[0113] (1) It does not depend on the absolute value of the received light intensity, thus fundamentally avoiding misjudgment caused by the light intensity increasing instead of decreasing due to the reflection / interference of pollutants;

[0114] (2) The interference fringes caused by pollutants are transformed from "interference sources" in traditional methods into "useful signals" for pollution detection;

[0115] (3) The fluctuation of the background baseline is quantitatively characterized by the standard deviation K value of the segmented variance, thus realizing an objective and quantitative judgment on the contamination status of the optical lens;

[0116] (4) It provides a reliable lens contamination detection method for TDLAS gas concentration sensors, especially for remote sensing and weak beam signal applications, ensuring the long-term stable operation of the sensor.

[0117] In one embodiment, a TDLAS optical lens contamination detection system is provided, comprising:

[0118] The signal acquisition module is used to acquire the absorption curve signal of the TDLAS gas concentration sensor;

[0119] The baseline extraction module is used to subtract the gas absorption portion from the absorption curve signal according to the gas absorption line model to obtain a background baseline signal that includes the non-gas absorption region.

[0120] The calculation and processing module is used to segment the background baseline, calculate the discreteness function value of the signal value in each segment, and determine whether there is contamination in the optical lens based on the fluctuation of the discreteness function value.

[0121] The output module is used to output the pollution determination result.

[0122] This application also provides a TDLAS optical lens contamination detection system based on baseline variance fluctuation. The system includes:

[0123] The signal acquisition module is used to acquire the absorption curve signal of the TDLAS gas concentration sensor. This module includes a detector, a preamplifier, and an analog-to-digital converter. The output signal of the detector is amplified by the preamplifier and then converted into a digital signal by the analog-to-digital converter before being sent to the processor.

[0124] The baseline extraction module is used to subtract the gas absorption portion from the absorption curve signal based on the gas absorption line model, obtaining a background baseline signal that includes the non-gas absorption region. This module is implemented in the processor and is accomplished by executing the Lorentz line fitting algorithm and subtraction operations.

[0125] The computational processing module is used to segment the background baseline, calculate the dispersion function value of the signal values ​​within each segment, and determine whether there is contamination in the optical lens based on the fluctuation of the dispersion function value. This module is implemented in the processor and is accomplished by executing segmentation, variance / standard deviation calculation, and K-value calculation algorithms.

[0126] The output module is used to output the pollution determination result. This module includes a display screen, indicator lights, or an audible and visual alarm to present the pollution status and warning information to the operator.

[0127] In one specific implementation, the functions of each module of the system are implemented by loading and executing corresponding computer programs by the embedded processor or host computer processor of the TDLAS gas concentration sensor. The specific implementation details of each module can be found in the above description of a TDLAS optical lens contamination detection method, and will not be repeated here.

[0128] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A TDLAS optical lens contamination detection method, characterized in that, include: Step S1: Acquire the absorption curve signal of the TDLAS gas concentration sensor; Step S2: Based on the gas absorption line model, subtract the gas absorption portion from the absorption curve signal to obtain a background baseline signal that includes the non-gas absorption region. Step S3: Divide the background baseline into segments and calculate the discreteness function value of the signal value in each segment; determine whether there is contamination in the optical lens based on the fluctuation of the discreteness function value.

2. The method according to claim 1, characterized in that, Step S3 specifically includes: Extract the front-end background baseline and the back-end background baseline from the signal containing the background baseline of the non-gas absorption region, and calculate the piecewise variance of the front-end background baseline and the back-end background baseline respectively; Based on the calculated piecewise variance, determine the K value that characterizes the degree of baseline fluctuation; The K value is compared with a preset contamination threshold to determine the contamination status of the optical lens.

3. The method according to claim 2, characterized in that, The K value is the maximum value among the piecewise variances of the front-end background baseline and the back-end background baseline.

4. The method according to claim 2 or 3, characterized in that, The specific method for determining the contamination status of the optical lens is as follows: when the K value is greater than the preset contamination determination threshold, the optical lens is determined to be contaminated, and a maintenance warning message is issued.

5. The method according to claim 2, characterized in that, Calculate the segmented standard deviations of the front-end background baseline and the back-end background baseline, respectively; Based on the calculated segmented standard deviation, the K value, which characterizes the degree of baseline fluctuation, is determined.

6. The method according to claim 1, characterized in that, The gas absorption line model is a Lorentz line model. The process of subtracting the gas absorption portion from the absorption curve signal specifically involves: constructing an absorption curve model based on the Beer-Lambert law, and using the Lorentz line function to fit and subtract the gas absorption signal from the absorption curve that satisfies a fixed line width and shape.

7. The method according to claim 2, characterized in that, The preset pollution determination threshold is determined based on the actual pollution data corresponding to different pollutant types.

8. The method according to claim 1, characterized in that, The absorption curve signal was acquired under the condition of a fixed number of sampling points and uniform wavelength sampling; The segmentation of the background baseline specifically involves: Extract the background baseline of the non-gas absorption region in the absorption curve, and divide the background baseline into equal-number segments according to the segment boundaries of the absorption curve, so that each segment contains the same number of sampling points.

9. The method according to claim 1, characterized in that, The method is applicable to TDLAS gas sensing applications using telemetry or weak through-beam signals.

10. A TDLAS optical lens contamination detection system, characterized in that, include: The signal acquisition module is used to acquire the absorption curve signal of the TDLAS gas concentration sensor; The baseline extraction module is used to subtract the gas absorption portion from the absorption curve signal according to the gas absorption line model to obtain a background baseline signal that includes the non-gas absorption region. The calculation and processing module is used to segment the background baseline, calculate the discreteness function value of the signal value in each segment, and determine whether there is contamination in the optical lens based on the fluctuation of the discreteness function value. The output module is used to output the pollution determination result.

Images