A method for monitoring gas content in underground coal mines for safety production monitoring.

Abstract

This application relates to the field of gas content detection technology, specifically to a method for monitoring gas content in underground coal mines for safety production monitoring. The method includes: acquiring the infrared spectrum of the gas to be measured in the coal mine and a sample set; calculating the importance of each wavenumber point using the sample set; obtaining the sensitivity coefficient, inhibition factor, baseline fluctuation, and adjustment coefficient for each wavenumber point in the infrared spectrum to be measured; correcting the smoothing parameters of the baseline calibration algorithm twice to obtain the final corrected smoothing parameters; and using the baseline calibration algorithm to perform baseline calibration on the infrared spectrum to be measured, thereby quantitatively detecting the gas content in the coal mine. This application can effectively remove baseline drift while preserving important spectral features, more accurately extracting gas characteristic peaks, and improving the quantitative detection accuracy of gas content.

Description

Technical Field

[0001] This application relates to the field of gas content detection technology, specifically to a method for monitoring gas content for underground safety production monitoring in coal mines. Background Technology

[0002] In the underground environment of coal mines, methane is one of the main harmful gases. Excessive concentration of methane can easily cause explosions or asphyxiation accidents. Therefore, the detection of methane gas is crucial for ensuring the safety of miners and preventing methane explosion accidents.

[0003] When using infrared spectroscopy gas analysis to detect methane, the complex underground environment and environmental changes during spectral acquisition can cause baseline drift. For example, fluctuations in the light source temperature can cause baseline drift of varying magnitudes, and angular deviations of the moving mirror during spectral scanning can also lead to baseline drift. Traditional airPLS algorithms typically use fixed smoothing parameters to control the smoothness of the baseline, which is difficult to adapt to changes in spectral characteristics across different wavenumber ranges. In characteristic bands with dense and rapidly changing absorption peaks, excessive smoothing can lead to signal distortion and loss of quantitative information. Conversely, in baseline bands with fewer and relatively stable absorption peaks, insufficient smoothing fails to effectively filter out low-frequency baseline drift caused by environmental interference, resulting in poor baseline calibration and affecting the accurate quantitative analysis of methane content. Summary of the Invention

[0004] To address the aforementioned technical problems, a method for monitoring gas content in underground coal mines is provided to solve existing issues.

[0005] The solution to the technical problem presented in this application is to provide a method for monitoring gas content in underground coal mines for safety production monitoring, comprising the following steps:

[0006] To obtain the infrared spectrum of the target gas in underground coal mine and a sample set, wherein the sample set contains multiple infrared spectra from historical periods;

[0007] The importance of each wavenumber point is calculated by analyzing the intensity characteristics of absorbance at the same wavenumber point in different infrared spectra within the sample set.

[0008] For the infrared spectrum to be measured, the absorbance peak characteristics contained in the local band of each wavenumber point are analyzed to obtain the sensitivity coefficient of each wavenumber point. Combined with the importance, the inhibition factor of each wavenumber point is determined.

[0009] The fluctuation of baseline variation within a local band is analyzed, the baseline fluctuation at each wavenumber point is calculated, and the adjustment coefficient at each wavenumber point is obtained by combining the suppression factor. The smoothing parameters of the baseline calibration algorithm are then initially corrected.

[0010] The distance relationship and importance of each wavenumber point to its local wavenumber points are evaluated. The smoothing parameters after the initial correction are then corrected a second time to obtain the final corrected smoothing parameters for each wavenumber point. The baseline calibration algorithm is used to perform baseline calibration on the infrared spectrum to be measured in order to quantitatively detect the gas content in underground coal mines.

[0011] Preferably, the calculation of the importance of each wavenumber point includes: normalizing the absorbance of all wavenumber points in each infrared spectrum in the sample set, and positively fusing the normalized absorbance of the same wavenumber point in all infrared spectra in the sample set as the importance of each wavenumber point.

[0012] Preferably, obtaining the sensitivity coefficient for each wavenumber point includes:

[0013] Obtain the peaks and troughs of absorbance corresponding to all wavenumber points in the infrared spectrum to be measured; analyze the number of peaks contained in the local band of each wavenumber point and calculate the information richness of each wavenumber point; analyze the peak width level of the peaks in the local band and calculate the local immunity of each wavenumber point.

[0014] The sensitivity coefficient is positively correlated with information richness, but negatively correlated with local immunity.

[0015] Preferably, the information richness calculation process is as follows: count the number of peaks contained in the local band of each wavenumber point, and take the ratio between the number of peaks and the total number of all wavenumber points in the local band as the information richness of each wavenumber point.

[0016] Preferably, the calculation process of the local immunity is as follows: the difference between the wavenumbers of the two wavenumber points corresponding to the two wavenumber points adjacent to each wavenumber is taken as the peak width of each wavenumber; the peak widths of all wavenumbers in the local band of each wavenumber point are positively fused to take as the local immunity of each wavenumber point.

[0017] Preferably, the inhibition factor is the result of a positive fusion of the sensitivity coefficient and the importance.

[0018] Preferably, the calculation of the baseline variability at each wavenumber point includes:

[0019] A baseline calibration algorithm is used, and the infrared spectrum to be tested is calibrated with preset initial smoothing parameters to extract the initial baseline;

[0020] For the initial baseline, the average difference between each wavenumber point in the local band and the baseline value of the corresponding two adjacent wavenumber points is taken as the average change of each wavenumber point.

[0021] The baseline volatility is the result of positive fusion of all average changes within a local band.

[0022] Preferably, the adjustment coefficient is the result of positively fusing the suppression factor and the baseline volatility.

[0023] Preferred, the first Each wavenumber point corresponds to the smoothing parameter after the initial correction. The calculation formula is: ,in, For the first Adjustment coefficient for each wavenumber point The preset initial smoothing parameters are: This is a preset value.

[0024] Preferably, obtaining the final corrected smoothing parameter corresponding to each wavenumber point includes: for any wavenumber point's local band, calculating the difference in wavenumber between each wavenumber point in the local band and the arbitrary wavenumber point, denoted as the relative difference; using the normalized result of the ratio of the importance of each wavenumber point in the local band to the relative difference as the weighting factor for each wavenumber point; and using the weighting factor as the weight, performing a weighted summation of the initial corrected smoothing parameters corresponding to all wavenumber points in the local band, which is used as the final corrected smoothing parameter corresponding to the arbitrary wavenumber point.

[0025] This application has at least the following beneficial effects:

[0026] This application calculates the importance of each wavenumber point using a sample set. Its advantage lies in considering the prevalent importance of each wavenumber point across different infrared spectra within the gas sample set, reflecting the criticality of the gas component characteristic information contained at that wavenumber point, thus identifying characteristic bands that significantly contribute to gas detection stability. It also obtains the sensitivity coefficient for each wavenumber point, which considers the peak density and sharpness within local bands, reflecting the complexity of the spectral information characteristics contained within those local bands and their vulnerability to noise and other interference, thereby evaluating the role of local bands in basic... The smoothing degree during line calibration; determining the suppression factor for each wavenumber point, the beneficial effect of which is to evaluate the degree of smoothing suppression based on the general importance of the wavenumber point and the peak shape characteristics of local bands in the infrared spectrum to be measured, so as to reduce the smoothing intensity in bands with sharp and drastic absorption peaks, effectively preventing real gas characteristic peaks from being treated as noise or distorted by excessive smoothing of the baseline during baseline fitting; calculating the baseline variability for each wavenumber point, the beneficial effect of which is to extract the baseline using traditional baseline calibration methods while considering the variability of the baseline, avoiding the use of larger smoothing parameters in bands with drastic baseline fluctuations. This excessive smoothing can prevent the baseline from tracking real, rapid changes, leading to underfitting. Adjustment coefficients are obtained for each wavenumber point, and the smoothing parameters of the baseline calibration algorithm are initially corrected. This initial correction combines the suppression of smoothing in local bands with the baseline's volatility, ensuring the baseline closely matches the real background trend while retaining key absorption features, effectively avoiding distortion or loss of useful signals. A second correction is then applied to the initially corrected smoothing parameters, yielding the final corrected smoothing parameters for each wavenumber point. This second correction leverages distance attenuation and importance... The weighted smoothing parameter is corrected twice to ensure its spatial continuity at different wavenumber points, avoiding baseline jumps or step effects, and ensuring a smooth and continuous final fitted baseline. A baseline calibration algorithm is used to calibrate the infrared spectrum to be measured for quantitative detection of gas content in coal mines. Its advantages lie in the fact that baseline calibration using dynamic smoothing parameters adjusts the smoothing degree according to local spectral characteristics, avoiding over-smoothing. This effectively removes baseline drift while preserving important spectral features, more accurately extracting gas characteristic peaks and improving the quantitative detection accuracy of gas content. Attached Figure Description

[0027] The following description, in conjunction with the accompanying drawings, provides a more detailed explanation of a gas content monitoring method for underground safety production monitoring in coal mines, based on this application.

[0028] Figure 1 A flowchart illustrating the steps of a gas content monitoring method for underground safety production monitoring in coal mines, provided in this application embodiment;

[0029] Figure 2 A flowchart illustrating the steps of a method for determining the suppression factor at each wavenumber point, as provided in an embodiment of this application. Detailed Implementation

[0030] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description, in conjunction with the accompanying drawings and implementation examples, provides a method for monitoring gas content in underground coal mine safety production. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0031] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0032] Please see Figure 1 The diagram illustrates a flowchart of a method for monitoring gas content in underground coal mines, according to an embodiment of this application. The method includes the following steps:

[0033] Step 1: Obtain the infrared spectrum of the gas to be measured in the coal mine and the sample set, wherein the sample set contains multiple infrared spectra from historical periods.

[0034] Mine fires and gas explosions have always seriously threatened the safe production of my country's coal mining industry. In coal mine gas detection, gas contains a variety of components, such as methane, ethane, propane, isobutane, and n-butane. However, the main component of gas is methane. In most coal seams, methane usually accounts for more than 83% to 89% of the gas volume, making it the absolute dominant component. Its concentration basically determines the boundary of a gas explosion. Therefore, accurate, rapid, and real-time monitoring and early warning of methane gas concentration is an effective means of controlling gas disasters and plays an important role in the safe operation of coal mines, the personal safety of miners, and environmental protection.

[0035] Based on the above analysis, infrared spectra of underground coal mine gas were collected using an infrared spectrometer as the infrared spectra to be measured.

[0036] In this embodiment, the resolution of the infrared spectral detector is set to 1 cm. -1 The wavenumber scan range is 400cm. -1 ~4000cm -1 Therefore, in the infrared spectrum, the horizontal axis represents wavenumber and the vertical axis represents absorbance, thus obtaining the absorbance corresponding to each wavenumber point. Since the resolution is 1 cm⁻¹... -1 The infrared spectrum contains absorbance corresponding to 3601 wavenumber points; all wavenumber points are arranged in ascending order of wavenumber.

[0037] Multiple infrared spectra of gas from historical periods were collected and, after baseline calibration, were used to form a sample set.

[0038] In this embodiment, the sample size in the sample set is 100. As for other implementation methods, the implementer can set the sample size according to the actual situation. The wavenumber range of the infrared spectra in the sample set is 400 cm⁻¹. -1 ~4000cm -1 Secondly, the baseline calibration process is as follows: baseline calibration is performed using a baseline calibration algorithm, specifically iteratively reweighted penalized least squares (airPLS). The smoothing parameter of the airPLS algorithm is set to 10. 5 As other implementation methods, implementers can set them according to their actual situation; the airPLS algorithm is a well-known technology and will not be described in detail here.

[0039] Thus, the infrared spectrum of the methane gas in the coal mine and its sample set were obtained.

[0040] Step 2: Calculate the importance of each wavenumber point by analyzing the intensity characteristics of absorbance at the same wavenumber point in different infrared spectra within the sample set; for the infrared spectrum to be measured, analyze the absorbance peak characteristics of each wavenumber point in the local band to obtain the sensitivity coefficient of each wavenumber point, and determine the inhibition factor of each wavenumber point by combining the importance.

[0041] Since gas contains multiple components, which have specific absorption peak characteristics in the infrared spectrum, by analyzing these characteristics, the importance of each wavenumber point can be determined. This allows for a focus on important wavenumber ranges during baseline calibration, reducing the effort required to correct less important wavenumber ranges and improving calibration effectiveness.

[0042] Based on the above analysis, the importance is calculated by examining the absorption characteristics of each wavenumber point in the sample set across different infrared spectra. Specifically:

[0043] The absorbance of all wavenumber points in each infrared spectrum in the sample set is normalized. The result of forward fusion of the normalized absorbance of the same wavenumber point in all infrared spectra in the sample set is used as the importance of each wavenumber point.

[0044] In this embodiment, the specific process of forward fusion is as follows: the mean of the normalized absorbance of the same wavenumber point in all infrared spectra within the sample set is used as the importance of each wavenumber point; as another implementation, the implementer can calculate the sum of the normalized absorbance of the same wavenumber point in all infrared spectra within the sample set as the importance of each wavenumber point; wherein, the specific process of normalization is as follows: the ratio of the absorbance of each wavenumber point in each infrared spectra within the sample set to the maximum absorbance in the infrared spectra is used as the normalized absorbance.

[0045] It should be noted that normalization eliminates the interference of different concentrations of methane gas on absorbance. The greater the importance, the higher the relative absorbance of the wavenumber point in the sample set. The higher the probability and stability of the wavenumber point as a significant absorption feature, the more critical the methane component feature information contained in the wavenumber point is in infrared spectroscopy analysis, and the greater the impact on the accuracy of methane component detection.

[0046] Secondly, in spectral analysis, spectral information is mainly found in absorption peaks. The richer the spectral information content, the more details and important information are contained in a certain wavenumber range. Therefore, when using the baseline calibration algorithm to perform baseline calibration for this wavenumber range, a smaller smoothing parameter is needed to control the smoothness during the calibration process and avoid overfitting and loss of important information.

[0047] Furthermore, the flowchart of the method for obtaining the suppression factor at each wavenumber point provided in the embodiments of this application is as follows: Figure 2 As shown.

[0048] First, by analyzing the peaks contained in local bands of the infrared spectrum to be measured, the information richness is calculated, specifically as follows:

[0049] Obtain the peaks and troughs of absorbance corresponding to all wavenumber points in the infrared spectrum to be measured;

[0050] In this embodiment, the AMPD algorithm (Automatic multiscale-based peak detection) is used to obtain the peaks and troughs. The AMPD algorithm is a well-known technology and will not be described in detail here.

[0051] Taking each wavenumber point in the infrared spectrum to be measured as the center, multiple wavenumber points adjacent to it are taken as local bands;

[0052] In this embodiment, the number of wavenumber points included in the local band is 51. As for other implementation methods, the implementer can set it according to the actual situation.

[0053] It should be noted that, since the local band contains 51 wavenumber points, therefore, for the _____ A local band with a wavenumber point has 25 wavenumber points on its left and 25 wavenumber points on its right. For wavenumber points at both ends, if the number of wavenumber points on one side is less than 25, the window boundary is adjusted to include only all wavenumber points on that side, thus forming an asymmetrical local band. For ease of understanding, let's assume the wavenumber point is 405cm. -1 The wavenumber of the wavenumber point to its left has a wavenumber range of 400cm. -1 ~405cm -1 The wavenumber range of the wavenumber point on the right is 450cm. -1 ~430cm -1 Therefore, 405cm -1 The wavenumber corresponds to a local band of 400cm. -1 ~430cm -1 .

[0054] The number of peaks contained in the local band of each wavenumber point is counted, and the ratio between the number of peaks and the total number of all wavenumber points in the local band is used as the information richness of each wavenumber point.

[0055] It should be noted that the more peaks there are, the richer the information obtained, indicating that the spectral information in the local band where the wavenumber point is located is richer and more complex, containing more absorption peaks, and requiring the use of smaller smoothing parameters.

[0056] Furthermore, in a spectrum, a wider peak typically contains more information and is less sensitive to noise. The shape and position of a wide peak are relatively stable in the spectrum and are less easily masked by noise; using a larger smoothing parameter can effectively remove noise. Conversely, narrow peaks have smaller widths in the spectrum and are more susceptible to noise interference, easily misinterpreted as baseline fluctuations. Therefore, for narrow peaks, a smaller smoothing parameter is needed to preserve their detailed information and avoid losing important spectral features. The analysis of peak widths within local bands and the calculation of local immunity are specifically as follows:

[0057] The difference between the wavenumbers of the two corresponding wavenumber points of the two wavenumber points adjacent to each wave crest is taken as the peak width of each wave crest;

[0058] In this embodiment, the absolute value of the difference between the wavenumbers of the corresponding wavenumber points of two wavenumber points adjacent to each wave peak is taken as the peak width of each wave peak.

[0059] The peak widths of all peaks within the local band at each wavenumber point are positively fused to form the local immunity at each wavenumber point.

[0060] In this embodiment, the specific process of forward fusion is as follows: the peak width of all peaks is normalized, and the average value of the normalized peak width of all peaks in the local band is used as the local immunity of each wavenumber point; wherein, the specific process of normalization is as follows: the ratio of the peak width of each peak to the maximum peak width of all peaks is used as the peak width of all peaks.

[0061] It should be noted that the smaller the peak width, the lower the local immunity, indicating that the spectral features in the local band where the wavenumber point is located are less resistant to noise and other interferences, requiring a smaller smoothing parameter to retain more detailed information and avoid losing important spectral features.

[0062] Based on this, the sensitivity coefficient is determined by information richness and local immunity, specifically as follows:

[0063] The sensitivity coefficient at each wavenumber point is positively correlated with information richness, but negatively correlated with local immunity.

[0064] It should be noted that a positive correlation means that the dependent variable increases as the independent variable increases and decreases as the independent variable decreases; a negative correlation means that the dependent variable decreases as the independent variable increases and increases as the independent variable decreases.

[0065] In this embodiment, the ratio of information richness to local immunity is used as the sensitivity coefficient for each wavenumber point. In other implementations, the implementer can calculate the difference between information richness and local immunity. To avoid the difference being less than 0, the difference is positively mapped as the sensitivity coefficient for each wavenumber point. The positive mapping process is as follows: the result of an exponential function with the natural constant as the exponent and the difference as the exponent is used as the result of the positive mapping.

[0066] It should be noted that the larger the sensitivity coefficient, the more sensitive the local band where the wavenumber point is located is, the more sensitive the region is, which is highly information-dense but has weak resistance to noise interference. When using the baseline calibration algorithm for baseline calibration, a smaller smoothing parameter is needed to control the smoothness during the calibration process and avoid overfitting and loss of important information.

[0067] Furthermore, based on the sensitivity coefficient and importance, the inhibition factor is determined, specifically:

[0068] The sensitivity coefficient and importance are positively integrated and used as the suppression factor for each wavenumber point;

[0069] In this embodiment, the specific process of forward fusion is as follows: the product of the sensitivity coefficient and the importance is used as the suppression factor for each wavenumber point. In other implementation methods, the implementer can calculate the sum of the sensitivity coefficient and the importance as the suppression factor for each wavenumber point.

[0070] It should be noted that the greater the importance, the higher the stability and prevalence of the wavenumber point as a strong absorption feature in the sample set, reflecting that the wavenumber point is the core characteristic peak position of the key gas component. When performing baseline calibration, a smaller smoothing parameter should be used to retain more important spectral features. The larger the suppression factor, the stronger the smoothing suppression needs to be applied to the wavenumber point in order to reduce the smoothing parameter and avoid smoothing out important spectral information.

[0071] Thus, the suppression factor for each wavenumber point in the infrared spectrum to be measured is obtained.

[0072] Step 3: Analyze the baseline variation fluctuation of each wavenumber point in the infrared spectrum to be tested within the local band, calculate the baseline fluctuation of each wavenumber point, and obtain the adjustment coefficient of each wavenumber point by combining the suppression factor, and make the initial correction to the smoothing parameter of the baseline calibration algorithm.

[0073] Furthermore, during baseline calibration, if the baseline in the infrared spectrum to be measured changes too rapidly in a local area, its fluctuations due to interference are significant. To avoid baseline underfitting, a smaller smoothing parameter needs to be used for the local bands where the baseline changes drastically to reflect the true baseline variation. Therefore, the fluctuation of the baseline in the infrared spectrum to be measured is analyzed, and the cumulative fluctuation is calculated, specifically as follows:

[0074] A baseline calibration algorithm is used, and the infrared spectrum to be tested is calibrated with preset initial smoothing parameters to extract the initial baseline;

[0075] In this embodiment, the baseline calibration algorithm is the airPLS algorithm, and its preset initial smoothing parameter is set to 10. 5 The initial baseline extraction process is as follows: a baseline calibration algorithm is used, and the infrared spectrum to be tested is calibrated with preset initial smoothing parameters to obtain the calibrated infrared spectrum. The difference in absorbance at the same wavenumber point between the original infrared spectrum to be tested and the infrared spectrum to be tested after initial calibration is used as the baseline value for each wavenumber point, thereby obtaining the initial baseline.

[0076] On the initial baseline, the average difference between each wavenumber point in a local band and the baseline value corresponding to two adjacent wavenumber points is taken as the average change of each wavenumber point.

[0077] In this embodiment, the average of the absolute values ​​of the differences between each wavenumber point and the baseline values ​​of two adjacent wavenumber points within a local band is taken as the average change of each wavenumber point.

[0078] The average changes in all local bands at each wavenumber point in the infrared spectrum to be measured are positively fused to form the baseline variability of each wavenumber point.

[0079] In this embodiment, the mean of all average changes within the local band of each wavenumber point in the infrared spectrum to be measured is taken as the baseline variability of each wavenumber point.

[0080] It should be noted that the larger the average change, the more significant the change in the baseline at this wavenumber point. The greater the baseline fluctuation, the more drastic the baseline change within the local band, indicating an unstable state. This reflects that the local band where the wavenumber point is located needs to reduce the smoothing parameter to avoid over-smoothing, which would cause the over-smoothed baseline to not correspond to the real rapid baseline change, thus leading to inaccurate correction results.

[0081] The suppression factor at each wavenumber point in the infrared spectrum to be measured is positively fused with the baseline variability and used as the adjustment coefficient for each wavenumber point.

[0082] In this embodiment, the normalized result of the suppression factor and the mean of the baseline fluctuation of each wavenumber point in the infrared spectrum to be measured is used as the adjustment coefficient of each wavenumber point. In this embodiment, the maximum-minimum normalization method is used for normalization processing. The maximum-minimum normalization method is a well-known technique and will not be described in detail here.

[0083] The formula for calculating the smoothing parameter corresponding to each wavenumber point in the infrared spectrum to be measured after the initial correction is as follows:

[0084]

[0085] in, The first one in the infrared spectrum to be measured Each wavenumber point corresponds to the smoothing parameters after the initial correction. The first one in the infrared spectrum to be measured Adjustment coefficient for each wavenumber point The preset initial smoothing parameters are: This is a preset value;

[0086] In this embodiment, the preset initial smoothing parameter The value is 10 5 , preset value The value is set to 1.5 to balance the relationship between the adjustment coefficient and the initial smoothing parameter, ensuring that the adjustment range of the smoothing parameter is at most 1.5. The minimum is 0.5 Therefore, when the adjustment coefficient is greater than 0.5, the wavenumber points are smoothed and suppressed. At this time, the smoothing degree should be reduced, and the smoothing parameter after the initial correction will decrease. When the adjustment coefficient is greater than 0.5, the smoothing parameter should be increased to effectively remove noise.

[0087] Thus, the smoothing parameters corresponding to each wavenumber point in the infrared spectrum to be measured are obtained after the initial correction.

[0088] Step 4: Evaluate the distance relationship and importance of each wavenumber point to its local wavenumber points, perform a second correction on the smoothing parameters after the initial correction, and obtain the final corrected smoothing parameters for each wavenumber point. Use the baseline calibration algorithm to perform baseline calibration on the infrared spectrum to be measured in order to quantitatively detect the gas content in coal mines.

[0089] Furthermore, during baseline calibration, if the smoothing parameters of adjacent wavenumber points are not similar, it may lead to baseline jumps, i.e., discontinuous changes in the baseline. To avoid baseline jumps, the smoothing parameters of each wavenumber point need to be weighted and corrected, specifically as follows:

[0090] For any wavenumber point in a local band, calculate the difference in wavenumber between each wavenumber point in the local band and the given wavenumber point, and denot it as the relative difference.

[0091] In this embodiment, the absolute value of the difference between the wavenumber of each wavenumber point within a local band and any wavenumber point is calculated and denoted as the relative difference.

[0092] The normalized result of the ratio of the importance of each wavenumber point to its relative difference within a local band is used as the weighting factor for each wavenumber point.

[0093] In this embodiment, the normalization process is as follows: the ratio of the importance of each wavenumber point within a local band to its relative difference is recorded as the relative ratio; the sum of the relative ratios of all wavenumber points within the local band is calculated; and the ratio of the relative ratio to the sum is used as the weighting factor.

[0094] It should be noted that the smaller the relative difference, the closer this wavenumber point is to any other wavenumber point. The greater the importance and the closer it is to any other wavenumber point, the greater the influence this wavenumber point has on the smoothing parameter adjustment of any other wavenumber point during the baseline calibration process, and thus the larger its weighting factor.

[0095] Using the weighting factor as the weight, the smoothing parameters corresponding to the initial correction of all wavenumber points in the local band are weighted and summed to obtain the final corrected smoothing parameters corresponding to any wavenumber point.

[0096] Based on the final corrected smoothing parameters, the baseline calibration algorithm is used to perform baseline calibration on the original infrared spectrum to be measured, and the final baseline-calibrated infrared spectrum is obtained.

[0097] In this embodiment, the baseline calibration algorithm is the airPLS algorithm.

[0098] The gas content in underground coal mines was detected using the infrared spectrum after final baseline calibration, specifically as follows:

[0099] Since the main component of gas is methane, the absorption peak region corresponding to methane is extracted from the infrared spectrum after final baseline calibration, its integral area is calculated, and the gas content is quantitatively analyzed using Beer-Lambert's law.

[0100] It should be noted that the Beer-Lambert law is a well-known technique and will not be elaborated upon here.

[0101] It should be understood that, although Figure 1 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 1 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0102] 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.

[0103] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application. Therefore, any simple modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of this application, without departing from the content of the technical solution of this application, shall fall within the protection scope of the technical solution of this application.

Claims

1. A method for monitoring gas content in underground coal mines for safety production monitoring, characterized in that, The method includes the following steps: To obtain the infrared spectrum of underground coal mine gas and a sample set, the sample set containing multiple infrared spectra from historical periods; The importance of each wavenumber point is calculated by analyzing the intensity characteristics of absorbance at the same wavenumber point in different infrared spectra within the sample set. For the infrared spectrum to be measured, the absorbance peak characteristics contained in the local band of each wavenumber point are analyzed to obtain the sensitivity coefficient of each wavenumber point. Combined with the importance, the inhibition factor of each wavenumber point is determined. The fluctuation of baseline variation within a local band is analyzed, the baseline fluctuation at each wavenumber point is calculated, and the adjustment coefficient at each wavenumber point is obtained by combining the suppression factor. The smoothing parameters of the baseline calibration algorithm are then initially corrected. The distance relationship and importance of each wavenumber point to its local wavenumber points are evaluated. The smoothing parameters after the initial correction are then corrected a second time to obtain the final corrected smoothing parameters for each wavenumber point. The baseline calibration algorithm is used to perform baseline calibration on the infrared spectrum to be measured in order to quantitatively detect the gas content in coal mines. The process of obtaining the final corrected smoothing parameter for each wavenumber point includes: for any wavenumber point in a local band, calculating the difference in wavenumber between each wavenumber point in the local band and the given wavenumber point, denoted as the relative difference; using the normalized result of the ratio of the importance of each wavenumber point in the local band to the relative difference as the weighting factor for each wavenumber point; and using the weighting factor as the weight, performing a weighted summation of the initial corrected smoothing parameters for all wavenumber points in the local band, which is then used as the final corrected smoothing parameter for the given wavenumber point.

2. The method for monitoring gas content in underground coal mine safety production as described in claim 1, characterized in that, The calculation of the importance of each wavenumber point includes: normalizing the absorbance of all wavenumber points in each infrared spectrum in the sample set, and positively fusing the normalized absorbance of the same wavenumber point in all infrared spectra in the sample set as the importance of each wavenumber point.

3. The method for monitoring gas content in underground coal mine safety production as described in claim 1, characterized in that, The process of obtaining the sensitivity coefficient at each wavenumber point includes: Obtain the peaks and troughs of absorbance corresponding to all wavenumber points in the infrared spectrum to be measured; analyze the number of peaks contained in the local band of each wavenumber point and calculate the information richness of each wavenumber point; analyze the peak width level of the peaks in the local band and calculate the local immunity of each wavenumber point. The sensitivity coefficient is positively correlated with information richness, but negatively correlated with local immunity.

4. The method for monitoring gas content in underground coal mine safety production as described in claim 3, characterized in that, The information richness calculation process is as follows: count the number of peaks contained in the local band of each wavenumber point, and take the ratio between the number of peaks and the total number of all wavenumber points in the local band as the information richness of each wavenumber point.

5. A method for monitoring gas content in underground coal mine safety production as described in claim 3, characterized in that, The calculation process of the local immunity is as follows: the difference between the wavenumbers of the two wavenumber points corresponding to the two wavenumber points adjacent to each wavenumber is taken as the peak width of each wavenumber; the peak widths of all wavenumbers in the local band of each wavenumber point are positively fused to take as the local immunity of each wavenumber point.

6. The method for monitoring gas content in underground coal mine safety production as described in claim 1, characterized in that, The inhibition factor is the result of a positive fusion of the sensitivity coefficient and importance.

7. The method for monitoring gas content in underground coal mine safety production as described in claim 1, characterized in that, The calculation of the baseline variability at each wavenumber point includes: A baseline calibration algorithm is used, and the infrared spectrum to be tested is calibrated with preset initial smoothing parameters to extract the initial baseline; For the initial baseline, the average difference between each wavenumber point in the local band and the baseline value of the corresponding two adjacent wavenumber points is taken as the average change of each wavenumber point. The baseline volatility is the result of positive fusion of all average changes within a local band.

8. A method for monitoring gas content in underground coal mine safety production as described in claim 1, characterized in that, The adjustment coefficient is the result of a positive fusion of the suppression factor and the baseline volatility.

9. A method for monitoring gas content in underground coal mine safety production as described in claim 1, characterized in that, No. Each wavenumber point corresponds to the smoothing parameter after the initial correction. The calculation formula is: ,in, For the first Adjustment coefficient for each wavenumber point The preset initial smoothing parameters, This is a preset value.

Images