Compressed air self-rescue breathing adaptive adjustment method and related devices

Abstract

This invention provides a method and related device for adaptive adjustment of compressed air self-rescue breathing. It acquires ambient sound signals and air pressure fluctuation signals from the compressed air self-rescue supply pipeline in the underground coal mine environment, couples and reduces noise to extract a pure breathing signal; identifies the user's breathing state and generates an air supply correction command based on the breathing state; drives the regulating valve of the compressed air self-rescue system to match the actual air supply of the system with the user's real-time breathing needs. This invention ensures the accuracy of breathing feature extraction, avoids irrelevant interference affecting the control logic, achieves real-time matching of air supply with the user's breathing needs, effectively improves the emergency response capability and adaptability of the compressed air self-rescue system, ensures the breathing safety of underground workers, reduces air waste, enhances the intelligence and practicality of the compressed air self-rescue system, and meets the emergency breathing protection needs under complex working conditions in underground coal mines.

Description

Technical Field

[0001] This invention relates to the field of underground safety protection technology in coal mines, and in particular to a self-rescue breathing adaptive adjustment method and related device using compressed air. Background Technology

[0002] Coal mine underground operations are characterized by enclosed spaces and complex environmental conditions, making them prone to sudden emergencies such as gas exceeding limits, roof collapses, and hazardous gas leaks. Compressed air self-rescue devices, as crucial safety equipment providing emergency breathing air to workers in case of underground accidents, play an irreplaceable role in ensuring the safety of underground personnel. Traditional compressed air self-rescue systems generally adopt a preset fixed-flow air supply mode. After leaving the factory, the air supply volume is mostly maintained at a constant output, lacking dynamic adaptive adjustment capabilities. In practical applications, different workers have individual differences in physiological constitution and respiratory rhythm, and when workers are in danger underground, experiencing panic, or feeling oxygen deficiency, their breathing frequency and intensity will change significantly. A constant air supply method easily leads to a mismatch between the air supply volume and the actual breathing needs of the personnel. Insufficient air supply can cause breathing difficulties and exacerbate oxygen deficiency; excessive air supply can easily create an airflow impact sensation and cause unnecessary consumption of compressed air resources. Furthermore, the underground working environment is filled with mechanical operating noise, airflow disturbance noise, and pipeline airflow noise. Traditional compressed air self-rescue devices struggle to accurately separate effective human breathing signals from complex environmental clutter, resulting in low accuracy in identifying breathing rhythm and breathing status. Summary of the Invention

[0003] This invention provides a compressed air self-rescue breathing adaptive adjustment method and related device to solve the defects of existing compressed air self-rescue technology, which lacks the ability to dynamically and adaptively adjust the air supply and cannot accurately identify individual breathing signals.

[0004] This invention provides a method for adaptive adjustment of compressed air self-rescue breathing, comprising: Acquire environmental acoustic signals in the underground coal mine environment and air pressure fluctuation signals in the compressed air self-rescue gas supply pipeline; The ambient sound signal and the air pressure fluctuation signal are coupled and denoised to extract a pure breathing signal; The user's breathing state is identified based on the pure breathing signal, and an air supply correction command is generated based on the breathing state. The regulating valve of the compressed air self-rescue system is activated by the air supply correction command so that the actual air supply of the compressed air self-rescue system matches the user's real-time breathing needs.

[0005] According to the self-rescue breathing adaptive adjustment method for compressed air provided by the present invention, the step of coupling and denoising the ambient sound wave signal and the air pressure fluctuation signal to extract a pure breathing signal includes: The ambient sound wave signal and the air pressure fluctuation signal are input into a pre-constructed wavelet packet attention dual-source coupled noise reduction network to output preliminary clean respiratory signal data. Feature enhancement extraction is performed on the preliminary purified respiratory signal data, and the core frequency components of the respiratory signal are analyzed and determined. Substitute the core frequency components into the dynamic model of the respiratory signal to calculate the respiratory characteristic parameters; The respiratory characteristic parameters are integrated into fluctuation characteristic parameters, and the respiratory signal is determined to be within a preset normal range based on the fluctuation characteristic parameters. If so, the preliminary pure breathing signal is determined to be a pure breathing signal.

[0006] According to the self-rescue breathing adaptive adjustment method for compressed air provided by the present invention, the step of inputting the ambient sound wave signal and the air pressure fluctuation signal into a pre-constructed wavelet packet attention dual-source coupled noise reduction network and outputting preliminary clean breathing signal data includes: Wavelet packet decomposition is performed on the ambient sound wave signal and the air pressure fluctuation signal respectively, decomposing each signal into multiple sub-band signals of different frequency bands; The subband signals are weighted using an attention mechanism to calculate the correlation weight between each subband signal and the respiratory signal. The correlation weight is used to enhance the subband components related to respiratory characteristics and suppress the subband components related to noise interference. The sub-band signals of the weighted environmental sound wave signal and the air pressure fluctuation signal are coupled and fused to generate a dual-source coupling feature matrix; Wavelet packet reconstruction is performed on the dual-source coupling feature matrix to output the preliminary purified respiratory signal data.

[0007] According to the self-rescue breathing adaptive adjustment method for compressed air provided by the present invention, the step of substituting the core frequency component into the dynamic model of the respiratory signal, calculating the respiratory characteristic parameters, and integrating the respiratory characteristic parameters into fluctuation characteristic parameters includes: Establish a time-frequency coupled dynamic model of respiratory signals based on core frequency components: ; in, This represents the wavelet packet time-frequency decomposition operator. This represents the frequency domain analytic operator. Represents the time-frequency distribution function of the respiratory signal. Represents a time variable. Represents frequency variables; Solve the time-frequency coupled dynamic model of the respiratory signal to obtain respiratory characteristic parameters, including respiratory modal period, time-frequency amplitude divergence, and respiratory transient rate of change. , ; ; ; in, Indicates the respiratory modal period, This represents the extreme values ​​of the core frequency mode of the respiratory signal. This represents the magnitude of the time-frequency distribution function of the respiratory signal. Indicates the time-frequency amplitude divergence. This represents the mean function of the time-frequency distribution of respiratory signals. Indicates the transient rate of change of respiration; According to respiratory modal period Time-frequency amplitude divergence and respiratory transient rate of change Respiratory fluctuation feature parameters are generated through adaptive weighted fusion: ; in, Represents the characteristic parameters of respiratory fluctuations. , , This represents the weighting coefficients for adaptive learning of the downhole environment. The time integral of the absolute value of the transient rate of change of respiration.

[0008] The compressed air self-rescue breathing adaptive adjustment method provided by the present invention further includes: If the fluctuation characteristic parameter determines that the respiratory signal exceeds the preset normal range, then a secondary correction process is performed on the preliminary pure respiratory signal to obtain the pure respiratory signal.

[0009] According to the self-rescue breathing adaptive adjustment method for compressed air provided by the present invention, the step of identifying the user's breathing state based on the pure breathing signal and generating a gas supply correction command based on the breathing state includes: The time-domain data stream corresponding to the pure breathing signal is obtained, and the peak and valley values ​​of the time-domain data stream are detected by an adaptive threshold detection algorithm to obtain a set of extreme points including inspiratory peak points and expiratory valley points. The respiratory cycle sequence is calculated based on the time interval between adjacent peak points and valley points in the set of extreme points; Based on the amplitude difference between the peak inspiratory point and the trough inspiratory point within the same respiratory cycle in the set of extreme points, the amplitude correlation parameter is extracted; By combining the respiratory cycle sequence with the amplitude correlation parameters, a target waveform feature vector of a unified dimension is constructed; A respiratory intensity identifier is generated based on the amplitude correlation parameter in the target waveform feature vector, and the dynamic change trend of respiratory intensity is obtained based on the respiratory cycle sequence in the target waveform feature vector. When the respiratory intensity indicator exceeds the preset respiratory intensity threshold range, an abnormal fluctuation marker for the respiratory signal is generated; Based on the target waveform feature vector and the abnormal fluctuation marker of the respiratory signal, a preset air supply correction rule is matched to correct the baseline air supply parameters and generate the air supply correction command.

[0010] According to the self-rescue adaptive breathing adjustment method for compressed air provided by the present invention, the step of correcting the reference air supply parameters by matching a preset air supply correction rule based on the target waveform feature vector and the abnormal fluctuation marker of the breathing signal, and generating the air supply correction command, includes: The target waveform feature vector is subjected to waveform peak feature extraction, and the dynamic change data of the respiratory waveform is obtained by parsing. Based on the dynamic change data of the respiratory waveform and the abnormal fluctuation marker of the respiratory signal, the frequency of abnormal respiratory fluctuations is counted. If it is determined that the frequency of abnormal respiratory fluctuations exceeds a preset frequency threshold, an emergency adjustment signal for the air supply is generated. Using the aforementioned emergency air supply adjustment signal, combined with preset air supply correction rules, the baseline air supply parameters are corrected and updated using a hyperbolic tangent nonlinear correction formula to determine an air supply control command that is suitable for the user's current abnormal breathing fluctuations.

[0011] According to the self-rescue breathing adaptive adjustment method for compressed air provided by the present invention, the hyperbolic tangent nonlinear correction formula is as follows: ; in, This indicates the revised final reference gas supply parameters. This indicates the baseline gas supply parameters retrieved during the initial matching process. This represents the adaptive correction coefficient for emergency gas supply in downhole wells. This represents the hyperbolic tangent nonlinear constraint function. This indicates the measured frequency of abnormal respiratory fluctuations. This indicates the preset threshold for abnormal respiratory frequency.

[0012] According to the adaptive adjustment method for compressed air self-rescue breathing provided by the present invention, the adaptive threshold detection algorithm dynamically updates the judgment threshold based on the baseline level of the data stream. The threshold update satisfies: ; in, Indicates a real-time adaptive threshold. Indicates a fixed adjustment coefficient. This represents the average signal amplitude within a short time window.

[0013] The present invention also provides a compressed air self-rescue breathing adaptive adjustment device, comprising: The acquisition module is used to acquire ambient sound wave signals in the underground coal mine environment and air pressure fluctuation signals in the compressed air self-rescue gas supply pipeline. The extraction module is used to couple and reduce noise from the ambient sound wave signal and the air pressure fluctuation signal to extract a pure breathing signal; The generation module is used to identify the user's breathing state based on the pure breathing signal and generate an air supply correction command based on the breathing state. The drive module is used to drive the regulating valve of the compressed air self-rescue system to operate using the air supply correction command, so that the actual air supply of the compressed air self-rescue system matches the user's real-time breathing needs.

[0014] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the compressed air self-rescue breathing adaptive adjustment method as described in any of the preceding claims.

[0015] The present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the compressed air self-rescue breathing adaptive adjustment method described in any of the preceding claims.

[0016] The present invention also provides a computer program product, including a computer program, which, when executed by a processor, implements the compressed air self-rescue breathing adaptive adjustment method described in any of the above claims.

[0017] The present invention provides a method and related device for adaptive adjustment of compressed air self-rescue breathing. This method acquires ambient sound signals and air pressure fluctuation signals from the compressed air self-rescue supply pipeline in a coal mine environment. It couples and reduces noise from the ambient sound signals and air pressure fluctuation signals to extract a pure breathing signal. Based on the pure breathing signal, it identifies the user's breathing state and generates an air supply correction command. The air supply correction command drives the regulating valve of the compressed air self-rescue system to match the actual air supply of the system with the user's real-time breathing needs. By first acquiring both ambient sound and pipeline air pressure fluctuation signals and then performing noise reduction processing to obtain a pure breathing signal, the method ensures proper breathing. The accuracy of feature extraction avoids irrelevant interference affecting the control logic; by constructing a target waveform feature vector, the breathing state is accurately captured, and combined with anomaly markers, the air supply parameters are targeted for correction. The generated adjustment commands can directly drive the regulating valve to achieve real-time matching of air supply with the user's breathing needs; it not only solves the shortcomings of traditional compressed air self-rescue devices with fixed air supply and inability to adapt to individual breathing differences, but also ensures air supply stability through precise control, effectively improving the emergency response capability and adaptability of the compressed air self-rescue system, ensuring the breathing safety of underground workers, while reducing air waste, improving the intelligence and practicality of the compressed air self-rescue system, and meeting the emergency respiratory protection needs under complex working conditions in coal mines. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is a flowchart of the compressed air self-rescue breathing adaptive adjustment method provided in the embodiments of the present invention; Figure 2 This is a functional structure diagram of the compressed air self-rescue breathing adaptive adjustment device provided in an embodiment of the present invention; Figure 3 This is a functional structure diagram of the electronic device provided in the embodiments of the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0021] Figure 1 A flowchart of the compressed air self-rescue breathing adaptive adjustment method provided in the embodiments of the present invention is shown below. Figure 1 As shown, the compressed air self-rescue breathing adaptive adjustment method provided in this embodiment of the invention includes: Step 101: Acquire the ambient sound wave signal in the underground coal mine environment and the air pressure fluctuation signal in the compressed air self-rescue gas supply pipeline; This invention utilizes acoustic sensors pre-installed in the underground coal mine work area to collect ambient acoustic signals in real time. These signals include environmental interference noise such as underground machinery operation and airflow disturbance, as well as acoustic signals generated when the user breathes. Additionally, pressure sensors installed on the inner wall of the compressed air self-rescue breathing supply equipment pipeline collect real-time air pressure fluctuation signals within the pipeline. These signals are caused by the user's breathing actions; the air pressure decreases during inhalation and increases during exhalation.

[0022] Step 102: Couple and denoise the ambient sound signal and the air pressure fluctuation signal to extract the pure breathing signal; In this embodiment of the invention, the collected ambient sound wave signal and air pressure fluctuation signal are used as dual inputs and simultaneously fed into a pre-trained noise reduction processing network. This network is used to filter out irrelevant interference components in the two signals. Mechanical noise and wind noise in the ambient sound wave signal, as well as pipeline vibration interference and airflow clutter in the air pressure fluctuation signal, are effectively suppressed. Through feature separation and signal purification processing of the network, after removing interference signals, the effective signals directly related to the user's breathing actions are extracted, and finally a pure target breathing signal is output. This target breathing signal can truly reflect the user's core physiological parameters such as breathing rhythm and breathing intensity.

[0023] Step 103: Identify the user's breathing state based on the pure breathing signal, and generate an air supply correction command based on the breathing state; This invention uses the output pure target respiratory signal as a basis, performs time-series analysis and feature extraction on it, extracts key respiratory parameters such as respiratory cycle, respiratory amplitude, and respiratory rate, normalizes these parameters, and integrates them to construct a target waveform feature vector of a unified dimension. This vector can comprehensively represent the user's current respiratory state. At the same time, a respiratory intensity identifier is extracted from the target waveform feature vector. This identifier is used to quantify the user's current respiratory intensity and compares it with a preset respiratory intensity threshold range. If the respiratory intensity identifier exceeds the preset range, it indicates that the user's respiratory state is abnormal. At this time, an abnormal fluctuation marker of the respiratory signal is automatically generated.

[0024] Step 104: Use the air supply correction command to drive the regulating valve of the compressed air self-rescue system to match the actual air supply of the compressed air self-rescue system with the user's real-time breathing needs.

[0025] In this embodiment of the invention, air supply correction rules are pre-stored in the system. These rules relate the matching relationship between the respiratory state corresponding to the target waveform feature vector and the air supply parameters, and also include an emergency correction strategy for air supply when there are abnormal fluctuations in breathing. First, a preset baseline air supply parameter is retrieved, which is the basic air supply adapted to normal breathing conditions. Then, the target waveform feature vector (reflecting real-time respiratory rhythm and intensity) and the abnormal fluctuation marker of the respiratory signal (reflecting whether breathing is abnormal) are combined to match the corresponding air supply correction rules and dynamically correct the baseline air supply parameter. The corrected air supply parameter can adapt to the user's current breathing state, especially to achieve emergency air supply compensation when breathing is abnormal. Finally, based on the corrected air supply parameter, an adjustment command for regulating the air supply of the compressed air self-rescue system is generated. This command includes the control parameter for the opening of the regulating valve.

[0026] In this embodiment of the invention, the generated adjustment command is transmitted to the regulating valve control unit of the compressed air self-rescue system. The control unit drives the regulating valve to perform corresponding opening adjustment actions according to the opening control parameters in the adjustment command. The opening of the regulating valve is positively correlated with the air supply volume; the larger the opening, the larger the air supply volume, and the smaller the opening, the smaller the air supply volume. By adjusting the opening of the regulating valve in real time, the air supply volume output by the compressed air self-rescue system can be dynamically adjusted according to the changes in the user's breathing state. When the user's breathing intensity increases or abnormal breathing occurs, the air supply volume increases synchronously to meet the emergency breathing needs. When the user's breathing state returns to normal, the air supply volume is adjusted to the normal level. Finally, the target air supply volume is output that precisely matches the user's real-time breathing needs, realizing adaptive control of the air supply volume, ensuring smooth breathing for the user, and avoiding air waste.

[0027] Traditional compressed air self-rescue systems generally adopt a constant flow air supply mode with preset fixed levels. After leaving the factory, the air supply volume remains constant, lacking dynamic adaptive adjustment capabilities. In practical applications, different workers have individual differences in physiological conditions and respiratory rhythms. Furthermore, when workers are in distress underground, experiencing panic or oxygen deficiency, their breathing frequency and intensity can change significantly. A constant air supply method easily leads to a mismatch between the supplied air volume and the actual breathing needs of the personnel. Simultaneously, the underground working environment is filled with mechanical operating noise, airflow disturbance noise, and pipeline airflow clutter. Traditional compressed air self-rescue equipment lacks effective signal filtering and interference suppression methods, making it difficult to accurately separate effective human breathing signals from complex environmental clutter, resulting in low accuracy in identifying breathing rhythm and status. Existing conventional solutions rely solely on simple pipeline pressure thresholds for coarse air supply control, failing to effectively perceive and analyze the characteristics of personnel's breathing waveforms and abnormal breathing fluctuations, and lacking the ability to dynamically adjust the air supply volume based on the real-time physiological state of the personnel's breathing.

[0028] In addition, existing compressed air self-rescue control methods are mostly purely mechanical set-value adjustment or simple electronic control gear switching. The control logic is simple and the response is lagging. There is no correlation and matching mechanism between breathing characteristics and air supply volume. It is impossible to carry out emergency air supply compensation adjustment for abnormal breathing fluctuations of personnel. The overall protection adaptability and intelligence level are low, making it difficult to meet the personalized and real-time emergency respiratory protection needs of workers in complex underground accident scenarios.

[0029] The adaptive adjustment method for compressed air self-rescue breathing provided in this invention acquires ambient sound wave signals and air pressure fluctuation signals in the compressed air self-rescue air supply pipeline in the underground coal mine environment; couples and denoises the ambient sound wave signals and the air pressure fluctuation signals to extract a pure breathing signal; identifies the user's breathing state based on the pure breathing signal, and generates an air supply correction command based on the breathing state; uses the air supply correction command to drive the regulating valve of the compressed air self-rescue system to match the actual air supply of the compressed air self-rescue system with the user's real-time breathing needs. By first acquiring dual signals of ambient sound waves and pipeline air pressure fluctuations, and then obtaining a pure breathing signal through noise reduction processing, the method ensures that the breathing characteristics are correct. The system ensures accurate feature extraction and avoids irrelevant interference affecting the control logic. By constructing a target waveform feature vector, it accurately captures the breathing state and combines it with anomaly markers to achieve targeted correction of the air supply parameters. The generated adjustment commands can directly drive the regulating valve to achieve real-time matching of air supply with the user's breathing needs. This not only solves the shortcomings of traditional compressed air self-rescue devices, such as fixed air supply and inability to adapt to individual breathing differences, but also ensures air supply stability through precise control. This effectively improves the emergency response capability and adaptability of the compressed air self-rescue system, protects the breathing safety of underground workers, reduces air waste, enhances the intelligence and practicality of the compressed air self-rescue system, and meets the emergency respiratory protection needs under complex working conditions in coal mines.

[0030] Based on any of the above embodiments, the step of coupling and denoising the ambient sound signal and the air pressure fluctuation signal to extract a clean breathing signal includes: Step 201: Input the ambient sound wave signal and the air pressure fluctuation signal into a pre-constructed wavelet packet attention dual-source coupling noise reduction network to output preliminary clean breathing signal data; Step 202: Perform feature enhancement extraction on the preliminary purified respiratory signal data, and analyze and determine the core frequency components of the respiratory signal; This invention employs a Fourier transform algorithm to perform spectral analysis on the preliminary purified respiratory signal. The calculation formula is as follows: ; in, The spectrum function representing the respiratory signal. This indicates the initial, clean breathing signal output. Represents frequency variables. Represents a time variable.

[0031] This formula can be used to obtain the spectral distribution of the respiratory signal, thereby determining the core frequency components. This involves screening out the characteristic frequencies corresponding to the respiratory action, eliminating irrelevant frequency interference, and ensuring that the extracted respiratory features are targeted. This provides accurate frequency data support for the subsequent construction of a dynamic model of the respiratory signal and analysis of the respiratory state, realizing the transformation from a preliminary pure signal to an effective characteristic signal.

[0032] Step 203: Substitute the core frequency components into the dynamic model of the respiratory signal to calculate the respiratory characteristic parameters; Step 204: Integrate the respiratory characteristic parameters into fluctuation characteristic parameters, and determine whether the respiratory signal is within a preset normal range based on the fluctuation characteristic parameters; Step 205: If yes, determine that the preliminary pure breathing signal is a pure breathing signal; Step 206: If the fluctuation characteristic parameter determines that the respiratory signal exceeds the preset normal range, then perform secondary correction processing on the preliminary pure respiratory signal to obtain the pure respiratory signal.

[0033] In this embodiment of the invention, a preset respiratory characteristic threshold is used as the judgment standard to compare the fluctuation characteristic parameter with the preset threshold. If the fluctuation characteristic parameter is within the preset normal range, no secondary correction is required, and the respiratory signal is directly used as the target respiratory signal. If the fluctuation characteristic parameter exceeds the preset normal range, it indicates that the respiratory signal is abnormal. At this time, a secondary correction process is performed, and the abnormal part is adjusted by a signal correction algorithm to eliminate the signal deviation and finally obtain a pure and accurate target respiratory signal.

[0034] During the correction process, the air supply parameters are adjusted based on the frequency of abnormal fluctuations and the characteristics of the respiratory waveform to ensure that the corrected air supply matches the current respiratory state. At the same time, corresponding air supply control commands are generated to determine the opening adjustment parameters of the regulating valve. The commands contain key information such as the air supply size and adjustment rate, providing a clear control basis for subsequent regulating valve actions and achieving precise matching between air supply and respiratory state.

[0035] This embodiment effectively filters out various interference signals in the complex underground environment by combining dual signal inputs (ambient sound wave + pipeline air pressure) with a dedicated noise reduction network, solving the problems of easy interference and signal distortion in traditional single signal acquisition. The signal enhancement and secondary correction process ensures the accuracy and stability of the target breathing signal, avoiding the impact of noise on subsequent air supply regulation. The construction and correction of the dynamic model makes the feature extraction of the breathing signal more accurate, providing reliable data support for air supply regulation and ensuring the rationality of subsequent air supply adaptation. The entire process is seamless, which not only improves the accuracy of breathing signal acquisition, but also provides a solid technical foundation for subsequent regulating valve control and air supply adaptation. It effectively solves the problems of large signal interference and poor adaptability in traditional compressed air self-rescue, while reducing signal processing errors, improving the stability and practicality of the overall system, and meeting the respiratory protection needs in the complex environment of underground coal mines.

[0036] Based on any of the above embodiments, the step of inputting the ambient sound signal and the air pressure fluctuation signal into a pre-constructed wavelet packet attention dual-source coupled noise reduction network to output preliminary clean respiratory signal data includes: Step 301: Perform wavelet packet decomposition on the ambient sound wave signal and the air pressure fluctuation signal respectively, and decompose each signal into multiple sub-band signals of different frequency bands; Step 302: The sub-band signals are weighted using an attention mechanism to calculate the correlation weight between each sub-band signal and the respiratory signal. The correlation weight is used to enhance the sub-band components related to respiratory characteristics and suppress the sub-band components related to noise interference. Step 303: Couple and fuse the sub-band signals of the weighted environmental sound wave signal and the air pressure fluctuation signal to generate a dual-source coupling feature matrix; Step 304: Perform wavelet packet reconstruction on the dual-source coupling feature matrix and output the preliminary purified respiratory signal data.

[0037] Based on any of the above embodiments, substituting the core frequency component into the dynamic model of the respiratory signal, calculating the respiratory characteristic parameters, and integrating the respiratory characteristic parameters into fluctuation characteristic parameters includes: Step 401: Establish a time-frequency coupled dynamic model of respiratory signals based on core frequency components: ; in, This represents the wavelet packet time-frequency decomposition operator. This represents the frequency domain analytic operator. Represents the time-frequency distribution function of the respiratory signal. Represents a time variable. Represents frequency variables; Step 402: Solve the time-frequency coupled dynamic model of the respiratory signal to obtain respiratory characteristic parameters, including respiratory modal period, time-frequency amplitude divergence, and respiratory transient rate of change. , ; ; ; in, Indicates the respiratory modal period, This represents the extreme values ​​of the core frequency mode of the respiratory signal. This represents the magnitude of the time-frequency distribution function of the respiratory signal. Indicates the time-frequency amplitude divergence. This represents the mean function of the time-frequency distribution of respiratory signals. Indicates the transient rate of change of respiration; Step 403: Based on the respiratory modal cycle Time-frequency amplitude divergence and respiratory transient rate of change Respiratory fluctuation feature parameters are generated through adaptive weighted fusion: ; in, Represents the characteristic parameters of respiratory fluctuations. , , This represents the weighting coefficients for adaptive learning of the downhole environment. The time integral of the absolute value of the transient rate of change of respiration.

[0038] This embodiment establishes a time-frequency coupled dynamic model of the respiratory signal, combining wavelet packet time-frequency decomposition and frequency domain analysis with dual operators to achieve precise fusion of the time-frequency dual-domain features of the respiratory signal. This solves the problem of insufficient accuracy in traditional single-domain feature analysis and ensures the accuracy of the respiratory signal's time-frequency distribution function. By solving three core feature parameters—respiratory modal period, time-frequency amplitude divergence, and respiratory transient rate of change—it comprehensively covers key information such as the period, amplitude dispersion, and transient changes of the respiratory signal, avoiding the drawback of a single feature parameter being unable to fully characterize the respiratory state. By adaptively weighting and fusing to generate respiratory fluctuation feature parameters, and using the underground environment to adaptively learn weighting coefficients, it can adapt to the dynamic changes of complex working conditions in coal mines, ensuring that the fluctuation feature parameters can truly reflect the respiratory state under different working conditions.

[0039] Based on any of the above embodiments, the step of identifying the user's breathing state based on the pure breathing signal and generating a gas supply correction command based on the breathing state includes: Step 501: Obtain the time-domain data stream corresponding to the pure breathing signal, and use an adaptive threshold detection algorithm to detect peak and valley values ​​in the time-domain data stream to obtain a set of extreme points including inspiratory peak points and expiratory valley points; Step 502: Calculate the respiratory cycle sequence based on the time interval between adjacent peak points and valley points in the extreme point set; In this embodiment of the invention, the set of extreme points is traversed, and the time difference between adjacent extreme points of the same type is used as a single respiratory cycle. The calculation formula is as follows: ; In the formula For the first One respiratory cycle, , The sampling time corresponds to two adjacent extreme points of the same attribute.

[0040] The effective time intervals for each group are solved sequentially, and all single respiratory cycles are arranged in chronological order to form a discretized respiratory cycle sequence.

[0041] Step 503: Extract amplitude correlation parameters based on the amplitude difference between the peak inspiratory point and the trough inspiratory point within the same respiratory cycle in the extreme point set; Perform mathematical statistical operations on discrete respiratory cycle sequences to calculate the mean and variance of the sequences: ; In the formula This is the average value of the respiratory cycle. The variance of the respiratory cycle, This represents the number of sequence samples.

[0042] The mean represents the overall average level of the respiratory cycle, and the variance represents the degree of dispersion of respiratory cycle fluctuations. At the same time, the sample proportion of different cycle intervals is statistically analyzed to obtain the distribution characteristics of the respiratory cycle and quantify the stability of the overall respiratory rhythm.

[0043] Step 504: Combine the respiratory cycle sequence with the amplitude correlation parameters to construct a target waveform feature vector of a unified dimension; In this embodiment of the invention, the respiratory cycle distribution feature component and the real-time amplitude mean component of the target respiratory signal are extracted respectively. The two types of components are normalized and dimensionally unified to eliminate the differences in the numerical range of different physical quantities. According to the preset fixed dimension structure, the cycle feature component and the amplitude feature component are sequentially spliced ​​together to generate a target waveform feature vector with a unified dimension.

[0044] Step 505: Generate a respiratory intensity identifier based on the amplitude correlation parameter in the target waveform feature vector, and obtain the dynamic change trend of respiratory intensity based on the respiratory cycle sequence in the target waveform feature vector; The amplitude-related feature parameters are analyzed from the target waveform feature vector, and the quantization index is solved using a linear mapping relationship to obtain the standardized respiratory intensity indicator: ; In the formula For respiratory intensity indicators, For amplitude correlation parameters, , These are the preset mapping coefficients.

[0045] Multiple sets of respiratory intensity markers are obtained through continuous time-series sampling, forming a marker time-series sequence. By analyzing the increase and decrease amplitude and rate of change of the sequence values, the real-time rise and fall and fluctuation trend of respiratory intensity can be directly determined, and the dynamic evolution process of the intensity of breathing can be quantitatively characterized.

[0046] Step 506: When the respiratory intensity indicator exceeds the preset respiratory intensity threshold range, a respiratory signal abnormal fluctuation marker is generated; Pre-set upper and lower limits of respiratory intensity thresholds corresponding to normal breathing conditions. The system compares the real-time calculated respiratory intensity marker with the threshold range; if the respiratory intensity marker is greater than the upper limit or less than the lower limit, it determines that the current respiratory intensity deviates from the normal range and the respiratory state is abnormal; the system immediately generates a fixed-format respiratory signal abnormal fluctuation marker.

[0047] This invention first acquires the time-domain data stream corresponding to the target respiratory signal, and uses an adaptive threshold detection algorithm to detect peak and trough values ​​of the respiratory signal amplitude in the time-domain data stream, filtering out an extreme point set composed of inspiratory peak points and expiratory trough points; based on the time interval between adjacent peak points and trough points in the extreme point set, a respiratory cycle sequence is calculated; statistical analysis is performed on the respiratory cycle sequence to obtain respiratory cycle distribution characteristics; the respiratory cycle sequence is correlated and fused with the real-time amplitude value of the target respiratory signal to construct a target waveform feature vector of a unified dimension; amplitude correlation parameters are extracted from the target waveform feature vector, and a respiratory intensity identifier is determined based on this parameter; the dynamic change trend of respiratory intensity is determined according to the temporal change pattern of the respiratory intensity identifier; the real-time respiratory intensity identifier is compared with a preset respiratory intensity threshold range, and when the respiratory intensity identifier exceeds the threshold range, an abnormal fluctuation marker of the respiratory signal is generated.

[0048] Step 507: Based on the target waveform feature vector and the abnormal fluctuation marker of the respiratory signal, match the preset air supply correction rule, correct the reference air supply parameters, and generate the air supply correction command.

[0049] This invention, based on the time-domain data stream of the target respiratory signal, uses an adaptive threshold algorithm to detect extreme points, adapting to the dynamic fluctuation characteristics of respiratory signal amplitude and ensuring the standardization of extreme point identification. By calculating the respiratory cycle sequence and statistically analyzing its distribution characteristics using the time interval between adjacent extreme points, the invention objectively reconstructs the rhythmic patterns of human respiration. It integrates the respiratory cycle and real-time signal amplitude values ​​to construct a unified-dimensional feature vector, achieving standardized representation of the two core information types: respiratory rhythm and amplitude. By extracting respiratory intensity identifiers from the feature vector amplitude parameters and analyzing temporal changes, the evolution of respiratory intensity can be quantified. Abnormal fluctuation markers are generated based on fixed threshold comparisons, establishing a unified standard for judging respiratory anomalies.

[0050] Based on any of the above embodiments, the step of matching a preset air supply correction rule with the target waveform feature vector and the abnormal fluctuation marker of the respiratory signal to correct the baseline air supply parameters and generate the air supply correction instruction includes: Step 601: Extract waveform peak features from the target waveform feature vector and parse to obtain dynamic change data of the respiratory waveform; Step 602: Based on the dynamic change data of the respiratory waveform and the abnormal fluctuation marker of the respiratory signal, count the frequency of abnormal respiratory fluctuations. If it is determined that the frequency of abnormal respiratory fluctuations exceeds a preset frequency threshold, generate an emergency adjustment signal for the air supply. Step 603: Using the emergency air supply adjustment signal and the preset air supply correction rules, the reference air supply parameters are corrected and updated using the hyperbolic tangent nonlinear correction formula to determine the air supply control command that is suitable for the user's current abnormal breathing fluctuation state.

[0051] This invention, based on the target waveform feature vector and abnormal respiratory signal fluctuation markers, matches the system's preset mapping relationship between respiratory state and air supply parameters to accurately retrieve baseline air supply parameters that are initially adapted to the user's current respiratory state. Subsequently, it extracts peak waveform features from the target waveform feature vector, analyzing the dynamic changes in the respiratory waveform, such as peak value variations, amplitude fluctuations, and waveform slope, to comprehensively capture the real-time dynamic characteristics of the respiratory signal. Combining the analyzed dynamic changes in the respiratory waveform with the abnormal respiratory signal fluctuation markers, it statistically analyzes the frequency of abnormal respiratory fluctuations per unit time, and then uses the statistically derived respiratory abnormalities... The frequency of fluctuations is compared with a preset frequency threshold to determine whether it exceeds the threshold range. If the frequency of abnormal breathing fluctuations exceeds the preset frequency threshold, it indicates that the user's abnormal breathing condition is relatively obvious and emergency gas supply adjustment is required. At this time, an emergency gas supply adjustment signal is generated. Finally, this emergency gas supply adjustment signal is used in conjunction with the preset gas supply correction rule. The reference gas supply parameters that have been retrieved are dynamically corrected and updated through a hyperbolic tangent nonlinear correction formula. The correction process takes into account both the frequency of abnormal breathing fluctuations and the current dynamic characteristics of breathing, and finally determines the gas supply control command that is suitable for the user's current abnormal breathing fluctuation state.

[0052] This invention uses the target waveform feature vector and abnormal respiratory signal fluctuation markers as core inputs. It retrieves baseline air supply parameters through a preset mapping relationship to ensure the basic adaptability of air supply adjustment and avoid blind control. By extracting dynamic change data of the respiratory waveform, it accurately captures real-time changes in respiratory status, providing reliable data support for abnormal frequency statistics and air supply correction. Combining abnormal fluctuation frequency statistics with a preset threshold, it achieves quantitative judgment of the degree of respiratory abnormality, ensuring the accuracy and timeliness of emergency adjustment signal generation. The hyperbolic tangent nonlinear correction formula is used to correct the baseline air supply parameters, effectively limiting the air supply adjustment range and avoiding respiratory shock caused by sudden increases or decreases in air supply, while simultaneously achieving precise adaptation between air supply and abnormal respiratory fluctuation states.

[0053] In this embodiment of the invention, the hyperbolic tangent nonlinear correction formula is: ; in, This indicates the revised final reference gas supply parameters. This indicates the baseline gas supply parameters retrieved during the initial matching process. This represents the adaptive correction coefficient for emergency gas supply in downhole wells. This represents the hyperbolic tangent nonlinear constraint function. This indicates the measured frequency of abnormal respiratory fluctuations. This indicates the preset threshold for abnormal respiratory frequency.

[0054] This embodiment effectively limits the correction range of the gas supply by introducing a hyperbolic tangent constraint function, avoiding airflow impact caused by sudden increases or decreases in the gas supply during the correction process, ensuring the stability of the gas supply, and adapting to the safety requirements of emergency gas supply in underground compressed air self-rescue systems in coal mines. It effectively solves the problems of insufficient accuracy and excessive or insufficient adjustment range in traditional gas supply correction methods, provides a reliable basis for the generation of gas supply control commands, further improves the accuracy of compressed air self-rescue system gas supply control and emergency response capability, and ensures the breathing safety of underground workers.

[0055] Based on any of the above embodiments, the adaptive threshold detection algorithm dynamically updates the judgment threshold according to the data stream baseline level. The threshold update satisfies: ; in, Indicates a real-time adaptive threshold. Indicates a fixed adjustment coefficient. This represents the average signal amplitude within a short time window.

[0056] In this embodiment of the invention, the target respiratory signal is sampled in time to form a continuous time-domain data stream. The time sequence is arranged with time as the horizontal axis and signal amplitude as the vertical axis. The dynamic threshold is used as the judgment criterion. The time-domain data stream is traversed point by point. Local maxima are identified as inspiratory peak points and local minima are identified as expiratory valley points. False extreme points caused by noise are eliminated. Valid extreme points are retained and sorted in chronological order to form a complete set of extreme points.

[0057] The implementation environment of this invention includes an underground coal mine operation and refuge area, a compressed air self-rescue gas supply system, an acquisition module, an intelligent control and execution unit, and a ground monitoring and management terminal. The underground coal mine operation and refuge area encompasses mining faces, mining area escape routes, refuge chambers, supply stations, and refuge points in areas with complex hydrogeology. The area features well-supported roadways and a regular working space, providing stable conditions for the installation, use, and signal acquisition of the compressed air self-rescue device.

[0058] The compressed air self-rescue breathing adaptive adjustment method provided in this invention inputs the collected ambient sound wave signal and air pressure fluctuation signal into a preset noise reduction processing network for noise suppression processing, filtering out irrelevant interference signals such as downhole mechanical operation and airflow disturbance, and finally outputting a pure target breathing signal. Based on the target breathing signal, a target waveform feature vector containing key information such as breathing rhythm and intensity is constructed. Simultaneously, it is determined whether the breathing intensity indicator in the feature vector exceeds a preset threshold. If it does, an abnormal fluctuation mark of the breathing signal is generated. Combining the target waveform feature vector and the abnormal fluctuation mark of the breathing signal, a preset air supply correction rule is matched to correct the baseline air supply parameters and generate an adjustment command for regulating the air supply. According to the adjustment command, the regulating valve of the compressed air self-rescue system is controlled to perform opening adjustment, and finally outputs a target air supply that matches the user's real-time breathing needs. Dual signal acquisition and noise reduction processing ensure the accuracy of respiratory feature extraction. Combined with waveform feature vectors and anomaly markers, targeted correction of air supply is achieved. This solves the shortcomings of traditional compressed air self-rescue air supply being fixed and unable to adapt to individual respiratory differences. It improves the system's emergency response capability and adaptability, ensures the respiratory safety of personnel underground, reduces gas waste, enhances the system's intelligence and practicality, and meets the complex emergency respiratory protection needs underground.

[0059] The following describes the compressed air self-rescue breathing adaptive adjustment device provided by the present invention. The compressed air self-rescue breathing adaptive adjustment device described below can be referred to in correspondence with the compressed air self-rescue breathing adaptive adjustment method described above.

[0060] Figure 2 This is a functional structure diagram of the compressed air self-rescue breathing adaptive adjustment device provided in an embodiment of the present invention, as shown below. Figure 2 As shown, the compressed air self-rescue breathing adaptive adjustment device provided in this embodiment of the invention includes: The acquisition module 201 is used to acquire the ambient sound wave signal in the underground coal mine environment and the air pressure fluctuation signal in the compressed air self-rescue gas supply pipeline. Extraction module 202 is used to couple and reduce noise between the ambient sound wave signal and the air pressure fluctuation signal to extract a pure breathing signal; The generation module 203 is used to identify the user's breathing state based on the pure breathing signal and generate an air supply correction command based on the breathing state. The drive module 204 is used to drive the regulating valve of the compressed air self-rescue system to operate using the air supply correction command, so that the actual air supply of the compressed air self-rescue system matches the user's real-time breathing needs.

[0061] In the embodiment of the present invention, the compressed air self-rescue air supply system consists of a ground compressed air station, a metal main air supply pipeline and branch pipelines, an oil-water separation device, a drain valve, a special air supply valve and a compressed air self-rescue device. The air supply pipeline is laid along the intake shaft and the disaster avoidance route to all the refuge places and working points. The pipeline diameter is adapted to the air supply demand. The air supply valve is made of corrosion-resistant alloy material. The compressed air self-rescue device can be selected in the form of a protective bag or a protective face mask structure to provide basic emergency air supply guarantee for underground personnel.

[0062] The acquisition module includes an acoustic acquisition component and a pressure detection component. The acoustic acquisition component is arranged in the underground working environment, and the pressure detection component is installed on the inner wall of the compressed air self-rescue air supply pipeline. Both components adopt a structural design adapted to the special underground environment, can stably collect the environmental acoustic wave signal and the pipeline air pressure fluctuation signal, and have the service performance of anti-interference, dust-proof and moisture-proof. An intrinsically safe acoustic acquisition unit and a pressure detection unit adapted to the underground explosive environment are adopted. The sensing components are arranged in areas with good support and convenient operation and maintenance. The external cables and components are made of flame-retardant and antistatic materials, meeting the electrical and combustion performance requirements for safe use underground.

[0063] The extraction module integrates functions of signal processing, feature analysis, anomaly recognition and regulating valve driving. It receives the original signals transmitted by the acquisition unit, completes noise reduction and purification, feature construction, anomaly determination and air supply volume correction, and outputs control instructions to drive the regulating valve of the air supply pipeline to perform opening adjustment. The unit as a whole is adapted to the operation conditions of underground closed and multi-interference.

[0064] The ground monitoring and management terminal establishes data communication with the underground intelligent regulation and execution unit, and receives parameters such as air supply pressure, air supply volume, breathing state, and device operation state in real time, realizing remote monitoring, parameter configuration, operation record storage and regular verification management of the compressed air self-rescue adaptive regulation system, and ensuring the stable operation of the whole process of the system.

[0065] The extraction module and the generation module are integrated in a metal protection box. The box has reliable dust-proof protection ability, and the operation noise of the module is controlled within the preset limit to avoid interfering with the underground working environment.

[0066] The air supply valve supporting the driving module is made of corrosion-resistant alloy material. The air supply pipeline is laid along the mine disaster avoidance route, refuge facilities and mining operation areas in a standardized manner. The main pipeline and the pipeline of the mining face meet the corresponding diameter requirements respectively. Oil-water separation and drainage components are set at the key nodes of the pipeline to ensure the long-term stable operation of the air supply system. The system is equipped with special underground on-line monitoring instruments to collect key parameters such as air supply pressure, air supply volume, and device operation state in real time. Components such as protective bags, face masks, and air ducts all meet the mechanical performance and combustion performance requirements. The system performs regular inspections, ventilation checks and performance calibration in accordance with the underground safety equipment management specifications. Each module works together to complete signal acquisition, noise reduction and purification, abnormal breathing identification and adaptive air supply control, fully adapting to the entire process of installation, deployment, stable operation, maintenance and performance verification of underground compressed air self-rescue systems in coal mines.

[0067] In this embodiment, the various modules of the system work together to effectively solve the problems of fixed air supply and poor adaptability of traditional compressed air self-rescue systems. This improves the intelligence level and reliability of the compressed air self-rescue system, meets the respiratory protection needs of workers in the complex environment of underground coal mines, ensures the respiratory safety of workers, reduces air waste, and improves system operating efficiency.

[0068] The compressed air self-rescue breathing adaptive adjustment device provided in this invention acquires ambient sound wave signals and air pressure fluctuation signals in the compressed air self-rescue air supply pipeline in the underground coal mine environment; couples and reduces noise from the ambient sound wave signals and the air pressure fluctuation signals to extract a pure breathing signal; identifies the user's breathing state based on the pure breathing signal, and generates an air supply correction command based on the breathing state; uses the air supply correction command to drive the regulating valve of the compressed air self-rescue system to match the actual air supply of the compressed air self-rescue system with the user's real-time breathing needs. By first acquiring dual signals of ambient sound waves and pipeline air pressure fluctuations, and then obtaining a pure breathing signal through noise reduction processing, the device ensures the correct breathing characteristics. The system ensures accurate feature extraction and avoids irrelevant interference affecting the control logic. By constructing a target waveform feature vector, it accurately captures the breathing state and combines it with anomaly markers to achieve targeted correction of the air supply parameters. The generated adjustment commands can directly drive the regulating valve to achieve real-time matching of air supply with the user's breathing needs. This not only solves the shortcomings of traditional compressed air self-rescue devices, such as fixed air supply and inability to adapt to individual breathing differences, but also ensures air supply stability through precise control. This effectively improves the emergency response capability and adaptability of the compressed air self-rescue system, protects the breathing safety of underground workers, reduces air waste, enhances the intelligence and practicality of the compressed air self-rescue system, and meets the emergency respiratory protection needs under complex working conditions in coal mines.

[0069] Figure 3 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3As shown, the electronic device may include a processor 310, a communication interface 320, a memory 330, and a communication bus 340. The processor 310, communication interface 320, and memory 330 communicate with each other via the communication bus 340. The memory 330 includes computer programs, an operating system, and acquired data. The processor 310 can call logical instructions in the memory 330 to execute a compressed air self-rescue breathing adaptive adjustment method. This method includes: acquiring ambient sound wave signals and air pressure fluctuation signals in the compressed air self-rescue supply pipeline in the coal mine environment; coupling and denoising the ambient sound wave signals and the air pressure fluctuation signals to extract a pure breathing signal; identifying the user's breathing state based on the pure breathing signal; generating an air supply correction command based on the breathing state; and using the air supply correction command to drive the regulating valve of the compressed air self-rescue system to match the actual air supply of the compressed air self-rescue system with the user's real-time breathing needs.

[0070] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to related technologies, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0071] On the other hand, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon. When executed by a processor, the computer program implements the adaptive adjustment method for compressed air self-rescue breathing provided by the above methods. The method includes: acquiring ambient sound wave signals in the underground coal mine environment and air pressure fluctuation signals in the compressed air self-rescue air supply pipeline; coupling and denoising the ambient sound wave signals and the air pressure fluctuation signals to extract a pure breathing signal; identifying the user's breathing state based on the pure breathing signal; generating an air supply correction command based on the breathing state; and using the air supply correction command to drive the regulating valve of the compressed air self-rescue system to actuate, so that the actual air supply of the compressed air self-rescue system matches the user's real-time breathing needs.

[0072] In another aspect, embodiments of the present invention also provide a computer program product, including a computer program, which, when executed by a processor, implements the compressed air self-rescue breathing adaptive adjustment method described above.

[0073] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0074] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the parts that contribute to the related technology, can be embodied in the form of software products. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0075] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for adaptive adjustment of compressed air self-rescue breathing, characterized in that, include: Acquire environmental acoustic signals in the underground coal mine environment and air pressure fluctuation signals in the compressed air self-rescue gas supply pipeline; The ambient sound signal and the air pressure fluctuation signal are coupled and denoised to extract a pure breathing signal; The user's breathing state is identified based on the pure breathing signal, and an air supply correction command is generated based on the breathing state. The regulating valve of the compressed air self-rescue system is activated by the air supply correction command so that the actual air supply of the compressed air self-rescue system matches the user's real-time breathing needs.

2. The self-rescue breathing adaptive adjustment method for compressed air as described in claim 1, characterized in that, The process of coupling and denoising the ambient sound signal and the air pressure fluctuation signal to extract a clean breathing signal includes: The ambient sound wave signal and the air pressure fluctuation signal are input into a pre-constructed wavelet packet attention dual-source coupling noise reduction network, so that the wavelet packet attention dual-source coupling noise reduction network outputs preliminary clean respiratory signal data. Feature enhancement extraction is performed on the preliminary purified respiratory signal data, and the core frequency components of the respiratory signal are analyzed and determined. The core frequency components are substituted into the dynamic model of the respiratory signal for calculation to obtain respiratory characteristic parameters, and the respiratory characteristic parameters are integrated into fluctuation characteristic parameters. Based on the fluctuation characteristic parameters, determine whether the respiratory signal is within a preset normal range. If so, determine that the preliminary pure respiratory signal is a pure respiratory signal.

3. The self-rescue breathing adaptive adjustment method for compressed air according to claim 2, characterized in that, The step of inputting the ambient sound wave signal and the air pressure fluctuation signal into a pre-constructed wavelet packet attention dual-source coupled noise reduction network, so that the wavelet packet attention dual-source coupled noise reduction network outputs preliminary clean respiratory signal data, includes: Wavelet packet decomposition is performed on the ambient sound wave signal and the air pressure fluctuation signal respectively, decomposing each signal into multiple sub-band signals of different frequency bands; The subband signals are weighted using an attention mechanism to calculate the correlation weight between each subband signal and the respiratory signal. The correlation weight is used to enhance the subband components related to respiratory characteristics and suppress the subband components related to noise interference. The sub-band signals of the weighted environmental sound wave signal and the air pressure fluctuation signal are coupled and fused to generate a dual-source coupling feature matrix; Wavelet packet reconstruction is performed on the dual-source coupling feature matrix to output the preliminary purified respiratory signal data.

4. The self-rescue breathing adaptive adjustment method for compressed air according to claim 2, characterized in that, The step of substituting the core frequency components into the dynamic model of the respiratory signal for calculation to obtain respiratory characteristic parameters, and integrating the respiratory characteristic parameters into fluctuation characteristic parameters includes: Establish a time-frequency coupled dynamic model of respiratory signals based on core frequency components: ； in, This represents the wavelet packet time-frequency decomposition operator. This represents the frequency domain analytic operator. Represents the time-frequency distribution function of the respiratory signal. Represents a time variable. Represents frequency variables; Solve the time-frequency coupled dynamic model of the respiratory signal to obtain respiratory characteristic parameters, including respiratory modal period, time-frequency amplitude divergence, and respiratory transient rate of change. ， ； ； ； in, Indicates the respiratory modal period, This represents the extreme values ​​of the core frequency mode of the respiratory signal. This represents the magnitude of the time-frequency distribution function of the respiratory signal. Indicates the time-frequency amplitude divergence. This represents the mean function of the time-frequency distribution of respiratory signals. Indicates the transient rate of change of respiration; According to respiratory modal period Time-frequency amplitude divergence and respiratory transient rate of change Respiratory fluctuation feature parameters are generated through adaptive weighted fusion: ； in, Represents the characteristic parameters of respiratory fluctuations. , , This represents the weighting coefficients for adaptive learning of the downhole environment. The time integral of the absolute value of the transient rate of change of respiration.

5. The self-rescue breathing adaptive adjustment method for compressed air according to claim 2, characterized in that, Also includes: If the fluctuation characteristic parameter determines that the respiratory signal exceeds the preset normal range, then a secondary correction process is performed on the preliminary pure respiratory signal to obtain the pure respiratory signal.

6. The self-rescue breathing adaptive adjustment method for compressed air according to claim 1, characterized in that, The step of identifying the user's breathing state based on the pure breathing signal and generating a gas supply correction command based on the breathing state includes: The time-domain data stream corresponding to the pure breathing signal is obtained, and the peak and valley values ​​of the time-domain data stream are detected by an adaptive threshold detection algorithm to obtain a set of extreme points including inspiratory peak points and expiratory valley points. The respiratory cycle sequence is calculated based on the time interval between adjacent peak points and valley points in the set of extreme points; Based on the amplitude difference between the peak inspiratory point and the trough inspiratory point within the same respiratory cycle in the set of extreme points, the amplitude correlation parameter is extracted; By combining the respiratory cycle sequence with the amplitude correlation parameters, a target waveform feature vector of a unified dimension is constructed; A respiratory intensity identifier is generated based on the amplitude correlation parameter in the target waveform feature vector, and the dynamic change trend of respiratory intensity is obtained based on the respiratory cycle sequence in the target waveform feature vector. When the respiratory intensity indicator exceeds the preset respiratory intensity threshold range, an abnormal fluctuation marker for the respiratory signal is generated; Based on the target waveform feature vector and the abnormal fluctuation marker of the respiratory signal, a preset air supply correction rule is matched to correct the baseline air supply parameters and generate the air supply correction command.

7. The self-rescue breathing adaptive adjustment method for compressed air according to claim 6, characterized in that, The step of matching the target waveform feature vector and the abnormal fluctuation marker of the respiratory signal with a preset air supply correction rule to correct the baseline air supply parameters and generate the air supply correction command includes: The target waveform feature vector is subjected to waveform peak feature extraction, and the dynamic change data of the respiratory waveform is obtained by parsing. Based on the dynamic change data of the respiratory waveform and the abnormal fluctuation marker of the respiratory signal, the frequency of abnormal respiratory fluctuations is counted. If it is determined that the frequency of abnormal respiratory fluctuations exceeds a preset frequency threshold, an emergency adjustment signal for the air supply is generated. Using the aforementioned emergency air supply adjustment signal, combined with preset air supply correction rules, the baseline air supply parameters are corrected and updated using a hyperbolic tangent nonlinear correction formula to determine an air supply control command that is suitable for the user's current abnormal breathing fluctuations.

8. The self-rescue breathing adaptive adjustment method for compressed air according to claim 7, characterized in that, The hyperbolic tangent nonlinear correction formula is as follows: ； in, This indicates the revised final reference gas supply parameters. This indicates the baseline gas supply parameters retrieved during the initial matching process. This represents the adaptive correction coefficient for emergency gas supply in downhole wells. This represents the hyperbolic tangent nonlinear constraint function. This indicates the measured frequency of abnormal respiratory fluctuations. This indicates the preset threshold for abnormal respiratory frequency.

9. The self-rescue breathing adaptive adjustment method for compressed air according to claim 6, characterized in that, The adaptive threshold detection algorithm dynamically updates the judgment threshold based on the baseline level of the data stream. The threshold update satisfies: ； in, Indicates a real-time adaptive threshold. Indicates a fixed adjustment coefficient. This represents the average signal amplitude within a short time window.

10. A self-rescue breathing adaptive adjustment device using compressed air, characterized in that, include: The acquisition module is used to acquire ambient sound wave signals in the underground coal mine environment and air pressure fluctuation signals in the compressed air self-rescue gas supply pipeline. The extraction module is used to couple and reduce noise from the ambient sound wave signal and the air pressure fluctuation signal to extract a pure breathing signal; The generation module is used to identify the user's breathing state based on the pure breathing signal and generate an air supply correction command based on the breathing state. The drive module is used to drive the regulating valve of the compressed air self-rescue system to operate using the air supply correction command, so that the actual air supply of the compressed air self-rescue system matches the user's real-time breathing needs.

11. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the self-rescue breathing adaptive adjustment method for compressed air as described in any one of claims 1 to 9.

12. A non-transitory readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the self-rescue breathing adaptive adjustment method for compressed air as described in any one of claims 1 to 9.

13. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the self-rescue breathing adaptive adjustment method for compressed air as described in any one of claims 1 to 9.

Images