A method and system for detecting abnormal information of a tunneling face based on coal mining

Abstract

The present application relates to the technical field of coal mining, and discloses a tunneling working face abnormal information detection method and system based on coal mining, the method comprising: obtaining an acoustic emission signal generated by a tunneling machine when working on a tunneling working face, separating a pick-to-coal rock impact signal and a coal rock mass fracture signal based on the propagation path and waveform characteristics of the acoustic emission signal; identifying the pick wear state based on the frequency attenuation characteristics of the pick-to-coal rock impact signal, and extracting the coal rock fracture dominant frequency band based on the frequency spectrum mutation characteristics of the coal rock mass fracture signal; compensating the energy of the coal rock fracture dominant frequency band according to the pick wear state, and establishing a cooperative evolution relationship between the compensated energy and the frequency spectrum mutation characteristics in time sequence; the present application acutely captures the energy precursor abnormal fluctuation, thereby realizing early, accurate and targeted detection of potential abnormal information of the tunneling working face, and significantly improving the accuracy and timeliness of the early warning.

Description

Technical Field

[0001] This invention relates to the field of coal mining technology, and in particular to a method and system for detecting abnormal information in coal mining tunneling faces. Background Technology

[0002] Underground coal mine tunneling faces are prone to geological disasters such as abnormal gas outbursts and rock bursts due to mining activities. Real-time and accurate detection of abnormal geological information inside the working face is a key link in preventing safety accidents and ensuring mining efficiency. At present, the monitoring of anomalies in tunneling faces mainly relies on manual experience judgment, borehole detection, and contact sensor arrays. These methods have problems such as response lag, low spatial coverage, or interference with underground operations. They are also insufficient in capturing early micro-fracture precursor signals caused by stress concentration. In addition, some non-contact detection technologies based on acoustic emission identify anomalies by collecting rock fracture signals, but they are mostly concentrated on energy threshold early warning or single waveform parameter analysis, and lack decoupling processing of signal characteristics of the coupling effect between mining equipment conditions and geological response. Existing anomaly detection methods based on acoustic emission signals are ineffective in identifying early-stage hazards in tunneling faces. This is because the frequency bands of mining machinery impact noise and coal and rock fracture signals overlap, and the energy distribution of signals is continuously altered by equipment wear. Traditional feature extraction methods struggle to effectively separate genuine geological anomalies from equipment interference, resulting in high false alarm rates or missed critical risks. This also prevents stress concentration precursors from being detected in a timely manner, potentially increasing the probability of accidents such as water inrush and gas outbursts, thus impacting safe coal mine production. Summary of the Invention

[0003] This invention provides a method and system for detecting abnormal information in a coal mining face, in order to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides a method for detecting anomalies in a coal mining face, the method comprising: S1. Acquire the acoustic emission signal generated by the tunneling machine when it is working on the tunneling face, and separate the cutting tooth-coal-rock impact signal and the coal-rock mass fracture signal based on the propagation path and waveform characteristics of the acoustic emission signal. S2. Identify the wear state of the cutting teeth based on the frequency attenuation characteristics of the cutting teeth-coal-rock impact signal, and extract the dominant frequency band of coal-rock fracture based on the spectral abrupt change characteristics of the coal-rock fracture signal. S3. Based on the wear state of the cutting teeth, the energy of the dominant frequency band of coal and rock fracture is compensated, and the temporal co-evolution relationship between the compensated energy and the spectral mutation characteristics is established. S4. Based on the co-evolution relationship, identify abnormal fluctuations in energy precursors caused by stress concentration within the dominant frequency band of coal and rock fracture. S5. By integrating the spatial distribution patterns of energy precursor anomaly fluctuations and spectral mutation characteristics, comprehensive anomaly information is generated to characterize the location and hazard level of geological anomalies at the working face.

[0005] Preferably, the step of acquiring the acoustic emission signal generated by the tunneling machine during operation on the tunneling face, and separating the cutting tooth-coal-rock impact signal and the coal-rock mass fracture signal based on the propagation path and waveform characteristics of the acoustic emission signal includes: An array of acoustic emission sensors is placed at the coupling interface between the rock face of the tunneling face and the bearing seat of the tunneling machine's cutting teeth to collect raw acoustic emission signals containing longitudinal and transverse wave components. The acquisition parameters of the acoustic emission sensor array are adjusted by setting different threshold voltages and sampling frequencies to separate the direct wave and reflected wave in the original acoustic emission signal; The waveform steepness characteristics of the direct wave signal are extracted, and pulse waveforms with millisecond-level rising edges are selected as the cutting tooth-coal-rock impact signal. Identify the waveform oscillation attenuation characteristics of the reflected wave signal and select continuous waveforms with kilohertz frequency band resonance peaks as coal and rock mass fracture signals.

[0006] Preferably, the identification of the cutter wear state based on the frequency attenuation characteristics of the cutter-coal-rock impact signal includes: The impact signal of the cutting tooth-coal-rock is input into a set of bandpass filters with different center frequencies to separate the energy components of the impact signal in different frequency bands. The amplitude of each energy component is measured and compared with the preset reference amplitude spectrum under the corresponding sharp cutting tooth state to generate the amplitude attenuation ratio of each frequency band. If the amplitude attenuation ratio of the high-frequency energy component exceeds a preset threshold, the cutting tooth is determined to be in a wear state, and a cutting tooth wear index is generated.

[0007] Preferably, the extraction of the dominant frequency band of coal and rock fracture based on the spectral abrupt change features of the coal and rock fracture signal includes: A fast Fourier transform is performed on the coal and rock mass fracture signal to generate an initial spectrum; The initial spectrum is passed through a notch filter to filter out the inherent harmonic interference bands introduced by the chipper-coal-rock impact signal; In the filtered spectrum, identify resonance peaks whose amplitude increases sharply per unit time and whose duration exceeds the critical duration; The frequency band where the resonance peak that meets the above conditions is located is marked as the dominant frequency band for coal and rock fracturing.

[0008] Preferably, the compensation for the energy of the dominant frequency band of coal and rock fracturing based on the wear state of the cutting teeth includes: The wear index of the cutting tooth is input into the gain controller to generate a gain compensation coefficient that is negatively correlated with the degree of wear. A programmable amplifier is used to apply a gain compensation coefficient to the original energy signal of the dominant frequency band of coal and rock fracturing, and the compensated energy signal is output. The amplitude of the compensated energy signal is maintained within the preset energy reference range.

[0009] Preferably, establishing the temporal co-evolutionary relationship between the compensated energy and spectral mutation characteristics includes: The amplitude of the compensated energy signal over time is time-aligned with the center frequency offset trajectory of the resonance peak in the spectral abrupt change characteristics. Identify the pulse peaks appearing in the amplitude curve and the step points occurring in the center frequency offset trajectory; When the time difference between the occurrence of the pulse peak and the step point is less than a preset tolerance, and the intensity changes of the two are positively correlated, it is marked as a co-evolution event. The frequency of co-evolution events per unit time is statistically analyzed to generate a co-evolution index.

[0010] Preferably, the identification of energy precursor anomalies caused by stress concentration within the dominant frequency band of coal and rock fracturing based on co-evolutionary relationships includes: Within the dominant frequency band of coal and rock fracture, an envelope detector is used to extract the amplitude envelope of the energy signal in real time, and the amplitude detection threshold is dynamically adjusted based on the co-evolution index. When the amplitude envelope exceeds the amplitude detection threshold, the high-speed data acquisition module is triggered to record the energy signal segment for that period. Time-frequency analysis of energy signal segments was performed to verify whether the compression of the resonance peak width and the blue shift of the center frequency occurred synchronously with the co-evolution event in time, and whether the compression ratio of the resonance peak width exceeded the critical compression ratio, thereby confirming the abnormal fluctuation of the energy precursor.

[0011] Preferably, the spatial distribution pattern of the fused energy precursor anomalous fluctuations and spectral abrupt change characteristics includes: The location of the abnormal fluctuations in the energy precursor is spatially superimposed with the distribution location of the resonance peaks in the spectral mutation characteristics to generate a fused spatial distribution map. Spatial clustering techniques are used to identify high-density clusters of anomalous events in a fused spatial distribution map; The spatial correlation between the geometry of high-density anomalous event clusters and the geological structural direction was verified, and anomalous areas distributed along the main geological structural lines were screened out and marked as candidate areas for geological anomalies.

[0012] Preferably, the generation of comprehensive anomaly information for characterizing the location and hazard level of geological anomalies at the working face includes: The energy precursor anomaly fluctuation amplitude pattern and spectral mutation feature evolution sequence of the geological anomaly candidate region are matched with the historical precursor patterns stored in the coal and rock instability precursor knowledge base; Based on the matching results, the geological anomaly type and hazard level corresponding to the candidate geological anomaly region are determined; The output includes comprehensive anomaly information such as the spatial location, type, and hazard level of the geological anomaly.

[0013] A system for detecting anomalies in a coal mining tunneling face, characterized in that the system comprises: To address the aforementioned problems, the present invention also provides an anomaly detection system for tunneling faces in coal mining, the system comprising: The signal acquisition module is used to acquire the acoustic emission signals generated by the tunneling machine when it is working on the tunneling face, and to separate the cutting tooth-coal-rock impact signal and the coal-rock mass fracture signal based on the propagation path and waveform characteristics of the acoustic emission signal. The feature extraction module is used to identify the wear state of the cutting teeth based on the frequency attenuation characteristics of the cutting teeth-coal-rock impact signal, and to extract the dominant frequency band of coal-rock fracture based on the spectral abrupt change features of the coal-rock fracture signal. The co-evolution module is used to compensate for the energy of the dominant frequency band of coal and rock fracture based on the wear state of the cutting teeth, and to establish the co-evolution relationship between the compensated energy and the spectral mutation characteristics in time series. Anomaly identification module is used to identify energy precursor anomalies caused by stress concentration within the dominant frequency band of coal and rock fracturing, based on co-evolutionary relationships. The results generation module is used to integrate the spatial distribution patterns of energy precursor anomaly fluctuations and spectral mutation characteristics to generate comprehensive anomaly information to characterize the location and hazard level of geological anomalies at the working face.

[0014] Beneficial effects Compared with the prior art, the present invention has the following beneficial effects: 1. By innovatively separating the cutter-coal-rock impact signal from the coal-rock fracture signal and introducing a dynamic identification and compensation mechanism for cutter wear state, the interference of equipment operating conditions on geological information detection is effectively eliminated. Based on this, the established acoustic-energy co-evolution relationship can keenly capture energy precursor anomalies caused by stress concentration that are difficult to identify using traditional methods. This enables early, accurate, and targeted detection of potential anomalies at the tunneling face, significantly improving the accuracy and timeliness of early warning.

[0015] 2. By relying on the refined analysis of signal propagation and waveform characteristics, the extraction of multi-dimensional features in the time and frequency domains, and the fusion of spatial distribution patterns, the accuracy of locating geological anomalies is effectively enhanced. At the same time, by matching with the knowledge base of coal and rock instability precursors, the types and hazard levels of anomalies are clarified, providing comprehensive and reliable safety guidance for tunneling operations and helping to ensure the safety and continuity of mining operations. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a method for detecting abnormal information in a coal mining tunneling face according to an embodiment of the present invention. Figure 2 A functional block diagram of an anomaly detection system for a coal mining tunneling face, provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0017] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0018] This application provides a method for detecting anomalies in a coal mining tunneling face. The execution entity of this method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for detecting anomalies in a coal mining tunneling face can be executed by software or hardware installed on a terminal device or a server device. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0019] Example 1, referring to Figure 1 The diagram shown is a flowchart illustrating a method for detecting anomalies in a coal mining tunneling face according to an embodiment of the present invention. In this embodiment, the method for detecting anomalies in a coal mining tunneling face includes: S1. Acquire the acoustic emission signal generated by the tunneling machine when it is working on the tunneling face, and separate the cutting tooth-coal-rock impact signal and the coal-rock mass fracture signal based on the propagation path and waveform characteristics of the acoustic emission signal. S2. Identify the wear state of the cutting teeth based on the frequency attenuation characteristics of the cutting teeth-coal-rock impact signal, and extract the dominant frequency band of coal-rock fracture based on the spectral abrupt change characteristics of the coal-rock fracture signal. S3. Based on the wear state of the cutting teeth, the energy of the dominant frequency band of coal and rock fracture is compensated, and the temporal co-evolution relationship between the compensated energy and the spectral mutation characteristics is established. S4. Based on the co-evolution relationship, identify abnormal fluctuations in energy precursors caused by stress concentration within the dominant frequency band of coal and rock fracture. S5. By integrating the spatial distribution patterns of energy precursor anomaly fluctuations and spectral mutation characteristics, comprehensive anomaly information is generated to characterize the location and hazard level of geological anomalies at the working face.

[0020] In a preferred embodiment, acquiring the acoustic emission signal generated by the tunneling machine during operation on the tunneling face, and separating the cutting tooth-coal-rock impact signal and the coal-rock mass fracture signal based on the propagation path and waveform characteristics of the acoustic emission signal includes: An array of acoustic emission sensors is placed at the coupling interface between the rock face of the tunneling face and the bearing seat of the tunneling machine's cutting teeth to collect raw acoustic emission signals containing longitudinal and transverse wave components. The acquisition parameters of the acoustic emission sensor array are adjusted by setting different threshold voltages and sampling frequencies to separate the direct wave and reflected wave in the original acoustic emission signal; The waveform steepness characteristics of the direct wave signal are extracted, and pulse waveforms with millisecond-level rising edges are selected as the cutting tooth-coal-rock impact signal. Identify the waveform oscillation attenuation characteristics of the reflected wave signal and select continuous waveforms with kilohertz frequency band resonance peaks as coal and rock mass fracture signals.

[0021] Specifically, at the coupling interface between the face rock wall and the bearing seat of the tunneling machine's cutting teeth, each sensor of the acoustic emission sensor array is tightly fitted to the interface and firmly fixed to ensure that the sensor and the interface have no gap contact, directly capturing the original acoustic emission signal containing longitudinal and transverse wave components generated by the interaction between coal and rock during tunneling machine operation.

[0022] Based on the difference in propagation characteristics between the direct wave and the reflected wave in the original acoustic emission signal, the acquisition parameters of the acoustic emission sensor array are manually adjusted. By setting a fixed threshold voltage and sampling frequency, the faster-propagating direct wave and the slower-propagating reflected wave are clearly captured by the sensor array, thereby achieving their separation.

[0023] The waveform of the separated direct wave signal is observed, its waveform steepness characteristics are extracted, the rising edge duration of the direct wave signal is identified one by one, and the pulse waveform with a rising edge duration of milliseconds is accurately screened out. This waveform is the cutting tooth-coal rock impact signal.

[0024] For the separated reflected wave signal, observe the oscillation attenuation law of its waveform, identify the resonance peak frequency band in the waveform, and specifically screen out the continuous waveform with resonance peak in the kilohertz frequency band. This waveform is the coal and rock mass fracture signal.

[0025] In summary, by deploying sensor arrays at key locations, the system comprehensively collects raw acoustic emission signals containing both longitudinal and transverse waves, ensuring signal integrity. By adjusting the acquisition parameters to separate direct and reflected waves, and then precisely filtering the two types of signals based on waveform characteristics, the system effectively solves the problem of overlapping frequency bands between mining machinery noise and coal and rock fracture signals, achieving decoupling between equipment operating signals and geological response signals. This provides a clean data foundation for subsequent identification of cutter wear status and extraction of the dominant frequency band of coal and rock fracture, reducing misjudgments caused by interference, improving the accuracy of subsequent anomaly detection, and laying a solid foundation for early detection of precursor signals of geological hazards.

[0026] In a preferred embodiment, identifying the wear state of the cutting tooth based on the frequency attenuation characteristics of the cutting tooth-coal-rock impact signal includes: The impact signal of the cutting tooth-coal-rock is input into a set of bandpass filters with different center frequencies to separate the energy components of the impact signal in different frequency bands. The amplitude of each energy component is measured and compared with the preset reference amplitude spectrum under the corresponding sharp cutting tooth state to generate the amplitude attenuation ratio of each frequency band. If the amplitude attenuation ratio of the high-frequency energy component exceeds a preset threshold, the cutting tooth is determined to be in a wear state, and a cutting tooth wear index is generated.

[0027] Specifically, the impact signal of the cutting tooth-coal-rock is fed into a set of bandpass filters with different center frequencies one by one. Each filter only allows the signal in the frequency band corresponding to the center frequency to pass through, while the signals in other frequency bands are blocked, thereby separating the independent energy components of the impact signal in different frequency bands.

[0028] The actual amplitude of the energy components of each frequency band is directly read using an amplitude measurement device. The actual amplitude of each frequency band is compared one by one with the reference amplitude spectrum of the corresponding frequency band under the pre-determined sharpness of the cutting teeth. The amplitude attenuation ratio of each frequency band is obtained through the difference between the actual amplitude and the reference amplitude.

[0029] The amplitude attenuation ratio corresponding to the separated high-frequency energy component is selected and directly compared with the preset judgment threshold. When the attenuation ratio exceeds the threshold, the cutting tooth is directly judged to have entered the wear state. The cutting tooth wear index is generated based on the specific value of the amplitude attenuation ratio of the high-frequency band.

[0030] In this embodiment of the invention, the dominant frequency band of coal and rock fracture is extracted based on the spectral abrupt change characteristics of the coal and rock fracture signal, including: A fast Fourier transform is performed on the coal and rock mass fracture signal to generate an initial spectrum; The initial spectrum is passed through a notch filter to filter out the inherent harmonic interference bands introduced by the chipper-coal-rock impact signal; In the filtered spectrum, identify resonance peaks whose amplitude increases sharply per unit time and whose duration exceeds the critical duration; The frequency band where the resonance peak that meets the above conditions is located is marked as the dominant frequency band for coal and rock fracturing.

[0031] Specifically, the time-domain waveform of the coal and rock mass fracture signal is converted into frequency-domain data according to the frequency decomposition law. By corresponding the time-domain signal and frequency components one by one, the initial spectrum containing each frequency and corresponding amplitude is obtained. The conversion effect is verified by observing whether the spectrum completely presents the signal frequency distribution.

[0032] The notch filter is aligned with the inherent harmonic frequency range of the cutter-coal-rock impact signal, blocking only signals within this range while allowing other frequencies of coal-rock fracture signals to transmit normally. By comparing the amplitude of the interference band in the spectrum before and after filtering, it is confirmed that the interference has been effectively filtered out.

[0033] The amplitude changes of each resonance peak in the filtered spectrum are characterized one by one. The increase in amplitude and duration per unit time are recorded. Resonance peaks with rapid amplitude increases and durations exceeding the critical duration are selected. The change data of the resonance peak are repeatedly checked to ensure accurate identification.

[0034] Identify the frequency range covered by the resonance peaks that meet the conditions, and directly mark this range as the dominant frequency band for coal and rock fracturing. Verify the marking results by confirming that the resonance peaks in this frequency band are the main energy concentration areas in the spectrum.

[0035] In summary, by analyzing the frequency attenuation characteristics of impact signals and generating a wear index by comparing with a reference amplitude spectrum, the working condition of the cutting teeth can be accurately monitored in real time. This avoids changes in signal energy distribution due to cutting tooth wear, eliminates interference from equipment status on geological information detection, and provides a basis for subsequent energy compensation.

[0036] First, harmonic interference from the impact of the cutting teeth is filtered out, and then the resonance peak frequency band with a sudden increase in amplitude and a sustained supercritical duration is locked, which can purify the core signal of coal and rock fracture. Combining these two methods to decouple the equipment from the geological signal lays a reliable data foundation for subsequent identification of abnormal fluctuations and improves the accuracy of geological anomaly detection.

[0037] In a preferred embodiment, energy compensation for the dominant frequency band of coal and rock fracturing based on the wear state of the cutting teeth includes: The wear index of the cutting tooth is input into the gain controller to generate a gain compensation coefficient that is negatively correlated with the degree of wear. A programmable amplifier is used to apply a gain compensation coefficient to the original energy signal of the dominant frequency band of coal and rock fracturing, and the compensated energy signal is output. The amplitude of the compensated energy signal is maintained within the preset energy reference range.

[0038] Specifically, the previously generated cutter wear index is directly input into the gain controller. The controller responds to the index size through its internal mechanical adjustment mechanism. The larger the cutter wear index, the larger the output gain compensation coefficient. By reading the controller's value display, it is confirmed that the coefficient is negatively correlated with the degree of cutter wear.

[0039] The original energy signal of the dominant frequency band of coal and rock fracturing is input into a programmable amplifier. By manually adjusting the control knob of the amplifier, the original signal is amplified according to the generated gain compensation coefficient. By comparing the signal amplitude before and after the amplifier input, it is confirmed that the gain compensation coefficient has been applied and the compensated energy signal is output.

[0040] The amplitude of the compensated energy signal is monitored in real time. If it is lower than the preset energy reference range, the amplifier is fine-tuned to increase the compensation coefficient. If it is higher, the compensation coefficient is decreased. By continuously observing the amplitude monitoring instrument, the amplitude of the compensated energy signal is ensured to be stable within the preset energy reference range.

[0041] In this embodiment of the invention, establishing the temporal co-evolutionary relationship between compensated energy and spectral abrupt change characteristics includes: The amplitude of the compensated energy signal over time is time-aligned with the center frequency offset trajectory of the resonance peak in the spectral abrupt change characteristics. Identify the pulse peaks appearing in the amplitude curve and the step points occurring in the center frequency offset trajectory; When the time difference between the occurrence of the pulse peak and the step point is less than a preset tolerance, and the intensity changes of the two are positively correlated, it is marked as a co-evolution event. The frequency of co-evolution events per unit time is statistically analyzed to generate a co-evolution index.

[0042] Specifically, using a unified time axis as a reference, the amplitude variation curve of the compensated energy signal over time is calibrated moment by moment with the center frequency offset trajectory of the resonance peak in the spectral mutation characteristics. At each time node, the data of the two are ensured to be completely matched. The timing alignment effect is verified by viewing the paired data of each node on the time axis.

[0043] Observe the amplitude change curve segment by segment, mark the part of the curve that suddenly rises and falls rapidly as pulse peak, and at the same time check the center frequency offset trajectory, identify the point in the trajectory that suddenly jumps and remains stable as step point, and confirm the identification result is correct by repeatedly reviewing the curve and trajectory segment.

[0044] Record the occurrence time of each pulse peak and step point, calculate the time difference between them, and observe the trend of pulse peak amplitude change and step point offset intensity change. When the time difference is less than the preset tolerance and both show a positive correlation of synchronous increase or decrease, mark the corresponding relationship as a co-evolution event. Verify the rationality of the marking by reviewing the time difference and the trend of change.

[0045] Set a fixed time interval, count the total number of co-evolutionary events marked within that interval, and directly convert that number into a co-evolutionary index. By verifying the division of the time interval and the number of events counted, ensure that the index accurately reflects the frequency of event occurrence.

[0046] In a preferred embodiment, the co-evolution index is generated by calculating the degree of co-evolution of each detected pulse peak and step point combination.

[0047] Specifically, an event coordination degree is calculated for each pair of pulse peaks and step points that occur at similar times, and the calculation formula is as follows:

[0048] in, This represents the time difference between the pulse peak and the step point at the moment of occurrence, and it directly corresponds to the determination that "the time difference is less than the preset tolerance". Indicates the preset maximum time tolerance. At that time, it was considered that the two were unrelated, and the degree of synergy was 0.

[0049] This indicates the magnitude of the pulse peak change. This represents the center frequency offset at the step point. This is a normalization constant used to keep the calculation results within a reasonable numerical range; its value can be determined based on typical historical data. and The product is used to determine this.

[0050] Subsequently, the co-evolution index By unit time window Internally, the degree of coordination for all events It is generated by accumulation, that is:

[0051] This calculation process is essentially a superior and more effective quantitative method for "statistically calculating the frequency of co-evolutionary events per unit time and generating a co-evolutionary index." It not only counts the number of events but also comprehensively considers the intensity and quality of each event, enabling the generated co-evolutionary index to more accurately and sensitively reflect the abnormal state of the tunneling face.

[0052] It should be noted that in the formula, the time difference comes from the difference between the occurrence time of the pulse peak marked in the amplitude change curve and the step point determined in the center frequency offset trajectory, the amplitude change comes from the actual amplitude fluctuation value of the pulse peak, the center frequency offset comes from the actual center frequency offset value of the step point, the maximum time tolerance is a pre-set time standard for determining whether the two are related, and the normalization constant comes from the product of the typical amplitude change and the center frequency offset in historical data.

[0053] This calculation is used to quantify the degree of coordination between each pair of pulse peaks and step points that are close in time. The degree of coordination is calculated only when the time difference is less than the maximum time tolerance and the amplitude change and center frequency offset are both positive, by combining the degree of time proximity with the product of the intensity changes of the two. Otherwise, the degree of coordination is 0. The cumulative result of all degrees of coordination per unit time is the co-evolution index, which realizes the comprehensive quantification of the frequency, intensity and quality of co-evolution events.

[0054] The smaller the time difference and the greater the amplitude change and the offset of the center frequency, the higher the event coordination degree. Conversely, the coordination degree is lower. When the conditions are not met, the coordination degree remains at 0. The more events with high coordination degree within a unit time window, the higher the coordination degree of a single event, the larger the coordination evolution index, and the more sensitively it can reflect the abnormal state of the tunneling face.

[0055] In summary, by generating a negative correlation gain coefficient based on the cutter wear index and maintaining the energy of the dominant frequency band of coal and rock fracture within a preset reference range through a programmable amplifier, signal energy distortion caused by cutter wear can be eliminated, and the interference of equipment operating condition changes on the true energy interpretation of geological signals can be avoided, thus ensuring the effectiveness of the energy characteristics of coal and rock fracture signals.

[0056] After time-series alignment compensation, the energy amplitude curve and the spectral resonance peak shift trajectory are identified, synchronous pulse peaks and step points are identified, and a co-evolution index is generated. This can establish a dynamic correlation between energy and spectral characteristics, break through the limitations of single-parameter analysis, and accurately reflect the inherent laws of coal and rock fracturing process.

[0057] The combination of these two methods further decouples the equipment from the geological signals, providing a reliable time-series correlation basis for the subsequent accurate identification of energy precursor anomalies caused by stress concentration, reducing misjudgments caused by signal interference or parameter isolation, and improving the accuracy of geological anomaly detection.

[0058] In a preferred embodiment, based on the co-evolutionary relationship, identifying energy precursor anomalies caused by stress concentration within the dominant frequency band of coal and rock fracturing includes: Within the dominant frequency band of coal and rock fracture, an envelope detector is used to extract the amplitude envelope of the energy signal in real time, and the amplitude detection threshold is dynamically adjusted based on the co-evolution index. When the amplitude envelope exceeds the amplitude detection threshold, the high-speed data acquisition module is triggered to record the energy signal segment for that period. Time-frequency analysis of energy signal segments was performed to verify whether the compression of the resonance peak width and the blue shift of the center frequency occurred synchronously with the co-evolution event in time, and whether the compression ratio of the resonance peak width exceeded the critical compression ratio, thereby confirming the abnormal fluctuation of the energy precursor.

[0059] Specifically, an envelope detector is connected to the energy signal transmission line of the dominant frequency band of coal and rock fracturing. The detector generates a continuous amplitude envelope by tracking the peak trajectory of the signal amplitude. At the same time, the co-evolution index is connected to the threshold adjustment component. When the index increases, the amplitude detection threshold is adjusted upwards in sync. By observing the correspondence between the envelope and the threshold in real time, it is verified whether the adjustment changes dynamically with the index.

[0060] The amplitude envelope is compared with the dynamic threshold in real time. When the envelope exceeds the threshold, the high-speed data acquisition module is immediately triggered to accurately record the energy signal segment within that period. The accuracy of the signal segment recording is verified by checking the consistency between the acquisition start time and the threshold trigger time.

[0061] Analyze energy signal segments based on the correlation between time and frequency, observe whether the resonance peak width shrinks and whether the center frequency shifts to higher frequencies, confirm that these two phenomena are completely synchronized with the occurrence time of the co-evolution event, and that the resonance peak width compression reaches the critical compression ratio. If all conditions are met, the abnormal fluctuation of the energy precursor is confirmed, and the reliability of the results is verified by repeating the time-frequency analysis process.

[0062] In summary, this scheme relies on the co-evolution index to dynamically adjust the amplitude detection threshold, avoiding the false alarms or missed alarms that are prone to occur with fixed thresholds. It allows the threshold to adapt to the real-time changes in coal and rock fracture signals, thereby improving the targeting of energy anomaly fluctuation identification.

[0063] When the amplitude envelope exceeds the threshold, high-speed acquisition is triggered, which can completely record the energy signal segment of the corresponding time period, avoid the loss of abnormal signals, and provide complete data support for subsequent verification.

[0064] Time-frequency analysis was used to verify the synchronicity of resonance peak width compression, center frequency blue shift, and co-evolution events. The compression ratio was confirmed to meet the standard, which can eliminate equipment interference, accurately identify abnormal fluctuations in energy precursors caused by stress concentration, and achieve accurate capture of early precursors of geological hazards, laying a solid foundation for subsequent risk warning.

[0065] In a preferred embodiment, the spatial distribution pattern that integrates energy precursor anomalous fluctuations and spectral abrupt changes includes: The location of the abnormal fluctuations in the energy precursor is spatially superimposed with the distribution location of the resonance peaks in the spectral mutation characteristics to generate a fused spatial distribution map. Spatial clustering techniques are used to identify high-density clusters of anomalous events in a fused spatial distribution map; The spatial correlation between the geometry of high-density anomalous event clusters and the geological structural direction was verified, and anomalous areas distributed along the main geological structural lines were screened out and marked as candidate areas for geological anomalies.

[0066] Using the three-dimensional spatial coordinate system of the tunneling face as a unified benchmark, the coordinates of each occurrence location of the energy precursor abnormal fluctuation are matched one by one with the corresponding distribution location coordinates of the resonance peak in the spectrum mutation characteristics and marked on the same spatial map to form a fused spatial distribution map. The integrity of the superposition is verified by checking whether each location in the map simultaneously presents two feature markers.

[0067] The number of anomalous events in each region of the fused spatial distribution map was manually counted. Areas with significantly more events than the surrounding areas were designated as high-density anomalous event clusters. By repeatedly counting the number of event points in each region, it was confirmed that the event density in the clusters was indeed higher than in other regions.

[0068] The boundary contours of high-density anomaly clusters are visually compared with known geological structural directions to observe whether the clusters extend along the main geological structural lines. Anomaly areas that perfectly match are selected and marked as candidate areas for geological anomalies. The accuracy of the marking is verified by checking the consistency of the extension direction of the structural lines and the candidate areas.

[0069] In this embodiment of the invention, generating comprehensive anomaly information to characterize the location and hazard level of geological anomalies at the working face includes: The energy precursor anomaly fluctuation amplitude pattern and spectral mutation feature evolution sequence of the geological anomaly candidate region are matched with the historical precursor patterns stored in the coal and rock instability precursor knowledge base; Based on the matching results, the geological anomaly type and hazard level corresponding to the candidate geological anomaly region are determined; The output includes comprehensive anomaly information such as the spatial location, type, and hazard level of the geological anomaly.

[0070] Specifically, various historical precursor patterns stored in the coal and rock instability precursor knowledge base are extracted, and the energy precursor abnormal fluctuation amplitude patterns and spectral mutation feature evolution sequences of the geological anomaly candidate areas are compared with the historical precursor patterns one by one. The consistency of amplitude change patterns and feature evolution order is checked point by point. By confirming that the core features of the two are completely consistent, the effectiveness of the matching results is verified.

[0071] Based on the matching results of the previous step, if the current feature completely matches the historical precursor pattern corresponding to the fault fracture zone in the knowledge base, the geological anomaly type is determined to be a fault fracture zone. At the same time, the hazard level of the current area is determined according to the hazard level marked by the historical pattern. The accuracy of the judgment result is verified by reviewing the correspondence between the pattern and the type and level in the knowledge base.

[0072] Organize the three-dimensional spatial coordinates of candidate geological anomaly areas, the identified geological anomaly types and hazard levels, summarize them into comprehensive anomaly information in a unified format and output them. By checking whether the output information contains all three core contents, ensure that the comprehensive anomaly information fully represents the geological anomaly situation of the working face.

[0073] In summary, by spatially superimposing the distribution locations of energy precursor anomaly fluctuations and spectral mutation characteristics, and combining this with spatial clustering to identify high-density anomaly areas, we can focus on core anomaly regions, avoid being misled by scattered interference signals, significantly improve the accuracy of locating geological anomalies, and solve the problems of low spatial coverage and ambiguous positioning in traditional methods.

[0074] By verifying the geometry and geological structure of high-density areas, anomaly zones distributed along tectonic lines are selected, and false signals caused by non-geological factors are eliminated to ensure that candidate areas are genuine geological anomalies.

[0075] By matching the knowledge base of coal and rock instability precursors, the types and hazard levels of abnormal bodies are determined, and comprehensive information including location, type, and hazard level is generated. This provides accurate safety guidance for tunneling operations, helps to avoid risks such as water inrush and gas outburst in advance, and ensures mining safety and efficiency.

[0076] Example 2, as Figure 2 The diagram shown is a functional block diagram of a reference information generation system based on artificial intelligence and smart home provided in an embodiment of the present invention.

[0077] This invention discloses an anomaly detection system 100 for a coal mining tunneling face, which can be installed in an electronic device. Depending on the functions implemented, the system 100 may include a signal acquisition module 101, a feature extraction module 102, a collaborative evolution module 103, an anomaly identification module 104, and a result generation module 105. The module in this invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and perform a fixed function, stored in the memory of the electronic device.

[0078] In this embodiment, the functions of each module / unit are as follows: The signal acquisition module 101 is used to acquire the acoustic emission signal generated by the tunneling machine when it is working on the tunneling face, and to separate the cutting tooth-coal-rock impact signal and the coal-rock mass fracture signal based on the propagation path and waveform characteristics of the acoustic emission signal. The feature extraction module 102 is used to identify the wear state of the cutting tooth based on the frequency attenuation characteristics of the cutting tooth-coal-rock impact signal, and to extract the dominant frequency band of coal-rock fracture based on the spectral abrupt change features of the coal-rock fracture signal. The co-evolution module 103 is used to compensate for the energy of the dominant frequency band of coal and rock fracture based on the wear state of the cutting teeth, and to establish the co-evolution relationship between the compensated energy and the spectral mutation characteristics in time series. Anomaly identification module 104 is used to identify energy precursor anomalies caused by stress concentration within the dominant frequency band of coal and rock fracture based on the co-evolution relationship. The result generation module 105 is used to integrate the spatial distribution patterns of energy precursor anomaly fluctuations and spectral mutation characteristics to generate comprehensive anomaly information to characterize the location and hazard level of geological anomalies at the working face.

[0079] In the several embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the system embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0080] The modules described as separate components may or may not be physically separate. The components shown as modules 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.

[0081] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0082] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0083] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0084] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for detecting abnormal information of a tunneling face based on coal mining, characterized in that, The method includes: S1. Acquire the acoustic emission signal generated by the tunneling machine when it is working on the tunneling face, and separate the cutting tooth-coal-rock impact signal and the coal-rock mass fracture signal based on the propagation path and waveform characteristics of the acoustic emission signal. S2. Identify the wear state of the cutting teeth based on the frequency attenuation characteristics of the cutting teeth-coal-rock impact signal, and extract the dominant frequency band of coal-rock fracture based on the spectral abrupt change characteristics of the coal-rock fracture signal. S3. Based on the wear state of the cutting teeth, the energy of the dominant frequency band of coal and rock fracture is compensated, and the temporal co-evolution relationship between the compensated energy and the spectral mutation characteristics is established. S4. Based on the co-evolution relationship, identify the abnormal fluctuations in energy precursors caused by stress concentration within the dominant frequency band of coal and rock fracture; S5. By integrating the spatial distribution patterns of energy precursor anomaly fluctuations and spectral mutation characteristics, comprehensive anomaly information is generated to characterize the location and hazard level of geological anomalies at the working face. The energy compensation for the dominant frequency band of coal and rock fracturing based on the wear state of the cutting teeth includes: The wear index of the cutting tooth is input into the gain controller to generate a gain compensation coefficient that is negatively correlated with the degree of wear. A programmable amplifier is used to apply a gain compensation coefficient to the original energy signal of the dominant frequency band of coal and rock fracturing, and the compensated energy signal is output. The amplitude of the compensated energy signal is maintained within the preset energy reference range; The establishment of the temporal co-evolutionary relationship between compensated energy and spectral abrupt change characteristics includes: The amplitude of the compensated energy signal over time is time-aligned with the center frequency offset trajectory of the resonance peak in the spectral abrupt change characteristics. Identify the pulse peaks appearing in the amplitude curve and the step points occurring in the center frequency offset trajectory; When the time difference between the occurrence of the pulse peak and the step point is less than a preset tolerance, and the intensity changes of the two are positively correlated, it is marked as a co-evolution event. The frequency of co-evolution events per unit time is counted to generate a co-evolution index; The method of identifying energy precursor anomalies caused by stress concentration within the dominant frequency band of coal and rock fracturing based on co-evolutionary relationships includes: Within the dominant frequency band of coal and rock fracture, an envelope detector is used to extract the amplitude envelope of the energy signal in real time, and the amplitude detection threshold is dynamically adjusted based on the co-evolution index. When the amplitude envelope exceeds the amplitude detection threshold, the high-speed data acquisition module is triggered to record the energy signal segment for that period. Time-frequency analysis of energy signal segments was performed to verify whether the compression of the resonance peak width and the blue shift of the center frequency occurred synchronously with the co-evolution event in time, and whether the compression ratio of the resonance peak width exceeded the critical compression ratio, thereby confirming the abnormal fluctuation of the energy precursor.

2. A method of detecting abnormal information at a coal mining-based heading face according to claim 1, characterized in that, The acquisition of acoustic emission signals generated by the tunneling machine during its operation on the tunneling face, and the separation of the cutting tooth-coal-rock impact signal and the coal-rock mass fracture signal based on the propagation path and waveform characteristics of the acoustic emission signals, includes: An array of acoustic emission sensors is placed at the coupling interface between the rock face of the tunneling face and the bearing seat of the tunneling machine's cutting teeth to collect raw acoustic emission signals containing longitudinal and transverse wave components. The acquisition parameters of the acoustic emission sensor array are adjusted by setting different threshold voltages and sampling frequencies to separate the direct wave and reflected wave in the original acoustic emission signal; The waveform steepness characteristics of the direct wave signal are extracted, and pulse waveforms with millisecond-level rising edges are selected as the cutting tooth-coal-rock impact signal. Identify the waveform oscillation attenuation characteristics of the reflected wave signal and select continuous waveforms with kilohertz frequency band resonance peaks as coal and rock mass fracture signals.

3. A method of detecting abnormal information at a coal mining-based heading face according to claim 1, characterized in that, The method for identifying the wear state of the cutting teeth based on the frequency attenuation characteristics of the cutting tooth-coal-rock impact signal includes: The impact signal of the cutting tooth-coal-rock is input into a set of bandpass filters with different center frequencies to separate the energy components of the impact signal in different frequency bands. The amplitude of each energy component is measured and compared with the preset reference amplitude spectrum under the corresponding sharp cutting tooth state to generate the amplitude attenuation ratio of each frequency band. If the amplitude attenuation ratio of the high-frequency energy component exceeds a preset threshold, the cutting tooth is determined to be in a wear state, and a cutting tooth wear index is generated.

4. A method of detecting abnormal information at a coal mining based excavation face according to claim 3, characterized in that, The extraction of the dominant frequency band of coal and rock fracture based on the spectral abrupt change features of coal and rock fracture signals includes: A fast Fourier transform is performed on the coal and rock mass fracture signal to generate an initial spectrum; The initial spectrum is passed through a notch filter to filter out the inherent harmonic interference bands introduced by the chipper-coal-rock impact signal; In the filtered spectrum, identify resonance peaks whose amplitude increases sharply per unit time and whose duration exceeds the critical duration; The frequency band where the resonance peak that meets the above conditions is located is marked as the dominant frequency band for coal and rock fracturing.

5. The method for detecting abnormal information in a tunneling face based on coal mining as described in claim 1, characterized in that, The spatial distribution pattern of the fusion energy precursor anomalous fluctuations and spectral mutation characteristics includes: The location of the abnormal fluctuations in the energy precursor is spatially superimposed with the distribution location of the resonance peaks in the spectral mutation characteristics to generate a fused spatial distribution map. Spatial clustering techniques are used to identify high-density clusters of anomalous events in a fused spatial distribution map; The spatial correlation between the geometry of high-density anomalous event clusters and the geological structural direction was verified, and anomalous areas distributed along the main geological structural lines were screened out and marked as candidate areas for geological anomalies.

6. The method for detecting abnormal information in a tunneling face based on coal mining as described in claim 1, characterized in that, The generation of comprehensive anomaly information to characterize the location and hazard level of geological anomalies at the working face includes: The energy precursor anomaly fluctuation amplitude pattern and spectral mutation feature evolution sequence of the geological anomaly candidate region are matched with the historical precursor patterns stored in the coal and rock instability precursor knowledge base; Based on the matching results, the geological anomaly type and hazard level corresponding to the candidate geological anomaly region are determined; The output includes comprehensive anomaly information such as the spatial location, type, and hazard level of the geological anomaly.

7. A system for detecting abnormal information in a coal mining tunneling face, characterized in that, The system, applicable to the method for detecting anomalies in a coal mining face as described in any one of claims 1-6, comprises: The signal acquisition module is used to acquire the acoustic emission signals generated by the tunneling machine when it is working on the tunneling face, and to separate the cutting tooth-coal-rock impact signal and the coal-rock mass fracture signal based on the propagation path and waveform characteristics of the acoustic emission signal. The feature extraction module is used to identify the wear state of the cutting teeth based on the frequency attenuation characteristics of the cutting teeth-coal-rock impact signal, and to extract the dominant frequency band of coal-rock fracture based on the spectral abrupt change features of the coal-rock fracture signal. The co-evolution module is used to compensate for the energy of the dominant frequency band of coal and rock fracture based on the wear state of the cutting teeth, and to establish the co-evolution relationship between the compensated energy and the spectral mutation characteristics in time series. Anomaly identification module is used to identify energy precursor anomalies caused by stress concentration within the dominant frequency band of coal and rock fracturing, based on co-evolutionary relationships. The results generation module is used to fuse the spatial distribution patterns of energy precursor anomaly fluctuations and spectral mutation characteristics to generate comprehensive anomaly information to characterize the location and hazard level of geological anomalies at the working face. The energy compensation for the dominant frequency band of coal and rock fracturing based on the wear state of the cutting teeth includes: The wear index of the cutting tooth is input into the gain controller to generate a gain compensation coefficient that is negatively correlated with the degree of wear. A programmable amplifier is used to apply a gain compensation coefficient to the original energy signal of the dominant frequency band of coal and rock fracturing, and the compensated energy signal is output. The amplitude of the compensated energy signal is maintained within the preset energy reference range; The establishment of the temporal co-evolutionary relationship between compensated energy and spectral abrupt change characteristics includes: The amplitude of the compensated energy signal over time is time-aligned with the center frequency offset trajectory of the resonance peak in the spectral abrupt change characteristics. Identify the pulse peaks appearing in the amplitude curve and the step points occurring in the center frequency offset trajectory; When the time difference between the occurrence of the pulse peak and the step point is less than a preset tolerance, and the intensity changes of the two are positively correlated, it is marked as a co-evolution event. The frequency of co-evolution events per unit time is counted to generate a co-evolution index; The method of identifying energy precursor anomalies caused by stress concentration within the dominant frequency band of coal and rock fracturing based on co-evolutionary relationships includes: Within the dominant frequency band of coal and rock fracture, an envelope detector is used to extract the amplitude envelope of the energy signal in real time, and the amplitude detection threshold is dynamically adjusted based on the co-evolution index. When the amplitude envelope exceeds the amplitude detection threshold, the high-speed data acquisition module is triggered to record the energy signal segment for that period. Time-frequency analysis of energy signal segments was performed to verify whether the compression of the resonance peak width and the blue shift of the center frequency occurred synchronously with the co-evolution event in time, and whether the compression ratio of the resonance peak width exceeded the critical compression ratio, thereby confirming the abnormal fluctuation of the energy precursor.

Images