A mine chute state automatic monitoring method based on acoustic and vibration coupling

Abstract

The application discloses a mine chute state automatic monitoring method based on acoustic and vibration coupling, and relates to the technical field of intelligent sensing, which comprises the following steps: collecting acoustic signals and shaft wall vibration signals in a shaft, and performing anti-aliasing processing, band-pass filtering and detrending; dividing the acoustic signals and the shaft wall vibration signals into acoustic analysis segments and vibration analysis segments through an event triggering mode; performing correlation analysis on the acoustic analysis segments and the vibration analysis segments in corresponding frequency bands and time windows, extracting acoustic-vibration coupling features, combining and mapping the acoustic-vibration coupling features, and forming acoustic-vibration coupling fingerprint data; calculating fingerprint drift amounts of the acoustic-vibration coupling fingerprint data and normal working condition baseline fingerprint data, and performing trend accumulation on continuous fingerprint drift amounts to form an acoustic-vibration coupling drift feature sequence; and the application realizes automatic identification and continuous monitoring of the chute operation condition, and improves the reliability of working condition determination.

Description

Technical Field

[0001] This invention relates to the field of intelligent sensing technology, and in particular to an automated monitoring method for the status of mine ore passes based on acoustic and vibration coupling. Background Technology

[0002] In underground gold mining, the pass, as a crucial infrastructure connecting the mining area and the transportation system, directly impacts ore transport efficiency and underground operational safety. As gold mines expand to deeper and larger areas, the underground environment exhibits characteristics such as high noise, high impact, and multi-physics coupling, making traditional manual inspections or single-point monitoring methods insufficient for refined management. Against this backdrop, pass operation monitoring technology based on intelligent sensing systems has gradually been applied and developed. Existing technologies typically involve deploying acoustic and vibration sensors inside or on the shaft wall to collect acoustic signals and vibration signals from the shaft wall. Signal preprocessing methods such as anti-aliasing processing, bandpass filtering, and detrending are then used to extract characteristic parameters reflecting ore flow, impact behavior, and structural response.

[0003] Existing technologies often focus on independent analysis of single signals or simple comparisons of multi-source signals, lacking a systematic correlation analysis mechanism within corresponding frequency bands and time windows. This makes it difficult to fully reflect the intrinsic relationship between acoustic behavior and structural vibration. Furthermore, changes in the operating state of well chutes often exhibit cumulative and gradual characteristics, making it difficult to reveal the evolution trend of operating conditions from a single analysis. Current methods are still imperfect in the organization and mapping of acoustic and vibration characteristics, as well as in cross-timescale stability analysis, resulting in a need for further improvement in the ability to continuously discriminate and finely identify operating conditions. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides an automated monitoring method for the status of mine ore passes based on acoustic and vibration coupling, which solves the problems of insufficient correlation between acoustic and vibration multi-source signals and difficulty in characterizing the long-term evolution of the operating status of mine ore passes in the prior art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: This invention provides an automated monitoring method for the condition of mine ore passes based on acoustic and vibration coupling. The method includes: acquiring acoustic signals from inside the mine and vibration signals from the mine wall, performing anti-aliasing processing, bandpass filtering, and detrending; segmenting the signals into acoustic analysis segments and vibration analysis segments via event triggering; performing correlation analysis on the acoustic and vibration analysis segments within corresponding frequency bands and time windows to extract acoustic-vibration coupling features, and combining and mapping these features to form acoustic-vibration coupling fingerprint data; calculating the fingerprint drift between the acoustic-vibration coupling fingerprint data and the baseline fingerprint data under normal operating conditions, and accumulating the continuous fingerprint drift to form an acoustic-vibration coupling drift feature sequence; calculating the acoustic-vibration consistency index between the acoustic and vibration analysis segments; and when the acoustic-vibration consistency index meets the validity criteria, performing multi-timescale correlation reconstruction and coupling stability evolution on the acoustic-vibration coupling drift feature sequence and the acoustic-vibration coupling fingerprint data to obtain the operating conditions of the mine ore pass.

[0007] As a preferred embodiment of the automated monitoring method for mine ore pass status based on acoustic and vibration coupling described in this invention, the specific steps for acquiring acoustic signals from inside the mine and vibration signals from the mine wall, and performing anti-aliasing processing, bandpass filtering, and detrending are as follows. Collect acoustic signals generated inside the well pass and vibration signals generated by the well wall; Anti-aliasing processing is used to suppress the aliasing interference between acoustic signals and wellbore vibration signals in the well, and bandpass filtering is used to retain the effective signal components related to the well pass operation status, outputting filtered acoustic signals and filtered vibration signals. The slowly varying drift components in the filtered acoustic and vibration signals are removed by linear trend term elimination to obtain the detrended acoustic and vibration signals.

[0008] As a preferred embodiment of the automated monitoring method for mine ore pass status based on acoustic and vibration coupling described in this invention, the step of segmenting the data into acoustic analysis segments and vibration analysis segments via event triggering is as follows: Based on the detrended acoustic signal and the detrended vibration signal, the signal amplitude is calculated in a continuous time window. When the signal amplitude exceeds the event trigger threshold, the corresponding time is determined as the trigger time of the well pass operation event. Centered on the trigger time of the well operation event, the acoustic signal and vibration signal after detrended are extracted according to the preset time window to obtain acoustic analysis segments and vibration analysis segments.

[0009] As a preferred embodiment of the automated monitoring method for mine chute conditions based on acoustic-vibration coupling described in this invention, the specific steps for performing correlation analysis on acoustic analysis segments and vibration analysis segments within corresponding frequency bands and time windows to extract acoustic-vibration coupling features are as follows. The acoustic analysis segment and the vibration analysis segment are time-aligned and divided into continuous time windows. Continuous wavelet transform is performed on the acoustic analysis segment and the vibration analysis segment in each time window to obtain the initial time-frequency representation. The dispersed time-frequency energy in the initial time-frequency representation is re-aggregated along the true frequency trajectory to obtain a high-resolution time-frequency representation. The high-resolution time-frequency representation is then compressed and reconstructed through a synchronous squeezing operation to form an acoustic time-frequency diagram and a vibration time-frequency diagram. The acoustic time-frequency diagram and the vibration time-frequency diagram are represented in two-dimensional coordinate form to form a time-frequency plane, and the time-frequency plane is divided into multiple frequency bands and time windows to form a two-dimensional time-frequency grid. The time-frequency coefficients corresponding to the acoustic time-frequency map and vibration time-frequency map are extracted from the two-dimensional time-frequency grid, and the correlation analysis between the acoustic signal in the well and the vibration signal of the well wall is performed to output the acoustic-vibration correlation information. The acoustic-vibration coupling characteristics are formed by collection and characterization.

[0010] As a preferred embodiment of the automated monitoring method for mine chute conditions based on acoustic-vibration coupling described in this invention, the steps for combining and mapping acoustic-vibration coupling features according to feature organization rules to form acoustic-vibration coupling fingerprint data are as follows: The acoustic-vibration coupling features are grouped according to a three-layer index structure of frequency band, time window, and coupling, and then spliced ​​together in the order of time window within each group to form a hierarchical coupling feature table; The hierarchical coupling feature table is mapped to discrete symbols through a fixed codebook and output in a fixed order according to the three-layer index structure of frequency band-time window-coupling to form acoustic-vibration coupling fingerprint data.

[0011] As a preferred embodiment of the automated monitoring method for mine chute status based on acoustic and vibration coupling described in this invention, the frequency band-time window-coupling three-layer index structure is formed by assigning frequency band indexes to corresponding regions according to the frequency band intervals divided in the time-frequency plane, assigning time window indexes to corresponding regions according to the time window, and identifying acoustic-vibration correlation information as coupling indexes.

[0012] As a preferred embodiment of the automated monitoring method for mine ore pass conditions based on acoustic and vibration coupling described in this invention, the specific steps for calculating the fingerprint drift between the acoustic-vibration coupling fingerprint data and the baseline fingerprint data under normal operating conditions are as follows: The acoustic-vibration coupling fingerprint data under normal operating conditions during historical operation are collected to form the normal operating condition baseline fingerprint data. Using a three-layer index structure of frequency band, time window, and coupling as the matching key, the acoustic-vibration coupled fingerprint data is matched item by item with the baseline fingerprint data under normal operating conditions to form fingerprint pairs; Based on the fingerprint pairs, the differences between the acoustic-vibration coupled fingerprint data and the baseline fingerprint data under normal operating conditions are calculated item by item, and the differences are summarized to obtain the fingerprint drift.

[0013] As a preferred embodiment of the automated monitoring method for mine ore pass conditions based on acoustic and vibration coupling described in this invention, the specific steps for accumulating the continuous fingerprint drift to form an acoustic-vibration coupling drift feature sequence are as follows. The fingerprint drift values ​​are sequentially concatenated over time to form a continuous fingerprint drift value sequence. The first drift value in the continuous fingerprint drift sequence is used as the initial cumulative value. The fingerprint drift values ​​are read one by one in chronological order. This process is repeated until all fingerprint drift values ​​are processed to obtain the acoustic-vibration coupling drift feature sequence.

[0014] As a preferred embodiment of the automated monitoring method for mine ore pass conditions based on acoustic and vibration coupling described in this invention, the specific steps for calculating the acoustic-vibration consistency index between the acoustic analysis segment and the vibration analysis segment are as follows: Under the same time alignment, extract the time-frequency raster representation of the corresponding frequency band and time window from the time-synchronized acoustic analysis segment and vibration analysis segment; The time-frequency grid representations are compared item by item, and the degree of synchronization consistency is calculated. The results are then summarized to form an acoustic-vibration consistency index.

[0015] As a preferred embodiment of the automated monitoring method for mine ore pass status based on acoustic and vibration coupling described in this invention, the specific steps for obtaining the operating conditions of the ore pass are as follows: When the acoustic-vibration consistency index does not meet the validity judgment criteria, it is determined that the coupling relationship between the current acoustic analysis segment and the vibration analysis segment is insufficient, and the current analysis process is terminated. When the acoustic-vibration consistency index meets the validity judgment condition, the acoustic-vibration coupling relationship between the current acoustic analysis segment and the vibration analysis segment is determined to be valid, and it is identified as a valid analysis object. Starting with the effective analysis object, the acoustic-vibration coupling fingerprint data is scaled and rearranged according to the preset multi-time scale rules to form an acoustic-vibration fingerprint association sequence. The acoustic-vibration coupling drift feature sequence is then scaled and continuous to obtain the acoustic-vibration drift evolution sequence. The consistency of the acoustic fingerprint association sequence and the acoustic drift evolution sequence was checked at multiple time scales to obtain the operating conditions of the ore pass.

[0016] The beneficial effects of this invention are as follows: by synchronously acquiring and analyzing acoustic signals and vibration signals from the wellbore, acoustic-vibration coupling features are constructed and acoustic-vibration coupling fingerprint data is formed, so that the operation behavior of the well pass can be stably characterized; further, relying on the intelligent sensing system, the fingerprint drift and its evolution trend are continuously analyzed, and combined with acoustic-vibration consistency and multi-timescale correlation reconstruction, the automated identification and continuous monitoring of the well pass operation conditions are realized, thereby improving the reliability of the operation condition judgment. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of an automated monitoring method for the condition of mine ore passes based on acoustic and vibration coupling.

[0019] Figure 2 This is a flowchart of signal acquisition and preprocessing.

[0020] Figure 3 This is a flowchart for extracting acoustic-vibration coupling features.

[0021] Figure 4 A flowchart for obtaining the operating conditions of the ore pass.

[0022] Figure 5 This is a diagram showing the correlation distribution of acoustic and vibration signal energy.

[0023] Figure 6 A comparison diagram of the evolution of acoustic-vibration coupling characteristics under different coupling strengths. Detailed Implementation

[0024] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0025] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0026] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0027] Reference Figures 1-6 This is one embodiment of the present invention, which provides an automated monitoring method for the condition of mine ore passes based on acoustic and vibration coupling, comprising the following steps: S1. Acquire acoustic signals and wellbore vibration signals from inside the well, and perform anti-aliasing processing, bandpass filtering, and detrending. Then, segment the signals into acoustic analysis segments and vibration analysis segments by event triggering.

[0028] S1.1 Collect the acoustic signals generated inside the well and the vibration signals generated by the well wall; It should be noted that an acoustic sensor is installed inside the well pass to collect acoustic signals generated inside the well pass, and a vibration sensor is fixedly installed at a designated location on the well wall to collect vibration signals generated on the well wall. The acoustic sensor inside the well and the vibration sensor on the well wall are configured for synchronous sampling, and the sampling frequency, sampling duration and data timestamp rules are uniformly set. The acoustic signal inside the well and the vibration signal on the well wall are collected, and the acoustic signal inside the well and the vibration signal on the well wall are converted into digital sampling sequences and written to the storage according to the timestamp.

[0029] It should also be noted that data timestamp rules refer to the uniformly set time stamp rules to ensure that data from multiple sensors can be accurately aligned. Data timestamp rules are acquired through external synchronization signals. The sensors automatically generate timestamps each time they sample, using a uniform time base (such as UTC time or GPS time) to ensure that data from different sensors can be accurately aligned, and that the timestamp interval matches the sampling frequency.

[0030] S1.2. Anti-aliasing processing is used to suppress the aliasing interference between acoustic signals and vibration signals in the wellbore, and bandpass filtering is used to retain the effective signal components related to the wellbore operation status, and the filtered acoustic signal and filtered vibration signal are output. It should be noted that anti-aliasing processing is performed on the acoustic signal inside the well and the vibration signal inside the wellbore, respectively. This is achieved by setting an anti-aliasing low-pass filter before signal sampling or in the digital domain to suppress high-frequency components above the Nyquist frequency to avoid spectral aliasing. After completing the anti-aliasing processing, bandpass filtering is performed on the acoustic signal inside the well and the vibration signal inside the wellbore, respectively, according to the target frequency band range of the relevant signals of the well pass operation status. The effective signal components within the target frequency band are retained, while low-frequency drift and high-frequency noise outside the target frequency band are filtered out. The filtered acoustic signal and the filtered vibration signal are then output as inputs for subsequent detrending processing.

[0031] It should be noted that the anti-aliasing low-pass filter is used to limit the effective frequency range of acoustic signals and wellbore vibration signals in the well. This effectively attenuates frequency components in the signal that are higher than the Nyquist frequency of the sampling system, thereby preventing high-frequency components from folding and aliasing into the low-frequency region during the sampling process, which would affect the authenticity and stability of subsequent acoustic and vibration analysis results.

[0032] The cutoff frequency of the anti-aliasing low-pass filter is set according to the sampling frequency of the acoustic signal in the well and the vibration signal of the well wall. The cutoff frequency is set to a frequency less than half of the sampling frequency (for example, when the sampling frequency is 1000Hz, the cutoff frequency is set to 500Hz), and adjusted in combination with the highest effective frequency range of the relevant signals of the well pass operation status. During the setting process, by ensuring that the filter has sufficient attenuation amplitude above the cutoff frequency, the high-frequency interference components are fully suppressed before entering the sampling or resampling stage, thereby meeting the anti-aliasing processing requirements.

[0033] S1.3. Remove the slow-drift components from the filtered acoustic signal and the filtered vibration signal by linear trend term elimination to obtain the detrended acoustic signal and the detrended vibration signal. It should be noted that the filtered acoustic signal and the filtered vibration signal are processed in segments according to a preset time length. Within each processing segment, the slope and intercept of the linear trend term are calculated using the least squares linear fitting method to generate the corresponding linear trend sequence. Subsequently, the generated linear trend sequence is subtracted point by point from the filtered acoustic signal and the filtered vibration signal within the corresponding processing segment, and the detrending results of each processing segment are spliced ​​together in the original time order to form a continuous detrended acoustic signal and a detrended vibration signal.

[0034] It should be noted that the time length is set based on the time scale of the slow drift component in the well acoustic signal and well wall vibration signal. The time range is selected to be significantly longer than the duration of the well pass operation event but shorter than the overall monitoring time, so as to ensure that the linear trend term can stably reflect the slow drift without affecting the fast change signal corresponding to the well pass operation event.

[0035] S1.4 Based on the detrended acoustic signal and the detrended vibration signal, calculate the signal amplitude within a continuous time window. When the signal amplitude exceeds the event trigger threshold, determine the corresponding time as the trigger time of the chute operation event. It should be noted that continuous time windows are divided using a sliding method for both the detrended acoustic signal and the detrended vibration signal. Within each continuous time window, the corresponding signal amplitude index is calculated, and the calculated acoustic signal amplitude index is compared with the acoustic event trigger threshold, and the vibration signal amplitude index is compared with the vibration event trigger threshold. When the acoustic signal amplitude index of the detrended acoustic signal exceeds the acoustic event trigger threshold or the vibration signal amplitude index of the detrended vibration signal exceeds the vibration event trigger threshold, the start time or center time of the time window that first meets the event trigger threshold condition is recorded, and this time is determined as the trigger time of the well pass operation event.

[0036] It should be noted that the acoustic event trigger threshold is set based on the statistical analysis of the detrended acoustic signal collected when the ore is in a normal background state. It is obtained by calculating the long-term statistical characteristics of the acoustic signal amplitude under the condition of no ore operation event, and setting a fixed threshold higher than the background fluctuation level based on the statistical characteristics. It is used to distinguish the significant acoustic changes caused by the ore operation event from the environmental background noise. Based on the statistical characteristics of the background amplitude of the acoustic signal after the downward trend in the no-ore operation state and the amplitude distribution of the typical acoustic response corresponding to the normal operation event of the ore, the range of the acoustic event triggering threshold is determined to be an amplitude range that is greater than the upper limit of the statistical background amplitude of the acoustic signal after the downward trend in the no-ore operation state and less than the lower limit of the amplitude of the typical acoustic response corresponding to the normal operation event of the ore (for example, if the upper limit of the statistical background amplitude is 0.05 and the lower limit of the typical acoustic response is 0.2, then the range of the acoustic event triggering threshold is 0.05 to 0.2).

[0037] The vibration event trigger threshold is set based on the statistical analysis of the detrended vibration signal collected when the ore is in a normal background state. It is obtained by calculating the statistical distribution characteristics of the vibration signal amplitude under the condition of no ore operation events, and setting a fixed threshold that can reflect the response changes of the well wall structure based on the statistical distribution characteristics. It is used to identify effective vibration events generated during the operation of the ore and suppress random vibration interference in the non-operation state. Based on the statistical characteristics of the background vibration amplitude of the vibration signal after the downward trend in the no-access operation state and the amplitude distribution of the typical wellbore vibration response corresponding to the normal operation event of the access well, the range of the vibration event trigger threshold is determined to be an amplitude range that is greater than the upper limit of the statistical background vibration amplitude of the vibration signal after the downward trend in the no-access operation state and less than the lower limit of the amplitude of the typical wellbore vibration response corresponding to the normal operation event of the access well (for example, if the upper limit of the statistical background vibration amplitude is 0.1 and the lower limit of the typical wellbore vibration response is 0.5, then the range of the vibration event trigger threshold is 0.1 to 0.5).

[0038] like Figure 5The figure shows the correlation distribution between acoustic signal energy and wellbore vibration signal energy during well pass operation events. The scatter points in the figure reflect the synchronous change characteristics of acoustic response and structural vibration response under the same operational event triggering conditions. It can be seen that the acoustic energy and vibration energy exhibit a relatively concentrated correlation distribution trend overall, indicating that they originate from the same operational behavior and have a good physical coupling basis. This result verifies the rationality of introducing synchronous acoustic-vibration acquisition and joint analysis in well pass condition monitoring, and provides physical consistency support for further constructing acoustic-vibration coupling characteristics and acoustic-vibration coupling fingerprint data.

[0039] S1.5. Taking the trigger time of the well operation event as the center, the acoustic signal and vibration signal after detrended are extracted according to the preset time window to obtain the acoustic analysis segment and vibration analysis segment. It should be noted that the trigger time of the well operation event is used as the time intercept center. According to the preset time window, forward and backward time intercepts are performed on the detrended acoustic signal and the detrended vibration signal respectively to obtain acoustic analysis segments and vibration analysis segments with consistent time ranges and mutual alignment, which are used for subsequent correlation analysis.

[0040] It should also be noted that the time window is set based on the typical duration and signal characteristics of the well pass operation event. Generally, the length of the time window is selected to be greater than the duration of the well pass operation event, but less than the overall monitoring duration, in order to ensure that the complete changes in the signal before and after the event can be captured, while avoiding interference with other non-event signals.

[0041] S2. Correlation analysis is performed on the acoustic analysis segment and the vibration analysis segment within the corresponding frequency band and time window to extract acoustic-vibration coupling features. The acoustic-vibration coupling features are then combined and mapped to form acoustic-vibration coupling fingerprint data.

[0042] S2.1. Align the acoustic analysis segment and the vibration analysis segment in time and divide them into continuous time windows. Perform continuous wavelet transform on the acoustic analysis segment and the vibration analysis segment in each time window to obtain the initial time-frequency representation. It should be noted that the acoustic analysis segment and the vibration analysis segment are time-aligned with the same start timestamp and the same sampling time reference to establish a one-to-one correspondence between the time indices of the acoustic analysis segment and the vibration analysis segment. Continuous time windows are then divided on the time-aligned acoustic analysis segment and the time-aligned vibration analysis segment according to a preset window length and a preset sliding step size. Within each continuous time window, continuous wavelet transforms are performed on the acoustic analysis segment window signal and the vibration analysis segment window signal, respectively, to obtain the acoustic continuous wavelet coefficient matrix and the vibration continuous wavelet coefficient matrix within the corresponding time window, thereby forming the initial time-frequency representation of acoustics and the initial time-frequency representation of vibration.

[0043] It should be noted that the sliding step size is set according to the window length, and a time interval smaller than the window length is selected as the sliding distance between adjacent consecutive time windows to ensure that there is a time overlap between adjacent consecutive time windows, thereby realizing the continuous characterization of the time-frequency change process of acoustic analysis segments and vibration analysis segments on the time axis. The window length is set based on the typical duration of chute operation events and the frequency characteristics of the signal, and is usually selected to capture the time period that can effectively capture changes in signal characteristics.

[0044] S2.2. The dispersed time-frequency energy in the initial time-frequency representation is re-aggregated along the true frequency trajectory to obtain a high-resolution time-frequency representation. The high-resolution time-frequency representation is then compressed and reconstructed through a synchronous squeezing operation to form an acoustic time-frequency diagram and a vibration time-frequency diagram. It should be noted that, based on the continuous wavelet coefficient matrices in the initial acoustic and vibration time-frequency representations, the instantaneous frequency information corresponding to each time position and each scale position of the continuous wavelet coefficient matrix is ​​calculated, and the true frequency trajectory of the time-frequency energy in the continuous wavelet coefficient matrix is ​​determined using the instantaneous frequency information. The dispersed time-frequency energy in the initial acoustic time-frequency representation is re-aggregated along the true frequency trajectory to obtain the high-resolution acoustic time-frequency representation. Simultaneously, the dispersed time-frequency energy in the initial vibration time-frequency representation is re-aggregated along the true frequency trajectory to obtain the high-resolution vibration time-frequency representation. A synchronous compression operation is performed on the high-resolution acoustic time-frequency representation to compress it in the frequency axis direction and complete the time-frequency representation reconstruction, outputting the acoustic time-frequency graph. A synchronous compression operation is also performed on the high-resolution vibration time-frequency representation to compress it in the frequency axis direction and complete the time-frequency representation reconstruction, outputting the vibration time-frequency graph.

[0045] It should be noted that the expression for calculating the instantaneous frequency information corresponding to each time position and each scale position of the continuous wavelet coefficient matrix is ​​as follows: ; in, At scale position and time location The instantaneous frequency information, in Hertz, is calculated from the continuous wavelet coefficient matrix corresponding to the detrended acoustic signal or the detrended vibration signal. It is an imaginary part extraction operation, used to extract the imaginary part information corresponding to the phase change of the continuous wavelet coefficient matrix; It is a continuous wavelet coefficient matrix Regarding time location The partial derivatives are used to represent the rate at which the continuous wavelet coefficient matrix changes with time position; It is a time position parameter The differential component represents the small change on the time axis in the continuous wavelet transform; It is the scaling parameter in continuous wavelet transform, used to control the analysis frequency range and correspond to different frequency components; It is the time position parameter in continuous wavelet transform, used to indicate the corresponding position of the continuous wavelet coefficient matrix on the time axis; It is the detrended acoustic signal or the detrended vibration signal. After continuous wavelet transform, at the scale position and time location The continuous wavelet coefficient matrix obtained at the point; It is the time-domain representation of the detrended acoustic signal or the detrended vibration signal.

[0046] S2.3 Represent the acoustic time-frequency diagram and the vibration time-frequency diagram in two-dimensional coordinate form to form a time-frequency plane, and divide the time-frequency plane into multiple frequency bands and time windows to form a two-dimensional time-frequency grid. It should be noted that the acoustic time-frequency diagram and the vibration time-frequency diagram are respectively converted into a two-dimensional coordinate representation with time as the horizontal axis and frequency as the vertical axis. Under the same time and frequency range, the acoustic time-frequency diagram is compared... Figure 2 3D coordinate representation and vibration time frequency Figure 2 The 2D coordinate system performs unified coordinate alignment to form a corresponding time-frequency plane. On the time-frequency plane, the frequency axis is divided into multiple frequency bands according to a preset frequency band division rule, and the time axis is divided into multiple time windows according to a preset time division rule. The intersection of the frequency bands and the time windows forms several grid cells. All grid cells together form a 2D time-frequency grid, which is used to extract the corresponding time-frequency coefficients of the acoustic time-frequency map and the vibration time-frequency map from the fixed frequency bands and fixed time window positions.

[0047] It should be noted that the frequency band division rules are set based on the frequency range of energy distribution during the operation of the chute in the acoustic time-frequency diagram and the vibration time-frequency diagram. By dividing the complete frequency range into frequency intervals with equal bandwidth or functionally related frequency intervals, each frequency band interval can cover the main frequency components corresponding to the changes in the chute's operating state, while avoiding significant overlap of frequency characteristics in different operating states. The time division rules are set based on the duration of the acoustic analysis segment and the vibration analysis segment, as well as the typical time scale of the well pass operation event. By continuously dividing the analysis time range according to a fixed time length, each time window can fully contain the local characteristics of the acoustic response and well wall vibration changes during the well pass operation, and ensure the continuity between adjacent time windows to support subsequent time series correlation analysis.

[0048] S2.4 Extract the corresponding time-frequency coefficients from the acoustic time-frequency map and vibration time-frequency map from the two-dimensional time-frequency grid, and perform correlation analysis on the acoustic signal in the well and the vibration signal of the well wall to output acoustic-vibration correlation information. Through collection and characterization, acoustic-vibration coupling characteristics are formed. It should be noted that, in the two-dimensional time-frequency grid, the intersection grids of frequency bands and time windows are traversed one by one. The acoustic time-frequency coefficient set within the corresponding grid coverage area is extracted from the acoustic time-frequency map, and the vibration time-frequency coefficient set within the same frequency band and time window coverage area is extracted from the vibration time-frequency map. Correlation analysis is performed on each pair of acoustic and vibration time-frequency coefficient sets. By pairing the acoustic and vibration time-frequency coefficient sets element by element and calculating the correlation metric, acoustic-vibration correlation information corresponding one-to-one with frequency bands and time windows is output. The acoustic-vibration correlation information corresponding to all grid units is collected and characterized. By organizing the acoustic-vibration correlation information according to the order of frequency bands and time windows and forming a structured set, the acoustic-vibration coupling characteristics are obtained, reflecting their coupling relationship under different frequency bands and time windows, which is used to characterize the characteristics of the well chute operation state.

[0049] It should be noted that the expression for calculating the relevance metric is: ; in, It is an acoustic-vibration correlation measure calculated based on acoustic time-frequency coefficients and vibration time-frequency coefficients within the same frequency band and the same time window. It is used to characterize the correlation strength between the acoustic signal in the well and the vibration signal of the well wall in the corresponding time-frequency region. It is extracted from the acoustic time-frequency diagram and lies within a specified frequency band and time window. One acoustic time-frequency coefficient; Within the same frequency band and the same time window, all acoustic time-frequency coefficients The arithmetic mean; It is extracted from the vibration time-frequency diagram and lies within the same frequency band and time window corresponding to the acoustic time-frequency coefficient. One vibration time-frequency coefficient; It refers to the time-frequency coefficients of all vibrations within the same frequency band and the same time window. The arithmetic mean; It is the index number when acoustic time-frequency coefficients and vibration time-frequency coefficients are paired one by one in the corresponding frequency band and time window; It refers to the number of pairs of acoustic time-frequency coefficients and vibration time-frequency coefficients that participate in the acoustic-vibration correlation measurement calculation within the corresponding frequency band and time window. The acoustic time-frequency coefficients are obtained by selecting grid cells corresponding to a specified frequency band and a specified time window in a two-dimensional time-frequency grid, and extracting them point by point from the acoustic time-frequency diagram within the corresponding frequency and time ranges. The vibration time-frequency coefficients are obtained by selecting grid cells with the same frequency band and the same time window as the acoustic time-frequency coefficients in a two-dimensional time-frequency grid, and extracting them point by point from the vibration time-frequency diagram within the corresponding frequency and time ranges.

[0050] like Figure 6 As shown, this figure illustrates the variation of acoustic-vibration coupling characteristics with operational event numbers under fixed noise levels and fingerprint drift rates, corresponding to different acoustic-vibration coupling intensities. Multiple curves in the figure represent the evolution of acoustic-vibration coupling characteristics under different coupling intensities; the upper part provides a global overview, while the lower part shows a magnified comparison of key difference regions. It can be seen that with increasing coupling intensity, the overall level of acoustic-vibration coupling characteristics significantly improves, maintaining good continuity and stability over time. The magnified local plots further demonstrate a significant distinguishing effect at the characteristic peak positions and points of maximum difference under different coupling intensities. These results indicate that by constructing acoustic-vibration coupling characteristics and forming acoustic-vibration coupling fingerprint data, a stable characterization of well pass operation behavior can be achieved, providing a reliable feature basis for subsequent operational condition identification and state evolution analysis.

[0051] S2.5. Group the acoustic-vibration coupling features according to the three-layer index structure of frequency band-time window-coupling, and then splice them in the order of time window within each group to form a layered coupling feature table; It should be noted that each acoustic-vibration coupling feature is read and labeled with its corresponding frequency band index, time window index, and coupling index. The acoustic-vibration coupling features are grouped according to the three-layer index structure of frequency band, time window, and coupling. Acoustic-vibration coupling features with the same frequency band index and the same coupling index are grouped together. Within each group, the acoustic-vibration coupling features are arranged according to the order of the time window index and spliced ​​to form a time-continuous coupling feature sequence. All coupling feature sequences are summarized and organized in tabular form to obtain a hierarchical coupling feature table.

[0052] It should be noted that the frequency band-time window-coupling three-layer index structure is formed by assigning frequency band indexes to corresponding regions based on the frequency band intervals divided in the time-frequency plane, assigning time window indexes to corresponding regions based on the time window, and identifying acoustic-vibration correlation information as coupling indexes.

[0053] S2.6 Map the hierarchical coupling feature table to discrete symbols through a fixed codebook, and output them in a fixed order according to the frequency band-time window-coupling three-layer index structure to form acoustic-vibration coupling fingerprint data; It should be noted that acoustic-vibration coupling features are read item by item from the hierarchical coupling feature table according to the frequency band index, time window index, and coupling index, and mapped to corresponding discrete symbols through a fixed codebook. According to the fixed order of the three-layer index structure of frequency band, time window, and coupling, the signal features are classified according to the frequency band range, and then the signals in each frequency band are divided according to the time window. Finally, the signals are sorted according to the coupling relationship between acoustic and vibration signals, and the discrete symbols under each index are concatenated in this order to form a complete acoustic-vibration coupling fingerprint data, which is used to characterize the operating status of the ore pass and help identify and monitor possible abnormal behaviors or faults during the operation of the ore pass.

[0054] It should also be noted that the fixed codebook is used to map continuous acoustic-vibration coupling features to discrete symbols, containing discrete symbols with different ranges of eigenvalues, ensuring that each eigenvalue can be mapped to a unique symbol.

[0055] The fixed codebook is obtained by analyzing historical acoustic and vibration signal data collected from the well under normal operating conditions. Based on the characteristic range of the acoustic and vibration signals, these characteristic value intervals are divided into multiple discrete symbol intervals, and the symbol corresponding to each interval is determined according to the data distribution. This ensures that the features can be represented stably and consistently in subsequent data processing and analysis.

[0056] S3. Calculate the fingerprint drift between the acoustic-vibration coupled fingerprint data and the normal operating condition baseline fingerprint data, and accumulate the continuous fingerprint drift to form an acoustic-vibration coupled drift feature sequence.

[0057] S3.1 Collect the acoustic-vibration coupling fingerprint data under normal operating conditions during historical operation to form normal operating condition baseline fingerprint data; It should be noted that the operating time periods of the ore pass under normal operating conditions are selected from the historical operation records, and the acoustic-vibration coupling fingerprint data generated in the corresponding time periods are collected. The acoustic-vibration coupling fingerprint data are aligned and statistically analyzed according to the fixed order of the frequency band-time window-coupling three-layer index structure. Representative discrete symbols are formed at each index position and combined in a fixed order to obtain the baseline fingerprint data of normal operating conditions.

[0058] S3.2 Using the frequency band-time window-coupling three-layer index structure as the matching key, the acoustic vibration coupled fingerprint data is matched item by item with the normal operating condition baseline fingerprint data to form fingerprint pairs; It should be noted that a mapping relationship between a three-layer index structure of frequency band, time window, and coupling is established for the acoustic-vibration coupled fingerprint data and the normal operating condition baseline fingerprint data, respectively. The frequency band index, time window index, and coupling index are used as matching keys. At each index combination position, the discrete symbols of the acoustic-vibration coupled fingerprint data are paired item by item with the discrete symbols of the normal operating condition baseline fingerprint data to form a fingerprint pair set.

[0059] S3.3 Based on the fingerprint pair, calculate the difference between the acoustic-vibration coupled fingerprint data and the normal operating condition baseline fingerprint data item by item, and summarize the differences to obtain the fingerprint drift amount. It should be noted that the fingerprint pair set is read item by item in a fixed order according to the three-layer index structure of frequency band-time window-coupling. In each fingerprint pair, the discrete symbols of the acoustic-vibration coupling fingerprint data and the discrete symbols of the normal operating condition baseline fingerprint data are extracted and the corresponding difference is calculated. The difference is assigned as a matching difference or a non-matching difference by determining whether the discrete symbols of the acoustic-vibration coupling fingerprint data and the discrete symbols of the normal operating condition baseline fingerprint data are consistent. The difference values ​​of all index positions are summarized by summation or weighted summation to obtain the fingerprint drift value, which is used to characterize the degree of deviation of the acoustic-vibration coupling fingerprint data from the normal operating condition baseline fingerprint data.

[0060] It should also be noted that the matching difference refers to the fact that when the discrete symbols of the acoustic-vibration coupled fingerprint data and the discrete symbols of the normal operating condition baseline fingerprint data are consistent under the same frequency band, time window and coupling index position, the difference is set to zero, indicating that the two fingerprint data are highly similar.

[0061] The mismatch difference refers to the discrepancy between the discrete symbols of the acoustic-vibration coupled fingerprint data and the discrete symbols of the normal operating condition baseline fingerprint data under the same frequency band, time window, and coupling index position. The difference is set to a fixed value (e.g., set to 1), indicating that there is a significant difference between the two fingerprint data, reflecting that the system may have changed or is abnormal.

[0062] S3.4. Concatenate the fingerprint drift values ​​in time sequence to form a continuous fingerprint drift value sequence; It should be noted that each fingerprint drift is bound to a corresponding timestamp of the sluice run event triggering time, and the fingerprint drifts are sorted and sequentially concatenated according to the chronological order of the sluice run event triggering time to form a continuous fingerprint drift sequence with timestamp as index and fingerprint drift as sequence element, which is used for subsequent trend accumulation processing.

[0063] S3.5. Take the first drift value of the continuous fingerprint drift value sequence as the initial cumulative value, read the fingerprint drift values ​​one by one in time order, and repeat the process until all fingerprint drift values ​​are processed to obtain the acoustic-vibration coupling drift feature sequence. It should be noted that the first fingerprint drift is read from the continuous fingerprint drift sequence and assigned the initial cumulative value. At the same time, the initial cumulative value is written to the first sequence position of the acoustic-vibration coupling drift feature sequence. Subsequently, the subsequent fingerprint drifts are read one by one according to the time order of the continuous fingerprint drift sequence. The cumulative sum operation is performed on the fingerprint drift read each time and the cumulative value at the previous moment to obtain the current cumulative value. The current cumulative value is written into the acoustic-vibration coupling drift feature sequence according to the corresponding timestamp. The loop operation of "reading fingerprint drift, cumulative sum operation, and writing into the acoustic-vibration coupling drift feature sequence" is repeated until all fingerprint drifts are processed, and finally the acoustic-vibration coupling drift feature sequence arranged continuously in time is obtained.

[0064] S4. Calculate the acoustic-vibration consistency index between the acoustic analysis segment and the vibration analysis segment. When the acoustic-vibration consistency index meets the validity judgment condition, perform multi-time-scale correlation reconstruction and coupling stability evolution of the acoustic-vibration coupling drift characteristic sequence and acoustic-vibration coupling fingerprint data to obtain the well operation condition.

[0065] It should be noted that the validity judgment criteria are used to determine whether the acoustic-vibration coupling relationship between the acoustic analysis segment and the vibration analysis segment is reliable. They are defined by setting a corresponding threshold range for the acoustic-vibration consistency index. The validity determination criteria are based on the statistical distribution characteristics of the acoustic and vibration consistency index of the acoustic analysis segments and vibration analysis segments during the historical operation of the ore pass, under the condition that the coupling relationship is established. Based on the statistical distribution characteristics of the acoustic and vibration consistency index, a threshold range that can distinguish between effective and ineffective acoustic and vibration coupling relationships is determined.

[0066] S4.1 Extract time-frequency raster representations within the corresponding frequency band and time window from the acoustic analysis segment and vibration analysis segment after time synchronization, under the same time alignment relationship; It should be noted that, while maintaining the time alignment between the acoustic analysis segment and the vibration analysis segment, according to the frequency band division rules and time division rules, the acoustic time-frequency coefficient set and the vibration time-frequency coefficient set within the same frequency band interval and the same time window are extracted from the acoustic time-frequency diagram corresponding to the acoustic analysis segment after time synchronization and the vibration time-frequency diagram corresponding to the vibration analysis segment after time synchronization, respectively. They are then organized according to the frequency band interval index and the time window index to form a time-frequency raster representation.

[0067] S4.2. Compare the time-frequency grid representation item by item, calculate the degree of synchronization consistency, and summarize to form an acoustic-vibration consistency index. It should be noted that for each grid cell in the time-frequency grid representation determined by the frequency band index and the time window index, the acoustic time-frequency coefficient set and the vibration time-frequency coefficient set are read item by item. An item-by-item comparison operation is performed on the acoustic time-frequency coefficient set and the vibration time-frequency coefficient set. The synchronization consistency value is calculated for the acoustic time-frequency coefficient set and the vibration time-frequency coefficient set within the same grid cell and recorded as the grid consistency result. The grid consistency results corresponding to all frequency bands and all time windows are summarized, and the acoustic-vibration consistency index is obtained by summing or averaging. This index is used to characterize the overall synchronization consistency of the acoustic analysis segment and the vibration analysis segment on the two-dimensional time-frequency grid after time synchronization.

[0068] S4.3 When the acoustic-vibration consistency index does not meet the validity judgment condition, it is determined that the coupling relationship between the current acoustic analysis segment and the vibration analysis segment is insufficient, and the current analysis process is terminated. It should be noted that the acoustic-vibration consistency index is compared with the validity judgment condition. When the acoustic-vibration consistency index is lower than the threshold range corresponding to the validity judgment condition, the judgment result of insufficient coupling relationship between the acoustic analysis segment and the vibration analysis segment is output, and the acoustic analysis segment and the vibration analysis segment are marked as invalid analysis objects. After the invalid analysis objects are marked, the subsequent processing steps such as acoustic-vibration coupling fingerprint data generation, fingerprint drift calculation and acoustic-vibration coupling drift feature sequence update are stopped, the current analysis process is terminated and the analysis process corresponding to the next well run event trigger time is entered.

[0069] S4.4 When the acoustic-vibration consistency index meets the validity judgment condition, the acoustic-vibration coupling relationship between the current acoustic analysis segment and the vibration analysis segment is determined to be valid, and it is identified as a valid analysis object; It should be noted that the acoustic-vibration consistency index is compared with the validity judgment condition. When the acoustic-vibration consistency index falls within the threshold range corresponding to the validity judgment condition, the judgment result that the acoustic-vibration coupling relationship between the acoustic analysis segment and the vibration analysis segment is established is output, and the acoustic analysis segment and the vibration analysis segment are marked as valid analysis objects. After the valid analysis objects are marked, subsequent processing steps such as acoustic-vibration coupling fingerprint data generation, fingerprint drift calculation, and acoustic-vibration coupling drift feature sequence update are allowed to carry out well operation condition identification.

[0070] S4.5 Starting with the effective analysis object, the acoustic-vibration coupling fingerprint data is scaled and rearranged according to the preset multi-time scale rules to form an acoustic-vibration fingerprint association sequence. The acoustic-vibration coupling drift feature sequence is then scaled and continuous to obtain the acoustic-vibration drift evolution sequence. It should be noted that, taking the trigger time of the well operation event corresponding to the effective analysis object as the starting benchmark, multiple time scales and time alignment boundaries corresponding to each time scale are determined according to the preset multi-time scale rules. The acoustic-vibration coupling fingerprint data corresponding to the effective analysis object and subsequent adjacent effective analysis objects are segmented according to the time scale alignment boundaries. Within each time scale, the acoustic-vibration coupling fingerprint data are sequentially rearranged and concatenated according to the time order to form an acoustic-vibration fingerprint association sequence. The acoustic-vibration coupling drift feature sequence is read under the same time scale division, and the acoustic-vibration coupling drift feature sequence is aggregated according to the time scale alignment boundaries. By taking the sum of the acoustic-vibration coupling drift feature sequences within each time scale and maintaining the continuity of the time order, the aggregated acoustic-vibration coupling drift feature sequence is continuously processed and output to obtain an acoustic-vibration drift evolution sequence with the same time scale as the acoustic-vibration fingerprint association sequence.

[0071] It should be noted that the multi-timescale rule is a time organization rule used to jointly analyze the operation process of the ore pass at different time resolutions; it is set based on the typical duration of the ore pass operation event, the event trigger interval, and the cumulative characteristics of the acoustic-vibration coupling drift characteristics over time.

[0072] S4.6 Perform multi-timescale consistency checks on the acoustic fingerprint association sequence and acoustic drift evolution sequence to obtain the operating conditions of the ore pass; It should be noted that, under the same time alignment boundary, the acoustic and vibration fingerprint association sequence and the acoustic and vibration drift evolution sequence are matched segment by segment according to the time scale. The changes in acoustic and vibration coupling fingerprint and the trend of acoustic and vibration coupling drift within each time scale are extracted. Consistency checks are performed on the degree of change of discrete symbols and the cumulative drift amplitude within each time scale. The consistency check results of each time scale are summarized and the well operation conditions are output.

[0073] In summary, this invention achieves stable characterization of well pass operation behavior by simultaneously acquiring and analyzing acoustic signals and vibration signals from the wellbore and constructing acoustic-vibration coupling features and forming acoustic-vibration coupled fingerprint data. Furthermore, it relies on an intelligent sensing system to continuously analyze fingerprint drift and its evolution trend, and combines acoustic-vibration consistency with multi-timescale correlation reconstruction to realize automated identification and continuous monitoring of well pass operation conditions, thereby improving the reliability of operation condition determination.

[0074] 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, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An automated monitoring method for the condition of mine ore passes based on acoustic and vibration coupling, characterized in that: include, Acoustic signals from inside the well and vibration signals from the wellbore are acquired and subjected to anti-aliasing, bandpass filtering and detrending processing. The signals are then segmented into acoustic analysis segments and vibration analysis segments by event triggering. The acoustic and vibration analysis segments are correlated within corresponding frequency bands and time windows to extract acoustic-vibration coupling features. These features are then combined and mapped to form acoustic-vibration coupling fingerprint data. The specific steps are as follows. The acoustic analysis segment and the vibration analysis segment are time-aligned and divided into continuous time windows. Continuous wavelet transform is performed on the acoustic analysis segment and the vibration analysis segment in each time window to obtain the initial time-frequency representation. The dispersed time-frequency energy in the initial time-frequency representation is re-aggregated along the true frequency trajectory to obtain a high-resolution time-frequency representation. The high-resolution time-frequency representation is then compressed and reconstructed through a synchronous squeezing operation to form an acoustic time-frequency diagram and a vibration time-frequency diagram. The acoustic time-frequency diagram and the vibration time-frequency diagram are represented in two-dimensional coordinate form to form a time-frequency plane, and the time-frequency plane is divided into multiple frequency bands and time windows to form a two-dimensional time-frequency grid. The time-frequency coefficients corresponding to the acoustic time-frequency map and vibration time-frequency map are extracted from the two-dimensional time-frequency grid, and the correlation analysis between the acoustic signal in the well and the vibration signal of the well wall is performed to output the acoustic-vibration correlation information. The acoustic-vibration coupling characteristics are formed by collection and characterization. The acoustic-vibration coupling features are grouped according to a three-layer index structure of frequency band, time window, and coupling, and then spliced ​​together in the order of time window within each group to form a hierarchical coupling feature table; The hierarchical coupling feature table is mapped to discrete symbols through a fixed codebook and output in a fixed order according to the three-layer index structure of frequency band-time window-coupling to form acoustic-vibration coupling fingerprint data; Calculate the fingerprint drift between the acoustic-vibration coupled fingerprint data and the normal operating condition baseline fingerprint data, and accumulate the continuous fingerprint drift to form an acoustic-vibration coupled drift feature sequence. The acoustic-vibration consistency index between the acoustic analysis segment and the vibration analysis segment is calculated. When the acoustic-vibration consistency index meets the validity judgment condition, the acoustic-vibration coupling drift characteristic sequence and acoustic-vibration coupling fingerprint data are reconstructed and coupled stability evolved over multiple time scales to obtain the operating conditions of the ore pass.

2. The automated monitoring method for mine ore pass status based on acoustic and vibration coupling as described in claim 1, characterized in that: The acquisition of acoustic signals from inside the well and vibration signals from the wellbore, followed by anti-aliasing processing, bandpass filtering, and detrending, involves the following specific steps: Collect acoustic signals generated inside the well pass and vibration signals generated by the well wall; Anti-aliasing processing is used to suppress the aliasing interference between acoustic signals and wellbore vibration signals in the well, and bandpass filtering is used to retain the effective signal components related to the well pass operation status, outputting filtered acoustic signals and filtered vibration signals. The slowly varying drift components in the filtered acoustic and vibration signals are removed by linear trend term elimination to obtain the detrended acoustic and vibration signals.

3. The automated monitoring method for mine ore pass status based on acoustic and vibration coupling as described in claim 1, characterized in that: The process of segmenting the data into acoustic analysis and vibration analysis segments via event triggering follows these steps: Based on the detrended acoustic signal and the detrended vibration signal, the signal amplitude is calculated in a continuous time window. When the signal amplitude exceeds the event trigger threshold, the corresponding time is determined as the trigger time of the well pass operation event. Centered on the trigger time of the well operation event, the acoustic signal and vibration signal after detrended are extracted according to the preset time window to obtain acoustic analysis segments and vibration analysis segments.

4. The automated monitoring method for mine ore condition based on acoustic and vibration coupling as described in claim 1, characterized in that: The frequency band-time window-coupling three-layer index structure is formed by assigning frequency band indexes to corresponding regions based on the frequency band intervals divided in the time-frequency plane, assigning time window indexes to corresponding regions based on the time window, and identifying acoustic-vibration correlation information as coupling indexes.

5. The automated monitoring method for mine ore pass status based on acoustic and vibration coupling as described in claim 1, characterized in that: The specific steps for calculating the fingerprint drift between the acoustic-vibration coupled fingerprint data and the baseline fingerprint data under normal operating conditions are as follows. The acoustic-vibration coupling fingerprint data under normal operating conditions during historical operation are collected to form the normal operating condition baseline fingerprint data. Using a three-layer index structure of frequency band, time window, and coupling as the matching key, the acoustic-vibration coupled fingerprint data is matched item by item with the baseline fingerprint data under normal operating conditions to form fingerprint pairs; Based on the fingerprint pairs, the differences between the acoustic-vibration coupled fingerprint data and the baseline fingerprint data under normal operating conditions are calculated item by item, and the differences are summarized to obtain the fingerprint drift.

6. The automated monitoring method for mine ore pass status based on acoustic and vibration coupling as described in claim 5, characterized in that: The specific steps for accumulating the trend of continuous fingerprint drift to form an acoustic-vibration coupling drift feature sequence are as follows. The fingerprint drift values ​​are sequentially concatenated over time to form a continuous fingerprint drift value sequence. The first drift value in the continuous fingerprint drift sequence is used as the initial cumulative value. The fingerprint drift values ​​are read one by one in chronological order. This process is repeated until all fingerprint drift values ​​are processed to obtain the acoustic-vibration coupling drift feature sequence.

7. The automated monitoring method for mine ore pass status based on acoustic and vibration coupling as described in claim 1, characterized in that: The specific steps for calculating the acoustic-vibration consistency index between the acoustic analysis segment and the vibration analysis segment are as follows. Under the same time alignment, extract the time-frequency raster representation of the corresponding frequency band and time window from the time-synchronized acoustic analysis segment and vibration analysis segment; The time-frequency grid representations are compared item by item, and the degree of synchronization consistency is calculated. The results are then summarized to form an acoustic-vibration consistency index.

8. The automated monitoring method for mine ore pass status based on acoustic and vibration coupling as described in claim 1, characterized in that: The specific steps for obtaining the operating conditions of the well pass are as follows. When the acoustic-vibration consistency index does not meet the validity judgment criteria, it is determined that the coupling relationship between the current acoustic analysis segment and the vibration analysis segment is insufficient, and the current analysis process is terminated. When the acoustic-vibration consistency index meets the validity judgment condition, the acoustic-vibration coupling relationship between the current acoustic analysis segment and the vibration analysis segment is determined to be valid, and it is identified as a valid analysis object. Starting with the effective analysis object, the acoustic-vibration coupling fingerprint data is scaled and rearranged according to the preset multi-time scale rules to form an acoustic-vibration fingerprint association sequence. The acoustic-vibration coupling drift feature sequence is then scaled and continuous to obtain the acoustic-vibration drift evolution sequence. The consistency of the acoustic fingerprint association sequence and the acoustic drift evolution sequence was checked at multiple time scales to obtain the operating conditions of the ore pass.

Images