A method and system for monitoring the condition of mining equipment based on the Industrial Internet of Things

Abstract

This invention relates to the field of data processing, specifically to a method and system for monitoring the condition of mine equipment based on the Industrial Internet of Things (IIoT). It includes: acquiring time-series pressure data; extracting low-frequency and high-frequency component signals and calculating spectral energy distribution characteristic parameters; calculating the confidence level of mechanical attributes characterizing whether the signal source has mechanical periodicity; if the confidence level of mechanical attributes corresponding to the current time window is higher than a preset confidence threshold, then determining that the time-series pressure data within the current time window is a mechanical construction interference signal, and acquiring a pure static load pressure curve after removing high-frequency interference; generating a comprehensive risk index for equipment condition; calculating the effective stress change characteristics of the support equipment based on the pure static load pressure curve, determining the equipment health status level, and triggering corresponding level early warning commands. This invention can improve the accuracy of mine equipment condition monitoring.

Description

Technical Field

[0001] This invention relates to the field of data processing, and specifically to a method and system for monitoring the status of mining equipment based on the Industrial Internet of Things. Background Technology

[0002] With the advancement of smart mine construction, real-time monitoring of mine support equipment and surrounding rock conditions has become a core means of ensuring safe coal mine production. Currently, monitoring systems generally employ terminal devices such as displacement sensors and pressure sensors deployed in roadways to assess the stability of the support system by sensing changes in physical indicators.

[0003] In existing mine monitoring systems, abnormal signals are mainly identified by monitoring the displacement, pressure, and evolution characteristics (such as rate of change, acceleration, and fluctuation amplitude) of the roof strata. When drastic fluctuations or sudden displacement changes are detected in these physical indicators, the system automatically increases the sampling frequency of the sensors to capture key changes.

[0004] However, mine roadways are complex environments with multiple coupled loads. Heavy construction machinery such as tunneling machines and coal mining machines generate strong mechanical vibrations and periodic stress fluctuations during operation. Due to a lack of awareness of the construction conditions, existing technologies often misjudge such normal construction disturbances as roof instability risks or support failure warnings, resulting in an extremely high false alarm rate. Summary of the Invention

[0005] This invention provides a method and system for monitoring the status of mining equipment based on the Industrial Internet of Things (IIoT) to solve existing problems.

[0006] The present invention provides a method for monitoring the status of mine equipment based on the Industrial Internet of Things, which adopts the following technical solution:

[0007] One embodiment of the present invention provides a method for monitoring the condition of mining equipment based on the Industrial Internet of Things (IIoT), the method comprising the following steps:

[0008] Acquire time-series pressure data collected by multi-channel high-frequency pressure sensors deployed on mine support equipment, and construct a time-series pressure dataset;

[0009] The time-frequency transformation process is performed on the time-series pressure dataset to extract the low-frequency component signal characterizing the influence of mechanical vibration and the high-frequency component signal characterizing the impact of rock fracture. The spectral energy distribution characteristic parameters are calculated based on the energy ratio of the low-frequency component signal to the high-frequency component signal.

[0010] Based on the jump point sequence of spectral energy distribution characteristic parameters, combined with the theoretical operating cycle parameters of the tunneling machinery, the confidence level of the mechanical attribute characterizing whether the signal source has mechanical periodicity is calculated.

[0011] If the confidence level of the mechanical property corresponding to the current time window is higher than the preset confidence threshold, then the time-series pressure data in the current time window is determined to be a mechanical construction interference signal, and the pure static load pressure curve after removing high-frequency interference is obtained based on the low-frequency component signal in the time window.

[0012] By integrating spectral energy distribution characteristic parameters with mechanical property confidence levels, a comprehensive risk index for equipment status is generated.

[0013] The effective stress change characteristics of the support equipment are calculated based on the pure static load pressure curve. Based on the comprehensive risk index of the equipment status and the effective stress change characteristics, the health status level of the equipment is determined and the corresponding level of early warning instruction is triggered.

[0014] Optionally, the time-series pressure dataset is subjected to time-frequency transformation processing to extract low-frequency component signals characterizing the influence of mechanical vibration and high-frequency component signals characterizing the impact of rock fracture, specifically including:

[0015] Wavelet packet decomposition was performed on each single-channel time-series stress data in the time-series stress dataset to obtain sub-band nodes covering different frequency ranges;

[0016] At least one first sub-band node whose frequency range covers the first preset frequency band is selected from the sub-band nodes, and the coefficients of the selected first sub-band nodes are subjected to wavelet inverse transform to reconstruct the low-frequency component signals of all sampling time points. The first preset frequency band corresponds to the mechanical vibration frequency band generated by the tunneling machinery construction.

[0017] At least one second sub-band node whose frequency range covers the second preset frequency band is selected from the sub-band nodes, and the coefficients of the selected second sub-band nodes are subjected to wavelet inverse transform to reconstruct the high-frequency component signals of all sampling time points; wherein, the second preset frequency band corresponds to the frequency band of high-frequency impact signals generated by rock strata fracturing.

[0018] Optionally, the spectral energy distribution characteristic parameters are calculated based on the energy ratio of the low-frequency components to the high-frequency components, specifically including:

[0019] Within a preset time window, the amplitude of each sampling point of the low-frequency component signal within the current time window is nonlinearly amplified and then accumulated to obtain the low-frequency energy parameter corresponding to the current time window. Similarly, the amplitude of each sampling point of the high-frequency component signal within the current time window is nonlinearly amplified and then accumulated to obtain the high-frequency energy parameter corresponding to the current time window.

[0020] The ratio of the high-frequency energy parameter and the low-frequency energy parameter corresponding to the current time window is calculated to generate the spectral energy distribution characteristic parameter corresponding to the current time window.

[0021] The time window of a preset length is moved according to a preset step size to obtain the spectral energy distribution characteristic parameters corresponding to different time windows;

[0022] The nonlinear amplification process involves squaring the amplitude.

[0023] Optionally, based on the jump point sequence of spectral energy distribution characteristic parameters, combined with the theoretical operating cycle parameters of the tunneling machinery, the confidence level of the mechanical attribute characterizing whether the signal source has mechanical periodicity is calculated, specifically including:

[0024] Based on the spectral energy distribution characteristic parameters corresponding to different time windows, their mean and standard deviation are calculated, and a judgment threshold is generated based on the mean and standard deviation.

[0025] The time points when the characteristic parameters of the spectral energy distribution are greater than the judgment threshold are marked as jump points, and the jump point time series is obtained.

[0026] The rated speed parameter and the equivalent impact structure number parameter are obtained from the theoretical operating cycle parameters of the tunneling machine, and the ratio of the two is determined as the theoretical mechanical impact cycle.

[0027] Calculate the time interval between adjacent jump points in the jump point time series, and calculate the deviation of each time interval from the theoretical mechanical impact period;

[0028] The confidence level of mechanical properties is calculated based on the average level of deviation across all time intervals and the consistency of fluctuations among the deviation levels. The average level of deviation is negatively correlated with the confidence level of mechanical properties, while the consistency of fluctuations is positively correlated with the confidence level of mechanical properties.

[0029] Optionally, the pure static load pressure curve after removing high-frequency interference is obtained based on the low-frequency component signal within this time window, specifically including:

[0030] Extract high-frequency component signals within the current time window;

[0031] The high-frequency component signal is subtracted from the time-series pressure data within the current time window to obtain preliminary denoised pressure data.

[0032] Obtain the pressure value at the last time point of the adjacent time window before the current time window, and the pressure value at the first time point of the adjacent time window after the current time window, as interpolation endpoints;

[0033] Using the interpolation endpoints as boundaries, the preliminary denoised pressure data within the current time window is interpolated to reconstruct a pure static load pressure curve, which reflects the actual rock load variation borne by the support equipment.

[0034] Optionally, by integrating spectral energy distribution characteristic parameters and mechanical attribute confidence levels, a comprehensive equipment condition risk index is generated, specifically including:

[0035] The ratio of the spectral energy distribution characteristic parameter of the current time window to the dynamic background threshold is calculated, and the ratio is logarithmically compressed to obtain the energy transition contribution value. The spectral energy distribution characteristic parameter and the energy transition contribution value are positively correlated.

[0036] The deviation between the confidence level of the mechanical properties in the current time window and the preset ratio of the mechanical background benchmark value is calculated. Based on the deviation, an exponential transformation is performed to obtain the physical property adjustment coefficient. The physical property adjustment coefficient is positively correlated with the deviation.

[0037] The energy transition contribution value is multiplied by the physical property adjustment coefficient to generate the comprehensive risk index of the equipment status corresponding to the current time window.

[0038] Optionally, the effective stress variation characteristics of the support equipment are calculated based on the pure static load pressure curve, specifically including:

[0039] For each transition point within the current time window, obtain the pure static load pressure curve within the first time window adjacent to the transition point, calculate its average value as the reference pressure value of the window before the transition point, obtain the pure static load pressure curve within the second time window adjacent to the transition point, calculate its average value as the reference pressure value of the window after the transition point.

[0040] The difference between the reference pressure values ​​of the front window and the reference pressure values ​​of the rear window is determined as the effective stress increment parameter at the jump point.

[0041] The jump point where the effective stress increment parameter is greater than the preset stress increment threshold is marked as a real rock stratum fracture event;

[0042] The number of jump points marked as real rock strata fracture events within the current time window is counted and used as the cumulative frequency parameter of jump points;

[0043] Calculate the pressure rise rate parameter per unit time based on the pure static load pressure curve of the current time window.

[0044] The pressure rise rate parameter and the cumulative frequency parameter of the jump point are determined as the effective stress change characteristics.

[0045] Optionally, based on the comprehensive risk index of equipment condition and the characteristics of effective stress changes, the equipment health status level is determined and a corresponding level of early warning instruction is triggered, specifically including:

[0046] The rate risk factor is obtained by calculating the ratio of the pressure rise rate parameter in the effective stress change characteristics to the preset maximum safe pressure rise rate.

[0047] The frequency risk factor is obtained by calculating the ratio of the cumulative frequency parameter of the jump points in the effective stress change characteristics to the preset rock stratum fracture frequency limit.

[0048] The rate risk factor and frequency risk factor are accumulated, and a negative exponential transformation is performed based on the accumulation result to obtain the comprehensive probability of the equipment's unhealthy state.

[0049] If the overall probability of the equipment being in an unhealthy state is greater than the preset probability threshold, the support equipment is determined to be in an unhealthy state and an early warning command is triggered; otherwise, the support equipment is determined to be in a healthy state and no early warning command is triggered.

[0050] Optionally, a decision threshold is generated based on the mean and standard deviation, specifically including:

[0051] The decision threshold is obtained by linearly combining the mean and standard deviation.

[0052] This invention proposes a mine equipment condition monitoring system based on the Industrial Internet of Things (IIoT), including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of the mine equipment condition monitoring method based on the Industrial Internet of Things.

[0053] The beneficial effects of the technical solution of the present invention are:

[0054] In this embodiment of the invention, low-frequency and high-frequency component signals are extracted from the time-series pressure dataset through time-frequency transformation processing. The spectral energy distribution characteristic parameters are calculated based on the energy ratio of the two components, and the mechanical attribute confidence level is calculated in conjunction with the theoretical operating cycle parameters of the tunneling machinery, thereby accurately identifying the mechanical periodicity of the signal source. When the mechanical attribute confidence level is higher than a preset confidence threshold, the current signal is determined to be mechanical construction interference, and a pure static load pressure curve is reconstructed based on the low-frequency component signal, effectively eliminating high-frequency interference components. Furthermore, the spectral energy distribution characteristic parameters and mechanical attribute confidence level are integrated to generate a comprehensive risk index for equipment status. Simultaneously, the effective stress change characteristics are calculated based on the pure static load pressure curve. Based on the comprehensive analysis of both, the equipment health status level is determined, and corresponding early warning commands are triggered. Compared to existing technologies, this method can accurately distinguish between mechanical construction interference and actual rock fracture signals, significantly reducing the false alarm rate, achieving multi-dimensional comprehensive assessment and graded early warning of the support equipment status, and improving the reliability and intelligence level of mine safety monitoring. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are 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.

[0056] Figure 1 A flowchart illustrating a method for monitoring the condition of mining equipment based on the Industrial Internet of Things (IIoT) according to an embodiment of the present invention;

[0057] Figure 2 This is a structural diagram of a mine equipment condition monitoring system based on the Industrial Internet of Things, provided as an embodiment of the present invention. Detailed Implementation

[0058] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a mine equipment status monitoring method based on the Industrial Internet of Things proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0059] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0060] The following description, in conjunction with the accompanying drawings, details a specific scheme for a mine equipment status monitoring method based on the Industrial Internet of Things provided by this invention.

[0061] This invention provides a method and system for monitoring the condition of mine equipment based on the Industrial Internet of Things (IIoT). Please refer to [link / reference]. Figure 1 The diagram illustrates a flowchart of a method for monitoring the condition of mining equipment based on the Industrial Internet of Things (IIoT) according to an embodiment of the present invention. The method includes the following steps:

[0062] S101. Obtain time-series pressure data collected by multi-channel high-frequency pressure sensors deployed on mine support equipment, and construct a time-series pressure dataset.

[0063] In this embodiment, multi-channel high-frequency pressure sensors are deployed on the hydraulic support group at the mine's longwall or tunneling face. The sensors are installed at a preset interval, specifically one pressure sensor is installed in the lower cavity of each hydraulic support column, or one sensor is installed every three supports according to the support density requirements, forming a pressure monitoring array covering the entire working face.

[0064] All sensors are connected to the edge computing gateway via industrial Ethernet, forming a monitoring network based on the Industrial Internet of Things (IIoT). The edge computing gateway uses Precision Time Protocol (PTP / NTP) to synchronize the time of all sensors in the array at the millisecond level, ensuring time consistency of data from all channels.

[0065] During tunneling machine operation, all sensors maintain a high-frequency sampling mode, with a sampling frequency set to 1000Hz, meaning one pressure data point is collected every millisecond. The edge computing gateway aligns and encapsulates sensor data from different physical locations within the same time period, constructing a time-series pressure dataset containing both temporal and spatial dimensions. This dataset is stored in the form of a multi-channel time-series matrix, where row indices correspond to sampling time points, column indices correspond to different sensor installation locations, and matrix elements represent the pressure values ​​at the corresponding time points and locations.

[0066] S102. Perform time-frequency transformation processing on the time-series pressure dataset to extract the low-frequency component signal characterizing the influence of mechanical vibration and the high-frequency component signal characterizing the impact of rock fracture, and calculate the spectral energy distribution characteristic parameters based on the energy ratio of the low-frequency component signal and the high-frequency component signal.

[0067] In this embodiment, the time-series pressure dataset is subjected to time-frequency transformation processing to extract low-frequency component signals characterizing the influence of mechanical vibration and high-frequency component signals characterizing the impact of rock fracture. Specifically, this includes:

[0068] Wavelet packet decomposition was performed on each single-channel time-series stress data in the time-series stress dataset to obtain sub-band nodes covering different frequency ranges;

[0069] At least one first sub-band node whose frequency range covers the first preset frequency band is selected from the sub-band nodes, and the coefficients of the selected first sub-band nodes are subjected to wavelet inverse transform to reconstruct the low-frequency component signals of all sampling time points. The first preset frequency band corresponds to the mechanical vibration frequency band generated by the tunneling machinery construction.

[0070] At least one second sub-band node whose frequency range covers the second preset frequency band is selected from the sub-band nodes, and the coefficients of the selected second sub-band nodes are subjected to wavelet inverse transform to reconstruct the high-frequency component signals of all sampling time points; wherein, the second preset frequency band corresponds to the frequency band of high-frequency impact signals generated by rock strata fracturing.

[0071] The characteristic parameters of spectral energy distribution are calculated based on the energy ratio of low-frequency components to high-frequency components, specifically including:

[0072] Within a preset time window, the amplitude of each sampling point of the low-frequency component signal within the current time window is nonlinearly amplified and then accumulated to obtain the low-frequency energy parameter corresponding to the current time window. Similarly, the amplitude of each sampling point of the high-frequency component signal within the current time window is nonlinearly amplified and then accumulated to obtain the high-frequency energy parameter corresponding to the current time window.

[0073] The ratio of the high-frequency energy parameter and the low-frequency energy parameter corresponding to the current time window is calculated to generate the spectral energy distribution characteristic parameter corresponding to the current time window.

[0074] The time window of a preset length is moved according to a preset step size to obtain the spectral energy distribution characteristic parameters corresponding to different time windows;

[0075] The nonlinear amplification process involves squaring the amplitude.

[0076] For example, when using time-series pressure data from support equipment for rock fracture early warning, the construction disturbances generated by the tunnel boring machine's cutting head breaking the rock often overlap with the rock fracture signals, severely interfering with the identification of true precursor information. Therefore, after obtaining the multi-channel synchronous time-series pressure matrix, feature extraction of the signals is required to distinguish between load signals from two different sources.

[0077] The mechanical vibration load generated during tunnel boring machine construction and the impact release load generated by rock fracturing show significant differences in the evolution logic of pressure signals:

[0078] The first category is the disturbance characteristics of tunneling machine construction, which belongs to interference signals. When the tunneling machine's cutting head is breaking rock, the pressure signal exhibits obvious periodic pulsations due to the cyclic cutting of the cutting teeth. In the time domain, this signal presents as equally spaced envelope fluctuations; in the frequency domain, its energy is highly concentrated near the operating frequency of the cutting motor and its harmonics, such as 20Hz to 100Hz and its harmonic range, exhibiting typical narrowband characteristics.

[0079] The second category consists of rock strata fracturing or roof instability characteristics, which fall under the category of target signals. Rock strata fracturing (such as microseismic events or delamination) is a process of instantaneous stress release, manifesting as a sudden pulse in pressure values. Its rise is extremely steep, typically reaching its peak within milliseconds, followed by a significant stress redistribution process. The pressure value cannot return to its initial level but instead forms a new step. Unlike the narrow-band characteristics of mechanical vibrations, rock strata fracturing signals exhibit broadband characteristics, especially showing a significant energy surge in the high-frequency range above 500Hz—a physical characteristic that mechanical rotation cannot simulate.

[0080] The time-series pressure data of a single support channel is extracted from the multi-channel time-series matrix as the analysis object. To balance the response speed of real-time monitoring with the ability to capture transient features, this embodiment uses a sliding time window of 100ms to extract the time-series pressure data of each single channel. The physical significance of this window length is to lock in the instantaneous features of pressure changes within a millisecond time scale, and to avoid the elimination of weak pulse signals induced by rock fractures through long-term averaging processing via short-time window analysis.

[0081] The sliding window step size is set to 50ms, with a 50% overlap between adjacent analysis windows. The technical effect of this redundant sampling mechanism is to ensure that any sudden stress release signal in the rock strata, even if it happens to occur at the window boundary, can be completely preserved in at least one analysis sequence, thereby maintaining the continuity and integrity of the behavioral analysis.

[0082] Let x(t) be the set of discrete pressure data points within the current window. To accurately pinpoint the mechanical vibration frequency band and the high-frequency impact frequency band, the number of wavelet packet decomposition layers is determined based on the sampling frequency of 1000Hz. Wavelet packet decomposition is then performed on the single-channel time-series pressure data within the current window to obtain multiple sub-band nodes covering different frequency ranges.

[0083] After wavelet packet decomposition, at least one first sub-band node covering a first preset frequency band is selected from the sub-band nodes. This first preset frequency band corresponds to the frequency band of mechanical vibration generated during tunneling machinery construction, specifically the low-frequency range from 0Hz to 100Hz, and is usually located in the leftmost branch of the wavelet packet decomposition tree. The coefficients of the selected first sub-band nodes are then subjected to inverse wavelet transform to reconstruct the low-frequency component signals corresponding to all sampling time points. This signal, stripped of high-frequency noise, purely reflects the mechanical vibration envelope transmitted to the hydraulic support during the rotation and rock-breaking process of the tunneling machine's cutting head.

[0084] Simultaneously, at least one second sub-band node whose frequency range covers a second preset frequency band is selected from the sub-band nodes. This second preset frequency band corresponds to the frequency band of high-frequency impact signals generated by rock strata fracturing, specifically the highest frequency band obtainable after wavelet packet decomposition, usually located in the rightmost branch of the wavelet packet decomposition tree. The coefficients of the selected second sub-band nodes are then subjected to inverse wavelet transform to reconstruct the high-frequency component signals corresponding to all sampling time points. The signal mainly contains components of sudden high-frequency impacts generated by rock strata fracturing.

[0085] Optionally, in this embodiment, the length and step size of the sliding time window can be adjusted according to the actual sampling frequency, or set according to specific monitoring requirements and operating conditions. The 100ms window length and 50ms sliding step size used in the above embodiment are only preferred implementation methods, used to adapt to a 1000Hz sampling frequency and achieve a 50% overlap rate, and do not constitute specific limitations in practical applications. Those skilled in the art can adjust the window length and sliding step size accordingly based on changes in the sampling frequency, the required frequency resolution, and the response requirements of real-time monitoring.

[0086] Furthermore, the number of wavelet packet decomposition layers, n, also depends on the sampling frequency and the target frequency band division requirements. In this embodiment, to accurately separate the mechanical vibration frequency band from 0Hz to 100Hz with the highest frequency impact frequency band, the corresponding number of decomposition layers was selected based on the sampling frequency of 1000Hz. In practical applications, if the sampling frequency changes, or if different target frequency band ranges need to be extracted, the number of decomposition layers should also be adjusted accordingly to ensure that the target frequency band can be completely included in the corresponding sub-frequency band nodes. Those skilled in the art can determine the appropriate number of wavelet packet decomposition layers based on the sampling theorem and the target frequency band width.

[0087] Furthermore, in this embodiment, the low-frequency energy parameter The calculation formula can be:

[0088]

[0089] Where N represents the total number of sampling points in the time window.

[0090] High-frequency energy parameters The calculation formula can be:

[0091]

[0092] Where N represents the total number of sampling points in the time window.

[0093] Spectral energy distribution characteristic parameters The calculation formula can be:

[0094]

[0095] in, This indicates a very small number set to prevent the denominator from being 0.

[0096] In this embodiment, the window length =100ms, sampling frequency =1000Hz, therefore This means that each time window contains 100 sampling points. Furthermore, the physical significance of using square operations for accumulation lies in quantifying the "energy" of the signal. During normal cutting by the tunneling machine, the mechanical vibration energy is continuously concentrated in the low-frequency component signal, manifesting as... It remains at a high level; however, when rock strata fracture, energy is instantly injected into the high-frequency component signal within a very short time (such as one or two milliseconds), leading to... An explosive growth occurs. The squaring operation can significantly amplify such sudden, minute impacts, ensuring that the early, weak signals of rock strata fracturing are not drowned out by background noise.

[0097] Optionally, the squaring operation used in this embodiment as a specific implementation of nonlinear amplification processing is only a preferred embodiment. In practical applications, those skilled in the art can select other nonlinear amplification processing methods, such as fourth-power operations or exponential operations, according to the signal characteristics to adapt to different signal dynamic range and sensitivity requirements. Similarly, an anti-overflow constant... The specific value can be adjusted according to the actual calculation accuracy and the signal background noise level; no specific restrictions are imposed here.

[0098] S103. Based on the jump point sequence of spectral energy distribution characteristic parameters, combined with the theoretical operating cycle parameters of the tunneling machinery, calculate the confidence level of the mechanical attribute characterizing whether the signal source has mechanical periodicity.

[0099] In this embodiment, based on the jump point sequence of spectral energy distribution characteristic parameters and combined with the theoretical operating cycle parameters of the tunneling machinery, the confidence level of the mechanical attribute characterizing whether the signal source has mechanical periodicity is calculated, specifically including:

[0100] Based on the spectral energy distribution characteristic parameters corresponding to different time windows, their mean and standard deviation are calculated, and a judgment threshold is generated based on the mean and standard deviation.

[0101] The time points when the characteristic parameters of the spectral energy distribution are greater than the judgment threshold are marked as jump points, and the jump point time series is obtained.

[0102] The rated speed parameter and the equivalent impact structure number parameter are obtained from the theoretical operating cycle parameters of the tunneling machine, and the ratio of the two is determined as the theoretical mechanical impact cycle.

[0103] Calculate the time interval between adjacent jump points in the jump point time series, and calculate the deviation of each time interval from the theoretical mechanical impact period;

[0104] The confidence level of mechanical properties is calculated based on the average level of deviation across all time intervals and the consistency of fluctuations among the deviation levels. The average level of deviation is negatively correlated with the confidence level of mechanical properties, while the consistency of fluctuations is positively correlated with the confidence level of mechanical properties.

[0105] The decision threshold is generated based on the mean and standard deviation, specifically including:

[0106] The decision threshold is obtained by linearly combining the mean and standard deviation.

[0107] For example, in an industrial IoT monitoring environment, the proportion of high-frequency energy... The surge in [something] does not directly equate to a rock strata fracturing event. Metal collisions cause [something]. The pseudo-jump is essentially due to the spectral characteristics of physical shocks.

[0108] After completing a cutting cycle, the tunneling machine needs to advance forward or adjust the position of the scraper conveyor at the rear of the machine. During this process, the metal casing of the tunneling machine or the steel joints of the hydraulic lines are highly susceptible to rubbing against the U-shaped steel supports or metal support netting already installed on the roadway sidewalls. Even if the rock strata are stable at this time, the collision energy is concentrated in the high-frequency band, affecting the spectral energy distribution characteristics calculated by the system. False jumps similar to rock layer delamination can still occur.

[0109] However, the tunneling machine cutting head, as a high-speed rotating body, if... The jump is caused by metal sweep, which necessarily exhibits strict periodicity on the time axis. Based on this physical characteristic, this method establishes a sliding first-in-first-out buffer with a capacity of 50 frames. The sequence was analyzed, and the time span corresponding to this buffer was approximately 2 seconds. The rationale for selecting this length is that a 2-second time span can cover at least one complete rotation cycle of the tunnel boring machine's cutting head under low-speed conditions, thereby ensuring the accuracy of the extraction of periodic feature parameters; this duration conforms to the statistical update frequency of the adaptive behavior benchmark, and can offset the background noise drift caused by changes in geological conditions in real time.

[0110] Within this sliding buffer, by extracting By identifying local peak points of the indicator, a time series of jump points is constructed, and the time interval between adjacent jump points is calculated to determine the source of the signal's mechanical properties.

[0111] First, establish a decision threshold to identify transition points. Decision Threshold Based on the current buffer The statistical distribution of the indicators is generated in real time, and the calculation formula is as follows:

[0112]

[0113] in, and These represent the current 50 frames in the buffer. The mean and standard deviation are calculated. The mean represents the average vibration level under the current tunneling conditions. When the tunneling machine is cutting smoothly in the coal seam, this value is relatively stable; when entering the hard rock area, the background noise increases, and the mean shifts upward smoothly, thus dynamically updating the system's baseline. The standard deviation defines the normal fluctuation range caused by mechanical vibration. When the mechanical impact is large, its value increases accordingly, automatically raising the judgment threshold to ensure that the system does not generate false alarms due to normal mechanical vibration.

[0114] Will Time points exceeding the threshold are marked as transition points, and a time series of these transition points is constructed. The time interval between adjacent transition points is then calculated. To verify whether this interval exhibits mechanical periodicity, a mechanical theoretical period deviation coefficient is introduced. (That is, the degree of deviation between the time interval and the theoretical mechanical impact period):

[0115]

[0116] in, Indicates the rated speed of the cutting head. It represents the equivalent number of impact teeth or blades, that is, the number of cutting teeth or blades on the cutting head that are distributed in the same circumferential phase and are most likely to produce strong high-frequency impacts. This represents the theoretical mechanical impact period. This deviation coefficient is used to measure the degree of agreement between the measured time interval and the theoretical period. This indicates the time interval between adjacent transition points.

[0117] Obtain the time series of jump points within the current 50-frame sliding window. Then, the volatility consistency of the time series at the jump point is calculated. ,in, The standard deviation of the time series at the jump point. This is the mean of the time series at the point of transition.

[0118] When the fluctuation consistency of the time series of jump points is small, it indicates that the signal exhibits high consistency. That is, each jump point not only conforms to the theoretical period, but its deviation from the theoretical value is also extremely stable. This physically corresponds to the mechanical impact generated by the cutting head of the tunneling machine at a constant speed.

[0119] Confidence level of mechanical properties The calculation formula can be:

[0120]

[0121] in, The measured time interval is the same as the theoretical mechanical impact period. When the measured time interval perfectly matches the theoretical period... As it approaches 0 and the pace becomes extremely stable, at this point... Approaching 0, Approaching 1; when the signal is random and relatively noisy, It will decay to below 0.1.

[0122] S104. If the confidence level of the mechanical attribute corresponding to the current time window is higher than the preset confidence threshold, then the time-series pressure data in the current time window is determined to be a mechanical construction interference signal, and the pure static load pressure curve after removing high-frequency interference is obtained based on the low-frequency component signal in the time window.

[0123] In this embodiment, the pure static load pressure curve after removing high-frequency interference is obtained based on the low-frequency component signal within the time window, specifically including:

[0124] Extract high-frequency component signals within the current time window;

[0125] The high-frequency component signal is subtracted from the time-series pressure data within the current time window to obtain preliminary denoised pressure data.

[0126] Obtain the pressure value at the last time point of the adjacent time window before the current time window, and the pressure value at the first time point of the adjacent time window after the current time window, as interpolation endpoints;

[0127] Using the interpolation endpoints as boundaries, the preliminary denoised pressure data within the current time window is interpolated to reconstruct a pure static load pressure curve, which reflects the actual rock load variation borne by the support equipment.

[0128] Optionally, in this embodiment, the calculation method for the pre-set confidence threshold can be: during the no-load operation of the tunneling machinery, collect confidence samples of mechanical attributes from multiple time windows and calculate their statistical mean.

[0129] For example, the confidence level of the mechanical attributes in the current time window is compared with the mechanical background benchmark value (a preset confidence threshold). When the confidence level of the mechanical attributes is lower than 50% of the mechanical background benchmark value, it is determined that the current periodicity is disrupted, meaning the signal no longer has obvious mechanical periodic characteristics. This 50% ratio is set to allow sufficient tolerance for non-stationary vibrations generated by the tunneling machine in complex rock formations, such as instantaneous load fluctuations.

[0130] Optionally, the specific values ​​used in this embodiment, such as the coefficient of the judgment threshold, 3, the buffer capacity of 50 frames, and the baseline ratio of 50%, are all preferred implementations. In practical applications, those skilled in the art can adjust parameters such as the buffer capacity, dynamic threshold coefficient, and baseline ratio according to the sampling frequency, tunneling machine model, geological conditions, and real-time monitoring response requirements; no specific limitations are imposed here.

[0131] If the time-series pressure data within the current time window is determined to be a mechanical construction interference signal, this embodiment does not simply discard the original data within that window. This is because these high-frequency spikes are interference components superimposed on the static load pressure of the support; directly removing them would cause "holes" in the data on the time axis, disrupting the continuity and integrity of the monitoring data.

[0132] Therefore, this embodiment reconstructs the signal based on the low-frequency component signal within this time window to obtain a clean static load pressure curve after removing high-frequency interference. The specific implementation process is as follows:

[0133] First, extract the high-frequency component signal within the current time window. This high-frequency component signal is identified as the mechanically periodic interference component, mainly including periodic high-frequency burrs generated by the rotation of the tunneling machine's cutting head and metal sweeping.

[0134] Subsequently, the high-frequency component signal is subtracted from the original time-series pressure data within the current time window to obtain preliminary denoised pressure data. This step removes high-frequency interference with mechanical periodicity from the original signal while retaining the low-frequency static load pressure component.

[0135] To eliminate potential data discontinuities or edge distortions introduced by subtraction operations, interpolation is further employed to correct the initial denoised pressure data. Specifically, the pressure value at the last time point of the adjacent time window before the current time window and the pressure value at the first time point of the adjacent time window after the current time window are obtained and used as interpolation endpoints. Using the pressure trends of adjacent windows as constraints, interpolation is performed on the initial denoised pressure data within the current time window to reconstruct a continuous, smooth, and pure static load pressure curve.

[0136] Through the above reconstruction process, the periodic high-frequency burrs superimposed on the static load pressure are effectively removed, restoring the true rock load variation curve borne by the support. This pure static load pressure curve reflects the actual rock load variation borne by the support equipment, providing an accurate data basis for subsequent effective stress variation characteristic calculation and risk assessment.

[0137] Optionally, the interpolation method used in this embodiment can be linear interpolation or spline interpolation. The specific interpolation method can be selected according to the data characteristics and smoothing requirements. Those skilled in the art can adjust the interpolation method according to the actual application scenario, and no specific limitations are made here.

[0138] S105. By integrating the characteristic parameters of the spectral energy distribution with the confidence level of mechanical properties, a comprehensive risk index for equipment status is generated.

[0139] In this embodiment, the comprehensive risk index of equipment status is generated by fusing spectral energy distribution characteristic parameters and mechanical property confidence levels, specifically including:

[0140] The ratio of the spectral energy distribution characteristic parameter of the current time window to the dynamic background threshold is calculated, and the ratio is logarithmically compressed to obtain the energy transition contribution value. The spectral energy distribution characteristic parameter and the energy transition contribution value are positively correlated.

[0141] The deviation between the confidence level of the mechanical properties in the current time window and the preset ratio of the mechanical background benchmark value is calculated. Based on the deviation, an exponential transformation is performed to obtain the physical property adjustment coefficient. The physical property adjustment coefficient is positively correlated with the deviation.

[0142] The energy transition contribution value is multiplied by the physical property adjustment coefficient to generate the comprehensive risk index of the equipment status corresponding to the current time window.

[0143] For example, within each time window, the comprehensive risk index of the equipment status is calculated, and the calculation formula can be:

[0144]

[0145] in, To determine the threshold, The energy transition contribution value is used to measure the spectral energy distribution characteristics of the current time window. Compared to the degree to which the judgment threshold is exceeded, the technical effect of using natural logarithm compression is that the dynamic range of signals in industrial settings is extremely large, and when a violent impact occurs, It can reach tens of times the background value. Logarithmic operations can prevent numerical explosion while maintaining sufficient sensitivity to small early anomalies, ensuring that early and weak risk signals are not ignored.

[0146] This represents the mechanical background baseline value (preset confidence threshold). This represents the physical property adjustment coefficient, used to adjust the confidence level of mechanical properties based on the current time window. The risk index is dynamically adjusted. When the value is relatively large, meaning the current signal exhibits obvious mechanical periodicity, the calculation result within the parentheses becomes negative, and the exponential function value decreases rapidly, thus suppressing the comprehensive risk index. The physical meaning is that although the energy transition contribution value of the current window is large, the signal rhythm is too stable, conforming to the periodic characteristics of mechanical interference. Therefore, it should be judged as construction noise such as cutting head friction or metal sweeping, rather than a real risk of rock strata fracturing.

[0147] Multiply the above two items to generate the comprehensive risk index of equipment status corresponding to the current time window. This index achieves a joint determination of "impact intensity" and "source attribute," taking into account both the degree of high-frequency energy transition and introducing a signal periodicity adjustment factor, which can effectively distinguish between mechanical construction interference and real rock fracture signals.

[0148] Optionally, the method used in this embodiment is... As a benchmark for the confidence level of mechanical properties, and a combination of the natural logarithm and the exponential function, both are preferred implementation methods. In practical applications, those skilled in the art can adjust the coefficients of the benchmark and the functional form of the nonlinear transformation according to the monitoring sensitivity requirements, signal dynamic range, and field operating conditions; no specific limitations are imposed here.

[0149] S106. Calculate the effective stress change characteristics of the support equipment based on the pure static load pressure curve, and determine the equipment health status level and trigger the corresponding level of early warning instruction based on the comprehensive risk index of equipment status and the effective stress change characteristics.

[0150] In this embodiment, the effective stress variation characteristics of the support equipment are calculated based on the pure static load pressure curve, specifically including:

[0151] For each transition point within the current time window, obtain the pure static load pressure curve within the first time window adjacent to the transition point, calculate its average value as the reference pressure value of the window before the transition point, obtain the pure static load pressure curve within the second time window adjacent to the transition point, calculate its average value as the reference pressure value of the window after the transition point.

[0152] The difference between the reference pressure values ​​of the front window and the reference pressure values ​​of the rear window is determined as the effective stress increment parameter at the jump point.

[0153] The jump point where the effective stress increment parameter is greater than the preset stress increment threshold is marked as a real rock stratum fracture event;

[0154] The number of jump points marked as real rock strata fracture events within the current time window is counted and used as the cumulative frequency parameter of jump points;

[0155] Calculate the pressure rise rate parameter per unit time based on the pure static load pressure curve of the current time window.

[0156] The pressure rise rate parameter and the cumulative frequency parameter of the jump point are determined as the effective stress change characteristics.

[0157] Based on the comprehensive risk index of equipment condition and the characteristics of effective stress changes, the equipment health status level is determined and corresponding early warning instructions are triggered, specifically including:

[0158] The rate risk factor is obtained by calculating the ratio of the pressure rise rate parameter in the effective stress change characteristics to the preset maximum safe pressure rise rate.

[0159] The frequency risk factor is obtained by calculating the ratio of the cumulative frequency parameter of the jump points in the effective stress change characteristics to the preset rock stratum fracture frequency limit.

[0160] The rate risk factor and frequency risk factor are accumulated, and a negative exponential transformation is performed based on the accumulation result to obtain the comprehensive probability of the equipment's unhealthy state.

[0161] If the overall probability of the equipment being in an unhealthy state is greater than the preset probability threshold, the support equipment is determined to be in an unhealthy state and an early warning command is triggered; otherwise, the support equipment is determined to be in a healthy state and no early warning command is triggered.

[0162] For example, after determining that the signal is a genuine surrounding rock risk, this embodiment establishes two time windows to extract effective stress increment parameters. The first time window is 100ms before the jump moment t on the time axis, i.e., the first time window adjacent to the jump point. The second time window is 200ms to 300ms after the jump moment on the time axis, ensuring that the high-frequency impact energy has been fully released.

[0163] For each transition point within the current time window, obtain the pure static load pressure curve within the first time window adjacent to the transition point, and calculate its average value as the reference pressure value of the preceding window for that transition point; obtain the pure static load pressure curve within the second time window adjacent to the transition point, and calculate its average value as the reference pressure value of the following window for that transition point; determine the difference between the reference pressure value of the preceding window and the reference pressure value of the following window as the effective stress increment parameter of that transition point.

[0164] If the effective stress increment parameter is close to 0, it indicates that the impact is merely vibration and has not caused permanent changes to the rock structure. If the effective stress increment parameter is significantly greater than 0, it indicates that the roof rock layer has undergone physical displacement or fracture, causing the support to bear a new, permanent load increment, which is the effective stress increment. Jump points where the effective stress increment parameter exceeds a preset stress increment threshold are marked as actual rock rupture events. The number of jump points marked as actual rock rupture events within the current time window is counted and used as the cumulative frequency parameter for jump points.

[0165] Simultaneously, based on the pure static load pressure curve of the current time window, the pressure rise rate parameter per unit time is calculated. The calculation formula can be:

[0166]

[0167] in, This represents the change in the pure static load pressure curve per unit time. This represents the duration of the current time window. This indicator reflects the urgency of the roof subsidence pressing on the support structure.

[0168] Jump point cumulative frequency parameter The total number of transition points that are identified as real rock strata fracturing events within the current 2-second sliding buffer is counted, reflecting the intensity of rock strata fracturing activity.

[0169] Subsequently, the two parameters were standardized to obtain the rate risk factor. The calculation formula can be:

[0170]

[0171] and frequency risk factors The calculation formula can be:

[0172]

[0173] in, This refers to the maximum safe pressure increase rate allowed by the support frame, which is the technical parameter specified by the manufacturer. This represents the limit of the frequency of rock strata fracturing per unit time.

[0174] The rate risk factor and frequency risk factor are summed, and a negative exponential transformation is performed based on the summation result to obtain the comprehensive probability of the equipment's unhealthy state. The calculation formula can be:

[0175]

[0176] The higher the probability value, the faster the equipment load increases, and the higher the risk of failure.

[0177] In the tiered early warning process, the overall probability of an unhealthy equipment state is compared with a preset probability threshold. In this embodiment, the preset probability threshold is set to 0.6. When the overall probability of an unhealthy equipment state is less than 0.6, the support equipment is determined to be in a normal load-bearing state or under construction fluctuation, which is considered a normal healthy state, and no early warning command is triggered. When the overall probability of an unhealthy equipment state is greater than or equal to 0.6, the support equipment is determined to be in an unhealthy state, and an early warning command is triggered, including audible and visual alarms, equipment shutdown, or personnel evacuation commands.

[0178] The 0.6 threshold is set based on the fact that when the cumulative effect of the two risk indicators reaches "unit load", that is... hour, The calculated value is 0.63. The rate risk factor and frequency risk factor together occupy 100% of the system's design safety capacity, indicating that the equipment has just reached the edge of the composite safety envelope. At this point, triggering the warning command has a clear physical meaning.

[0179] Optionally, the specific values ​​used in this embodiment, such as the 100ms pre-trigger window, the 200-300ms stabilization window, the 2-second sliding buffer, and the 0.6 probability threshold, as well as parameters such as the preset stress increment threshold, the maximum safe pressure increase rate, and the rock strata fracture frequency limit, are all preferred implementations. In practical applications, those skilled in the art can adjust the above parameters accordingly based on factors such as on-site working conditions, equipment parameters, and safety requirements; no specific limitations are imposed here.

[0180] In summary, in this embodiment of the invention, low-frequency and high-frequency component signals are extracted from the time-series pressure dataset through time-frequency transformation processing. The spectral energy distribution characteristic parameters are calculated based on the energy ratio of the two components, and the mechanical attribute confidence level is calculated in conjunction with the theoretical operating cycle parameters of the tunneling machinery, thereby accurately identifying the mechanical periodicity of the signal source. When the mechanical attribute confidence level is higher than a preset confidence threshold, the current signal is determined to be mechanical construction interference, and a pure static load pressure curve is reconstructed based on the low-frequency component signal, effectively eliminating high-frequency interference components. Furthermore, the spectral energy distribution characteristic parameters and the mechanical attribute confidence level are integrated to generate a comprehensive equipment status risk index. Simultaneously, effective stress increment parameters, pressure rise rate parameters, and cumulative frequency parameters of jump points are calculated based on the pure static load pressure curve as effective stress change characteristics. The equipment health status level is comprehensively determined based on the comprehensive equipment status risk index and the effective stress change characteristics, and corresponding early warning commands are triggered. Compared with existing technologies, this method can accurately distinguish between mechanical construction interference and real rock strata fracture signals, significantly reducing the false alarm rate, realizing multi-dimensional comprehensive assessment and graded early warning of the status of support equipment, and preserving the integrity of real load data through signal reconstruction technology, thereby improving the reliability and intelligence level of mine safety monitoring.

[0181] This invention also proposes a mine equipment condition monitoring system based on the Industrial Internet of Things (IIoT). Please refer to [link / reference]. Figure 2 The diagram shows a structural diagram of a mine equipment status monitoring system based on the Industrial Internet of Things provided in an embodiment of the present invention. The system includes: a data acquisition module 101, a data processing module 102, and an anomaly warning module 103.

[0182] Data acquisition module 101 is used to acquire time-series pressure data collected by multi-channel high-frequency pressure sensors deployed on mine support equipment and construct a time-series pressure dataset;

[0183] Data processing module 102 is used to perform time-frequency transformation processing on time-series pressure dataset, extract low-frequency component signals characterizing the influence of mechanical vibration and high-frequency component signals characterizing rock fracture impact, and calculate spectral energy distribution characteristic parameters based on the energy ratio of low-frequency component signals to high-frequency component signals.

[0184] Based on the jump point sequence of spectral energy distribution characteristic parameters, combined with the theoretical operating cycle parameters of the tunneling machinery, the confidence level of the mechanical attribute characterizing whether the signal source has mechanical periodicity is calculated.

[0185] If the confidence level of the mechanical property corresponding to the current time window is higher than the preset confidence threshold, then the time-series pressure data in the current time window is determined to be a mechanical construction interference signal, and the pure static load pressure curve after removing high-frequency interference is obtained based on the low-frequency component signal in the time window.

[0186] By integrating spectral energy distribution characteristic parameters with mechanical property confidence levels, a comprehensive risk index for equipment status is generated.

[0187] The abnormal warning module 103 is used to calculate the effective stress change characteristics of the support equipment based on the pure static load pressure curve, and determine the health status level of the equipment and trigger the corresponding level of warning instruction according to the comprehensive risk index of the equipment status and the effective stress change characteristics.

[0188] It should be noted that the system provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the computer equipment can be divided into different functional modules to complete all or part of the functions described above. In addition, the mine equipment condition monitoring system based on industrial Internet of Things and the mine equipment condition monitoring method based on industrial Internet of Things provided in the above embodiments belong to the same concept, and their specific implementation process can be found in the method embodiment, which will not be repeated here.

[0189] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0190] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0191] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for monitoring the condition of mining equipment based on the Industrial Internet of Things, characterized in that, include: Acquire time-series pressure data collected by multi-channel high-frequency pressure sensors deployed on mine support equipment, and construct a time-series pressure dataset; The time-frequency transformation process is performed on the time-series pressure dataset to extract the low-frequency component signal characterizing the influence of mechanical vibration and the high-frequency component signal characterizing the impact of rock fracture. The spectral energy distribution characteristic parameters are calculated based on the energy ratio of the low-frequency component signal to the high-frequency component signal. Based on the jump point sequence of spectral energy distribution characteristic parameters, combined with the theoretical operating cycle parameters of the tunneling machinery, the confidence level of the mechanical attribute characterizing whether the signal source has mechanical periodicity is calculated. If the confidence level of the mechanical property corresponding to the current time window is higher than the preset confidence threshold, then the time-series pressure data in the current time window is determined to be a mechanical construction interference signal, and the pure static load pressure curve after removing high-frequency interference is obtained based on the low-frequency component signal in the time window. By integrating spectral energy distribution characteristic parameters with mechanical property confidence levels, a comprehensive risk index for equipment status is generated. The effective stress change characteristics of the support equipment are calculated based on the pure static load pressure curve. Based on the comprehensive risk index of the equipment status and the effective stress change characteristics, the health status level of the equipment is determined and the corresponding level of early warning instruction is triggered.

2. The method for monitoring the condition of mine equipment based on the Industrial Internet of Things according to claim 1, characterized in that, The process of performing time-frequency transformation on the time-series pressure dataset to extract low-frequency component signals characterizing the influence of mechanical vibration and high-frequency component signals characterizing the impact of rock fracture specifically includes: Wavelet packet decomposition was performed on each single-channel time-series stress data in the time-series stress dataset to obtain sub-band nodes covering different frequency ranges; At least one first sub-band node whose frequency range covers the first preset frequency band is selected from the sub-band nodes, and the coefficients of the selected first sub-band nodes are subjected to wavelet inverse transform to reconstruct the low-frequency component signals of all sampling time points. The first preset frequency band corresponds to the mechanical vibration frequency band generated by the tunneling machinery construction. At least one second sub-band node whose frequency range covers the second preset frequency band is selected from the sub-band nodes, and the coefficients of the selected second sub-band nodes are subjected to wavelet inverse transform to reconstruct the high-frequency component signals of all sampling time points; wherein, the second preset frequency band corresponds to the frequency band of high-frequency impact signals generated by rock strata fracturing.

3. The method for monitoring the condition of mine equipment based on the Industrial Internet of Things according to claim 1, characterized in that, The calculation of spectral energy distribution characteristic parameters based on the energy ratio of low-frequency components to high-frequency components specifically includes: Within a preset time window, the amplitude of each sampling point of the low-frequency component signal within the current time window is nonlinearly amplified and then accumulated to obtain the low-frequency energy parameter corresponding to the current time window. Similarly, the amplitude of each sampling point of the high-frequency component signal within the current time window is nonlinearly amplified and then accumulated to obtain the high-frequency energy parameter corresponding to the current time window. The ratio of the high-frequency energy parameter and the low-frequency energy parameter corresponding to the current time window is calculated to generate the spectral energy distribution characteristic parameter corresponding to the current time window. The time window of a preset length is moved according to a preset step size to obtain the spectral energy distribution characteristic parameters corresponding to different time windows; The nonlinear amplification process involves squaring the amplitude.

4. The method for monitoring the condition of mine equipment based on the Industrial Internet of Things according to claim 1, characterized in that, The jump point sequence based on spectral energy distribution characteristic parameters, combined with the theoretical operating cycle parameters of the tunneling machinery, is used to calculate the confidence level of the mechanical attribute characterizing whether the signal source has mechanical periodicity. Specifically, this includes: Based on the spectral energy distribution characteristic parameters corresponding to different time windows, their mean and standard deviation are calculated, and a judgment threshold is generated based on the mean and standard deviation. The time points when the characteristic parameters of the spectral energy distribution are greater than the judgment threshold are marked as jump points, and the jump point time series is obtained. The rated speed parameter and the equivalent impact structure number parameter are obtained from the theoretical operating cycle parameters of the tunneling machine, and the ratio of the two is determined as the theoretical mechanical impact cycle. Calculate the time interval between adjacent jump points in the jump point time series, and calculate the deviation of each time interval from the theoretical mechanical impact period; The confidence level of mechanical properties is calculated based on the average level of deviation across all time intervals and the consistency of fluctuations among the deviation levels. The average level of deviation is negatively correlated with the confidence level of mechanical properties, while the consistency of fluctuations is positively correlated with the confidence level of mechanical properties.

5. The method for monitoring the condition of mine equipment based on the Industrial Internet of Things according to claim 1, characterized in that, The process of obtaining the pure static load pressure curve after removing high-frequency interference based on the low-frequency component signal within this time window specifically includes: Extract high-frequency component signals within the current time window; The high-frequency component signal is subtracted from the time-series pressure data within the current time window to obtain preliminary denoised pressure data. Obtain the pressure value at the last time point of the adjacent time window before the current time window, and the pressure value at the first time point of the adjacent time window after the current time window, as interpolation endpoints; Using the interpolation endpoints as boundaries, the preliminary denoised pressure data within the current time window is interpolated to reconstruct a pure static load pressure curve, which reflects the actual rock load variation borne by the support equipment.

6. The method for monitoring the condition of mine equipment based on the Industrial Internet of Things according to claim 1, characterized in that, The fusion of spectral energy distribution characteristic parameters and mechanical attribute confidence levels generates a comprehensive equipment status risk index, specifically including: The ratio of the spectral energy distribution characteristic parameter of the current time window to the dynamic background threshold is calculated, and the ratio is logarithmically compressed to obtain the energy transition contribution value. The spectral energy distribution characteristic parameter and the energy transition contribution value are positively correlated. The deviation between the confidence level of the mechanical properties in the current time window and the preset ratio of the mechanical background benchmark value is calculated. Based on the deviation, an exponential transformation is performed to obtain the physical property adjustment coefficient. The physical property adjustment coefficient is positively correlated with the deviation. The energy transition contribution value is multiplied by the physical property adjustment coefficient to generate the comprehensive risk index of the equipment status corresponding to the current time window.

7. The method for monitoring the condition of mine equipment based on the Industrial Internet of Things according to claim 1, characterized in that, The effective stress variation characteristics of the support equipment calculated based on the pure static load pressure curve specifically include: For each transition point within the current time window, obtain the pure static load pressure curve within the first time window adjacent to the transition point, calculate its average value as the reference pressure value of the window before the transition point, obtain the pure static load pressure curve within the second time window adjacent to the transition point, calculate its average value as the reference pressure value of the window after the transition point. The difference between the reference pressure values ​​of the front window and the reference pressure values ​​of the rear window is determined as the effective stress increment parameter at the jump point. The jump point where the effective stress increment parameter is greater than the preset stress increment threshold is marked as a real rock stratum fracture event; The number of jump points marked as real rock strata fracture events within the current time window is counted and used as the cumulative frequency parameter of jump points; Calculate the pressure rise rate parameter per unit time based on the pure static load pressure curve of the current time window. The pressure rise rate parameter and the cumulative frequency parameter of the jump point are determined as the effective stress change characteristics.

8. The method for monitoring the condition of mine equipment based on the Industrial Internet of Things according to claim 1, characterized in that, The process of determining the equipment health status level and triggering corresponding early warning commands based on the comprehensive risk index of equipment status and effective stress change characteristics specifically includes: The rate risk factor is obtained by calculating the ratio of the pressure rise rate parameter in the effective stress change characteristics to the preset maximum safe pressure rise rate. The frequency risk factor is obtained by calculating the ratio of the cumulative frequency parameter of the jump points in the effective stress change characteristics to the preset rock stratum fracture frequency limit. The rate risk factor and frequency risk factor are accumulated, and a negative exponential transformation is performed based on the accumulation result to obtain the comprehensive probability of the equipment's unhealthy state. If the overall probability of the equipment being in an unhealthy state is greater than the preset probability threshold, the support equipment is determined to be in an unhealthy state and an early warning command is triggered; otherwise, the support equipment is determined to be in a healthy state and no early warning command is triggered.

9. A method for monitoring the condition of mine equipment based on the Industrial Internet of Things according to claim 4, characterized in that, The step of generating the judgment threshold based on the mean and standard deviation specifically includes: The decision threshold is obtained by linearly combining the mean and standard deviation.

10. A mine equipment condition monitoring system based on the Industrial Internet of Things, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the computer program is executed by the processor, it implements the steps of a method for monitoring the condition of mining equipment based on the Industrial Internet of Things as described in any one of claims 1-9.

Images