A mine rigid guide fault diagnosis method decoupled from multiple source factors

Abstract

This invention discloses a multi-source factor decoupling method for fault diagnosis of rigid guideways in mines, belonging to the field of equipment fault diagnosis technology. This invention separates the dynamic oscillation component from the wire rope tension signal and quantifies its intensity; by comparing the measured oscillation level with a normal operating condition benchmark, a stationary reliability index is obtained to evaluate the reliability of the tension data; stationary sample points unaffected by oscillation contamination are selected for load back-calculation to obtain a reliable real-time load sequence; finally, an adjustment weight is jointly constructed using the real-time load level and the stationary reliability index to adaptively correct the fault judgment threshold, making the diagnostic threshold dynamically change with the load state, thus achieving accurate diagnosis and location of guideway faults. This invention overcomes the technical difficulties of high false alarm and false negative rates in fixed threshold diagnosis methods under variable load conditions, improving the accuracy and reliability of fault diagnosis for rigid guideways in mines.

Description

Technical Field

[0001] This application relates to the field of equipment fault diagnosis technology, and in particular to a fault diagnosis method for rigid mine tunnels with multi-source factor decoupling. Background Technology

[0002] Mine hoisting systems mainly consist of hoists, wire ropes, guide rails, hoisting containers, and shafts. Rigid guide rails, composed of guide rails laid along the shaft depth, constrain the hoisting container's vertical movement within the shaft. During long-term service, rigid guide rails are prone to faults such as joint misalignment, localized bending, wear, and loose bolts, which can lead to serious accidents like container jamming or falling. Therefore, accurate and reliable online fault diagnosis of rigid guide rails is of paramount importance for ensuring the safe operation of mine hoisting systems.

[0003] Existing fault diagnosis methods for rigid conveyor systems mostly employ vibration detection. This involves using accelerometers mounted on the lifting vessel to collect longitudinal vibration signals as it runs along the rigid conveyor, extracting time-domain or frequency-domain features, and comparing these signals with preset fixed thresholds to determine if a fault exists. However, under varying load conditions, fixed threshold diagnostic methods suffer from several drawbacks. If the threshold is set too low, normal vibration fluctuations under light loads can easily exceed it, resulting in numerous false alarms. Conversely, if the threshold is set too high, weak impacts caused by actual defects in the conveyor under heavy loads may be missed because they fail to reach the threshold. While some existing methods acknowledge the load interference problem, they typically employ simple static correction strategies, which are insufficient to effectively handle the changes in vibration response characteristics caused by dynamic loads in actual operating conditions. Therefore, they cannot fundamentally solve the problem of high false alarm and missed detection rates under varying load conditions. Summary of the Invention

[0004] Therefore, it is necessary to provide a fault diagnosis method for rigid mine tunnels that decouples multiple factors to address the above-mentioned technical problems.

[0005] The present invention adopts the following technical solution: This invention provides a fault diagnosis method for rigid mine tunnels based on multi-source factor decoupling, comprising: Acquire time-synchronized and spatially aligned wire rope tension signals, longitudinal vibration signals of the hoisting container, and hoisting displacement signals; Dynamic tension components are extracted from the wire rope tension signal. Based on a sliding analysis window set along the lifting direction, the instantaneous oscillation intensity of each sample point is calculated according to the dynamic tension components of each sample point within the window. The instantaneous oscillation intensity characterizes the degree of strong oscillation in the wire rope. The instantaneous oscillation intensity is compared with the oscillation benchmark under no-load low-speed uniform working condition to obtain the stability confidence of each sample point, which represents the degree of stable vibration of the wire rope. Stable sample points are selected based on the stability confidence. The real-time load is calculated based on the tension data of the stable sample points. For non-stable sample points, the real-time load of the nearest adjacent stable sample point is used to obtain a real-time load sequence that is not contaminated by the oscillation of the wire rope. An adjustment weight is constructed based on the real-time load of each sample point and the stability confidence level. The preset basic fault threshold is then adaptively adjusted point by point using the adjustment weight to obtain a dynamic fault threshold that changes with the load state. Based on the amplitude of the longitudinal vibration signal of the container, a vibration anomaly factor reflecting the fault characteristic level at each sample point is obtained. The vibration anomaly factor is compared with the dynamic fault threshold. The effective fault section is confirmed based on the number of consecutive anomaly points. The effective fault section is mapped to the wellbore depth coordinates based on the lifting displacement signal, and the fault location of the tank passage is output.

[0006] Preferably, the calculation of the instantaneous oscillation intensity of each sample point includes: Centered on each sample point, a sliding analysis window with a fixed spatial length along the lifting direction is taken, and the dynamic tension components of all sample points within the sliding analysis window, as well as the average value of the absolute values ​​of the dynamic tension components of all sample points, are obtained. Calculate the ratio of the absolute value of the dynamic tension component of each sample point within the window to the average of the absolute values ​​of the dynamic tension components of all sample points, and take the arithmetic mean of the ratios of all sample points within the window. Use the arithmetic mean as the instantaneous oscillation intensity of the sample point at the center of the window.

[0007] Preferably, the method for obtaining the oscillation reference includes: Using no-load and low-speed uniform conditions as a reference condition, the full-length tension data of the wire rope during multiple runs of the mine hoist under the reference condition are obtained. The instantaneous oscillation intensity of each sample point is calculated based on each run, and the instantaneous oscillation intensity of the same sample point during multiple runs under the reference condition is used as multiple oscillation benchmarks for that sample point.

[0008] Preferably, obtaining the stationary confidence level of each sample point includes: The instantaneous oscillation intensity of each sample point is subtracted from the average value of multiple oscillation benchmarks corresponding to that sample point to obtain the difference. The ratio of the difference to the upper limit of the oscillation benchmark is calculated and normalized to obtain the stability confidence. The stability confidence characterizes the degree to which the wire rope at that sample point is in stable vibration.

[0009] Preferably, the method for obtaining the upper limit of the oscillation reference includes: The upper limit of the oscillation benchmark is the product of the average value of the oscillation benchmark and the preset tolerance coefficient and the standard deviation of the oscillation benchmark.

[0010] Preferably, the construction of the adjustment weights includes: The ratio of the real-time load to the rated maximum lift load at each sample point is used as the load factor. The adjustment weight is obtained by mapping the product of the load factor and the stability confidence through an inverse proportional normalization function.

[0011] Preferably, the calibration method for the preset basic fault threshold includes: The hoisting system is carried through the entire shaft at a constant speed under no-load, low-speed, and stable wire rope conditions, and the entire guideway is in good and defect-free condition. The standard working condition vibration anomaly factor of all sample points under this standard working condition is obtained, and the standard working condition vibration anomaly factor is used as the vibration anomaly benchmark. The average value of the vibration anomaly benchmark is added to the product of the preset margin coefficient and the standard deviation of the vibration anomaly benchmark, and the sum is used as the preset basic fault threshold.

[0012] Preferably, obtaining the dynamic fault threshold that varies with load state includes: The product of the preset margin coefficient and the standard deviation of the vibration anomaly benchmark is used as the benchmark fluctuation amount; The average value of the vibration anomaly benchmark, plus the benchmark fluctuation amount, plus the product of the adjustment weight and the benchmark fluctuation amount, are used as the dynamic fault threshold.

[0013] Preferably, obtaining the vibration anomaly factor includes: A sliding diagnostic window is preset along the lifting direction. The maximum absolute value of the longitudinal vibration data of the lifting container in each sliding diagnostic window is obtained as the peak value. The root mean square value of the amplitude of all longitudinal vibration data in the window is obtained as the effective value. The quotient obtained by dividing the peak value by the effective value is used as the vibration anomaly factor of the center sample point of the window.

[0014] Preferably, the step of confirming the effective fault section based on the number of consecutive abnormal points includes: A preset number of valid faults is set. When the number of abnormal points with vibration abnormality factors greater than or equal to the dynamic fault threshold along the lifting direction reaches the preset number of valid faults, the existence of a fault in the tank passage is confirmed, and the actual location and coverage of the fault are determined based on the lifting displacement data of all abnormal points.

[0015] The above-mentioned technical solution adopted in this invention can achieve the following beneficial effects: In the multi-source factor decoupling fault diagnosis method for rigid mine hoisting systems provided by this invention, a unified spatiotemporal reference multidimensional signal data stream is first established to ensure strict alignment of the wire rope tension signal, the longitudinal vibration signal of the hoisting container, and the hoisting displacement information in the time and space dimensions. Dynamic oscillation components are separated from the original tension signal, and the instantaneous oscillation intensity of each sample point is quantified through a sliding analysis window. Based on this, the measured instantaneous oscillation intensity is compared with the statistical characteristics under no-load reference conditions to obtain a stable reliability index characterizing the reliability of the tension data. Based on this, stable sample points uncontaminated by wire rope oscillations are selected for load back-calculation to obtain a reliable real-time load sequence. An adjustment weight is constructed using the real-time load level and the stable reliability index to adaptively correct the preset basic fault threshold, making the fault judgment threshold dynamically change with the load state. The vibration anomaly factor of the hoisting container is compared with the dynamic threshold to achieve accurate diagnosis and location of hoisting system faults.

[0016] This invention achieves deep decoupling between load and vibration, enabling the fault determination threshold to adapt to different load conditions. Under light load conditions, it effectively suppresses false alarms caused by normal vibration fluctuations, and under heavy load conditions, it ensures that weak impact characteristics are not missed. This overcomes the technical challenge of high false alarm and missed detection rates in fixed threshold diagnostic methods under variable load conditions, and effectively improves the accuracy and reliability of fault diagnosis in mine rigid guideways. Attached Figure Description

[0017] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0018] Figure 1 This document presents a flowchart illustrating a multi-source factor decoupling method for diagnosing faults in rigid mine tunnels. Figure 2 This is a schematic diagram of the lifting and lowering directions provided in this manual; Figure 3 This is a schematic diagram illustrating the process of obtaining real-time load based on wire rope tension, as provided in this manual. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this application will be clearly and completely described below with reference to specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments in the specification without creative effort are within the scope of protection of this application.

[0020] In existing technologies, fault diagnosis of rigid hoisting systems in mines mainly employs vibration detection methods. This involves installing accelerometers to collect longitudinal vibration signals from the hoisting container, extracting time-domain or frequency-domain features, and comparing them with fixed thresholds to determine if faults such as joint misalignment, bending, or wear exist. However, in actual hoisting operations, the load varies significantly depending on the loading amount for each run and is also affected by the mass of the tail rope. Load variations alter the system's natural frequency, thus affecting the response characteristics of the vibration signal used to detect hoisting system faults. While existing diagnostic methods acknowledge the load interference problem, they typically employ simple static correction strategies, such as using the no-load signal as a reference for differential processing or normalizing the vibration amplitude using fixed coefficients. These static correction strategies are ill-suited to effectively handle the changes in vibration response characteristics caused by dynamic loads in actual operating conditions. Under heavy load conditions, the system vibration is more stable, and static correction may miss weak impact characteristics, leading to false alarms or missed faults. Under light load conditions, the system vibration is more intense, and static correction can easily misjudge normal vibrations as faults, seriously affecting the safe operation of the mine hoisting system.

[0021] The technical solutions provided by the various embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] Figure 1 This is a flowchart illustrating a multi-source factor decoupling method for fault diagnosis of rigid mine guideways, as described in this specification. The method specifically includes the following steps: S101: Acquire time-synchronized and spatially aligned wire rope tension signals, longitudinal vibration signals of the hoisting container, and hoisting displacement signals.

[0023] To perform correlation analysis on the stress state and vibration response of the hoisting system at the same time and location in the wellbore, it is necessary to obtain time-synchronized and spatially consistent information on wire rope tension, longitudinal vibration of the hoisting container, and motion of the hoisting system. To avoid reduced load decoupling accuracy and chaotic fault location caused by time mismatch or positional misalignment of various signals, this step establishes a multi-dimensional signal data stream with unified spatiotemporal reference.

[0024] Specifically, multi-channel data is collected and labeled with a unified timestamp tag. The multi-channel data includes at least: Wire rope tension data: Install wire rope tension sensors at the main tie rod or wedge rope ring suspension device of the hoisting container to directly acquire the time sequence of the total axial tension signal of the wire rope on the hoisting container, in Newtons.

[0025] Longitudinal vibration data of the lifting container: A triaxial accelerometer is installed on the lower surface of the top plate of the lifting container to collect the time sequence of triaxial vibration acceleration signals of the lifting container, in meters per square second. This invention mainly focuses on analyzing the longitudinal vibration response characteristics of the lifting container due to defects in the tank passage, and therefore mainly obtains the longitudinal vibration signal of the lifting container.

[0026] Well depth data during hoisting process: Using an absolute encoder installed at the main shaft end of the hoist drum, a real-time position pulse signal is generated to obtain the displacement time sequence of the hoisting container, in meters.

[0027] Furthermore, based on the actual lifting or lowering process of the mine hoist during operation, the starting position of the lifting or lowering process is calibrated as the zero point, with the lifting direction as the positive direction of the displacement data, such as... Figure 2 The diagram shows the lifting and lowering directions. It should be noted that the steps in this embodiment are symmetrically applicable to both the lifting and lowering processes. This embodiment describes the lifting process as an example. The lifting process is resampled into several equidistant sample points on the lifting coordinate axis using a preset fixed step size. Here, the fixed step size is 0.02 meters as an example, which can be adjusted according to the specific implementation situation. This embodiment does not impose a specific limitation on this step size.

[0028] For sample points where actual collected data is missing, interpolation algorithms such as cubic spline interpolation are used to interpolate the tension and vibration data at each sample point from nearby points where actual collected data exists. This results in spatiotemporally aligned multi-channel data. After alignment, each sample point contains the following data: the sampling timestamp of the sample point, the lifting displacement data of the sample point, the wire rope tension data of the sample point, and the longitudinal vibration data of the lifting container of the sample point.

[0029] Thus, we have acquired multidimensional signal data and established a unified spatiotemporal reference for the multidimensional signal data stream, providing a data foundation for subsequent analysis.

[0030] S102: Extract the dynamic tension component from the wire rope tension signal, and calculate the instantaneous oscillation intensity of each sample point based on the dynamic tension component of each sample point in the window, according to the sliding analysis window set along the lifting direction. The instantaneous oscillation intensity characterizes the degree of strong oscillation in the wire rope.

[0031] It should be noted that the measured tension of the wire rope in the mine hoisting system during the operation of the hoisting system is not only due to the static load component generated by the self-weight of the hoisting container and the material load, but also due to factors such as the periodic longitudinal vibration caused by the uneven rotation and roundness error of the hoisting drum, the axial impact caused by the inertial force of the container's acceleration and deceleration, the transverse vortex-induced vibration of the wire rope itself, and the parametric resonance generated by the coupling with the sheave and the derrick structure. These factors generate tension-relaxation fluctuations inside the wire rope, which directly result in a significant dynamic fluctuation component superimposed on the measured tension signal.

[0032] If the tension signal is used directly to calculate the load increase without separation, the dynamic oscillation component will be incorrectly converted into changes in material load, resulting in significant high-frequency fluctuations and distortions in the load estimate. This, in turn, leads to deviations in the adaptive adjustment of tank duct fault diagnosis based on dynamic load. Therefore, this step extracts the dynamic tension component and quantifies its oscillation intensity.

[0033] Specifically, the total tension sequence of the wire rope is filtered using a zero-phase low-pass digital filter with an extremely low cutoff spatial frequency. The filtered signal is then processed using acceleration feedforward compensation. The processed result is recorded as the static tension sequence. Here, the cutoff spatial frequency should be chosen to be much lower than the spatial frequency corresponding to the periodic wavelength of the gap in the guide joint or the bending, to ensure that only signals that change slowly in space and reflect the gradual changes in gravity load and suspended weight are retained. The digital low-pass filtering method and acceleration feedforward compensation are well-known techniques and will not be described in detail in this embodiment.

[0034] After obtaining the static tension sequence, for each sample point on the lifting coordinate axis, the difference between the original wire rope tension value collected at that sample point and the corresponding static tension value is the dynamic tension component of that sample point.

[0035] Furthermore, a sliding analysis window with a fixed length along the lifting direction is defined. The spatial length of the sliding analysis window is set to 25 spatial sampling points. In this embodiment, the corresponding actual physical length is usually 0.5 meters. It can be adjusted according to the typical spatial scale of wire rope oscillation and the physical interval of sampling points in the actual application scenario. This embodiment does not make a specific limitation on this. When the sliding analysis window is taken with any sampling point as the center, if the number of sampling points on one side of the sampling point is insufficient, the maximum number of sampling points that can be taken is taken. At this time, although the sliding analysis window is not symmetrical according to the sampling point, the sampling point is still recorded as the center of the sliding analysis window.

[0036] The instantaneous oscillation intensity of the sample point corresponding to the center of the sliding analysis window is obtained by considering the oscillation level of the dynamic tension components of all sample points within the sliding analysis window centered on each sample point. Taking the k-th sample point as the center of the sliding analysis window as an example (hereinafter referred to as the k-th sliding analysis window), the instantaneous oscillation intensity of the k-th sample point is... The calculation methods include:

[0037] (1) in, This represents the total number of sample points contained in the k-th sliding analysis window. This represents the dynamic tension component at the h-th sample point in the k-th sliding analysis window. This represents the average of the absolute values ​​of the dynamic tension components at all sample points.

[0038] It should be noted that the average value of the sliding analysis window is used here to represent the instantaneous oscillation intensity of the sample points. This can provide a more robust estimate of the oscillation level in the presence of occasional large-amplitude pulses, avoiding misjudgment of the overall operating status of the wire rope due to a single abnormal event. The greater the instantaneous oscillation intensity, the stronger the oscillation of the wire rope. In this case, a large amount of energy in the signal collected by the tension sensor comes from dynamic disturbances rather than actual load changes. If this is directly used for load back calculation, it will produce significant errors.

[0039] The static tension components and instantaneous oscillation intensities of all sample points in the current diagnostic process are obtained. The static tension components provide the data basis for subsequent load back calculation, and the instantaneous oscillation intensities reflect the reliability of the tension data of the sample points.

[0040] S103: Compare the instantaneous oscillation intensity with the oscillation benchmark under no-load, low-speed, uniform operating conditions to obtain the stability confidence of each sample point, which represents the degree of stable vibration of the wire rope. Based on the stability confidence, select stable sample points, calculate the real-time load based on the tension data of the stable sample points, and use the real-time load of the nearest adjacent stable sample point for non-stable sample points to obtain a real-time load sequence that is not contaminated by wire rope oscillation.

[0041] It should be noted that since the tension signal collected by the wire rope may contain significant dynamic fluctuation components, if the tension signal of the entire stroke is used for load back calculation without discrimination, the calculated load curve will be superimposed with false high-frequency fluctuations caused by the oscillation of the wire rope. This makes it impossible to identify the true load state at the sample points affected by the dynamic fluctuation components. When making adaptive adjustments for tank tunnel fault diagnosis based on load, the stability of fault diagnosis will be affected by load jumps.

[0042] Therefore, to achieve high-precision fault diagnosis, a clean and accurate real-time load sequence is required. This step uses instantaneous oscillation intensity as the signal quality evaluation criterion, performing load inverse calculation in high-confidence sections; while in sections with severe oscillations, the load values ​​are maintained using the most recent reliable inverse calculation results. This constructs a high-quality real-time load sequence that accurately reflects the load change trend without being contaminated by wire rope oscillations. Figure 3 The diagram shows a process for obtaining real-time load based on wire rope tension.

[0043] Specifically, taking no-load and low-speed uniform speed as reference conditions, the full tension data of the wire rope of the mine hoist during multiple runs under the reference conditions are obtained, and the instantaneous oscillation intensity at each sample point during multiple no-load runs under the reference conditions is calculated based on formula (1), and used as multiple oscillation benchmarks for each sample point.

[0044] The instantaneous oscillation intensity of the measured tension data at each sample point is compared with the corresponding oscillation benchmark. Based on the comparison results, the stability reliability of each sample point is evaluated, characterizing the degree of stable vibration of the wire rope. The stability reliability of the k-th sample point... The calculation methods include:

[0045] (2) in, This represents the instantaneous oscillation intensity of the k-th sample point. This represents any oscillation reference for the k-th sample point. This represents the preset tolerance coefficient, which is taken here. , This represents the function of taking the average value. Represents a linear normalization function. This represents the standard deviation function; where This represents the average of all oscillation benchmarks for the k-th sample point. This represents taking the standard deviation of all oscillation benchmarks for the k-th sample point, which is used here. Achieve normalization.

[0046] It should be noted that the purpose of multiple no-load and low-speed uniform operation is to eliminate the influence of random disturbances on the benchmark value during a single operation and obtain a statistically representative expected oscillation level.

[0047] A preset confidence threshold is set, here set to 0.25. All sample points with a stationary confidence level greater than the confidence threshold are selected and recorded as stationary sample points; other sample points are recorded as non-stationary sample points. Load inversion is performed based on the tension data of all stationary sample points to obtain the real-time load of each sample point. This real-time load reflects the load on the hoisting system under the influence of non-ideal factors such as material slippage, center of gravity shift, and changes in the swing posture of the hoisting container. The process of load inversion based on wire rope tension is a well-known technique and will not be described in detail in this embodiment. For non-stationary sample points, the real-time load of the nearest preceding stationary sample point is taken as the real-time load of that non-stationary sample point. If no preceding stationary sample point exists, the nearest subsequent stationary sample point is taken.

[0048] Thus, the real-time load of each sample point is obtained. Through the stability and reliability screening mechanism, the decoupling of the wire rope dynamic disturbance and the load back calculation process is realized, avoiding the impact of false load fluctuations caused by wire rope oscillation on the stability of fault detection. Furthermore, the load obtained through back calculation avoids the low reliability of the direct measurement device caused by the harsh downhole environment.

[0049] S104: Based on the real-time load of each sample point and the stability confidence level, an adjustment weight is constructed. The preset basic fault threshold is then adaptively adjusted point by point using the adjustment weight to obtain a dynamic fault threshold that changes with the load state.

[0050] It should be noted that, at the level of vibration signal analysis, the fault diagnosis of rigid mine hoisting systems essentially involves identifying the abnormal impact characteristics generated when the hoisting container passes through defects in the hoisting system and comparing them with a preset fault judgment threshold.

[0051] Ideally, a fault in the hoisting system is determined to exist as long as the vibration and impact index exceeds a fixed and uniform fault threshold. However, in actual mine hoisting operations, the material load on the hoisting container may vary in each hoisting cycle, ranging from empty to fully loaded.

[0052] When the lifting load is small, the contact normal pressure between the can ear and the can guide is insufficient, which easily leads to dynamic gaps. The lifting container will generate significant vibration during operation. If a uniform fixed threshold is used for fault diagnosis, normal vibration fluctuations caused by light loads are very likely to exceed the threshold, resulting in a large number of false alarms. Conversely, when the lifting load is large, the contact stiffness between the can ear and the can guide increases, and the overall vibration of the lifting container tends to be stable. The relative prominence of the actual local defects in the can guide in the overall vibration is weakened due to the reduction of background energy. If a uniform fixed threshold is used for fault diagnosis, some moderately severe fault impacts may be missed because they cannot reach the threshold.

[0053] In addition, the strong oscillation of the wire rope not only affects the accuracy of load back calculation, but also may transmit a large transient force to the lifting container through the suspension device when the wire rope oscillates strongly, which will show an amplitude spike in the vibration signal and also lead to false alarms.

[0054] Therefore, this step is based on the real-time load level of each sample point and combines the stability confidence of each sample point to construct adjustment weights, which are used to dynamically adjust the fixed threshold according to the working conditions to adapt to the vibration characteristics under different working conditions.

[0055] Specifically, the rated maximum lifting load of the current lifting container is obtained, and the real-time load level of the container is reflected by the ratio of the real-time load at each sample point to the rated maximum lifting load of the current lifting container. Based on the real-time load level and stability confidence of each sample point, adjustment weights are constructed to correct the fault judgment threshold in real time. The adjustment weight for the k-th sample point is... The calculation methods include:

[0056] (3) in, This represents the real-time load of the k-th sample point. Rated maximum lifting load, This represents the stationarity confidence level of the k-th sample point. This represents an exponential function with the natural constant as its base, which is used here. The model is then subjected to inverse proportional normalization.

[0057] It should be noted that This value characterizes the load level of the k-th sample point; the closer the value is to 0, the smaller the load. The value reflects the stability of the wire rope oscillation at each sample point, and is used here to avoid false alarms caused by large amplitude spikes when the wire rope oscillates.

[0058] Thus, a dynamic fault threshold is obtained, which is used as the threshold for diagnosing the presence of faults in the tank guide vibration at each sample point, in order to improve the vibration response characteristics of the container under different load conditions.

[0059] S105: Based on the amplitude of the longitudinal vibration signal of the container, obtain the vibration anomaly factor that reflects the fault characteristic level at each sample point, compare the vibration anomaly factor with the dynamic fault threshold, confirm the effective fault section based on the number of consecutive anomaly points, map the effective fault section to the wellbore depth coordinates based on the lifting displacement signal, and output the fault location of the tank passage.

[0060] It should be noted that if there are defects such as misalignment of joints, local bending, or surface peeling at a certain point in the hoistway, the tank lugs will experience a momentary impact when passing through this point. This will generate a short-duration, wide-bandwidth impact pulse train in the longitudinal vibration signal of the hoisting container, with a peak value significantly higher than the vibration background of the surrounding stable sections. This step extracts fault features from the longitudinal vibration signal of the hoisting container and compares them with a dynamic fault threshold constructed by adjusting the weights, thereby achieving online diagnosis and location of faults in the hoistway.

[0061] Specifically, a sliding diagnostic window is preset. The spatial length of the sliding diagnostic window should match the travel required for the lifting system to contact the tank passage and pass through the tank passage defect, so as to be able to completely capture a single impact event. In this embodiment, the spatial length of the sliding diagnostic window is 5 sampling points as an example. In this embodiment, the corresponding physical length is 0.1 meters, which can be adjusted according to the specific implementation. When the sliding diagnostic window is taken with any sampling point as the center, if the number of sampling points on one side of the sampling point is insufficient, then the maximum number of sampling points that can be taken are taken. At this time, although the sliding diagnostic window is not symmetrical with respect to the sampling point, the sampling point is still recorded as the center of the sliding diagnostic window.

[0062] Along the lifting direction, sliding diagnostic windows are sequentially selected, and complete vibration signals within each window are acquired. The peak values ​​of the vibration signals within each window are analyzed to calculate the vibration anomaly factor for each window. For any given sliding diagnostic window, the maximum absolute value of the longitudinal vibration data of the lifting container within that window is taken as the peak value of that window. The root mean square of the amplitudes of all longitudinal vibration data within that window is taken as the effective value of that window. The quotient obtained by dividing the peak value by the effective value is the vibration anomaly factor for that window. The vibration anomaly factor for each sliding diagnostic window is obtained and assigned to the sample point at the center of that window. The vibration anomaly factor reflects the fault characteristic level at each sample point.

[0063] Furthermore, since the longitudinal vibration response of the lifting container varies under different load conditions, a dynamic fault threshold for adaptive real-time load is constructed by calibrating the basic threshold and adjusting the weights, and fault over-limit judgment is performed to achieve fault diagnosis of the tank passage.

[0064] The hoisting system is operated under unloaded, low-speed, and stable wire rope conditions, traversing the entire shaft at a constant speed. Prior to this, the entire guideway is manually inspected and confirmed to be in good condition with no known defects. Under this standard operating condition, the vibration anomaly factor for all sample points is obtained and recorded as the vibration anomaly benchmark for each sample point. This benchmark is used to calibrate the basic threshold for guideway fault diagnosis. Based on the vibration anomaly benchmark and combined with adjusted weights, the dynamic fault threshold is obtained. The dynamic fault threshold for the k-th sample point is... The calculation methods include:

[0065] (4) This formula can also be expressed as: in, This represents the vibration anomaly benchmark for any sample point. This represents the adjusted weight of the k-th sample point. This represents the preset margin coefficient, which is taken here. , This indicates that the average value of the vibration anomaly benchmark is taken from all sample points. This represents the standard deviation of the vibration anomaly benchmark for all sample points.

[0066] It should be noted that This constitutes a basic threshold without considering load differences. Based on this, combined with adjustment weights, a dynamic adjustment amount under the influence of real-time load differences is added to achieve adaptive adjustment of the threshold based on different load states.

[0067] The vibration anomaly factor of each sample point is compared with the dynamic fault threshold. When the vibration anomaly factor is greater than or equal to the dynamic fault threshold, the sample point is marked as an anomaly point, and the lifting displacement data of the sample point is recorded for fault location.

[0068] It should be noted that, in order to eliminate isolated events such as occasional electromagnetic interference or instantaneous foreign object impacts, a spatial continuity verification strategy needs to be introduced. Real tank passage defects have a continuous spatial extension; when the tank lugs pass through, they continuously generate anomalies within several consecutive windows, while random interference manifests as individual, isolated anomaly points.

[0069] Specifically, the preset number of valid faults is 3. When the number of consecutive abnormal points is greater than or equal to the number of valid faults, the existence of the hoisting system fault is confirmed. The actual location and coverage of the fault are determined based on the lifting displacement data of all abnormal points. The timestamp corresponding to the sample point when the existence of the hoisting system fault is confirmed is output, and the length and location of the fault zone are output to guide the maintenance plan of the mine hoisting system.

[0070] Thus, the fault diagnosis of rigid guideways in mines is achieved, and this embodiment is complete.

[0071] based on Figure 1 This paper presents a multi-source factor decoupling fault diagnosis method for rigid hoisting systems in mines. By constructing a stable reliability evaluation mechanism, it filters high-confidence sections from tension signals contaminated by the dynamic oscillation of the wire rope for load back-calculation, obtaining a real-time load sequence that truly reflects the material load variation law, thus completely eliminating the interference of wire rope oscillation on load estimation. Based on this, a dynamic mapping relationship between load state and fault judgment threshold is established, realizing adaptive adjustment of the fault threshold. Under light load conditions, the threshold is appropriately raised to avoid false alarms caused by normal vibration fluctuations; under heavy load conditions, the threshold is appropriately lowered to ensure that weak fault impact characteristics can be effectively captured. This fundamentally overcomes the technical difficulties of high false alarm and missed detection rates in fixed threshold diagnosis methods under variable load conditions, significantly improving the operational safety of mine hoisting systems and providing reliable technical support for condition-based maintenance of rigid hoisting systems.

[0072] When applying the multi-source factor decoupling fault diagnosis method for rigid mine guideways provided in this specification, it is not necessary to consider... Figure 1 The steps shown are executed in sequence. The specific execution order of each step can be determined as needed, and this manual does not impose any restrictions on it.

[0073] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

Claims

1. A fault diagnosis method for rigid mine tunnels based on multi-source factor decoupling, characterized in that, include: Acquire time-synchronized and spatially aligned wire rope tension signals, longitudinal vibration signals of the hoisting container, and hoisting displacement signals; Dynamic tension components are extracted from the wire rope tension signal. Based on a sliding analysis window set along the lifting direction, the instantaneous oscillation intensity of each sample point is calculated according to the dynamic tension components of each sample point within the window. The instantaneous oscillation intensity characterizes the degree of strong oscillation in the wire rope. The instantaneous oscillation intensity is compared with the oscillation benchmark under no-load low-speed uniform working condition to obtain the stability confidence of each sample point, which represents the degree of stable vibration of the wire rope. Stable sample points are selected based on the stability confidence. The real-time load is calculated based on the tension data of the stable sample points. For non-stable sample points, the real-time load of the nearest adjacent stable sample point is used to obtain a real-time load sequence that is not contaminated by the oscillation of the wire rope. An adjustment weight is constructed based on the real-time load of each sample point and the stability confidence level. The preset basic fault threshold is then adaptively adjusted point by point using the adjustment weight to obtain a dynamic fault threshold that changes with the load state. Based on the amplitude of the longitudinal vibration signal of the container, a vibration anomaly factor reflecting the fault characteristic level at each sample point is obtained. The vibration anomaly factor is compared with the dynamic fault threshold. The effective fault section is confirmed based on the number of consecutive anomaly points. The effective fault section is mapped to the wellbore depth coordinates based on the lifting displacement signal, and the fault location of the tank passage is output.

2. The fault diagnosis method for rigid mine guideways with multi-source factor decoupling as described in claim 1, characterized in that, The calculation of the instantaneous oscillation intensity of each sample point includes: Centered on each sample point, a sliding analysis window with a fixed spatial length along the lifting direction is taken, and the dynamic tension components of all sample points within the sliding analysis window, as well as the average value of the absolute values ​​of the dynamic tension components of all sample points, are obtained. Calculate the ratio of the absolute value of the dynamic tension component of each sample point within the window to the average of the absolute values ​​of the dynamic tension components of all sample points, and take the arithmetic mean of the ratios of all sample points within the window. Use the arithmetic mean as the instantaneous oscillation intensity of the sample point at the center of the window.

3. The fault diagnosis method for rigid mine guideways with multi-source factor decoupling as described in claim 1, characterized in that, The method for obtaining the oscillation reference includes: Using no-load and low-speed uniform conditions as a reference condition, the full-length tension data of the wire rope during multiple runs of the mine hoist under the reference condition are obtained. The instantaneous oscillation intensity of each sample point is calculated based on each run, and the instantaneous oscillation intensity of the same sample point during multiple runs under the reference condition is used as multiple oscillation benchmarks for that sample point.

4. The fault diagnosis method for rigid mine guideways with multi-source factor decoupling as described in claim 1, characterized in that, The process of obtaining the stationary confidence level of each sample point includes: The instantaneous oscillation intensity of each sample point is subtracted from the average value of multiple oscillation benchmarks corresponding to that sample point to obtain the difference. The ratio of the difference to the upper limit of the oscillation benchmark is calculated and normalized to obtain the stability confidence. The stability confidence characterizes the degree to which the wire rope at that sample point is in stable vibration.

5. The fault diagnosis method for rigid mine guideways with multi-source factor decoupling as described in claim 4, characterized in that, The method for obtaining the upper limit of the oscillation reference includes: The upper limit of the oscillation benchmark is the product of the average value of the oscillation benchmark and the preset tolerance coefficient and the standard deviation of the oscillation benchmark.

6. The fault diagnosis method for rigid mine guideways with multi-source factor decoupling as described in claim 1, characterized in that, The construction of the adjusted weights includes: The ratio of the real-time load to the rated maximum lift load at each sample point is used as the load factor. The adjustment weight is obtained by mapping the product of the load factor and the stability confidence through an inverse proportional normalization function.

7. The fault diagnosis method for rigid mine guideways with multi-source factor decoupling as described in claim 1, characterized in that, The calibration method for the preset basic fault threshold includes: The hoisting system is carried through the entire shaft at a constant speed under no-load, low-speed, and stable wire rope conditions, and the entire guideway is in good and defect-free condition. The standard working condition vibration anomaly factor of all sample points under this standard working condition is obtained, and the standard working condition vibration anomaly factor is used as the vibration anomaly benchmark. The average value of the vibration anomaly benchmark is added to the product of the preset margin coefficient and the standard deviation of the vibration anomaly benchmark, and the sum is used as the preset basic fault threshold.

8. The fault diagnosis method for rigid mine guideways with multi-source factor decoupling as described in claim 7, characterized in that, The method for obtaining the dynamic fault threshold that varies with load state includes: The product of the preset margin coefficient and the standard deviation of the vibration anomaly benchmark is used as the benchmark fluctuation amount; The average value of the vibration anomaly benchmark, plus the benchmark fluctuation amount, plus the product of the adjustment weight and the benchmark fluctuation amount, are used as the dynamic fault threshold.

9. The fault diagnosis method for rigid mine guideways with multi-source factor decoupling as described in claim 1, characterized in that, The acquisition of vibration anomaly factors includes: A sliding diagnostic window is preset along the lifting direction. The maximum absolute value of the longitudinal vibration data of the lifting container in each sliding diagnostic window is obtained as the peak value. The root mean square value of the amplitude of all longitudinal vibration data in the window is obtained as the effective value. The quotient obtained by dividing the peak value by the effective value is used as the vibration anomaly factor of the center sample point of the window.

10. The method for fault diagnosis of rigid mine guideways with multi-source factor decoupling as described in claim 1, characterized in that, The step of identifying valid fault sections based on the number of consecutive anomalies includes: A preset number of valid faults is set. When the number of abnormal points with vibration abnormality factors greater than or equal to the dynamic fault threshold along the lifting direction reaches the preset number of valid faults, the existence of a fault in the tank passage is confirmed, and the actual location and coverage of the fault are determined based on the lifting displacement data of all abnormal points.

Images