Online monitoring method and system for wear state of bearing of flour mill

Abstract

The invention discloses an online monitoring method and system for the wear state of a bearing of a flour mill, particularly relates to the technical field of state monitoring of industrial equipment, and is used for solving the problem of misjudgment caused by signal characteristic deviation under a dynamic load working condition in an existing flour mill bearing vibration signal monitoring method. A grinding pressure parameter and a bearing vibration signal are collected in real time, the dynamic incidence relation of the grinding pressure parameter and the bearing vibration signal is analyzed, interference frequency band components are separated, the time-frequency ridge line track and curvature sudden change density of a characteristic frequency band are extracted, a dynamic reference standard is generated to adjust an amplitude-frequency threshold value in a self-adaptive mode, and finally a wear state evaluation result is output through amplitude-frequency joint analysis. Accurate monitoring of the wear state of the bearing under the complex working condition is achieved, and the reliability of an online monitoring system and the scientificity of maintenance decision are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of industrial equipment condition monitoring. More specifically, the present invention relates to an online monitoring method and system for the wear state of a flour mill bearing. Background Art

[0002] In the field of flour processing, as a core equipment, the long-term stable operation of the bearings of a flour mill directly affects production efficiency and equipment maintenance costs. At present, for the wear state monitoring of the flour mill bearings, a method combining vibration signal analysis and temperature detection is generally adopted. By collecting the bearing operation data in real time and comparing it with a preset threshold, abnormal early warning is realized. However, in actual working conditions, the flour mill needs to dynamically adjust the grinding pressure according to the raw material characteristics and processing requirements. The frequent change of this process parameter leads to a significant fluctuation in the internal stress distribution of the bearing, and then causes an unexpected shift in the vibration signal characteristics.

[0003] In the prior art, the bearing wear monitoring method based on vibration signals has insufficient adaptability under dynamic load conditions. Due to the strong coupling effect between the grinding pressure adjustment and the bearing stress distribution, the main frequency characteristics and amplitude of the vibration signal will deviate from the normal wear model with the change of process parameters, resulting in a significant increase in the risk of misjudgment. This problem is particularly prominent in industrial scenarios such as flour processing that require frequent load adjustment, seriously affecting the reliability of the online monitoring system. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides an online monitoring method and system for the wear state of a flour mill bearing to solve the problems raised in the above background art.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] An online monitoring method for the wear state of a flour mill bearing includes the following steps:

[0007] S1. Obtain the grinding pressure parameter and the bearing vibration signal during the operation of the flour mill;

[0008] S2. Analyze the corresponding relationship between the change of the grinding pressure parameter and the offset amount of the main frequency characteristics of the bearing vibration signal;

[0009] S3. Separate the interference frequency band components and characteristic frequency band components in the bearing vibration signal according to the offset amount of the main frequency characteristics;

[0010] S4. Perform time-frequency ridge tracking on the characteristic frequency band components, extract the ridge trajectory of the time-frequency energy distribution and calculate the curvature mutation density of the ridge trajectory, and determine the dynamic reference standard of the bearing wear characteristics according to the curvature mutation density;

[0011] S5. Adjust the amplitude threshold and frequency threshold of the bearing vibration signal according to the dynamic reference standard;

[0012] S6. Perform amplitude-frequency joint analysis on the components in the characteristic frequency band based on the adjusted amplitude threshold and frequency threshold, and output the evaluation result of the bearing wear state.

[0013] In a preferred embodiment, obtaining the grinding pressure parameter and the bearing vibration signal during the operation of the mill includes:

[0014] Collect the grinding pressure parameter through a pressure sensor installed on the grinding roller of the mill, and collect the bearing vibration signal through an acceleration sensor installed on the bearing housing of the mill;

[0015] Synchronously sample the grinding pressure parameter and the bearing vibration signal to generate a time-aligned grinding pressure parameter sequence and a bearing vibration signal sequence.

[0016] In a preferred embodiment, the acceleration sensor is arranged in the horizontal and vertical directions of the bearing housing, and the pressure sensor is connected to the hydraulic adjustment mechanism of the grinding roller.

[0017] In a preferred embodiment, analyzing the corresponding relationship between the change of the grinding pressure parameter and the main frequency characteristic offset of the bearing vibration signal includes:

[0018] Generate a dynamic time window based on the time-aligned grinding pressure parameter sequence, and the interval length of the dynamic time window is inversely proportional to the change rate of the grinding pressure parameter;

[0019] Extract the main frequency energy distribution and phase synchronization index of the bearing vibration signal sequence within the dynamic time window;

[0020] According to the joint change trend of the main frequency energy distribution and the phase synchronization index, construct a non-linear mapping rule between the change of the grinding pressure parameter and the main frequency characteristic offset;

[0021] The non-linear mapping rule is used to indicate the confidence interval of the main frequency characteristic offset under dynamic load fluctuations.

[0022] In a preferred embodiment, separating the interference frequency band components and characteristic frequency band components in the bearing vibration signal according to the main frequency characteristic offset includes:

[0023] Determine the dynamic frequency band division boundary based on the main frequency characteristic offset and the non-linear mapping rule;

[0024] Extract the frequency band with an energy ratio lower than the main frequency energy distribution threshold and a negative phase synchronization index in the bearing vibration signal as the interference frequency band component according to the dynamic frequency band division boundary;

[0025] Extract the frequency band with an energy extraction ratio higher than the main frequency energy distribution threshold and a positive phase synchronization index as the characteristic frequency band component;

[0026] Perform complementary filtering on the interference frequency band component and the characteristic frequency band component to generate a separated signal sequence aligned with the dynamic time window.

[0027] In a preferred embodiment, perform time-frequency ridge tracking on the characteristic frequency band component, extract the ridge trajectory of the time-frequency energy distribution, and calculate the curvature mutation density of the ridge trajectory. Determine the dynamic reference standard for bearing wear characteristics according to the curvature mutation density, including:

[0028] Perform time-frequency ridge tracking within the dynamic time window on the time-domain signal of the characteristic frequency band component to generate the ridge trajectory of the time-frequency energy distribution;

[0029] Perform cubic spline interpolation on the ridge trajectory, calculate the curvature change rate of the interpolated trajectory; count the number of mutation points where the curvature change rate exceeds the set mutation threshold within a unit time to generate the curvature mutation density;

[0030] Determine the dynamic reference standard for bearing wear characteristics within the current dynamic time window according to the mapping relationship between the curvature mutation density and the bearing wear degree in the historical wear data;

[0031] The dynamic reference standard includes the normal fluctuation range of the curvature mutation density and the abnormal determination threshold.

[0032] In a preferred embodiment, adjust the amplitude threshold and frequency threshold of the bearing vibration signal according to the dynamic reference standard, including:

[0033] Adjust the amplitude threshold of the bearing vibration signal according to the normal fluctuation range of the curvature mutation density in the dynamic reference standard;

[0034] Adjust the frequency threshold of the bearing vibration signal according to the abnormal determination threshold of the curvature mutation density in the dynamic reference standard;

[0035] Generate a dynamic amplitude-frequency joint determination rule based on the adjusted amplitude threshold and frequency threshold;

[0036] Perform real-time matching of the dynamic amplitude-frequency joint determination rule with the bearing vibration signal within the current dynamic time window, and update the effective interval of the amplitude threshold and frequency threshold.

[0037] In a preferred embodiment, perform amplitude-frequency joint analysis on the characteristic frequency band component based on the adjusted amplitude threshold and frequency threshold, and output the bearing wear state evaluation result, including:

[0038] Calculate the time-domain energy mean value and extract the frequency-domain main frequency offset of the characteristic frequency band component within the dynamic time window;

[0039] Determine whether the time-domain energy mean value exceeds the adjusted amplitude threshold and whether the frequency-domain main frequency offset exceeds the adjusted frequency threshold;

[0040] When both the time-domain energy mean value and the frequency-domain main frequency offset exceed the limit, generate a bearing wear mode according to the combined coding of the exceeded amplitude threshold and frequency threshold;

[0041] Based on the current dynamic time window and the cumulative result of the bearing wear mode in adjacent windows, output the bearing wear state evaluation result.

[0042] In a preferred embodiment, the bearing wear state evaluation result is the proportion of the number of occurrences of the same wear mode within three consecutive dynamic time windows.

[0043] On the other hand, the present invention provides an on-line monitoring system for the bearing wear state of a grinding mill, including the following modules:

[0044] A signal acquisition module, used to obtain the grinding pressure parameter and the bearing vibration signal when the grinding mill is running;

[0045] A dynamic correlation module, used to analyze the corresponding relationship between the change of the grinding pressure parameter and the main frequency characteristic offset of the bearing vibration signal;

[0046] A frequency band separation module, used to separate the interference frequency band component and the characteristic frequency band component in the bearing vibration signal according to the main frequency characteristic offset;

[0047] A curvature analysis module, used to perform time-frequency ridge line tracking on the characteristic frequency band component, extract the ridge line trajectory of the time-frequency energy distribution and calculate the curvature mutation density of the ridge line trajectory, and determine the dynamic reference standard of the bearing wear characteristics according to the curvature mutation density;

[0048] A threshold adjustment module, used to adjust the amplitude threshold and frequency threshold of the bearing vibration signal according to the dynamic reference standard;

[0049] A state evaluation module, used to perform amplitude-frequency joint analysis on the characteristic frequency band component based on the adjusted amplitude threshold and frequency threshold, and output the bearing wear state evaluation result.

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] 1. By dynamically correlating the grinding pressure parameter with the main frequency characteristic offset of the bearing vibration signal, a real-time mapping relationship between load changes and vibration characteristics is constructed, effectively overcoming the misjudgment problem caused by signal offset in traditional monitoring methods under dynamic loads. By separating the interference frequency band components and extracting the time-frequency ridge line trajectory of the characteristic frequency band, the identification accuracy of the wear characteristic components in the vibration signal is significantly improved, avoiding the interference of background noise and process fluctuations on the monitoring results. At the same time, based on the dynamic reference standard generated by the curvature mutation density, it can adapt to the wear characteristic changes under different load conditions, making the threshold adjustment more in line with the actual operating state, thus greatly improving the sensitivity and reliability of the monitoring system.

[0052] 2. Compared with static thresholds and single-signal analysis, the amplitude-frequency joint analysis strategy is adopted, combining the time-domain energy distribution and the frequency-domain harmonic attenuation characteristics to achieve a multi-dimensional comprehensive evaluation of the bearing wear state. By dynamically adjusting the amplitude and frequency thresholds, the system can accurately capture the weak characteristic signals of early wear and maintain a stable decision logic under complex working conditions. This closed-loop monitoring mechanism not only reduces the false alarm rate and missed alarm rate, but also provides a quantifiable decision-making basis for equipment maintenance, significantly extending the bearing service life and reducing the loss of unplanned downtime. Description of the Drawings

[0053] Figure 1 It is a flowchart of an online monitoring method for the wear state of a bearing of a grinding mill according to the present invention;

[0054] Figure 2 It is a schematic structural diagram of an online monitoring system for the wear state of a bearing of a grinding mill according to the present invention. Detailed Embodiments

[0055] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0056] Embodiment 1: Figure 1 An online monitoring method for the wear state of a bearing of a grinding mill according to the present invention is given, which includes the following steps:

[0057] S1. Obtain the grinding pressure parameter and the bearing vibration signal during the operation of the grinding mill;

[0058] S2. Analyze the corresponding relationship between the change of the grinding pressure parameter and the main frequency characteristic offset of the bearing vibration signal;

[0059] S3. Separate the interference frequency band components and characteristic frequency band components in the bearing vibration signal according to the main frequency characteristic offset;

[0060] S4. Perform time-frequency ridge tracking on the characteristic frequency band components, extract the ridge trajectory of the time-frequency energy distribution and calculate the curvature mutation density of the ridge trajectory, and determine the dynamic reference standard for bearing wear characteristics according to the curvature mutation density;

[0061] S5. Adjust the amplitude threshold and frequency threshold of the bearing vibration signal according to the dynamic reference standard;

[0062] S6. Perform amplitude-frequency joint analysis on the characteristic frequency band components based on the adjusted amplitude threshold and frequency threshold, and output the evaluation result of the bearing wear state.

[0063] S1. Obtain the grinding pressure parameter and the bearing vibration signal during the operation of the pulverizer. The specific implementation is as follows:

[0064] Collect the grinding pressure parameter through the pressure sensor installed on the grinding roller of the pulverizer. The pressure sensor is fixed at the flange interface of the oil inlet pipeline of the hydraulic adjustment mechanism of the grinding roller, and the flange interface is hermetically connected to the oil circuit of the hydraulic adjustment mechanism. The detection range of the pressure sensor covers the working pressure range of the pulverizer, and the output signal is converted into a digital signal through an analog-to-digital conversion circuit and then transmitted to the subsequent processing steps. The sampling frequency of collecting the grinding pressure parameter is 100 times per second. During the data preprocessing process, the grinding pressure parameters are arranged in chronological order according to the working cycle of the pulverizer to generate a grinding pressure parameter sequence.

[0065] Collect the bearing vibration signal through the acceleration sensor installed on the bearing housing of the pulverizer. The acceleration sensor includes a horizontal acceleration sensor and a vertical acceleration sensor. The horizontal acceleration sensor is installed at a position on the outer surface of the bearing housing parallel to the axis of the grinding roller, and the vertical acceleration sensor is installed at a position on the outer surface of the bearing housing perpendicular to the axis of the grinding roller. The acceleration sensor is fixed through a magnetic base, and the contact surface between the magnetic base and the bearing housing is coated with anti-slip damping glue to reduce the distortion of the vibration signal. The detection range of the acceleration sensor covers the typical vibration range of the pulverizer bearing, and the sampling frequency is 100 times per second. The vibration signal is transmitted to the subsequent processing steps after being processed by an anti-aliasing filter to generate a bearing vibration signal sequence.

[0066] Synchronously sample the grinding pressure parameters and the bearing vibration signals to generate a time-aligned grinding pressure parameter sequence and a bearing vibration signal sequence. Set up a global clock synchronization mechanism to send a synchronous trigger signal to the pressure sensor and the acceleration sensor. Based on the synchronous trigger signal, the pressure sensor and the acceleration sensor start sampling at the same moment. Each data point in the grinding pressure parameter sequence and the bearing vibration signal sequence contains the same timestamp label, and the grinding pressure parameter sequence and the bearing vibration signal sequence are aligned with millisecond-level precision through the timestamp label. After synchronous sampling, store the time-aligned grinding pressure parameter sequence and the bearing vibration signal sequence in the buffer for subsequent analysis.

[0067] Among them, acceleration sensors are arranged in the horizontal and vertical directions of the bearing housing. The detection direction of the horizontal acceleration sensor is parallel to the radial force direction of the grinding roller bearing, and the detection direction of the vertical acceleration sensor is parallel to the axial force direction of the grinding roller bearing. The pressure sensor is connected to the hydraulic adjustment mechanism of the grinding roller. The hydraulic oil pressure of the hydraulic adjustment mechanism is adjusted by a proportional valve. The detection end of the pressure sensor directly contacts the oil in the hydraulic oil pipeline to collect the hydraulic oil pressure value in real time as the grinding pressure parameter.

[0068] During the process of generating the grinding pressure parameter sequence and the bearing vibration signal sequence, perform data cleaning operations on the original signals. The data cleaning operations include smoothing the grinding pressure parameter sequence using a sliding window mean filter. The length of the sliding window is set to 10 sampling points, and the arithmetic mean of the data points within the window replaces the original value at the center point of the window. For the bearing vibration signal sequence, use the threshold comparison method to eliminate abnormal data points. The threshold in the threshold comparison method is determined by statistically analyzing the maximum amplitude of the vibration signals in the historical normal operation data. Data points exceeding the maximum amplitude are marked as outliers and replaced by linear interpolation.

[0069] Through the above implementation method, the acquisition and synchronization process of the grinding pressure parameters and the bearing vibration signals can accurately match the real-time working conditions under dynamic load changes. For example, when the pressure value in the grinding pressure parameter sequence rises from 5 MPa to 8 MPa within 1 second, the change range of the vibration signal amplitude and the frequency distribution characteristics at the corresponding moment in the synchronously collected bearing vibration signal sequence can fully reflect the actual operating state of the bearing, avoiding the failure of the association between load changes and vibration characteristics caused by time deviation.

[0070] S2. Analyze the corresponding relationship between the change of the grinding pressure parameter and the offset of the main frequency characteristic of the bearing vibration signal. The specific implementation is as follows:

[0071] Generate a dynamic time window based on time alignment for the sequence of grinding pressure parameters. The interval length of the dynamic time window is dynamically adjusted according to the change rate of the grinding pressure parameters, and the change rate is obtained by calculating the ratio of the absolute value of the difference between two adjacent grinding pressure parameter data points to the time interval. For example, when the grinding pressure parameter rises from 5 MPa to 7 MPa within 0.1 seconds, the change rate is (7 - 5) / 0.1 = 20 MPa / s, and at this time, the length of the dynamic time window is set to 0.05 seconds; if the change rate is 5 MPa / s, the window length is set to 0.2 seconds. The generation rule of the dynamic time window is: window length = reference window length × (reference change rate / current change rate), where the reference window length is set to 0.1 seconds, and the reference change rate is set to 10 MPa / s according to the average pressure change rate of historical operation data. Through the above rule, the length of the dynamic time window is inversely proportional to the change rate of the grinding pressure parameters, ensuring that the analysis frequency increases when the load fluctuates rapidly, and the analysis window extends when the load is smooth to capture long-term characteristics.

[0072] Extract the main frequency energy distribution and phase synchronization index of the bearing vibration signal sequence within the dynamic time window. The extraction process of the main frequency energy distribution includes: performing a fast Fourier transform on the bearing vibration signal sequence within the dynamic time window, and identifying the frequency component with the highest energy in the spectrum as the main frequency; calculating the proportion of the main frequency component in the total spectrum energy as the main frequency energy distribution value. The extraction process of the phase synchronization index includes: performing a Hilbert transform on the bearing vibration signal sequence within the dynamic time window to obtain the instantaneous phase sequence of the signal; calculating the sliding window covariance between the instantaneous phase sequence and the change rate of the grinding pressure parameters as the phase synchronization index. For example, the sliding window length is set to 1 / 5 of the dynamic time window length, and when the covariance calculation result is greater than zero, it indicates that the phase change is synchronized with the pressure change trend, and when it is less than zero, it indicates phase unlocking.

[0073] Construct a non-linear mapping rule for the change of grinding pressure parameters and the offset of the main frequency characteristics according to the joint change trend of the main frequency energy distribution and the phase synchronization index. The specific construction process includes: statistically analyzing the joint distribution range of the main frequency energy distribution and the phase synchronization index in historical normal operation data as the reference interval; during real-time monitoring, if the main frequency energy distribution value is lower than the lower limit of the reference interval and the phase synchronization index remains negative, it is determined that the offset of the main frequency characteristics exceeds the normal range; if the main frequency energy distribution value is within the reference interval and the phase synchronization index is positive, it is determined that the offset of the main frequency characteristics is normal fluctuation. The non-linear mapping rule is defined by a two-dimensional decision boundary, which is jointly determined by the main frequency energy distribution threshold and the phase synchronization index threshold, and the thresholds are dynamically adjusted according to the statistical quantiles of historical data. For example, the main frequency energy distribution threshold is set to the 10th percentile of historical data, and the phase synchronization index threshold is set to the 25th percentile of historical data.

[0074] The non - linear mapping rule is used to indicate the confidence interval of the main frequency feature offset under dynamic load fluctuations. The update logic of the confidence interval includes: when the change rate of the grinding pressure parameter is in a low - fluctuation state, expand the upper and lower limits of the confidence interval to allow a larger main frequency feature offset; when the change rate of the grinding pressure parameter is in a high - fluctuation state, compress the confidence interval to improve the detection sensitivity. For example, the low - fluctuation state is defined as a change rate less than 5 MPa / s, and at this time, the normal range of the main frequency feature offset is relaxed to ±15% of the historical reference value; the high - fluctuation state is defined as a change rate greater than 15 MPa / s, and at this time, the normal range is compressed to ±5% of the historical reference value.

[0075] When extracting the main frequency energy distribution within a dynamic time window, if the window length is too short to result in insufficient spectral resolution, the overlapping window analysis method is used for compensation. The step size of the overlapping window is set to 50% of the window length, and the data in the overlapping part of adjacent windows are fused by weighted average, and the weight coefficient is dynamically allocated according to the signal - to - noise ratio of the data within the window. For example, the window area with a signal - to - noise ratio higher than 20 dB is assigned a weight of 0.7, and the area lower than 20 dB is assigned a weight of 0.3 to suppress the interference of noise on the main frequency energy distribution.

[0076] In the process of constructing the non - linear mapping rule, trend fitting is performed on the joint change trend of the main frequency energy distribution and the phase synchronization index. The trend - fitting methods include: discretizing the relationship curve between the main frequency energy distribution and the phase synchronization index in historical data into multiple piece - wise linear intervals, and each linear interval corresponds to a specific load change stage; real - time data matches the corresponding linear interval according to the current change rate of the grinding pressure parameter, and calculates the predicted value of the main frequency feature offset. When the residual between the predicted value and the actual value exceeds the set threshold, the confidence interval update mechanism is triggered. For example, the residual threshold is set to 3 times the standard deviation of the historical data residuals, and when it exceeds this threshold, the reference interval is recalculated.

[0077] Through the above implementation methods, the corresponding relationship between the change of the grinding pressure parameter and the main frequency feature offset can adapt to the complex working conditions of dynamic loads. For example, in the stage of rapid rise of the grinding pressure parameter (change rate of 20 MPa / s), the dynamic time window is shortened to 0.05 s, the main frequency energy distribution value increases significantly due to the enhancement of high - frequency vibration, and the phase synchronization index shows a short - term negative value due to the sudden change of the load; at this time, the non - linear mapping rule will combine the joint trend of energy rise and phase unlocking to determine that the main frequency feature offset enters the transitional state confidence interval, avoiding misjudgment as abnormal wear.

[0078] S3. Separate the interference frequency band components and characteristic frequency band components in the bearing vibration signal according to the main frequency feature offset, and the specific implementation is as follows:

[0079] Determine the dynamic frequency band division boundary based on the main frequency characteristic offset and the non-linear mapping rule. The generation logic of the dynamic frequency band division boundary includes: divide the main frequency characteristic offset into a normal offset interval and an abnormal offset interval according to the confidence interval defined in the non-linear mapping rule. The range of the normal offset interval is determined by the statistical distribution of the main frequency characteristic offset of historical normal operation data. For example, take the 5th to 95th percentiles of historical data as the normal offset interval; the abnormal offset interval is the area outside the normal offset interval. When the main frequency characteristic offset is within the normal offset interval, the dynamic frequency band division boundary is set as the ±5% frequency band range of the main frequency component; when the main frequency characteristic offset is within the abnormal offset interval, the dynamic frequency band division boundary is extended to the ±10% frequency band range of the main frequency component. For example, if the non-linear mapping rule indicates that the confidence interval of the current main frequency characteristic offset is ±8%, the dynamic frequency band division boundary is the ±8% frequency band of the main frequency center frequency, and the frequency band outside this range is regarded as the interference or characteristic component to be separated. The update period of the dynamic frequency band division boundary is consistent with the length of the dynamic time window to ensure that the frequency band division within each time window adapts to the load change in real time.

[0080] Extract the frequency band with an energy proportion lower than the main frequency energy distribution threshold and a negative phase synchronization index in the bearing vibration signal according to the dynamic frequency band division boundary as the interference frequency band component. The extraction conditions of the interference frequency band component include: within the dynamic frequency band division boundary, divide the bearing vibration signal sequence into multiple sub-bands according to the frequency resolution, and calculate the energy proportion of each sub-band (sub-band energy / total energy). The calculation method of the energy proportion is: perform an integration operation on the vibration signal within the sub-band to obtain the sub-band energy value, and then divide the sub-band energy value by the total energy value. If the energy proportion of the sub-band is lower than the main frequency energy distribution threshold (such as the 10th percentile of historical data), and the phase synchronization index corresponding to this sub-band is negative (the phase change is out of lock with the load change trend), then determine that this sub-band is the interference frequency band component. For example, the energy proportion of a certain sub-band is 3% (lower than the threshold of 5%), and the phase synchronization index is -0.2, and this sub-band is classified as the interference frequency band component. The marking method of the interference frequency band component is: set the amplitude of the corresponding sub-band to zero in the frequency domain, or filter it out through a band-stop filter in the time domain.

[0081] Extract the frequency bands with energy proportion higher than the main frequency energy distribution threshold and positive phase synchronization index as the characteristic frequency band components. The extraction conditions for the characteristic frequency band components include: within the dynamic frequency band division boundary, if the energy proportion of a sub-frequency band is higher than the main frequency energy distribution threshold and the phase synchronization index is positive (the phase change is synchronized with the load change trend), then determine that the sub-frequency band is a characteristic frequency band component. For example, if the energy proportion of a sub-frequency band is 12% (higher than the threshold of 5%) and the phase synchronization index is 0.6, this sub-frequency band is classified as a characteristic frequency band component. The extraction method for the characteristic frequency band components is: retain the amplitude and phase information of the corresponding sub-frequency band in the frequency domain, or retain the target frequency band through a band-pass filter in the time domain. The phase synchronization index of the characteristic frequency band components is obtained by calculating the sliding covariance between the instantaneous phase of the sub-frequency band and the change rate of the grinding pressure parameter, and the length of the sliding window is the same as the length of the dynamic time window.

[0082] Perform complementary filtering on the interference frequency band components and the characteristic frequency band components to generate a separated signal sequence aligned with the dynamic time window. The implementation methods of complementary filtering include: subtract the signal component corresponding to the interference frequency band component from the original bearing vibration signal sequence and retain the signal component corresponding to the characteristic frequency band component. The specific steps are: perform inverse Fourier transform on the interference frequency band components to generate a time-domain interference signal, subtract the original vibration signal sequence and the interference signal sequence point by point in time to obtain the time-domain signal of the characteristic frequency band components. Or directly extract the characteristic frequency band components from the original signal and suppress the interference frequency band components, for example, set the amplitude of the interference frequency band components to zero in the frequency domain and then perform inverse Fourier transform. The signal sequence after complementary filtering is strictly aligned with the time stamps of the dynamic time window to ensure the time series consistency of subsequent analysis. The output of complementary filtering is the separated characteristic frequency band signal sequence and the interference frequency band signal sequence, and the two together constitute the complete separated signal sequence.

[0083] During the adjustment process of the dynamic frequency band division boundary, if the main frequency characteristic offset continuously exceeds the normal offset range within multiple consecutive dynamic time windows, trigger the boundary adaptive expansion mechanism. The boundary adaptive expansion mechanism includes: associate the expansion amplitude of the dynamic frequency band division boundary with the deviation degree of the main frequency characteristic offset, and the deviation degree is quantified by the absolute value of the difference between the current offset and the historical reference value. For example, if the main frequency characteristic offset exceeds the normal range for 3 consecutive windows and the average deviation value is 20% of the historical reference, then the dynamic frequency band division boundary is expanded to the ±15% frequency band of the main frequency component. The expanded boundary gradually shrinks in subsequent windows according to the situation of the offset returning to the normal range, and the shrinkage rate is 1% frequency band per window until it returns to the initial set range.

[0084] When extracting the interfering frequency band components and characteristic frequency band components, smooth transition processing is performed on the transition frequency band near the boundary. The implementation methods of smooth transition processing include: within the ±2% frequency band of the dynamic frequency band division boundary, linearly weighted fusion is performed on the energy ratios of the interfering frequency band and the characteristic frequency band, and the weight coefficient is dynamically allocated according to the distance between the frequency band and the boundary. For example, the weight of the sub-frequency band at the -2% frequency point from the boundary is 0.2 (the interference accounts for 80%), and the weight of the sub-frequency band at the +2% frequency point from the boundary is 0.8 (the characteristic accounts for 80%) to eliminate signal distortion caused by sudden changes in the frequency band.

[0085] The calculation formula for the energy ratio of the sub-frequency band after weighted fusion is: energy after fusion = interference energy × weight + characteristic energy × (1 - weight). This method ensures smooth transition of the signal energy at the frequency band boundary and avoids high-frequency noise caused by hard division.

[0086] Through the above implementation methods, the separation process of the interfering frequency band components and the characteristic frequency band components can adapt to the main frequency characteristic fluctuations under dynamic loads. For example, when the offset of the main frequency characteristic enters the abnormal range due to the drastic fluctuation of the grinding pressure parameter, the dynamic frequency band division boundary automatically expands to ensure effective separation of high-frequency interference components; at the same time, the positive and negative determination of the phase synchronization index combined with the energy ratio condition avoids misjudging transient load shocks as bearing wear characteristics. During the stable stage of the grinding pressure parameter, the dynamic frequency band division boundary shrinks, and the extraction accuracy of the characteristic frequency band components is improved, so as to accurately identify the early wear signs of the bearing.

[0087] S4. Perform time-frequency ridge line tracking on the characteristic frequency band components, extract the ridge line trajectory of the time-frequency energy distribution, and calculate the curvature mutation density of the ridge line trajectory. Determine the dynamic reference standard for bearing wear characteristics according to the curvature mutation density. The specific implementation is as follows:

[0088] Perform time-frequency ridge line tracking on the time-domain signal of the characteristic frequency band components within a dynamic time window to generate the ridge line trajectory of the time-frequency energy distribution. The implementation methods of time-frequency ridge line tracking include: within the dynamic time window, perform continuous wavelet transform on the time-domain signal of the characteristic frequency band components to generate a time-frequency energy distribution matrix; in the time-frequency energy distribution matrix, search for the frequency value corresponding to the maximum energy at each time point to form a time-frequency ridge line trajectory. The length of the dynamic time window is aligned with the time of the separated signal sequence generated in step S3 to ensure that the ridge line tracking range strictly matches the load fluctuation stage. For example, for a signal segment with a dynamic time window length of 0.1 seconds, traverse each time point at millisecond intervals, select the frequency with the highest energy in the wavelet energy spectrum as the ridge line point, and connect all the ridge line points to generate the time-frequency ridge line trajectory. The starting moment of the dynamic time window is synchronized with the key time points of the change of the grinding pressure parameter, such as the starting moment of sudden pressure increase or decrease, to ensure that the ridge line tracking covers the vibration characteristics of the load mutation stage.

[0089] Perform cubic spline interpolation on the ridge line trajectory and calculate the curvature change rate of the interpolated trajectory. The specific steps of cubic spline interpolation include: input the discrete point sequence of the time-frequency ridge line trajectory into the interpolation algorithm to generate a smooth and continuous curve. The parameter settings of the interpolation algorithm include the node spacing and the smoothing factor. The node spacing is set to 1% of the dynamic time window length. For example, when the window length is 0.1 seconds, the node spacing is 1 millisecond. The smoothing factor is dynamically adjusted according to the local curvature fluctuation amplitude of the ridge line trajectory. The larger the fluctuation amplitude, the smaller the smoothing factor. Sample at a fixed interval along the interpolated curve and calculate the curvature value of each sampling point. The calculation formula of the curvature value adopts the curvature definition in differential geometry and is realized by calculating the ratio of the tangent angle change rate of the curve at a certain point to the arc length. The calculation method of the curvature change rate is: take the difference between the curvature values of adjacent sampling points and then divide by the sampling interval time. For example, if the sampling interval is 1 millisecond and the curvature difference between two points is 0.05, then the curvature change rate is 0.05 / 0.001 = 50 per second.

[0090] Count the number of mutation points whose curvature change rate exceeds the set mutation threshold within a unit time to generate the curvature mutation density. The setting rules of the mutation threshold include: according to the statistical distribution of the curvature change rate in the historical normal operation data, take the 95th percentile as the benchmark mutation threshold; during real-time monitoring, the benchmark mutation threshold is dynamically adjusted according to the average value of the curvature change rate within the current dynamic time window, and the adjustment range is ±20%. For example, if the historical benchmark threshold is 40 per second and the average curvature change rate within the current window is 35 per second, then the real-time threshold is adjusted to 35×1.2 = 42 per second. The statistical method of the number of mutation points within a unit time is: within the dynamic time window, traverse all sampling points, count the number of points whose curvature change rate exceeds the threshold, and then divide by the window length. For example, if 5 mutation points are detected within a 0.1-second window, then the curvature mutation density is 5 / 0.1 = 50 per second. During the statistical process, if the same sampling point exceeds the threshold for multiple consecutive intervals, it is only counted as one mutation to avoid duplicate counting.

[0091] Based on the mapping relationship between the curvature mutation density and the bearing wear degree in the historical wear data, determine the dynamic reference standard for the bearing wear characteristics within the current dynamic time window. The process of establishing the mapping relationship includes: collecting the full-life cycle data of the historical bearing from normal to severe wear, and recording the curvature mutation density values at each stage; fitting the quantitative relationship between the curvature mutation density and the wear degree through linear regression analysis to generate a reference standard curve. For example, if the fitting result shows that for every 10 increases in the curvature mutation density per second, the wear depth increases by 0.01 mm, then the normal fluctuation range in the dynamic reference standard is set to ±3 times the standard deviation of the density under the historical normal working conditions, and the abnormal determination threshold is set to 1.5 times the upper limit of the normal range. During real-time monitoring, if the current curvature mutation density falls within the normal fluctuation range, it is determined that the bearing state is normal; if it exceeds the abnormal determination threshold, it is determined that there is a wear risk. The input data for the linear regression analysis needs to be normalized. The normalization method is: divide the curvature mutation density value by the historical maximum value, and divide the wear depth value by the maximum allowable wear amount of the bearing material to ensure consistent data scales.

[0092] The dynamic reference standard includes the normal fluctuation range and the abnormal determination threshold of the curvature mutation density. The update logic of the normal fluctuation range is: based on the curvature mutation density data of the most recent 100 dynamic time windows, recalculate the mean and standard deviation, and dynamically adjust the normal range. For example, if the mean of the most recent 100 windows is 30 per second and the standard deviation is 5 per second, then the normal range is updated to 30 ± 3×5 = 15 to 45 per second. The abnormal determination threshold is set according to the critical wear data in the equipment maintenance records. For example, take the maximum value of the curvature mutation density in the 10 windows before the occurrence of the historical critical wear event as the threshold. The update cycle of the dynamic reference standard is synchronized with the equipment maintenance cycle. For example, it is updated every 24 hours or after each grinding task is completed to ensure that the detection standard adapts to the equipment aging trend.

[0093] During the calculation of the curvature change rate, if there are local distortions in the interpolated trajectory, trigger the trajectory smoothing and correction mechanism. The smoothing and correction mechanism includes: detecting whether the mutation amplitude of the curvature change rate exceeds 2 times the historical maximum mutation amplitude; if it exceeds, perform median filtering on the curvature values of 5 sampling points before and after the distorted point to replace the original curvature value of the distorted point. For example, if the curvature change rate of a certain point is 120 per second (historical maximum 60 per second), then take the median of 11 points before and after it to eliminate the interference of abnormal mutations. The window length of the median filtering is adaptively adjusted according to the number of distorted points. If 3 consecutive points exceed the historical threshold, the window length is extended to 15 points to cover a longer abnormal interval. The corrected curvature change rate participates in the mutation point count again to ensure data reliability.

[0094] Through the above implementation, the curvature mutation density of the time-frequency ridge trajectory can effectively characterize the dynamic characteristics of bearing wear. For example, in the early stage of bearing wear, the time-frequency ridge trajectory exhibits high-frequency jitter due to local stress concentration, and the curvature mutation density rises from the normal value of 30 per second to 45 per second, triggering an early warning; in the severe wear stage, the ridge trajectory shows periodic fractures, and the curvature mutation density exceeds 60 per second, and it is determined that immediate shutdown maintenance is required. The adaptive update mechanism of the dynamic reference standard ensures that the detection standard is dynamically adjusted with the aging trend of the equipment to avoid misjudgment. In the stable stage of grinding pressure parameters, the dynamic time window is extended, the smoothness of the ridge trajectory is improved, and the curvature mutation density is stabilized within the normal range; in the stage of severe load fluctuations, the window is automatically shortened, and the high-frequency mutation characteristics are accurately captured, thereby achieving real-time and accurate assessment of the health status of the bearing.

[0095] S5. Adjust the amplitude threshold and frequency threshold of the bearing vibration signal according to the dynamic reference standard, which is specifically implemented as follows:

[0096] According to the normal fluctuation range of the curvature mutation density in the dynamic reference standard, the amplitude threshold of the bearing vibration signal is adjusted. The specific rules for adjusting the amplitude threshold include: mapping the upper limit of the normal fluctuation range to the upper limit of the amplitude threshold, and mapping the lower limit to the lower limit of the amplitude threshold. The upper limit and lower limit of the normal fluctuation range are obtained by fitting historical data. The fitting method is: in the historical normal operation stage, the maximum and minimum values ​​of the curvature mutation density are recorded as the initial values ​​of the upper limit and lower limit, respectively, and then dynamically adjusted according to the distribution of the curvature mutation density in the current dynamic time window. For example, if the historical normal fluctuation range is 15 to 45 seconds, and the curvature mutation density of the current window is concentrated between 20 and 40 seconds, the upper limit of the adjusted amplitude threshold corresponds to 40 seconds, and the lower limit corresponds to 20 seconds. The mapping relationship of the amplitude threshold is realized by linear interpolation. The input of the linear interpolation method is the curvature mutation density value, and the output is the vibration signal amplitude threshold.

[0097] According to the abnormal judgment threshold of the curvature mutation density in the dynamic reference standard, the frequency threshold of the bearing vibration signal is adjusted. The specific rules for adjusting the frequency threshold include: mapping the curvature mutation density value corresponding to the abnormal judgment threshold to the trigger point of the frequency threshold. The abnormal judgment threshold is obtained in the following way: in the accelerated wear test of the bearing, the main frequency offset of the vibration signal corresponding to the curvature mutation density reaching the preset critical value (such as 60 per second) is recorded and used as the frequency threshold. For example, when the curvature mutation density reaches 60 per second, the main frequency offset is usually ±50 Hz, and the frequency threshold is set to ±50 Hz. In real-time monitoring, the frequency threshold is adjusted synchronously with the dynamic update of the abnormal judgment threshold, and the adjustment range is determined according to the standard deviation of the main frequency offset in the current window. For example, if the standard deviation of the main frequency offset is 5 Hz, the frequency threshold is adjusted to ±(50+5)=±55 Hz.

[0098] Generate a dynamic amplitude-frequency combined determination rule based on the adjusted amplitude threshold and frequency threshold. The generation logic of the dynamic amplitude-frequency combined determination rule includes: defining the combined conditions of the amplitude threshold and the frequency threshold, including three determination scenarios: amplitude overrun, frequency overrun, and compound overrun. The amplitude overrun determination condition is that the amplitude of the vibration signal exceeds the upper limit or is lower than the lower limit of the amplitude threshold; the frequency overrun determination condition is that the main frequency offset exceeds the frequency threshold; the compound overrun determination condition is that both the amplitude and the frequency are overrun. For example, if the amplitude threshold is 0.8 to 2.5 mm / s² and the frequency threshold is ±50 Hz, the compound overrun condition is that the amplitude > 2.5 mm / s² and the main frequency offset > +50 Hz. The priority of the dynamic amplitude-frequency combined determination rule is set as: compound overrun > amplitude overrun > frequency overrun to ensure that high-risk states are alarmed first.

[0099] Match the dynamic amplitude-frequency combined determination rule with the bearing vibration signal in the current dynamic time window in real time, and update the effective range of the amplitude threshold and the frequency threshold. The update method of the effective range includes: setting the effective action time of the threshold according to the sliding step of the dynamic time window and the load change trend. For example, if the sliding step of the dynamic time window is 0.05 s and the grinding pressure is in the rising stage, the effective range is set to within 0.03 s after the end of the current window; if the pressure tends to be stable, the effective range is extended to 0.08 s. During the real-time matching process, the amplitude and the main frequency offset of the vibration signal are compared point by point with the threshold in each effective range, and the matching result is output. The update period of the effective range is strictly synchronized with the sliding step of the dynamic time window to ensure that the threshold action time is matched with the load fluctuation in real time.

[0100] During the process of adjusting the amplitude threshold, if the normal fluctuation range of the curvature mutation density changes suddenly, trigger the threshold smooth transition mechanism. The implementation method of the smooth transition mechanism includes: calculating the difference between the new and old thresholds, evenly distributing the difference according to the preset number of transition windows, and gradually transitioning to the new threshold. For example, if the upper limit of the amplitude threshold is adjusted from 2.5 mm / s² to 3.0 mm / s² and the number of transition windows is set to 5, the adjustment increment of each window is (3.0 - 2.5) / 5 = 0.1 mm / s², and it increases sequentially until the new threshold is reached. The number of transition windows is dynamically set according to the change rate of the curvature mutation density. The faster the change rate, the fewer the number of transition windows. For example, when the change rate exceeds 10 per second / second, the number of transition windows is set to 3; when the rate is lower than 5 per second / second, the number of windows is set to 10.

[0101] When generating the dynamic amplitude-frequency joint determination rule, the determination weights of the amplitude and frequency thresholds are dynamically allocated. The weight allocation rule includes: adjusting the determination weights of the amplitude and frequency according to the signal-to-noise ratio of the vibration signal within the current dynamic time window. The calculation method of the signal-to-noise ratio is: the ratio of the total signal energy to the noise energy (total energy - characteristic frequency band energy), and it is expressed in decibels after taking the logarithm. For example, when the signal-to-noise ratio is 20 dB, the amplitude weight is set to 0.6 and the frequency weight is set to 0.4; when the signal-to-noise ratio is 10 dB, the amplitude weight is reduced to 0.4 and the frequency weight is increased to 0.6. The weight coefficient is determined by the principle of minimizing the misjudgment rate in historical data. The specific method is: traverse different weight combinations in the historical dataset and select the weight with the lowest misjudgment rate as the current weight.

[0102] Through the above implementation, the dynamic adjustment of the amplitude threshold and frequency threshold can accurately adapt to the real-time changes of the bearing wear state. For example, in the stage of severe fluctuation of the grinding pressure parameter, the curvature mutation density rises rapidly, the upper limit of the amplitude threshold is increased to 3.0 mm / s² to tolerate transient shocks, and the frequency threshold is synchronously tightened to ±55 Hz to capture high-frequency anomalies; in the stage of stable pressure, the lower limit of the amplitude threshold is reduced to 0.5 mm / s² to improve the detection sensitivity, and the frequency threshold is relaxed to ±60 Hz to reduce false alarms. The real-time matching mechanism of the dynamic amplitude-frequency joint determination rule and the effective range ensures the stability and reliability of the online monitoring system under complex working conditions.

[0103] S6. Perform amplitude-frequency joint analysis on the characteristic frequency band components based on the adjusted amplitude threshold and frequency threshold, and output the bearing wear state evaluation result. The specific implementation is as follows:

[0104] Calculate the time-domain energy mean value and extract the frequency-domain main frequency offset of the characteristic frequency band components within the dynamic time window. The calculation method of the time-domain energy mean value includes: performing an integration operation on the time-domain signal of the characteristic frequency band components within the dynamic time window. The integration interval is from the start time to the end time of the current dynamic time window, and the integration result is divided by the window length to generate the time-domain energy mean value. For example, when the length of the dynamic time window is 0.1 s, the total energy obtained by integrating the amplitude of the time-domain signal of the characteristic frequency band components within the window is 0.25 (unit: mm² / s³), then the time-domain energy mean value is 0.25 / 0.1 = 2.5 mm / s². The extraction method of the frequency-domain main frequency offset includes: performing a fast Fourier transform on the characteristic frequency band components within the dynamic time window, setting the spectral resolution to the reciprocal of the window length, identifying the frequency component with the highest energy in the spectrum as the main frequency, and calculating the difference between the main frequency and the device calibration reference frequency as the main frequency offset. For example, the reference frequency is 1000 Hz and the current main frequency is 1050 Hz, then the main frequency offset is +50 Hz.

[0105] Determine whether the time-domain energy mean exceeds the adjusted amplitude threshold and whether the frequency-domain main frequency offset exceeds the adjusted frequency threshold. The judgment condition for the amplitude threshold is that the time-domain energy mean is greater than the upper limit of the adjusted amplitude threshold or less than the lower limit of the adjusted amplitude threshold; the judgment condition for the frequency threshold is that the absolute value of the main frequency offset is greater than the absolute value of the adjusted frequency threshold. For example, if the adjusted amplitude threshold is 0.8 to 2.5 mm / s² and the frequency threshold is ±50 Hz, and the time-domain energy mean is 3.0 mm / s² and the main frequency offset is +60 Hz, it is determined to be over-limit at the same time. In the judgment logic, the over-limit determination needs to meet the time persistence condition, that is, the over-limit state persists for more than 50% of the window length within the dynamic time window to avoid misjudgment caused by instantaneous interference. For example, within a window length of 0.1 s, the over-limit duration needs to exceed 0.05 s to trigger an effective determination.

[0106] When the time-domain energy mean and the frequency-domain main frequency offset are over-limit at the same time, generate a bearing wear mode according to the combined coding of the over-limit amplitude threshold and the frequency threshold. The generation rules of the combined coding include: combining the amplitude over-limit direction (exceeding the upper limit or lower than the lower limit) and the frequency over-limit direction (positive or negative) into a four-digit coding, and the coding format is "amplitude direction + frequency direction + threshold level". For example, when the amplitude exceeds the upper limit and the frequency is positively over-limit, the coding is "A1+"; when the amplitude is lower than the lower limit and the frequency is negatively over-limit, the coding is "B2-". The definition of the bearing wear mode includes: mapping according to the coding to a predefined wear type library, and the wear type library is obtained through training with historical fault data. The training method is: count the occurrence frequency of each coding in known fault cases, and the coding with a frequency exceeding the set threshold is bound to the fault type. For example, if the coding "A1+" appears in 90% of the outer ring wear cases, it is mapped to "outer ring wear".

[0107] Based on the cumulative results of the bearing wear modes in the current dynamic time window and adjacent windows, output the bearing wear state evaluation result. The statistical method of the cumulative results includes: using a sliding window counter to record the wear modes of the current window and the previous two windows, and the counter capacity is 3. When new data enters, the earliest data is overwritten. Count the proportion of the occurrence times of the same wear mode in the three windows, and the calculation formula is: occurrence times / 3 × 100%. For example, if the current window is "outer ring wear", and the previous two windows are "outer ring wear" and "inner ring wear" respectively, the proportion is 2 / 3 ≈ 66.7%. The output rule of the evaluation result is: if the proportion of the same mode exceeds 70%, it is determined to be a definite wear state; if the proportion is between 50% and 70%, it is determined to be a suspected wear state; if it is less than 50%, it is determined to be a normal fluctuation. The thresholds of 70% and 50% are determined by the distribution interval between the normal and abnormal states in the historical data. For example, under normal working conditions, the proportion of the same mode is usually less than 30%, and under abnormal working conditions, it is higher than 70%.

[0108] When generating bearing wear patterns, if multiple overlimit combinations occur within the same window, the priority determination rule is used to determine the final pattern. The definition of the priority rule includes: sorting the combination codes according to the severity of overlimits, where the severity is determined by the weighted sum of the amplitude overlimit and the frequency overlimit. The weight coefficients are 0.6 for amplitude and 0.4 for frequency. For example, for the code "A1+", the amplitude overlimit is 3.0 - 2.5 = 0.5 mm / s², the frequency overlimit is 60 - 50 = 10 Hz, and the severity score is 0.5×0.6 + 10×0.4 = 4.3; for the code "B2-", the score is 0.3×0.6 + 8×0.4 = 3.38, so the "A1+" with the higher score is selected as the current pattern. The weight coefficients of the priority rule are determined through regression analysis of historical failure data, with the regression objective of minimizing the pattern misjudgment rate.

[0109] The update and storage logic of the cumulative result includes: using a circular buffer to store the wear patterns of the last three dynamic time windows, and the buffer index is updated according to the first-in-first-out principle. For example, buffer index 0 is the earliest data, and index 2 is the latest data; when new data enters, the data at index 0 is overwritten, and indexes 1 and 2 are shifted forward in turn. The output period of the evaluation result is strictly synchronized with the sliding step of the dynamic time window. For example, if the window sliding step is 0.05 s, the evaluation result is updated every 0.05 s. The forms of the output results include: text description (such as "outer ring wear"), numerical grade (such as "wear grade 3"), and control signal (such as "shutdown instruction"), and the control signal is transmitted to the equipment control system through an industrial communication protocol.

[0110] Through the above implementation methods, the evaluation result of the bearing wear state can reflect the wear trend under dynamic load in real time. For example, within three consecutive dynamic time windows, if the "outer ring wear" pattern appears twice and the time-domain energy mean value continues to rise, the evaluation result outputs "the outer ring wear is intensifying, it is recommended to stop for inspection"; if the "outer ring wear", "inner ring wear", and "outer ring wear" appear in three windows respectively and the proportion is 66.7%, it outputs "compound wear warning, further diagnosis is required". The verification mechanism of the evaluation result includes: recording the actual wear state during equipment maintenance, comparing it with the system output result, calculating the misjudgment rate, and dynamically optimizing the threshold and weight coefficients. For example, if it is actually found during maintenance that the outer ring is worn but the system does not alarm, the threshold is adjusted from 70% to 65% to improve the sensitivity.

[0111] Embodiment 2: Figure 2 The structural schematic diagram of an on-line monitoring system for the wear state of a grinding mill bearing according to the present invention is given. An on-line monitoring system for the wear state of a grinding mill bearing includes the following modules:

[0112] A signal acquisition module, which is used to obtain the grinding pressure parameters and bearing vibration signals during the operation of the grinding mill;

[0113] A dynamic association module, which is used to analyze the corresponding relationship between the change of the grinding pressure parameter and the offset of the main frequency characteristic of the bearing vibration signal;

[0114] A frequency band separation module, which is used to separate the interference frequency band components and characteristic frequency band components in the bearing vibration signal according to the offset of the main frequency characteristic;

[0115] A curvature analysis module, which is used to perform time-frequency ridge line tracking on the characteristic frequency band components, extract the ridge line trajectory of the time-frequency energy distribution and calculate the curvature mutation density of the ridge line trajectory, and determine the dynamic reference standard of the bearing wear characteristics according to the curvature mutation density;

[0116] A threshold adjustment module, which is used to adjust the amplitude threshold and frequency threshold of the bearing vibration signal according to the dynamic reference standard;

[0117] A state evaluation module, which is used to perform amplitude-frequency joint analysis on the characteristic frequency band components based on the adjusted amplitude threshold and frequency threshold, and output the evaluation result of the bearing wear state.

[0118] The calculations involved in the embodiments are all dimensionless numerical calculations, and the preset parameters and threshold selections in the calculations are set by those skilled in the art according to the actual situation.

[0119] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product.

[0120] Those of ordinary skill in the art can realize that the modules and algorithm steps of each example described in combination with the embodiments disclosed in this article can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in hardware or software depends on the specific application of the technical solution and the invention constraints. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0121] In addition, the functional modules in each embodiment of this application can be integrated in a processing module, or each module can exist physically alone, or two or more modules can be integrated in one module.

[0122] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division. In actual implementation, there may be other division methods. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed coupling or direct coupling or communication connection to each other can be through some interfaces. The indirect coupling or communication connection of the devices or modules can be in electrical, mechanical, or other forms.

[0123] As described above, this is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0124] Finally: The above description is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for online monitoring of the wear state of a grinding mill bearing, characterized in that: The steps include: S1, obtaining grinding pressure parameters and bearing vibration signals when the mill is running; S2, analyzing the corresponding relationship between the change of grinding pressure parameters and the main frequency characteristic offset of the bearing vibration signal; S3, separating the interference frequency band component and the characteristic frequency band component in the bearing vibration signal according to the main frequency characteristic offset; S4, performing time-frequency ridge tracking on the characteristic frequency band components, extracting the ridge trajectory of the time-frequency energy distribution and calculating the curvature mutation density of the ridge trajectory, and determining the dynamic reference standard of the bearing wear characteristics according to the curvature mutation density; S5. adjusting the amplitude threshold and frequency threshold of the bearing vibration signal according to the dynamic reference standard; S6. Based on the adjusted amplitude threshold and frequency threshold, perform amplitude-frequency joint analysis on the characteristic frequency band components and output the bearing wear status assessment result.

2. The method for online monitoring of the wear state of a grinding mill bearing according to claim 1, characterized in that: Obtain grinding pressure parameters and bearing vibration signals during mill operation, including: Grinding pressure parameters are collected by a pressure sensor installed on the grinding roller of the mill, and bearing vibration signals are collected by an acceleration sensor installed on the bearing seat of the mill; The grinding pressure parameters and the bearing vibration signals are synchronously sampled to generate a time-aligned grinding pressure parameter sequence and a bearing vibration signal sequence.

3. A method for online monitoring of bearing wear status of a grinding mill according to claim 2, characterized in that: The acceleration sensor is arranged in the horizontal and vertical directions of the bearing seat, and the pressure sensor is connected with the hydraulic adjustment mechanism of the grinding roller.

4. The method for online monitoring of the wear state of a grinding mill bearing according to claim 1, characterized in that: Analyze the corresponding relationship between the change of grinding pressure parameters and the main frequency characteristic offset of the bearing vibration signal, including: A dynamic time window is generated based on the time-aligned grinding pressure parameter sequence, and the interval length of the dynamic time window is inversely proportional to the change rate of the grinding pressure parameter; Extract the main frequency energy distribution and phase synchronization index of the bearing vibration signal sequence within the dynamic time window; According to the joint change trend of the main frequency energy distribution and the phase synchronization index, a nonlinear mapping rule between the grinding pressure parameter change and the main frequency characteristic offset is constructed; The nonlinear mapping rule is used to indicate the confidence interval of the main frequency characteristic offset under dynamic load fluctuation.

5. The method for online monitoring of the wear state of a grinding mill bearing according to claim 1, characterized in that: The interference frequency band components and characteristic frequency band components in the bearing vibration signal are separated according to the main frequency characteristic offset, including: Determine the dynamic frequency band division boundary based on the main frequency characteristic offset and nonlinear mapping rules; According to the dynamic frequency band division boundary, the frequency band in the bearing vibration signal with an energy ratio lower than the main frequency energy distribution threshold and a negative phase synchronization index is extracted as the interference frequency band component; Extract the frequency band whose energy proportion is higher than the main frequency energy distribution threshold and whose phase synchronization index is positive as the characteristic frequency band component; Complementary filtering is performed on the interference frequency band components and the characteristic frequency band components to generate a separated signal sequence aligned with the dynamic time window.

6. The method for online monitoring of the wear state of a grinding mill bearing according to claim 1, characterized in that: Track the time-frequency ridges of the characteristic frequency band components, extract the ridge trajectory of the time-frequency energy distribution, and calculate the curvature mutation density of the ridge trajectory. Determine the dynamic reference standard of the bearing wear characteristics based on the curvature mutation density, including: Track the time-frequency ridge line of the time domain signal of the characteristic frequency band component in a dynamic time window to generate the ridge line trajectory of the time-frequency energy distribution; The ridge trajectory is processed by cubic spline interpolation to calculate the curvature change rate of the interpolated trajectory; the number of mutation points whose curvature change rate exceeds the set mutation threshold per unit time is counted to generate the curvature mutation density; According to the mapping relationship between the curvature mutation density and the bearing wear degree in the historical wear data, the dynamic reference standard of the bearing wear characteristics in the current dynamic time window is determined; The dynamic reference standards include the normal fluctuation range of curvature mutation density and the abnormal judgment threshold.

7. The method for online monitoring of the wear state of a grinding mill bearing according to claim 1, characterized in that: Adjust the amplitude threshold and frequency threshold of the bearing vibration signal according to the dynamic reference standard, including: According to the normal fluctuation range of the curvature mutation density in the dynamic reference standard, the amplitude threshold of the bearing vibration signal is adjusted; According to the abnormal judgment threshold of the curvature mutation density in the dynamic reference standard, the frequency threshold of the bearing vibration signal is adjusted; Generate a dynamic amplitude-frequency joint determination rule based on the adjusted amplitude threshold and frequency threshold; The dynamic amplitude-frequency joint judgment rule is matched with the bearing vibration signal in the current dynamic time window in real time, and the effective intervals of the amplitude threshold and the frequency threshold are updated.

8. The method for online monitoring of the bearing wear state of a grinding mill according to claim 1, characterized in that: Based on the adjusted amplitude threshold and frequency threshold, the characteristic frequency band components are analyzed by amplitude-frequency combination, and the bearing wear status assessment results are output, including: Calculate the time domain energy mean and extract the frequency domain main frequency offset of the characteristic frequency band components in the dynamic time window; Determine whether the time domain energy mean exceeds the adjusted amplitude threshold and whether the frequency domain main frequency offset exceeds the adjusted frequency threshold; When both the time-domain energy mean and the frequency-domain main frequency offset exceed the limit at the same time, the bearing wear pattern is generated according to the combined coding of the exceeding amplitude threshold and the frequency threshold; Based on the cumulative results of the bearing wear patterns in the current dynamic time window and adjacent windows, the bearing wear status assessment results are output.

9. The method for online monitoring of the bearing wear state of a grinding mill according to claim 8, characterized in that: The bearing wear status evaluation result is the proportion of the occurrence of the same wear pattern in three consecutive dynamic time windows.

10. An online monitoring system for the wear state of a mill bearing, used to implement an online monitoring method for the wear state of a mill bearing according to any one of claims 1 to 9, characterized in that: Includes the following modules: A signal acquisition module is used to obtain grinding pressure parameters and bearing vibration signals when the mill is running; Dynamic correlation module, used to analyze the corresponding relationship between the change of grinding pressure parameters and the offset of the main frequency characteristic of the bearing vibration signal; A frequency band separation module is used to separate the interference frequency band components and characteristic frequency band components in the bearing vibration signal according to the main frequency characteristic offset; The curvature analysis module is used to track the time-frequency ridges of the characteristic frequency band components, extract the ridge trajectory of the time-frequency energy distribution and calculate the curvature mutation density of the ridge trajectory, and determine the dynamic reference standard of the bearing wear characteristics based on the curvature mutation density; A threshold adjustment module, used to adjust the amplitude threshold and frequency threshold of the bearing vibration signal according to a dynamic reference standard; The state assessment module is used to perform amplitude-frequency joint analysis on characteristic frequency band components based on the adjusted amplitude threshold and frequency threshold, and output the bearing wear state assessment result.

Images