Intelligent bearing dustproof cover based on edge calculation method and state monitoring method

Abstract

The invention relates to the technical field of bearing state monitoring, and provides an intelligent bearing dustproof cover based on an edge calculation method and a state monitoring method. A temperature sensor, a rotating speed sensor and a vibration acceleration sensor are integrated on a bearing dust cover, and bearing operation parameters are collected; processing a vibration acceleration signal through an adaptive sampling algorithm based on a short-time RMS energy threshold method, and performing full sampling in a peak section and down-sampling in a stationary section; performing temperature compensation calculation on the processed vibration acceleration signal, and eliminating temperature drift by using a zero offset and sensitivity function; and analyzing the compensated signal through an AI model, extracting a waveform feature, comparing the waveform feature with a pre-stored health feature, and judging a bearing fault stage. According to the scheme, the sensor is integrated on the dustproof cover, state monitoring is achieved without changing the bearing structure, and destructive transformation of equipment in a traditional method is avoided. Edge calculation and a self-adaptive sampling strategy are adopted, the data transmission quantity is reduced, the network bandwidth requirement is lowered, accurate collection of key data is guaranteed, and the bearing state can be accurately determined in real time.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of bearing state monitoring, and in particular to an intelligent bearing dust cover based on an edge computing method and a state monitoring method. BACKGROUND

[0002] As a core basic component in modern industrial equipment, the running state of a bearing is directly related to the reliability, safety and service life of the entire mechanical system. In key fields such as wind power generation, high-speed rail and precision machine tools, sudden failure of bearings can cause huge economic losses and even safety accidents. Therefore, real-time and accurate monitoring of the running state of bearings to achieve predictive maintenance has always been the goal pursued by the industry. Traditional monitoring methods mainly rely on periodic manual inspection or offline analysis, which is not only inefficient, but also cannot capture transient and early fault features in the degradation process of bearings, and the warning capability is very limited.

[0003] The existing online bearing state monitoring technology implementation scheme usually needs to additionally install a special sensor on the bearing or its adjacent components, for example, a threaded hole is opened on the bearing seat to install a piezoelectric acceleration sensor, or a vortex sensor probe is aligned with the bearing outer ring to measure the rotating speed. Although these methods can obtain monitoring data, the installation process often inevitably requires mechanical processing and modification of the original bearing seat or bearing end cover and other related structures, such as drilling, tapping or milling installation planes. This modification not only increases the process complexity and installation cost, but more importantly, it can change the mechanical integrity of the bearing support structure, disturb its original stress distribution, and thus have unpredictable potential impact on the load capacity, dynamic characteristics and fatigue life of the bearing.

[0004] Therefore, the existing bearing state monitoring technology has the problem that in order to realize the state monitoring function, the original structure of the bearing and its associated components is often damaged or changed, which not only introduces additional installation complexity and cost, but also brings potential risks to the performance and reliability of the bearing body. SUMMARY

[0005] To solve the above-mentioned existing bearing state monitoring technology, in order to realize the state monitoring function, the original structure of the bearing and its associated components is often damaged or changed, which introduces additional installation complexity and cost, and brings potential risks to the performance and reliability of the bearing body.

[0006] One aspect of the present application provides an intelligent bearing state monitoring method based on an edge computing method, the method comprising: integrating a temperature sensor, a rotating speed sensor and a vibration acceleration sensor on a bearing dust cover to collect temperature signals, rotating speed signals and vibration acceleration signals of the bearing; The vibration acceleration signal collected by the vibration acceleration sensor is processed by an adaptive sampling algorithm. The adaptive sampling algorithm performs peak detection based on a short-time RMS energy threshold method, and uses full sampling rate to maintain the vibration acceleration signal in the identified peak section, and uses a sliding window mean sampling to perform down-sampling processing on the vibration acceleration signal in the smooth section. The vibration acceleration signal processed by the adaptive sampling is subjected to temperature compensation calculation, and a zero offset compensation function and a sensitivity function are used to eliminate the temperature drift effect. The vibration acceleration signal processed by the temperature compensation is analyzed by an AI model, waveform features are extracted from the vibration acceleration signal, and the waveform features are compared with pre-stored health state waveform features, and the bearing fault stage is judged based on the comparison result.

[0007] In a feasible implementation, the peak detection includes calculating a short-time RMS energy threshold of the vibration acceleration signal as a peak judgment reference, and setting a dynamic threshold to accurately capture transient events. The calculation formula of the short-time RMS energy threshold is: RMS[n]= ; In the formula, RMS[n] represents the root mean square value at time point n, is the sliding window length, is the original value of the vibration acceleration signal at time point k; A dynamic threshold θ is set based on the short-time RMS energy threshold, and the formula of the dynamic threshold θ is: ; In the formula, is a gain coefficient, is a smooth section historical RMS mean value, and β is a safety margin for compensating signal baseline drift.

[0008] In a feasible implementation, the peak detection further includes the following based on the dynamic threshold judgment: The rapid change characteristics of the vibration acceleration signal are captured by calculating the absolute change rate of the vibration acceleration signal; When the absolute change rate exceeds a preset change rate threshold, a pre-warning is triggered; The peak entry condition is that the RMS value continuously exceeds the dynamic threshold and the duration is greater than a preset minimum peak duration; The peak exit condition is that the RMS value continuously falls below the adjusted dynamic threshold and the duration is greater than a preset exit holding time; The calculation formula of the absolute change rate is: ΔX[n]=∣x[n]-x[n-1]∣; In the formula, ΔX[n] represents the absolute change rate at time point n, x[n] is the original value of the vibration acceleration signal at time point n, and x[n-1] is the original value of the vibration acceleration signal at the previous time point.

[0009] In a feasible implementation, the sliding window mean sampling adopts an overlapping window method in the stationary section processing, and the calculation formula is: ; Step=N / 2; In the formula, y[m] is the output signal after decimation, N is the window length, Step=N / 2 is the sliding step, and an overlap rate of 50% is achieved to reduce smoothing distortion, is the original signal sample, is the starting point index of the mth window.

[0010] In a feasible implementation, temperature compensation is further included for the vibration acceleration sensor, and the effective output value after compensation is calculated through the sensor output value, a zero offset compensation function, and a sensitivity function; The temperature compensation formula is: ; In the formula, Oc is the effective output value after compensation, O is the original output value of the sensor, Z(T) is the zero offset compensation function, and S(T) is the sensitivity compensation function. Z(T)=aT+b; S(T)=cT+d; In the formula, T is a temperature value, and a, b, c, and d are constants obtained through sensor calibration.

[0011] In a feasible implementation, the waveform feature extraction in the AI model analysis includes time domain features, frequency domain features, time-frequency features, implicit rotational speed, and implicit temperature. The time domain features are used to identify impact characteristics through kurtosis calculation, the frequency domain features are used to detect fault frequencies through FFT analysis, the time-frequency features are used to analyze modulation phenomena through wavelet transform, the implicit rotational speed is used to calibrate the rotational frequency through adaptive filtering, and the implicit temperature is used to evaluate thermal deformation through long-term trend analysis.

[0012] In a feasible implementation, the fault stage includes a healthy stage, an early fault stage, a medium fault stage, and a late fault stage. The judgment conditions of the healthy stage include that the kurtosis value of the time domain features is between 2.9 and 3.1, the peak signal-to-noise ratio of the frequency domain features is lower than 3 dB and the energy distribution is uniform, there is no irregular transient event in the time-frequency features, and the implicit rotational speed fluctuation is between 0 and 0.5%. The judgment condition of the early failure stage includes that the time domain feature appears periodic impact pulse and the kurtosis value is greater than 5, and the feature frequency signal-to-noise ratio of the frequency domain feature is less than 5dB, and the impact event of the time-frequency feature is synchronized with the rotation period; The judgment condition of the medium-term failure stage includes that the impact amplitude of the time domain feature increases and secondary impact appears, and the feature frequency signal-to-noise ratio of the frequency domain feature is between 5-15dB, and the modulation phenomenon of the time-frequency feature is significant and the envelope spectrum energy is concentrated; The judgment condition of the late failure stage includes that the time domain feature signal is saturated and periodic, and the wideband energy of the frequency domain feature increases and the feature frequency is blurred, and a large number of random impacts of the time-frequency feature and the modulation phenomenon disappear.

[0013] Another aspect of the present application provides an intelligent bearing dust cover based on an edge computing method, which is used to realize the intelligent bearing state monitoring method based on the edge computing method described in any one of the above aspects, comprising: a dust cover body, a circuit board and a magnet assembly; The dust cover body is clamped on the standard clamping groove of the bearing outer ring through the fixing structure of its edge, and the dust cover body contains a metal skeleton and a rubber sealing layer and is connected with the seal of the bearing; The circuit board is fixed on the side of the dust cover body facing the internal space of the bearing, and a Hall sensor, a temperature sensor, a vibration acceleration sensor and a microprocessor are arranged on the circuit board; The magnet assembly is fixed on the inner ring of the bearing; The Hall sensor is configured to generate a rotating speed signal of the bearing by sensing the magnetic field change generated when the magnet assembly rotates; The temperature sensor is configured to monitor the temperature signal of the bearing; The vibration acceleration sensor is configured to collect the vibration acceleration signal of the bearing; The microprocessor is configured to run an edge computing algorithm for real-time processing, AI model analysis and adaptive sampling processing of the rotating speed signal, the temperature signal and the vibration acceleration signal to determine whether the running state of the bearing is abnormal.

[0014] In a feasible implementation manner, the circuit board comprises: a micro 4G module and a circuit board body; The Hall sensor is arranged in the central region of the circuit board body; The temperature sensor is arranged on the edge of the circuit board body and close to the raceway of the bearing; The vibration acceleration sensor is arranged on the side of the circuit board body away from the center of the bearing; The micro 4G module is integrated on the circuit board body, and is used for data transmission between the Hall sensor, the temperature sensor, the vibration acceleration sensor and the microprocessor.

[0015] In a feasible implementation, the magnet assembly is a plurality of neodymium iron boron patch magnets uniformly fixed on the circumference of the bearing inner ring by an adhesive. The sensing end of the Hall sensor and the pole surface of the magnet assembly are aligned in the radial direction. The microprocessor realizes adaptive sampling and temperature compensation through an edge computing algorithm, and analyzes the vibration acceleration signal features in combination with an AI model to realize real-time judgment of the bearing failure stage.

[0016] From the above, the intelligent bearing dust cover and state monitoring method based on the edge computing method are provided, the sensor is integrated in the dust cover, the state monitoring can be realized without changing the bearing structure, and the destructive modification of the traditional method to the equipment is avoided. At the same time, the edge computing and the adaptive sampling strategy are adopted to reduce the data transmission amount and reduce the network bandwidth demand, and the accuracy of data acquisition is improved through temperature compensation. Through the high-frequency vibration acquisition and real-time analysis mechanism, the early failure signal of the bearing can be captured, potential problems can be warned in advance, the risk of unplanned shutdown is reduced, and the equipment operation and maintenance efficiency is improved. BRIEF DESCRIPTION OF DRAWINGS

[0017] The drawings incorporated into the specification and forming a part thereof show, in accordance with the embodiments of the present application, and, along with the specification, serve to explain the principles of the embodiments of the present application. It is apparent that the drawings described below are only some of the embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0018] Figure 1 The flowchart of the intelligent bearing state monitoring method based on the edge computing method shown in the exemplary embodiments of the present application; Figure 2 The overall structure diagram of the intelligent bearing dust cover based on the edge computing method shown in the exemplary embodiments of the present application; Figure 3 The cross-sectional diagram of the intelligent bearing dust cover based on the edge computing method shown in the exemplary embodiments of the present application after installation; Figure 4 The structure diagram of the circuit board shown in the exemplary embodiments of the present application.

[0019] Explanation of reference signs: 1-bearing; 2-dust cover body; 3-circuit board; 4-magnet assembly; 301-Hall sensor; 302-temperature sensor; 303-vibration acceleration sensor; 304-micro 4G module; 305-circuit board body. DETAILED DESCRIPTION

[0020] Example implementations will now be described more fully with reference to the accompanying drawings. Example implementations may, however, be implemented in many different forms and should not be construed as limited to the examples set forth herein; rather, these implementations are provided so that this disclosure will be thorough and complete, and will fully convey the scope of example implementations to those skilled in the art. The described features, acts, or characteristics can be combined in any suitable manner in one or more implementations. In the following description, numerous specific details are provided to give a thorough understanding of example implementations.

[0021] The existing bearing state monitoring technology generally needs to install additional sensors, which needs to destroy the bearing structure in the process, and has great limitations in adaptability, and generally needs to be customized according to the bearing state. Changing the bearing structure will destroy the stress distribution of the bearing, which may affect the service life of the bearing. In order to avoid this situation, a dust cover is designed, which is based on the original structure of the bearing dust cover, and without changing the connection mode with the bearing, the temperature, rotating speed, vibration acceleration and other sensors are integrated on the dust cover to realize the monitoring of the running state of the bearing.

[0022] The application provides an intelligent bearing state monitoring method based on an edge computing method, referring to FIG. 1, the method comprises the steps of: Figure 1 S100: integrating a temperature sensor, a rotating speed sensor and a vibration acceleration sensor on a bearing dust cover, and collecting a temperature signal, a rotating speed signal and a vibration acceleration signal of the bearing.

[0023] First, the temperature, rotating speed and vibration acceleration signals of the bearing are collected by using a specially designed bearing dust cover. The circuit board 3 is fixed inside the bearing dust cover, and the temperature sensor 302, the Hall sensor 301 (as a rotating speed sensor) and the vibration acceleration sensor 303 are integrated on the circuit board. The sensors are installed on the standard clamping groove of the bearing 1 through the fixing structure of the standard dust cover, without changing the original structure of the bearing, so as to avoid destroying the stress distribution of the bearing. The function of this step is to realize the real-time collection of the running parameters of the bearing, while maintaining the simplicity and universality of the installation.

[0024] ​S200: The vibration acceleration signal collected by the vibration acceleration sensor is processed by an adaptive sampling algorithm. The adaptive sampling algorithm performs peak detection based on a short-time RMS energy threshold method, and uses full sampling rate to maintain the vibration acceleration signal in the identified peak section, and uses sliding window mean sampling to reduce the sampling of the vibration acceleration signal in the smooth section.

[0025] Specifically, the microprocessor calculates the short-time RMS value of the vibration signal in real time to dynamically identify peak events. The peak section stores data at the original sampling rate (e.g., 10 kHz), and the smooth section outputs the mean value calculated by the sliding window to reduce the data volume, effectively reducing the redundant data generated by high-frequency sampling, and relieving the bandwidth pressure.

[0026] S300: Temperature compensation calculation is performed on the vibration acceleration signal processed by the adaptive sampling. Zero offset compensation function and sensitivity function are used to eliminate the temperature drift effect.

[0027] In this step, the temperature compensation calculation is performed on the down-sampled signal. The feedback value of the temperature sensor 302 is used to compensate the vibration signal for zero offset and sensitivity, eliminating the temperature drift effect of the piezoelectric acceleration sensor to improve the accuracy of vibration measurement.

[0028] S400: The vibration acceleration signal processed by the temperature compensation is analyzed by an AI model to extract the waveform features from the vibration acceleration signal, and the waveform features are compared with the pre-stored health state waveform features to determine the bearing fault stage based on the comparison result.

[0029] In this step, the AI model is run on the edge computing device to analyze the waveform in real time and quickly detect abnormalities.

[0030] This embodiment combines sensor integration with edge computing intelligent processing to solve the problems of complex installation, large data redundancy transmission, poor universality, etc. in traditional monitoring. Moreover, it does not need to change the bearing structure, protecting the mechanical integrity. The method greatly reduces the data transmission volume and improves the monitoring real-time performance through local adaptive sampling and AI analysis, and is suitable for standard bearings with strong universality.

[0031] In some embodiments, peak detection is achieved by the following steps: calculating the short-time energy of the vibration acceleration signal as a judgment criterion, and setting a dynamic threshold based on it to accurately capture transient events.

[0032] First, the short-time RMS energy of the signal is calculated. The microprocessor collects the vibration signal x[n] in real time, and sets a fixed length sliding window (e.g., corresponding to 10 ms of data points). For each time point n, its short-time RMS value is calculated according to the following formula: RMS[n]= RMS[n]= .

[0033] where RMS[n] represents the root mean square value at time point n, quantifying the local energy level of the signal within the window, providing a reference for subsequent peak detection. For the sliding window length, is the original value of the vibration acceleration signal at time point k.

[0034] Then, a dynamic threshold θ is set based on the short-time RMS energy threshold, and the formula of the dynamic threshold θ is: .

[0035] where, is the mean value of the history RMS of the stationary section, is a gain coefficient set between 2.5 and 3.5, which is used to effectively cover the background noise; β is a small safety margin, which is used to compensate for possible baseline drift of the signal.

[0036] During the detection process, the microprocessor continuously compares the RMS[n] calculated in real time with the dynamic threshold Once RMS[n] > θ, the microprocessor triggers a peak event.

[0037] The embodiment can reliably distinguish between real abnormal transient events and background noise through the dynamic threshold mechanism, effectively reducing false positives. This ensures that the subsequent adaptive sampling does not lose key feature information of bearing faults while significantly compressing the data volume, enhancing the robustness and accuracy of the entire monitoring system.

[0038] In some embodiments, to further improve the accuracy and reliability of transient event capture, the peak detection mechanism also introduces signal absolute rate analysis based on dynamic threshold judgment, and clearly defines the entering and exiting judgment conditions of the peak.

[0039] First, based on the rapid warning of the rate, the microprocessor calculates the absolute rate ΔX[n] of the vibration signal in real time, and sets a rate threshold (e.g. based on signal characteristics), when a warning is triggered, and a comprehensive judgment is made in combination with the RMS threshold.

[0040] where the formula for calculating the absolute rate is: ΔX[n]=∣x[n]-x[n-1]∣.

[0041] where ΔX[n] represents the absolute rate at time point n, x[n] is the original value of the vibration acceleration signal at time point n, and x[n-1] is the original value of the vibration acceleration signal at the previous time point. ​

[0042] Secondly, the system's determination of peak events is not instantaneous jump, but has a clear duration condition, in which the entry condition is that the RMS value is required to be continuously above the dynamic threshold and the duration is greater than the minimum peak duration (such as 5ms).

[0043] The exit condition is that the RMS value is continuously below the adjusted threshold / , =1.2-1.5) and the duration is greater than the exit holding time (such as 20ms).

[0044] This embodiment effectively filters transient interference by introducing a rate of change auxiliary judgment and strict duration determination, ensuring that the identified peak events have clear start and end boundaries and physical meaning, and enhancing the robustness of the adaptive sampling algorithm.

[0045] In some embodiments, when the signal is identified as a stationary segment, the microprocessor uses a sliding window mean method with a 50% overlap rate for downsampling processing to preserve the low-frequency trend while reducing the amount of data.

[0046] Specifically, the algorithm sets the window length to N and the sliding step to Step=N / 2, thereby achieving a 50% overlap rate. For each output point m, its value is calculated by the following formula: .

[0047] In the formula, y[m] is the downsampled output signal, N is the window length, is the original signal sample, and mstart is the starting point index of the mth window.

[0048] The calculation in this embodiment performs one mean filtering and downsampling. The main purpose of using an overlapping window design is to avoid signal distortion that may be caused by downsampling and smoothing, and to ensure the representativeness of the stationary segment data. This method is only activated in non-peak segments, thereby achieving effective data compression while maximizing the retention of macro trend information of the signal.

[0049] In some embodiments, to eliminate the influence of sensor temperature drift on measurement accuracy, the system will perform compensation calculation on the vibration signal based on the real-time temperature reading after completing adaptive sampling.

[0050] Specifically, the microprocessor reads the real-time value T of the temperature sensor 302 and combines it with the original output value O of the vibration acceleration sensor 303 to calculate the corrected effective output Oc through the following compensation formula: .

[0051] In the formula, Oc is the compensated effective output value, and O is the original output value of the sensor.

[0052] Wherein: Z(T) = aT + b is a compensation function of zero offset varying with temperature; S(T) = cT + d is a compensation function of sensitivity varying with temperature. Parameters a, b, c, d are constant coefficients obtained through pre-calibration of the sensor.

[0053] The compensation process of the embodiment can effectively suppress the zero offset and sensitivity drift of the piezoelectric acceleration sensor caused by the change of the ambient temperature, thereby significantly improving the measurement accuracy of the vibration signal and providing a high-quality data basis for the reliable analysis of the subsequent AI model.

[0054] In some embodiments, the AI model analysis link realizes the diagnosis of the bearing state through multi-dimensional waveform feature extraction. The waveform feature extraction covers time domain features, frequency domain features, time-frequency features, implicit rotating speed, and implicit temperature, and each dimension uses a special algorithm to fully mine the fault information in the vibration signal.

[0055] In the healthy stage, the bearing runs smoothly, and the waveform features show that the time domain is approximately Gaussian distributed, the kurtosis value is stable at about 3, the frequency domain energy is uniformly distributed and has no obvious peak value, the time-frequency feature has no regular transient event, the implicit rotating speed fluctuation is very small, usually less than 0.5%. These features collectively indicate that the bearing is in a damage-free state, providing a benchmark for subsequent fault judgment.

[0056] As the bearing begins to show early faults, such as damage less than 0.5mm, the time domain feature first shows periodic impact pulses, the kurtosis value increases significantly to more than 5, the frequency domain feature frequency energy appears initially but the signal-to-noise ratio is less than 5dB, the time-frequency feature shows that the impact event is synchronized with the rotation period, and the implicit rotating speed is basically stable but has a slight fluctuation, which indicates the presence of a fault.

[0057] In the medium-term fault stage, the bearing damage expands, the time domain feature further deteriorates, the impact amplitude increases and secondary impact appears, the frequency domain feature frequency and its sidebands become obvious, the signal-to-noise ratio increases to 5-15dB, the time-frequency feature modulation phenomenon is significant, the envelope spectrum energy is concentrated, and the implicit rotating speed fluctuation increases and is intensified by the load change. The coordinated changes in the multi-dimensional features in this stage provide key evidence for fault location.

[0058] When the fault progresses to the late stage, the damage exceeds the threshold, the time domain signal is saturated and periodic, the frequency domain wideband energy increases, the characteristic frequency is blurred, the time-frequency feature appears a large number of random impacts, the modulation phenomenon disappears, and the implicit rotating speed fluctuates violently and even loses lock. These features collectively point to the serious degradation of the bearing.

[0059] In this embodiment, the AI model can quickly respond and effectively prevent sudden failures by comprehensively judging the failure stage through real-time comparison of the above-mentioned features with the pre-stored health state waveform.

[0060] Another aspect of the present application provides an intelligent bearing dust cover based on an edge computing method, which is used to implement the intelligent bearing state monitoring method based on the edge computing method. Figures 2-4 As shown in the figure, it comprises a dust cover body 2, a circuit board 3 and a magnet assembly 4.

[0061] The dust cover body 2 is clamped to the standard clamping groove of the bearing 1 outer ring through the edge fixing structure, forming a non-invasive installation structure.

[0062] Further, the dust cover body 2 comprises a metal framework and a rubber sealing layer. The metal framework is interference-fitted with the clamping groove of the bearing 1 outer ring, and the rubber sealing layer is wrapped on the outer surface of the metal framework. This structure forms double sealing protection: the metal framework provides mechanical support, and the rubber sealing layer prevents contaminants from entering the bearing interior.

[0063] During installation, the metal framework is interference-fitted with the outer ring clamping groove to form a mechanical fixation, and the rubber sealing layer forms a sealing interface after compression. This avoids insufficient sealing performance of the dust cover and improves the dustproof and waterproof level through double sealing. At the same time, the mechanical strength of the metal framework ensures the installation stability and avoids the loosening problem caused by vibration.

[0064] The circuit board 3 is fixed to the side of the dust cover body 2 facing the interior space of the bearing 1, and a Hall sensor 301, a temperature sensor 302, a vibration acceleration sensor 303 and a microprocessor are integrated on it.

[0065] The Hall sensor 301 is configured to generate a bearing 1 rotation speed signal by sensing the magnetic field change generated when the magnet assembly 4 rotates; the temperature sensor 302 is configured to monitor the temperature signal of the bearing 1, and the vibration acceleration sensor 303 is configured to collect the vibration acceleration signal of the bearing 1.

[0066] In some embodiments, the circuit board 3 further comprises a circuit board body 305, the Hall sensor 301 is arranged in the central region, the temperature sensor 302 is located at the edge close to the bearing 1 raceway position, ensuring that the temperature sensor 302 can accurately monitor the raceway temperature signal of the bearing 1. The vibration acceleration sensor 303 is located on the side away from the center of the bearing 1, so that the vibration acceleration sensor 303 can efficiently capture the radial vibration acceleration signal. Each sensor is connected to the microprocessor through a printed circuit to form an integrated data acquisition module.

[0067] Further, the microprocessor is configured to run an edge computing algorithm for real-time processing, AI model analysis and adaptive sampling processing of the rotation speed signal, temperature signal and vibration acceleration signal to determine whether the running state of the bearing 1 is abnormal.

[0068] It can be understood that the circuit board 3 in the embodiment improves data accuracy by optimizing sensor positions, eliminates environmental influences by using a temperature compensation algorithm, and reduces the volume of the circuit board through integrated design, reduces the space occupied by the circuit board, and improves the system integration to adapt to the limited space of the dust cover.

[0069] The magnet assembly 4 is fixed to the inner ring of the bearing 1. In some embodiments, the magnet assembly 4 is composed of multiple neodymium-iron-boron patch magnets, which are uniformly bonded to the circumferential surface of the inner ring of the bearing 1 and aligned with the Hall sensor 301 in the radial direction. Surface treatment before bonding ensures bonding strength and forms a stable magnetic field signal source. The structure ensures that the magnetic field change is linearly related to the bearing speed, achieving accurate speed measurement. And through non-contact measurement, the bearing life is prolonged. At the same time, the use of neodymium-iron-boron material ensures the stability of the magnetic field strength, solving the problem of measurement error caused by low magnetic field strength.

[0070] Specifically, the process of the intelligent bearing dust cover based on the edge computing method of the present application for bearing state detection is as follows: when the bearing 1 rotates, the magnet assembly 4 rotates with the inner ring, the Hall sensor 301 senses the change of the magnetic field to generate a pulse signal, which is converted into rotation speed data by the microprocessor. The temperature sensor 302 monitors the temperature of the bearing 1 in real time, and the vibration acceleration sensor 303 collects vibration acceleration signals. The microprocessor processes the rotation speed signal, temperature signal and vibration acceleration signal in real time: corrects the temperature drift of the vibration acceleration sensor through the temperature signal, and analyzes the vibration characteristics combined with the rotation speed signal.

[0071] The structure of the intelligent bearing dust cover provided in the embodiment solves the problem of traditional technology that needs to damage the bearing structure to install sensors, avoiding the influence of stress distribution change on bearing life; through edge computing, local data processing is realized, solving the technical problem of delayed response caused by cloud transmission.

[0072] In some embodiments of the present application, the circuit board 3 further comprises a micro 4G module 304, which is welded to the upper surface of the circuit board body 305. The micro 4G module 304 is connected to the microprocessor through printed circuit to realize data transmission function. The size of the micro 4G module 304 is adapted to the internal space of the dust cover, which can ensure that it does not affect the layout of other sensors.

[0073] It can be understood that the data processed by the microprocessor is uploaded to the cloud system through the micro 4G module 304, and the wireless transmission improves the flexibility of the system. At the same time, the low-power design of the micro 4G module prolongs the battery life and ensures continuous monitoring, prolonging the service life of the system.

[0074] In combination with the above embodiments, the intelligent bearing dust cover and state monitoring method of the present application realizes real-time monitoring of the bearing state through non-invasive design, avoids destructive modification of the bearing structure in the traditional way, thereby maintaining the mechanical integrity and service life of the bearing. The use of edge computing strategy effectively reduces the data transmission amount and reduces the network bandwidth demand. At the same time, the accuracy of data acquisition is improved through adaptive sampling and temperature compensation. The scheme also enhances the environmental adaptability and reliability of the system, supports remote monitoring and intelligent decision-making, helps early fault warning, reduces the risk of unplanned downtime, and improves the efficiency of equipment operation and maintenance.

[0075] Other embodiments of the present disclosure will be readily apparent to those skilled in the art upon considering the disclosure herein, particularly the examples contained herein. The present application is intended to cover any variations, uses, or adaptations of the present disclosure that follow, in general, the principles of the present disclosure and include other known ins and outs in the art to which the present disclosure pertains. The specification and examples are considered exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

Claims

1. A smart bearing condition monitoring method based on edge computing, characterized in that, The method includes: Temperature sensor, speed sensor and vibration acceleration sensor are integrated into bearing dust cover to collect bearing temperature signal, speed signal and vibration acceleration signal; The vibration acceleration signal collected by the vibration acceleration sensor is processed by an adaptive sampling algorithm. The adaptive sampling algorithm is based on the short-time RMS energy threshold method for peak detection. In the identified peak segment, the vibration acceleration signal is maintained at full sampling rate. In the stable segment, the vibration acceleration signal is downsampled by sliding window mean sampling. Temperature compensation calculations are performed on the vibration acceleration signal after adaptive sampling processing, and the effects of temperature drift are eliminated by using a zero-bias compensation function and a sensitivity function. The vibration acceleration signal after temperature compensation is analyzed by an AI model. Waveform features are extracted from the vibration acceleration signal and compared with pre-stored healthy state waveform features. The bearing failure stage is determined based on the comparison results.

2. The intelligent bearing condition monitoring method based on edge computing as described in claim 1, characterized in that, The peak detection includes calculating the short-time RMS energy threshold of the vibration acceleration signal as the peak judgment benchmark, and setting a dynamic threshold to achieve accurate capture of transient events. The formula for calculating the short-time RMS energy threshold is as follows: RMS[n]= ; In the formula, RMS[n] represents the root mean square value at time point n. The length of the sliding window. This represents the original value of the vibration acceleration signal at time point k; A dynamic threshold θ is set based on the short-time RMS energy threshold. The formula for the dynamic threshold θ is: ； In the formula, This is the gain coefficient. β is the historical RMS mean value during the steady-state period, and β is the safety margin used to compensate for signal baseline drift.

3. The intelligent bearing condition monitoring method based on edge computing as described in claim 1 or 2, characterized in that, The peak detection, based on dynamic threshold judgment, also includes: The rapid change characteristics of the vibration acceleration signal are captured by calculating the absolute rate of change of the vibration acceleration signal; An alert is triggered when the absolute rate of change exceeds a preset rate of change threshold. The peak threshold condition is that the RMS value continuously exceeds the dynamic threshold and the duration is greater than the preset minimum peak duration. The peak exit condition is that the RMS value is continuously lower than the adjusted dynamic threshold and the duration is greater than the preset exit hold time; The formula for calculating the absolute rate of change is: ΔX[n]=|x[n]-x[n-1]|; In the formula, ΔX[n] represents the absolute rate of change at time point n, x[n] is the original value of the vibration acceleration signal at time point n, and x[n-1] is the original value of the vibration acceleration signal at the previous time point.

4. The intelligent bearing condition monitoring method based on edge computing as described in claim 1, characterized in that, The sliding window mean sampling employs an overlapping window method in the processing of the stationary segment, and its calculation formula is as follows: ； Step = N / 2; In the formula, y[m] is the downsampled output signal, N is the window length, and Step=N / 2 is the sliding step size, achieving a 50% overlap rate to reduce smoothing distortion. The original signal sample, This is the index of the starting point of the m-th window.

5. The intelligent bearing condition monitoring method based on edge computing as described in claim 1 or 2, characterized in that, It also includes temperature compensation for the vibration acceleration sensor, and calculating the effective output value after compensation using the sensor output value, zero bias compensation function and sensitivity function; The temperature compensation formula is as follows: ； In the formula, Oc is the effective output value after compensation, O is the original output value of the sensor, Z(T) is the zero bias compensation function, and S(T) is the sensitivity compensation function. Where Z(T) = aT + b; S(T) = cT + d; where T is the temperature value, and a, b, c, and d are constants obtained through sensor calibration.

6. The intelligent bearing condition monitoring method based on edge computing as described in claim 1, characterized in that, The waveform feature extraction in the AI ​​model analysis includes: time-domain features, frequency-domain features, time-frequency features, implicit rotational speed, and implicit temperature; The time-domain features identify impact characteristics through kurtosis calculation, the frequency-domain features detect fault frequencies through FFT analysis, the time-frequency features analyze modulation phenomena through wavelet transform, the implicit rotational speed calibrates the rotational frequency through adaptive filtering, and the implicit temperature evaluates thermal deformation through long-term trend analysis.

7. The intelligent bearing condition monitoring method based on edge computing as described in claim 6, characterized in that, The failure phases include: health phase, early failure phase, mid-term failure phase, and late failure phase; The criteria for determining the health stage include: the kurtosis value of the time-domain feature is between 2.9 and 3.1, the peak signal-to-noise ratio of the frequency-domain feature is less than 3dB and the energy distribution is uniform, the time-frequency feature has no irregular transient events, and the implicit rotational speed fluctuation is between 0 and 0.5%. The criteria for determining the early fault stage include: the time-domain feature exhibits periodic impact pulses with a kurtosis value greater than 5, the frequency-domain feature has a signal-to-noise ratio of less than 5 dB, and the time-frequency feature impact event is synchronized with the rotation period. The criteria for determining the intermediate fault stage include: the amplitude of the time-domain characteristic impact increases and a secondary impact occurs, the signal-to-noise ratio of the frequency domain characteristic is between 5 and 15 dB, and the time-frequency characteristic modulation phenomenon is significant and the envelope spectrum energy is concentrated. The criteria for determining the late-stage fault include: the time-domain characteristic signal is saturated and periodically disappears, the frequency-domain characteristic broadband energy increases and the characteristic frequency becomes blurred, and the time-frequency characteristics experience a large number of random impacts and the modulation phenomenon disappears.

8. A smart bearing dust cover based on edge computing, used to implement the smart bearing condition monitoring method based on edge computing as described in any one of claims 1-7, characterized in that, include: Dust cover body (2), circuit board (3) and magnet assembly (4); The dust cover body (2) is snapped onto the standard slot of the outer ring of the bearing (1) by the fixing structure of its edge. The dust cover body (2) includes a metal skeleton and a rubber sealing layer and is sealed to the bearing (1). The circuit board (3) is fixed to the side of the dust cover body (2) facing the internal space of the bearing (1), and a Hall sensor (301), a temperature sensor (302), a vibration acceleration sensor (303) and a microprocessor are provided on it; The magnet assembly (4) is fixed to the inner ring of the bearing (1); The Hall sensor (301) is configured to sense the change in magnetic field generated when the magnet assembly (4) rotates to generate a rotational speed signal of the bearing (1); The temperature sensor (302) is configured to monitor the temperature signal of the bearing (1); The vibration acceleration sensor (303) is configured to acquire the vibration acceleration signal of the bearing (1); The microprocessor is configured to run an edge computing algorithm to perform real-time processing, AI model analysis and adaptive sampling processing on the rotation speed signal, temperature signal and vibration acceleration signal, in order to determine whether there is an abnormality in the operating state of the bearing (1).

9. The intelligent bearing dust cover based on edge computing method according to claim 8, characterized in that, The circuit board (3) includes: a micro 4G module (304) and a circuit board body (305). The Hall sensor (301) is disposed in the central area of ​​the circuit board body (305); The temperature sensor (302) is located on the edge of the circuit board body (305) and close to the raceway of the bearing (1); The vibration acceleration sensor (303) is disposed on the side of the circuit board body (305) away from the center of the bearing; The micro 4G module (304) is integrated on the circuit board body (305) and is used for data transmission between the Hall sensor (301), temperature sensor (302), vibration acceleration sensor (303) and the microprocessor.

10. The intelligent bearing dust cover based on edge computing method according to claim 9, characterized in that, The magnet assembly (4) consists of multiple neodymium iron boron patch magnets, which are uniformly fixed to the circumference of the inner ring of the bearing (1) by adhesive. The sensing end of the Hall sensor (301) is aligned radially with the magnetic pole surface of the magnet assembly (4); The microprocessor uses edge computing algorithms to achieve adaptive sampling and temperature compensation, and combines AI models to analyze the characteristics of the vibration acceleration signal to determine the bearing failure stage in real time.

Images