A fault diagnosis system for equipment bearings based on the shock pulse method

Abstract

The utility model discloses a kind of fault diagnosis systems of bearing based on impact pulse method equipment, the fault diagnosis system uses modularization design, comprising: impact pulse sensor module, signal conditioning module, characteristic calculation module, visual diagnosis module, impact pulse sensor module, signal conditioning module, characteristic calculation module, visual diagnosis module each module between through standard industrial interface realizes data interaction and function cooperation, the probe of impact pulse sensor module is connected to the most sensitive position of bearing seat radial vibration, ensure the integrity of vibration transmission path;The utility model realizes accurate monitoring and early fault diagnosis to low-speed heavy load bearing operating state by multi-module cooperative work, the utility model has the characteristics such as simple installation, accurate diagnosis, early warning in time, can significantly improve the safety operation level of mine hoisting equipment.

Description

Technical Field

[0001] This utility model relates to a bearing fault diagnosis system, specifically a bearing fault diagnosis system based on the impact pulse method, which is particularly suitable for early warning of the bearing working status of low-speed heavy-load equipment in the auxiliary vertical shaft hoisting system of mines, and realizes the detection of early bearing faults. Background Technology

[0002] Currently, the fault diagnosis of equipment bearings used in mine auxiliary vertical shaft hoisting systems mainly includes the following:

[0003] 1. Fault diagnosis based on signal decomposition and entropy analysis: used for vibration signal denoising and feature extraction, combined with entropy analysis (such as sample entropy, fuzzy entropy, etc.) for fault detection; however, it has the following drawbacks: ① Modal aliasing problem (EMD / EEMD): different frequency components interfere with each other during signal decomposition, affecting the accuracy of feature extraction; ② Strong parameter dependence (VMD): the number of modes (K) and penalty factor (α) need to be manually set, and improper adjustment will lead to over-decomposition or under-decomposition; ③ High computational complexity: poor real-time performance, making it difficult to meet the online monitoring needs of mine hoisting systems;

[0004] 2. Machine learning-based classification (SVM, ELM, random forest, etc.): This method uses time-frequency domain features (such as kurtosis and envelope spectrum) to train classification models for fault identification. However, it has the following drawbacks: ① It relies on manual feature extraction: Expert experience is required to select effective features, resulting in poor versatility; ② It is sensitive to noise: In the high-noise environment of mines, features are easily interfered with, leading to a decrease in classification accuracy; ③ It is difficult to adapt to changing working conditions: When the training data does not match the actual operating conditions, the model performance drops sharply.

[0005] 3. Deep learning-based intelligent diagnosis (CNN, LSTM, Transformer, etc.): Automatically extracts fault features end-to-end, suitable for complex fault mode identification; however, it has the following drawbacks: ① Strong data dependence: Requires a large number of fault samples for training, but fault data is scarce in actual working conditions; ② High computational resource consumption: Training and inference require high-performance hardware, which is difficult to deploy on edge devices; ③ Poor interpretability: The model decision-making process is not transparent, which is not conducive to fault mechanism analysis.

[0006] 4. Multi-sensor data fusion: Combining information from multiple sources such as vibration, temperature, and acoustic emission improves diagnostic reliability; however, it has the following drawbacks: ① High hardware cost: It requires the deployment of multiple high-precision sensors, which are complex to install and maintain; ② Difficulty in data synchronization: Different sensors have inconsistent sampling frequencies, resulting in high complexity of the fusion algorithm; Redundant information interference; ③ Low correlation of some sensor data, which may reduce diagnostic efficiency. Utility Model Content

[0007] To address the problems existing in the prior art, this utility model provides a fault diagnosis system for equipment bearings based on the impact pulse method. This system achieves accurate monitoring and early fault diagnosis of the operating status of low-speed, heavy-load bearings through the collaborative work of multiple modules. It solves the problem of low accuracy in fault diagnosis of equipment bearings under low-speed, heavy-load, and high-noise conditions currently applied in mine auxiliary shaft systems. At the same time, it can assess the working status of the bearing and accurately determine the fault type, thus playing an early warning role.

[0008] To achieve the above objectives, the technical solution adopted by this utility model is: a fault diagnosis system for equipment bearings based on the impact pulse method. The fault diagnosis system adopts a modular design and includes: an impact pulse sensor module, a signal conditioning module, a feature calculation module, and a visualization diagnosis module. The impact pulse sensor module, signal conditioning module, feature calculation module, and visualization diagnosis module interact and coordinate functions with each other through a standard industrial interface. The probe of the impact pulse sensor module is connected to the most sensitive position of radial vibration of the bearing housing to ensure the integrity of the vibration transmission path.

[0009] Furthermore, the impact pulse sensor module is the data acquisition front end of the fault diagnosis system. It adopts an ICP-type piezoelectric accelerometer, whose core component is a PZT-5A lead zirconate titanate piezoelectric ceramic element with a resonant frequency of 32kHz±1kHz and a sensitivity of 100mV / g.

[0010] Furthermore, the piezoelectric accelerometer integrates a JFET impedance transformation circuit to convert high-impedance charge signals into low-impedance voltage signals.

[0011] Furthermore, the piezoelectric accelerometer housing adopts a stainless steel sealed design, with a protection level of IP68, which can adapt to the humid and dusty environment of mines.

[0012] Furthermore, the signal conditioning module includes a three-stage signal processing circuit. The first stage is an instrumentation amplifier, model AD620, with a gain set to 20dB, used to increase the signal amplitude. The second stage is an active bandpass filter based on OPA2188 op-amp, with cutoff frequencies set to 28kHz and 36kHz and an attenuation slope of 40dB / dec. The third stage is a precision rectifier circuit implemented using AD824, which converts bidirectional signals into unidirectional signals.

[0013] Furthermore, the signal sampling in the signal conditioning module adopts a 24-bit Σ-Δ ADC with a sampling rate of 100kS / s, which meets the requirements of the Nyquist sampling theorem. For low-speed operation, an RS422 standard incremental encoder interface is configured, and signal acquisition is triggered by Z pulse to ensure that the sampling window is synchronized with the phase of the shaft rotation.

[0014] Furthermore, the feature calculation module runs on an ARM Cortex-A8 embedded processor.

[0015] Furthermore, the visualization diagnostic module is software developed based on LabVIEW in an industrial control computer. It includes four main functional interfaces: real-time monitoring interface, 3D bearing model, alarm management, and data storage using an SQLite database.

[0016] Furthermore, the fault diagnosis system adopts an intrinsically safe design with an input voltage of 24VDC, which generates ±15V analog circuit and 3.3V digital circuit power supply through a DC-DC converter.

[0017] Furthermore, all circuit boards in the fault diagnosis system are treated to prevent moisture, mold, and salt spray, meeting the requirements for use in explosive atmospheres as per GB3836.1-2010.

[0018] The beneficial effects of this utility model are as follows: Through the collaborative work of multiple modules, this utility model achieves accurate monitoring and early fault diagnosis of the operating status of low-speed heavy-load bearings, solving the problem of the accuracy of fault diagnosis of bearings in low-speed heavy-load and high-noise operating conditions currently applied in mine auxiliary vertical shaft systems. At the same time, it can assess the working status of the bearings and accurately determine the fault type, playing an early warning role. This utility model has the characteristics of simple installation, accurate diagnosis, and timely early warning, which can significantly improve the safe operation level of mine hoisting equipment. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the structure of this utility model;

[0020] Figure 2 This is a schematic diagram of the arrangement structure of this utility model;

[0021] In the diagram: 1. Impact pulse sensor module; 2. Signal conditioning module; 3. Feature calculation module; 4. Visual diagnostic module; 5. Industrial control computer. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this utility model clearer, the present utility model will be further described in detail below with reference to the accompanying drawings and embodiments. However, it should be understood that the specific embodiments described herein are merely illustrative of the present utility model and are not intended to limit its scope.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention.

[0024] like Figure 1-2 As shown, a fault diagnosis system for equipment bearings based on the impact pulse method is presented. This system achieves accurate monitoring and early fault diagnosis of the operating status of low-speed heavy-load bearings through the collaborative work of multiple modules. The system adopts a modular design and consists of an impact pulse sensor module 1, a signal conditioning module 2, a feature calculation module 3, and a visualization diagnosis module 4. The modules interact with each other and coordinate functions through a standard industrial interface.

[0025] The impact pulse sensor module 1 is the data acquisition front end of the system. It adopts a specially designed ICP piezoelectric accelerometer, whose core component is a PZT-5A lead zirconate titanate piezoelectric ceramic element with a resonant frequency of 32kHz±1kHz and a sensitivity of 100mV / g. The piezoelectric accelerometer integrates a JFET impedance transformation circuit, which can convert high-impedance charge signals into low-impedance voltage signals. The piezoelectric accelerometer housing adopts a stainless steel sealed design with an IP68 protection rating, which can adapt to the humid and dusty environment of the mine. During installation, the piezoelectric accelerometer probe is connected to the most sensitive position of radial vibration on the bearing housing, usually located directly above the load area, to ensure the integrity of the vibration transmission path. The piezoelectric accelerometer operates in the 1-40kHz frequency band, and its response sensitivity at the 32kHz center frequency is more than 15dB higher than other frequency bands, which can effectively amplify the bearing fault characteristic signals.

[0026] Signal conditioning module 2 includes a three-stage signal processing circuit to achieve precise conditioning of the raw signal. The first stage is an instrumentation amplifier, model AD620, with a gain set to 20dB, used to increase the signal amplitude. The second stage is an active bandpass filter based on OPA2188 operational amplifier, with cutoff frequencies set to 28kHz and 36kHz and an attenuation slope of 40dB / dec. The third stage is a precision rectifier circuit implemented using AD824 to convert bidirectional signals into unidirectional signals. Signal sampling uses a 24-bit Σ-Δ ADC (ADS1256) with a sampling rate set to 100kS / s, meeting the requirements of the Nyquist sampling theorem. For low-speed operation, the system is equipped with an RS422 standard incremental encoder interface, which triggers signal acquisition via Z-pulse to ensure that the sampling window is synchronized with the phase of the shaft rotation. Data from 50 complete rotation cycles is collected in each analysis cycle, approximately 1 minute in length. Real-time Hilbert transform is performed using a TMS320F28335 DSP chip to calculate the signal envelope.

[0027] Feature calculation module 3 runs on an ARM Cortex-A8 embedded processor and includes three core algorithms: the HDm calculation algorithm performs peak detection on the envelope signal, uses a five-point cubic smoothing algorithm to eliminate glitches, retains the 50 peaks with the largest amplitude, takes the average value, and converts it to a decibel value; the HDc calculation algorithm divides 1 second of data into 200 5ms time intervals, uses the sliding window method to extract the maximum value of each time interval, and uses a quick sorting algorithm to find the minimum value; the HDi calculation algorithm reads the speed signal in the PLC through the Modbus protocol, combines it with the pre-stored bearing parameters, and calculates the correction value in real time; the calculation results are transmitted to the industrial control computer 5 through the CAN bus, with a transmission cycle of 10 seconds.

[0028] The visualization diagnostic module 4 is software developed using LabVIEW in the industrial control computer 5. It includes four main functional interfaces: the real-time monitoring interface displays time-domain waveforms, envelope spectra, and eigenvalue trend graphs with a sampling rate refreshed at 10Hz, where the trend graphs can trace back 24 hours of data; the 3D bearing model is rendered using OpenGL, mapping the fault location onto the bearing geometry model, with the inner ring, outer ring, rolling elements, and cage marked with different colors; the alarm management adopts a three-level alarm mechanism, where a yellow warning triggers an SMS notification, and a red alarm simultaneously activates the audible and visual alarms, including a 105dB buzzer and LED warning lights; data storage uses an SQLite database, recording a complete set of data every 10 seconds, including the original waveform, eigenvalues, and operating parameters, and supports exporting CSV format reports.

[0029] The fault diagnosis system adopts an intrinsically safe design, with an input voltage of 24VDC. It generates ±15V analog circuit power and 3.3V digital circuit power through a DC-DC converter. All circuit boards in the fault diagnosis system are treated for moisture resistance, mildew resistance, and salt spray resistance, meeting the requirements for use in explosive atmospheres (GB3836.1-2010). This fault diagnosis system features easy installation, accurate diagnosis, and timely early warning, significantly improving the safe operation of mine hoisting equipment.

[0030] The impact pulse sensor module 1 is connected to the feature calculation module 3 through the signal conditioning module 2. The feature calculation module 3 transmits data to the industrial control computer 5 via the CAN bus and is analyzed by the visualization diagnostic module 4.

[0031] During installation:

[0032] 1. Determine the specific location of the bearing based on the model of the key equipment in the mine's auxiliary vertical shaft hoisting system;

[0033] 2. Use adhesive substances such as glue to install the piezoelectric accelerometer probe at the bearing location, ensuring good contact between the piezoelectric accelerometer and the bearing surface to accurately obtain the bearing vibration signal;

[0034] 3. Connect the piezoelectric accelerometer, signal conditioning module 2, feature calculation module 3, and industrial control computer 5 together, and use the visualization diagnostic module 4 in the industrial control computer 5 to accurately detect the working status, damage degree, and fault type of the bearing.

[0035] Among them, the visualization diagnostic module 4 is the SPM Condmaster Ruby 2022.6.3 analysis software developed based on LabVIEW.

[0036] The above description is only a preferred embodiment of the present utility model and is not intended to limit the present utility model. Any modifications, equivalent substitutions or improvements made within the spirit and principles of the present utility model should be included within the protection scope of the present utility model.

Claims

1. A fault diagnosis system for equipment bearings based on the impact pulse method, characterized in that, The fault diagnosis system adopts a modular design, including: an impact pulse sensor module, a signal conditioning module, a feature calculation module, and a visualization diagnosis module. The impact pulse sensor module, signal conditioning module, feature calculation module, and visualization diagnosis module achieve data interaction and functional collaboration through a standard industrial interface. The probe of the impact pulse sensor module is connected to the most sensitive position of radial vibration of the bearing housing to ensure the integrity of the vibration transmission path.

2. The fault diagnosis system for equipment bearings based on the impact pulse method according to claim 1, characterized in that, The impact pulse sensor module is the data acquisition front end of the fault diagnosis system. It adopts an ICP type piezoelectric accelerometer, and its core component is a PZT-5A lead zirconate titanate piezoelectric ceramic element with a resonant frequency of 32kHz±1kHz and a sensitivity of 100mV / g.

3. The fault diagnosis system for equipment bearings based on the impact pulse method according to claim 2, characterized in that, The piezoelectric accelerometer integrates a JFET impedance transformation circuit to convert a high-impedance charge signal into a low-impedance voltage signal.

4. The fault diagnosis system for equipment bearings based on the impact pulse method according to claim 2, characterized in that, The piezoelectric accelerometer housing features a stainless steel sealed design with an IP68 protection rating, making it suitable for the humid and dusty environment of mines.

5. The fault diagnosis system for equipment bearings based on the impact pulse method according to claim 1, characterized in that, The signal conditioning module contains a three-stage signal processing circuit. The first stage is an instrumentation amplifier, model AD620, with a gain set to 20dB, used to increase the signal amplitude. The second stage is an active bandpass filter based on OPA2188 op-amp, with cutoff frequencies set to 28kHz and 36kHz and an attenuation slope of 40dB / dec. The third stage is a precision rectifier circuit implemented using AD824, which converts bidirectional signals into unidirectional signals.

6. The fault diagnosis system for equipment bearings based on the impact pulse method according to claim 5, characterized in that, The signal sampling in the signal conditioning module uses a 24-bit Σ-Δ ADC with a sampling rate of 100kS / s, which meets the requirements of the Nyquist sampling theorem. For low-speed operation, an RS422 standard incremental encoder interface is configured, and signal acquisition is triggered by Z pulse to ensure that the sampling window is synchronized with the phase of the shaft rotation.

7. The fault diagnosis system for equipment bearings based on the impact pulse method according to claim 1, characterized in that, The feature calculation module runs on an ARM Cortex-A8 embedded processor.

8. A fault diagnosis system for equipment bearings based on the impact pulse method according to claim 1, characterized in that, The visual diagnostic module is software developed based on LabVIEW in an industrial control computer. It includes four main functional interfaces: real-time monitoring interface, 3D bearing model, alarm management, and data storage using an SQLite database.

9. A fault diagnosis system for equipment bearings based on the impact pulse method according to claim 1, characterized in that, The fault diagnosis system adopts an intrinsically safe design with an input voltage of 24VDC. It generates ±15V analog circuit and 3.3V digital circuit power supply through a DC-DC converter.

10. A fault diagnosis system for equipment bearings based on the impact pulse method according to claim 1, characterized in that, All circuit boards in the fault diagnosis system are treated to prevent moisture, mildew, and salt spray, meeting the requirements for use in explosive atmospheres as per GB3836.1-2010.

Images