Track mechanical fault diagnosis method and system based on axle counting triggering

Abstract

The invention discloses a track mechanical fault diagnosis method and system based on axle counting triggering, and belongs to the technical field of track traffic equipment monitoring, and the method comprises the steps: collecting the vibration data and temperature data of a train through an event triggering and synchronous collection mode; preprocessing the vibration data and the temperature data; and feature extraction is performed on the vibration data, collaborative diagnosis is performed in combination with the temperature data, and a track mechanical fault diagnosis result is generated. Vibration acquisition is accurately triggered by the axle counting sensor, the track mechanical state is efficiently and accurately diagnosed in cooperation with the vibration sensor and the temperature sensor, and compared with an existing vibration monitoring system, continuous acquisition is changed into event triggering acquisition, so that the data volume is greatly reduced, the system efficiency is improved, and the system power consumption is reduced. Accurate space-time correlation between the vibration event and the wheel position is achieved, the fault positioning accuracy reaches the'axle counting point 'level, and the problems that an existing vibration monitoring system is low in diagnosis efficiency and low in accuracy are solved.

Description

Technical Field

[0001] This invention belongs to the field of rail transit equipment monitoring technology, specifically relating to a rail machinery fault diagnosis method and system based on axle counting triggering. Background Technology

[0002] Existing vibration monitoring systems mostly employ continuous acquisition or simple threshold triggering modes, resulting in large data volumes, numerous invalid data points, and simplistic processing methods. They largely rely on time-domain amplitude threshold judgments and lack the ability to deeply mine the characteristics of non-stationary and nonlinear vibration signals. This leads to low fault identification rates, difficulty in early warning, and an inability to accurately correlate vibration events with specific wheels, resulting in low diagnostic efficiency and inaccurate location. Axle counting systems have limited functionality, only used for counting, and their valuable train passage time information is not utilized to enhance condition monitoring. Summary of the Invention

[0003] In view of the above-mentioned shortcomings in the prior art, the track machinery fault diagnosis method and system based on axle counting triggering provided by the present invention solves the problems of low diagnostic efficiency and low accuracy of existing vibration monitoring systems.

[0004] To achieve the above-mentioned objectives, the technical solution adopted by this invention is: a fault diagnosis method for track machinery based on axle counting triggering, comprising the following steps: S1. Collect train vibration data and temperature data through event triggering and synchronous acquisition methods; S2. Preprocess the vibration and temperature data; S3. Extract features from vibration data and combine them with temperature data for collaborative diagnosis to generate fault diagnosis results for track machinery.

[0005] Furthermore: In S1, the method for collecting train vibration data through event triggering and synchronous acquisition is as follows: When the axle counting sensor detects the passing of a train wheel, it sends a trigger signal to the synchronous acquisition controller, which then controls the vibration acceleration sensor to collect the train's vibration data.

[0006] Furthermore, in S2, the specific method for preprocessing the vibration data is as follows: The vibration data is processed using an adaptive wavelet threshold denoising method. A wavelet basis that matches the impact component is selected to separate the background noise from the fault impact data. The specific method for preprocessing temperature data is as follows: Temperature data is filtered using a moving average and then combined with historical data for temperature gradient compensation.

[0007] Furthermore, in S3, the specific method for feature extraction of vibration data is as follows: Time-frequency domain feature extraction: Wavelet packet transform is performed on the vibration data to decompose it into independent frequency bands, and the energy entropy and normalized energy value of each frequency band are calculated as feature vectors; Demodulation analysis: Perform Hilbert transform on the vibration data and solve for its envelope spectrum; Nonlinear feature extraction: Calculate the nonlinear dynamic features of vibration data, including sample entropy and permutation entropy.

[0008] Furthermore: In S3, the method for collaborative diagnosis combining temperature data is as follows: Vibration data and temperature data correlation: Vibration data is correlated with the axle counting point location that triggered the vibration data acquisition, wheel sequence and temperature data to perform collaborative diagnosis, which includes fault location, trend analysis and stress state assessment. Fault location: Based on the trigger signal output by the axle counter sensor, the abnormal vibration characteristics are located to the specific axle counter point. Trend analysis: By comparing the vibration data triggered by different trains passing the same axle counting point, we can analyze its long-term trend and provide early warning of track condition deterioration. Stress state assessment: Based on temperature data analysis, the changes in track vibration characteristics under different thermal stress conditions are assessed to evaluate the risk of track bulging.

[0009] A track machinery fault diagnosis system based on axle counting triggering includes: Axle counter sensor is used to detect the time when the wheel flange passes by and output a trigger signal; Vibration acceleration sensors are used to collect vibration data from trains; Temperature sensors are used to collect temperature data from the train. The synchronous acquisition controller is used to receive trigger signals and control the vibration sensor and temperature sensor to synchronously acquire data before and after the train passes. The intelligent analysis center is used to extract features and perform collaborative diagnosis on vibration and temperature data to generate fault diagnosis results for track machinery.

[0010] The beneficial effects of this invention are as follows: This invention provides a method and system for fault diagnosis of track machinery based on axle counting triggering. It utilizes an axle counting sensor to accurately trigger vibration acquisition, and coordinates vibration sensors and multiple temperature sensors to perform efficient and accurate diagnosis of the track machinery condition. Compared with existing vibration monitoring systems, it has the following advantages: (1) High efficiency: The continuous acquisition is changed to event-triggered acquisition, which greatly reduces the amount of data, improves system efficiency and reduces system power consumption.

[0011] (2) High precision: It realizes the precise spatiotemporal correlation between vibration events and wheel positions, and the fault location accuracy reaches the level of "axle counting point".

[0012] (3) High reliability: The reliable triggering of the shaft counting sensor ensures the relevance and effectiveness of vibration data analysis.

[0013] (4) Strong anti-interference ability: The preprocessing technology such as adaptive wavelet denoising effectively suppresses the interference of complex environment on site and ensures the accuracy of feature extraction. Attached Figure Description

[0014] Figure 1 This is a flowchart of the fault diagnosis method for track machinery based on axle counting triggering according to the present invention.

[0015] Figure 2 This is a schematic diagram of the track machinery fault diagnosis system based on axle counting triggering according to the present invention. Detailed Implementation

[0016] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0017] like Figure 1 As shown, in one embodiment of the present invention, the track machinery fault diagnosis method based on axle counting triggering includes the following steps: S1. Collect train vibration data and temperature data through event triggering and synchronous acquisition methods; S2. Preprocess the vibration and temperature data; S3. Extract features from vibration data and combine them with temperature data for collaborative diagnosis to generate fault diagnosis results for track machinery.

[0018] In S1, the method for collecting train vibration data through event triggering and synchronous acquisition is as follows: When the axle counting sensor detects the passing of a train wheel, it sends a trigger signal to the synchronous acquisition controller, which then controls the vibration acceleration sensor to collect the train's vibration data.

[0019] In S2, the specific method for preprocessing vibration data is as follows: The vibration data is processed using an adaptive wavelet threshold denoising method. A wavelet basis that matches the impact component is selected to separate background noise from the fault impact data and improve the signal-to-noise ratio. The specific method for preprocessing temperature data is as follows: Temperature data is filtered using a moving average to eliminate instantaneous fluctuations, and temperature gradient compensation is performed by combining historical data to ensure data stability and representativeness.

[0020] In S3, the specific method for feature extraction from vibration data is as follows: Time-frequency domain feature extraction: Wavelet packet transform is performed on the vibration data to decompose it into independent frequency bands, and the energy entropy and normalized energy value of each frequency band are calculated as feature vectors. In this embodiment, the time-frequency domain feature extraction method can accurately capture the energy changes caused by different faults such as fastener loosening and corrugation in different frequency bands.

[0021] Demodulation analysis: Perform Hilbert transform on the vibration data and solve its envelope spectrum; the envelope spectrum can highlight the low-frequency fault characteristic frequencies (such as the fault frequencies of bearings and gears) modulated in the high-frequency resonance signal, which is suitable for early fault diagnosis of vehicle running gear.

[0022] Nonlinear feature extraction: Calculate the nonlinear dynamic characteristics of vibration data, including sample entropy and permutation entropy, to characterize the degree of chaos in the dynamic characteristics of the system caused by track bed compaction, subgrade settlement, etc.

[0023] In S3, the method for collaborative diagnosis by combining temperature data is as follows: Vibration data and temperature data correlation: Vibration data is correlated with the axle counting point location that triggered the vibration data acquisition, wheel sequence and temperature data to perform collaborative diagnosis, which includes fault location, trend analysis and stress state assessment. Fault location: Based on the trigger signal output by the axle counter sensor, the abnormal vibration characteristics are located to the specific axle counter point. In this embodiment, since the axle counting sensor installation point has clear coordinates, and the vibration data acquisition is triggered by the axle counting sensor outputting a trigger signal when it senses the vehicle, the axle counting point location can be associated with the vibration data, and the abnormal vibration characteristics can be directly located to the specific axle counting point location, achieving mileage-level accurate fault location.

[0024] Trend analysis: By comparing the vibration data triggered by different trains passing the same axle counting point, we can analyze its long-term trend and provide early warning of track condition deterioration. Track condition deterioration includes pre-loosening of fasteners and track bed settlement.

[0025] In this embodiment, to eliminate the inconsistency between vehicle speed (measured by the axle sensor) and vehicle type (distinguished by the number of wheels calculated by the axle sensor), vibration data of the same vehicle speed and vehicle type at the same location are recorded in the background and analyzed to improve the accuracy of assessing characteristics such as rail wear, rail surface defects, and rail joints.

[0026] Stress state assessment: Based on temperature data analysis, the changes in track vibration characteristics under different thermal stress conditions are assessed to evaluate the risk of track bulging.

[0027] In this embodiment, the current rail temperature stress is calculated using temperature data. When the condition is determined to be a high-temperature pressure state and the lateral stability index of the vibration data is lower than the threshold, a "high risk of rail expansion" warning is generated.

[0028] like Figure 2 As shown, the track machinery fault diagnosis system based on axle counting triggering includes: Axle counter sensor is used to detect the time when the wheel flange passes by and output a trigger signal; Vibration acceleration sensors are used to collect vibration data from trains; Temperature sensors are used to collect temperature data from the train. The synchronous acquisition controller is used to receive trigger signals and control the vibration sensor and temperature sensor to synchronously acquire data before and after the train passes. The intelligent analysis center is used to extract features and perform collaborative diagnosis on vibration and temperature data to generate fault diagnosis results for track machinery.

[0029] In the description of this invention, it should be understood that the terms "center," "thickness," "upper," "lower," "horizontal," "top," "bottom," "inner," "outer," and "radial," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying the relative importance or the number of technical features implicitly specified. Therefore, a feature defined by "first," "second," and "third" may explicitly or implicitly include one or more of that feature.

Claims

1. A fault diagnosis method for track machinery based on axle counting triggering, characterized in that, Includes the following steps: S1. Collect train vibration data and temperature data through event triggering and synchronous acquisition methods; S2. Preprocess the vibration and temperature data; S3. Extract features from vibration data and combine them with temperature data for collaborative diagnosis to generate fault diagnosis results for track machinery.

2. The method for fault diagnosis of track machinery based on axle counting triggering according to claim 1, characterized in that, In S1, the method for collecting train vibration data through event triggering and synchronous acquisition is as follows: When the axle counting sensor detects the passing of a train wheel, it sends a trigger signal to the synchronous acquisition controller, which then controls the vibration acceleration sensor to collect the train's vibration data.

3. The method for fault diagnosis of track machinery based on axle counting triggering according to claim 1, characterized in that, In S2, the specific method for preprocessing vibration data is as follows: The vibration data is processed using an adaptive wavelet threshold denoising method. A wavelet basis that matches the impact component is selected to separate the background noise from the fault impact data. The specific method for preprocessing temperature data is as follows: Temperature data is filtered using a moving average and then combined with historical data for temperature gradient compensation.

4. The method for fault diagnosis of track machinery based on axle counting triggering according to claim 1, characterized in that, In S3, the specific method for feature extraction from vibration data is as follows: Time-frequency domain feature extraction: Wavelet packet transform is performed on the vibration data to decompose it into independent frequency bands, and the energy entropy and normalized energy value of each frequency band are calculated as feature vectors; Demodulation analysis: Perform Hilbert transform on the vibration data and solve for its envelope spectrum; Nonlinear feature extraction: Calculate the nonlinear dynamic features of vibration data, including sample entropy and permutation entropy.

5. The method for fault diagnosis of track machinery based on axle counting triggering according to claim 4, characterized in that, In S3, the method for collaborative diagnosis by combining temperature data is as follows: Vibration data and temperature data correlation: Vibration data is correlated with the axle counting point location, wheel sequence and temperature data that triggered the vibration data acquisition to perform collaborative diagnosis, which includes fault location, trend analysis and stress state assessment. Fault location: Based on the trigger signal output by the axle counter sensor, the abnormal vibration characteristics are located to the specific axle counter point. Trend analysis: By comparing the vibration data triggered by different trains passing the same axle counting point, we can analyze its long-term trend and provide early warning of track condition deterioration. Stress state assessment: Based on temperature data analysis, the changes in track vibration characteristics under different thermal stress conditions are assessed to evaluate the risk of track bulging.

6. A track machinery fault diagnosis system based on axle counting triggering, applied to the track machinery fault diagnosis method based on axle counting triggering as described in any one of claims 1 to 5, characterized in that, include: Axle counter sensor is used to detect the time when the wheel flange passes by and output a trigger signal; Vibration acceleration sensors are used to collect vibration data from trains; Temperature sensors are used to collect temperature data from the train. The synchronous acquisition controller is used to receive trigger signals and control the vibration sensor and temperature sensor to synchronously acquire data before and after the train passes. The intelligent analysis center is used to extract features and perform collaborative diagnosis on vibration and temperature data to generate fault diagnosis results for track machinery.

Images