Railway vehicle body abnormal vibration diagnosis and identification method

Abstract

The invention discloses a railway vehicle body abnormal vibration diagnosis and recognition method. The method comprises the following steps that 1, an accelerometer sensor is used for collecting vehicle body vibration signals; step 2, using EMD to decompose the vibration signal x (t), wherein each mode IMFi (t) after decomposition corresponds to a certain frequency component in the vibration signal; step 3, after preprocessing the data, selecting modal data of the first 3-5 orders, and then performing fast Fourier transform on each modal IMFi (t) to obtain frequency spectrum information of each modal IMFi (t); step 4, smoothing processing is carried out on the frequency spectrum image; 5, calculating the energy ratio P of the vibration dominant frequency; step 6, calculating a performance index Q; and 7, judging whether the vehicle is in an abnormal state or not according to the P and the Q. According to the method, the vibration energy ratio index P and the performance index Q are used in cooperation, whether the stability, comfort and other performance of vehicle operation are good or not can be clearly known, and meanwhile whether the vehicle body or the bogie has obvious harmonic vibration or not can be judged.

Description

Technical Field

[0001] The invention belongs to the technical field of rail transportation, and in particular relates to a method for diagnosing and identifying abnormal vibration of a rail vehicle body. Background Art

[0002] With the continuous development of my country's rail transit sector, people's requirements for the safety, stability and comfort of train operation are also increasing. Rail vehicles that have been in service for a long time may experience abnormal vibrations and abnormal noises during operation due to reasons such as wheel-rail wear. These problems may affect the comfort of passengers and even driving safety, and accelerate the fatigue and aging of vehicle components. Therefore, it is crucial to check and identify abnormal conditions of rail vehicles in operation. Timely and effective identification of abnormal vibration of the vehicle body is conducive to protecting the life of the vehicle itself and the personal safety of passengers.

[0003] The performance indicators of rail vehicles mainly include stability index, comfort index, body vibration acceleration, and vehicle operation quality. These indicators not only determine the passenger's riding experience and satisfaction, but are also closely related to the safety, reliability and economy of the train. For example, the stability index directly affects the passenger's experience during the journey. Poor stability may cause discomfort to passengers during the train operation, affecting the overall quality of the trip. The comfort index comprehensively considers multiple factors such as vibration, noise, and temperature, and has an important impact on the passenger's physical and psychological experience. The vibration acceleration of the body directly reflects the external force applied to the train during operation. Excessive vibration will not only reduce the passenger's comfort, but also cause increased wear of the body parts, thereby affecting the safety and service life of the train. By optimizing these indicators, not only can the vibration and noise felt by passengers be effectively reduced, but the stability of the train can also be improved, providing a calmer and more comfortable travel environment. In addition, good stability and vehicle operation quality can reduce the failure rate of the train, thereby reducing downtime. Maintaining stability during operation means that its track wear and parts consumption will be more uniform, thereby extending the maintenance cycle and improving the punctuality of the vehicle. This not only improves the operating efficiency of trains, but also reduces operating costs, thereby improving the overall transportation capacity and economic benefits of the railway system. By setting and continuously optimizing these performance indicators, the railway system can provide higher quality services and improve the efficiency and benefits of the entire transportation system.

[0004] However, these evaluation indicators are only limiting values, which can be used to measure vehicle operating efficiency, safety, comfort and environmental impact, and belong to the evaluation standards. For example, in the Sperling index of passenger cars, when the index W is less than 2.50, it is excellent, 2.50 less than W less than 2.75 is good, and 2.75 less than W less than 3.00 is qualified. If the test result is 2.60, it can be clearly seen that the stability of the passenger car is good and the passenger comfort is moderate. However, these performance indicators are only an evaluation of the overall performance of the vehicle. The reason why the stability index cannot meet the excellent level may be caused by large random excitation of the track, or it may be that the car body or bogie has obvious harmonic vibration, and it is impossible to determine whether the car body is in an abnormal state. Whether the vehicle has abnormal vibration is the basis for vehicle parameter optimization, vibration control and condition repair.

[0005] The abnormal state of the vehicle body refers to the obvious harmonic vibration of the vehicle body caused by the excitation of its own mode during the operation of the rail vehicle. The current general method is to evaluate whether the vehicle has abnormal vibration from the perspective of vehicle performance indicators, but this method has a large error, and even if the performance indicators exceed the limit, it may not be the vehicle that has abnormal vibration, but the external aerodynamic excitation, random track excitation, bridges and other line conditions, and the diagnosis and identification results have a large error. Comprehensive analysis shows that the current evaluation system cannot accurately and efficiently determine whether the vehicle has abnormal vibration. In view of this, proposing a method that can accurately diagnose and identify abnormal vibration of rail vehicle bodies is a key technical problem that researchers in this field urgently need to solve. Summary of the invention

[0006] The purpose of the present invention is to overcome the shortcomings of the prior art and to provide a method for diagnosing and identifying abnormal vibration of a rail vehicle body by using a vibration energy proportion index P and a performance index Q in combination, which can not only clearly understand whether the vehicle's running stability, comfort and other performance are good, but also can determine whether the car body or bogie has obvious harmonic vibration.

[0007] The object of the present invention is achieved through the following technical scheme: A method for diagnosing and identifying abnormal vibration of a rail vehicle body comprises the following steps:

[0008] Step 1: Use an accelerometer sensor installed on the train body to collect body vibration signals in real time; these signals are accelerations that change over time, called time domain signals;

[0009] Step 2: Preprocess the collected time domain signal and filter it using the EMD method: record the obtained vibration signal as x(t), and use EMD to decompose the vibration signal x(t):

[0010]

[0011] IMFi (t) is the i-th mode after decomposition, r(t) is the remaining residual term; each mode IMF after decomposition i (t) corresponds to a certain frequency component in the vibration signal;

[0012] Step 3: After preprocessing the data, select the first 3-5 modal data and then perform IMF analysis on each modal. i (t) performing fast Fourier transform to obtain its spectrum information;

[0013] Step 4: Smoothing the spectrum image;

[0014] Step 5: Calculate the energy proportion P of the main vibration frequency: Obtain the main vibration frequency f in each data segment through the spectrum image after each mode smoothing, and calculate f±f e Energy within the range E main , where f e To set the analysis range; calculate the set analysis segment f a ~f b The vibration energy E total ; Get the energy proportion P of the main frequency of vibration:

[0015]

[0016] Step 6, obtaining performance index Q: the evaluation index Q includes a stability index, a comfort index, a root mean square value of vehicle body vibration acceleration, a maximum value of vehicle body vibration acceleration or vehicle running quality;

[0017] Step 7: judge whether the vehicle is in an abnormal state, define the upper and lower limits of P as k1 and k2 respectively; define the upper and lower limits of Q as h1 and h2; judge the vehicle state according to P and Q in each data segment respectively;

[0018] Four evaluation criteria are given for vehicle status, namely normal, possibly abnormal, highly likely abnormal, and alarm, corresponding to Level I, Level II, Level III, and Level IV respectively:

[0019] Level Ⅰ: P<k1&Q

[0020] Level II: P≤k2&Q≤h1; k1≤P <k2&Q

[0021] Level III: P ≥ k2 & h1 ≤ Q < h2; k1 ≤ P < k2 & Q ≥ h2; k1 ≤ P < k2 & h1 ≤ Q < h2

[0022] Level IV: P≥k2&Q≥h2.

[0023] The beneficial effects of the present invention are:​​

[0024] 1. Standardization enables more accurate detection of abnormalities: The method for diagnosing and identifying abnormal vibration of rail vehicle bodies described in the present invention overcomes the problem of lack of unified standards in previous body vibration detection. The invention uses the vibration energy proportion index P and the performance index Q together, which can not only clearly understand whether the vehicle's running stability, comfort and other performance are good, but also determine whether the body or bogie has obvious harmonic vibration.

[0025] Second, the operation is convenient, efficient and accurate: During the diagnosis process, the staff only needs to transfer the real-time data of the vehicle to the computer to directly obtain the values ​​of P and Q, quickly understand the current status level of the vehicle, and determine whether it needs to be repaired. Compared with traditional methods, this method is more intuitive and efficient, and the diagnostic accuracy is significantly improved, ensuring that abnormal problems can be discovered and handled in a timely manner.

[0026] 3. Various detection methods and higher fault tolerance: In actual detection, real-time diagnosis can be performed, or an overall assessment can be performed through a section of measured data. There are various methods, and the method can be selected according to the specific situation. Potential problems can be discovered more accurately, making vehicle maintenance more timely and accurate, providing reliable guarantee for improving rail transit safety and service quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a flow chart of the method for diagnosing and identifying abnormal vibration of a rail vehicle body according to the present invention;

[0028] Figure 2 Quadrant diagram for graded management of abnormal vibration status of vehicle body;

[0029] Figure 3 This is the regional division method of the evaluation indicators P and Q in this embodiment;

[0030] Figure 4 It is the hierarchical management of abnormal vehicle body vibration based on the evaluation indicators P and Q;

[0031] Figure 5 Real-time diagnosis and identification of abnormal vibration status of vehicle body;

[0032] Figure 6 The diagnostic identification results and corresponding kilometer mark of the vehicle's actual operating route;

[0033] Figure 7 This is the vehicle body abnormal vibration detection system of the present invention. DETAILED DESCRIPTION

[0034] In the field of rail transit, the diagnosis and identification of abnormal vibration phenomena of the vehicle body is an important task to ensure the smooth and safe operation of the train. The vibration in the track and train system may come from problems such as uneven track, unstable motor operation, structural resonance, etc., and these vibrations usually manifest as specific frequency components. How to judge whether the train has abnormal vibration and whether the train can continue to transport passengers, the present invention uses two evaluation indicators in combination, one evaluation indicator is the vibration energy proportion P, and the other evaluation indicator is the performance indicator Q. This method takes into account the detection of abnormal vibration energy and vehicle stability indicators, root mean square value or maximum acceleration and other performance indicators, which can better help staff judge the overall performance of the vehicle and ensure the running quality and driving safety of the train.

[0035] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0036] like Figure 1 As shown, a method for diagnosing and identifying abnormal vibration of a rail vehicle body of the present invention comprises the following steps:

[0037] Step 1, P is the vibration energy ratio. When diagnosing harmonic vibration, data collection must be performed first. The accelerometer sensor is installed on the train body, which can be located in the bolster, passenger compartment, driver's cab, etc., to collect the body vibration signal in real time. Generally, basic data such as the speed and vibration acceleration of the train are collected. These signals are accelerations that change with time, which are called time domain signals. Then the collected data is processed and analyzed by data processing software to obtain the time domain diagram of vibration acceleration.

[0038] Step 2: Preprocess the collected time domain signal. The collected vibration signal may contain interference signals and needs to be filtered to remove high-frequency or low-frequency irrelevant noise to improve the accuracy of the analysis. The measured signal can be preprocessed by different signal decomposition methods. Based on the results of a large amount of data analysis, it is recommended to use the EMD method for filtering: record the obtained vibration signal as x(t), and use EMD to decompose the vibration signal x(t):

[0039]

[0040] The core idea of ​​EMD is to decompose nonlinear and non-stationary signals into several intrinsic mode functions (IMFs). i (t) is the i-th mode after decomposition, r(t) is the remaining residual term; each mode IMF after decomposition i (t) corresponds to a certain frequency component in the vibration signal. Usually, the first few modes correspond to high-frequency components, and the latter modes correspond to low-frequency components.

[0041] Step 3: After preprocessing the data, select the first 3-5 modal data and then perform IMF analysis on each modal. i (t) Perform fast Fourier transform (FFT) to obtain its spectrum information; FFT plays a key role in this scenario. It can help convert the vibration signal in the time domain into the frequency domain, facilitate the identification and analysis of various frequency components, and thus diagnose the source and characteristics of the vibration.

[0042] Step 4: Smooth the spectrum image; the FFT spectrum of the vibration signal may contain multiple vibration modes or frequency components. Some frequency components may be very close and difficult to distinguish. Therefore, the image should be smoothed to reduce the noise and details in the image and make the overall outline of the spectrum clearer. Common smoothing methods include Gaussian smoothing, Savitzky-Golay filtering, moving average smoothing, etc. It is necessary to select a reasonable method based on the data characteristics. Generally, Gaussian smoothing is recommended, which focuses on local features, can effectively reduce high-frequency noise, and has a more natural smoothing effect.

[0043] Step 5: Calculate the energy proportion P of the main vibration frequency: Obtain the main vibration frequency f in each data segment through the spectrum image after each mode smoothing, and calculate f±f e Energy within the range E main , where f e To set the analysis scope; f e is a frequency bandwidth. If it is 0.25Hz, f±0.25 represents the vibration energy within the analysis range. It is considered that the vibration energy within this range is the vibration energy corresponding to the main vibration frequency f. Generally, it is reasonably selected according to the analysis needs. The larger the frequency to be examined, the smaller f e The larger the value is, the smaller the f e The general value range is 0.25~0.5Hz. When analyzing abnormal vibration of the vehicle body, usually only the set analysis segment f is considered. a ~f b The vibration response inside, f a 、f b are the set lower cutoff frequency and upper cutoff frequency respectively; for abnormal vibration of the vehicle body, whether it is longitudinal vibration, lateral vibration or vertical vibration, the natural modal frequencies of the vehicle body do not exceed 10Hz, generally within 0.5Hz~10Hz, here, f a Usually 0.5Hz, f b is 10 Hz, but for some special cases the frequency range may be wider. Calculate the analysis segment f a ~f b The vibration energy E total . The energy proportion P of the main vibration frequency is obtained:

[0044]

[0045] In order to diagnose and identify whether the vehicle has abnormal harmonic vibration, it can be achieved by dividing the limit value of the energy proportion P of the vibration main frequency. In order to determine the limit value of the energy proportion, generally speaking, it is the best choice to extract the limit value of the energy proportion P through a large amount of test data, but this method requires a lot of time, test resources and costs. Generally speaking, the time and cost of obtaining the limit value can be greatly shortened by means of simulation and test verification. First, a large number of dynamic simulation calculations are required to preliminarily formulate the limit value, and then apply it to the actual experimental circuit, and finally correct the established limit value based on the measured data.

[0046] Step 6, obtain the performance index Q: the evaluation index Q can be a stability index, a comfort index, the root mean square value of the body vibration acceleration, the maximum value of the body vibration acceleration, or the vehicle running quality, which is a performance index that characterizes whether the body is comfortable to ride. These data are commonly used indicators, and the specific acquisition methods can be found in the national standard GB / T5599-2019 "Specifications for the Evaluation and Test Identification of Locomotive and Vehicle Dynamics Performance". The division of its limits can be achieved through different methods. One is historical data. Based on a large amount of historical data, the distribution range of the performance index Q is obtained, and the distribution position of the real-time data can be judged according to the percentile of the performance index Q; the second is that in the absence of data or when the amount of data is very small, according to the limit recommended in the standard, the percentile distribution range of the preset limit Q is preset. After obtaining a certain amount of data, the preset limit is corrected and replaced with the percentile distribution range of the actual collected data.

[0047] Step 7. Judge whether the vehicle is in an abnormal state: After obtaining the two evaluation indicators P and Q, a reasonable judgment can be made on whether the vehicle is in an abnormal state. Each indicator has two limits. For the same vehicle and the same service environment, the limit levels of P and Q are fixed values. The upper and lower limits of P are defined as k1 and k2 respectively; the upper and lower limits of Q are defined as h1 and h2; for different service environments, the limits also fluctuate within a reasonable range, and the specific values ​​can be set according to the actual service environment, such as Figure 2 shown.

[0048] Four evaluation criteria are given for vehicle status, namely normal, possibly abnormal, highly likely abnormal, and alarm, corresponding to Level I, Level II, Level III, and Level IV respectively;

[0049] Level Ⅰ: P<k1&Q

[0050] Level II: P≤k2&Q≤h1; k1≤P <k2&Q

[0051] ​​Level III: P ≥ k2 & h1 ≤ Q < h2; k1 ≤ P < k2 & Q ≥ h2; k1 ≤ P < k2 & h1 ≤ Q < h2

[0052] Level IV: P≥k2&Q≥h2.

[0053] According to the above evaluation criteria, the distribution of indicators is divided into four levels and corresponds to the quadrant diagram, such as Figure 2 As shown in the figure, the evaluation indicators obtained by analysis are graded using the quadrant diagram. The data points are divided into different quadrants to help identify abnormal train vibration.

[0054] Figure 3 (a) is the distribution of the energy proportion of the main frequency of vibration in a certain period of time under the same working conditions. It can be clearly seen from the figure that the energy proportion of most data points is concentrated below k2, which indicates that under this working condition, the stability of the main frequency vibration energy is good. However, there are also a small number of data points whose energy proportion exceeds k2, which means that a small amount of abnormal harmonic vibration occurred during this period of time. The occurrence of these abnormal vibrations may be related to fluctuations in the operating state of the system or external interference factors, indicating potential system abnormalities or abnormal vibration sources. By observing the distribution of these energy proportions, the stability of the vibration system can be further analyzed, and a basis can be provided for subsequent status monitoring and abnormality troubleshooting.

[0055] Similarly, Figure 3 (b) is the evaluation of vehicle operating performance indicators over a period of time. Data less than h1 is excellent, data greater than h2 is unqualified, and data between h1 and h2 is good. This graph can reflect the operating performance status of the vehicle during this period of time.

[0056] Figure 4(a) shows the data measured on a test line within a specified time, which is used to evaluate whether there is abnormal vibration of the vehicle on this line. Each data has its corresponding limit value P and Q. The data are statistically put into the coordinate system as shown in the figure, where the limit values ​​are k1=0.35, k2=0.55, h1=2.5, h2=3.0. It is stipulated that P<0.35 and Q<2.5 are level Ⅰ (normal state), indicating that the vehicle performance is normal and there is no abnormal vibration. P≥0.55 and Q<2.5; 0.35≤P<0.55 and Q<2.5; P<0.35 and Q≥3.0; P<0.35 and 2.5≤Q<3.5 are all level Ⅱ (possible abnormal state). At this time, the staff should pay attention in advance and make preparations for maintenance. P≥0.55 and 2.5≤Q<3.0; 0.35≤P<0.55 and Q≥3.0; 0.35≤P<0.55 and 2.5≤Q<3.0 are all Level III (highly likely to be abnormal). These data are the focus of attention and timely maintenance should be carried out to prevent abnormal vehicle operation. P≥0.55 and Q≥3.0 are Level IV (warning state), which means that the vehicle has abnormal vibration and the performance indicators are not qualified. The staff needs to carry out maintenance and abnormality investigation.

[0057] Figure 4 (b) shows the position of the test data points in the coordinate system, which can more intuitively show the level of each data point. Most of them are distributed in levels I and II, and a small amount of IV data indicates abnormal vibration. The advantage of this figure is that it allows maintenance personnel to more intuitively see the level of each data point and determine the number of abnormal vibrations that occur in this section of the line.

[0058] like Figure 5 As shown in the figure, it is real-time diagnostic data. After analyzing the vehicle operation data of each second, the values ​​of P and Q can be obtained. After computer analysis, the level can be directly obtained. With 5 seconds as a data segment and 2.5 seconds as a slip step, the problems of the line and the car body can be analyzed more intuitively. For example, in the figure, the data points of level III and level IV are relatively concentrated around 80 seconds, indicating that there are frequent abnormal vibrations when the line is running. The staff can obtain the data parameters and operating position of the vehicle during this period through sensors. If the parameters of the car body are normal, it means that the abnormal vibration is caused by the line problem, and the line problem can be directly repaired.

[0059] Figure 6The figure also shows real-time diagnosis and identification. The collected vehicle running speed is integrated over time to obtain the real-time displacement of the vehicle. At this time, the vibration acceleration and various indicators of the corresponding position of the vehicle at each moment can be clearly known, and the abnormal vibration level of the vehicle body can be calculated at the same time, which is very efficient and accurate. It can be clearly seen in the figure that in the area of ​​10km to 11km, the vibration acceleration is very unstable, and there are many abnormal points of indicators P and Q. Therefore, the corresponding abnormal vibration classification of the vehicle body in the figure below is basically at level III and IV, indicating that the vehicle body at this position has obvious abnormal vibration, and the staff needs to conduct investigation and maintenance.

[0060] Figure 7 The figure shows the detection system of abnormal vibration of rail vehicles. First, sensors are installed on the train to collect basic data such as the speed and vibration acceleration of the train. Then, the collected data is processed and analyzed by data processing software to obtain the time domain diagram of vibration acceleration. Based on this, the evaluation indicators P and Q are obtained according to the calculation method of the evaluation indicators. Then, the evaluation indicators obtained by analysis are graded using the quadrant diagram. The quadrant diagram is used to divide the data points into different quadrants according to the combination of P and Q, so as to help identify abnormal train vibration. Finally, the results of the analysis are visualized through charts. The visualized chart can intuitively show the trend of train vibration over time or mileage marks, as well as the distribution of vibration levels, so that relevant personnel can evaluate and deal with abnormal vibrations and find the corresponding positions while mastering abnormal vibrations of the vehicle body.

[0061] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. A method for diagnosing and identifying abnormal vibration of a railway vehicle body, characterized in that: The following steps are involved: Step 1: Install an accelerometer sensor on the train body to collect body vibration signals in real time; These signals are accelerations that vary with time and are called time-domain signals; Step 2: Preprocess the collected time domain signal and filter it using the EMD method: record the obtained vibration signal as x(t), and use EMD to decompose the vibration signal x(t): IMF i (t) is the i-th mode after decomposition, r(t) is the remaining residual term; each mode IMF after decomposition i (t) corresponds to a certain frequency component in the vibration signal; Step 3: After preprocessing the data, select the first 3-5 modal data and then perform IMF analysis on each modal. i (t) performing fast Fourier transform to obtain its spectrum information; Step 4: Smoothing the spectrum image; Step 5: Calculate the energy proportion P of the main vibration frequency: Obtain the main vibration frequency f in each data segment through the spectrum image after each mode smoothing, and calculate f±f e Energy within the range E main , where f e To set the analysis range; calculate the set analysis segment f a ~f b The vibration energy E total ; Get the energy proportion P of the main frequency of vibration: Step 6, obtaining performance index Q: the evaluation index Q includes a stability index, a comfort index, a root mean square value of vehicle body vibration acceleration, a maximum value of vehicle body vibration acceleration or vehicle running quality; Step 7: Determine whether the vehicle is in an abnormal state, and define the upper and lower limits of P as k1 and k2 respectively; define the upper and lower limits of Q as h1 and h2; Four evaluation criteria are given for vehicle status, namely normal, possibly abnormal, highly likely abnormal, and alarm, corresponding to Level I, Level II, Level III, and Level IV respectively: Level Ⅰ: P<k1&Q<h1 Level II: P≤k2&Q≤h1; k1≤P <k2&Q<h1;P<k1&Q≥h2;P<k1&h1≤Q<h2 Level III: P ≥ k2 & h1 ≤ Q < h2; k1 ≤ P < k2 & Q ≥ h2; k1 ≤ P < k2 & h1 ≤ Q < h2 Level IV: P≥k2&Q≥h2.

Images