WO2011075896A1 - Procédé et dispositif permettant d'analyser une réponse d'accélération d'un véhicule ferroviaire - Google Patents

Procédé et dispositif permettant d'analyser une réponse d'accélération d'un véhicule ferroviaire Download PDF

Info

Publication number
WO2011075896A1
WO2011075896A1 PCT/CN2009/075963 CN2009075963W WO2011075896A1 WO 2011075896 A1 WO2011075896 A1 WO 2011075896A1 CN 2009075963 W CN2009075963 W CN 2009075963W WO 2011075896 A1 WO2011075896 A1 WO 2011075896A1
Authority
WO
WIPO (PCT)
Prior art keywords
speed
frequency
velocity
spectrum
power
Prior art date
Application number
PCT/CN2009/075963
Other languages
English (en)
Chinese (zh)
Inventor
曾宇清
甘敦文
董孝卿
倪纯双
于卫东
扈海军
刘秀波
王卫东
黄强
Original Assignee
中国铁道科学研究院机车车辆研究所
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中国铁道科学研究院机车车辆研究所 filed Critical 中国铁道科学研究院机车车辆研究所
Priority to PCT/CN2009/075963 priority Critical patent/WO2011075896A1/fr
Publication of WO2011075896A1 publication Critical patent/WO2011075896A1/fr

Links

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M17/00—Testing of vehicles
    • G01M17/08—Railway vehicles

Definitions

  • the present invention relates to a method for analyzing test data of a rail vehicle, and more particularly to a method and apparatus for accelerating response analysis of a rail vehicle.
  • test capability of railway vehicle dynamics test has been greatly improved in recent years.
  • analysis method of test data is still based on the time domain and parameterized processing of test data, supplemented by the simple frequency domain parameter-stationary index.
  • GB5599-85 Code for Dynamic Performance Evaluation and Test Identification of railway Rail Vehicles
  • the ability to analyze vehicle characteristics lags behind practical needs and needs to be improved.
  • the acceleration response of the vehicle is relatively easy to obtain, it is hoped that the above purpose can be achieved by the vehicle acceleration response.
  • the implementation of the method will be of great significance for the test, monitoring, overhaul and characteristic optimization of the rail vehicle, especially for the high-speed rail vehicle.
  • the object of the embodiments of the present invention is to provide a method and a device for analyzing the acceleration response of a rail vehicle, which can directly utilize the vehicle dynamics line test and the vehicle acceleration response during the actual operation of the vehicle to perform comprehensive, intuitive and effective rail vehicle characteristics and fixed wavelengths. Statistical assessment of vehicle inputs.
  • an embodiment of the present invention provides a method for analyzing acceleration response of a rail vehicle, the method comprising: a. acquiring an acceleration response signal of a rail vehicle and corresponding speed information; b. grouping the acceleration signals according to speed; c. obtaining a power spectrum or a power equivalent amplitude spectrum of the acceleration response signal corresponding to each speed level according to the periodogram method; d. expressing the result of step c with a three-dimensional velocity-spectral map; e. identifying the A local peak that does not vary with velocity in a three-dimensional velocity-spectral map and/or a local peak that varies linearly with velocity.
  • the embodiment of the present invention further provides an apparatus for analyzing acceleration response of a rail vehicle, comprising: an acceleration and speed acquiring unit, configured to acquire an acceleration response signal of the rail vehicle and corresponding speed information; and a grouping unit, configured to The speed is used to group the acceleration signals; a spectrum calculation unit is configured to obtain a power spectrum or a power equivalent amplitude spectrum of the acceleration response signal corresponding to each speed level according to a periodogram method; a three-dimensional display unit, configured to: The calculation result of the spectrum calculation unit is represented by a three-dimensional velocity-spectral map; the identification unit is configured to identify a local peak that does not vary with speed in the three-dimensional velocity-spectral map and/or a local peak that varies linearly with speed.
  • Method and device for analyzing acceleration response of rail vehicle capable of being comprehensive and straight Observing and effectively obtaining the characteristics of the rail vehicle and the fixed-wavelength vehicle input of interest, as long as the original signal band is sufficiently wide, the analysis result is not affected by the subsequent filtering frequency or the like;
  • the embodiment of the invention can be used not only for the manual identification of the acceleration signal
  • Qualitative analysis can also be used for quantitative evaluation of rail vehicle characteristics and inputs, and the algorithm is simple and easy to implement
  • the embodiment of the invention has strong engineering application for testing, monitoring, overhauling and characteristic optimization of rail vehicles, especially high-speed rail vehicles. value.
  • FIG. 1 is a schematic flow chart of a method for analyzing acceleration response of a rail vehicle according to an embodiment of the present invention
  • Fig. 2a - Fig. 2b are respectively a three-dimensional velocity-spectral map of the vertical vibration of the wheelset and its corresponding local peak automatic recognition result which does not change with the speed;
  • Figure 3a - Figure 3b are the three-dimensional velocity-spectral maps of the vertical vibration of the frame and their corresponding local peak automatic recognition results without speed changes;
  • Fig. 4a-4b are respectively a three-dimensional velocity-spectral map of the vertical vibration of the vehicle body and corresponding corresponding local peak automatic recognition results without changing with speed;
  • Fig. 5a-Fig. 5b are respectively a three-dimensional velocity-spectral map of the wheel-to-side lateral vibration and its corresponding local peak automatic recognition result which does not change with the speed;
  • Fig. 6a-Fig. 6b are respectively a three-dimensional velocity-spectral map of the lateral vibration of the architecture and corresponding corresponding local peak automatic recognition results without changing with speed;
  • Fig. 7a-7b are respectively a three-dimensional velocity-spectral map of the lateral vibration of the vehicle body and corresponding corresponding local peak automatic recognition results without changing with speed;
  • Figure 8 shows that the displacement signal of a spatial domain is twice as fast as the sampling frequency at the same time.
  • the displacement input power spectrum comparison chart obtained under the condition;
  • 9a-9d are diagrams showing periodic input results obtained by automatically recognizing a three-dimensional velocity-spectral image of a vertical acceleration vibration of a Beijing-Tianjin line according to an embodiment of the present invention.
  • FIG. 10A-lOb is a preliminary result diagram of a periodic input wavelength obtained by automatically recognizing a three-dimensional velocity-spectral image of a vertical acceleration vibration of a Beijing-Tianjin line according to an embodiment of the present invention
  • 1 la-l ld is a cycle input result diagram obtained by automatically recognizing a three-dimensional velocity-spectral image of a vertical acceleration vibration of a descending wheel of the Beijing-Tianjin line according to an embodiment of the present invention
  • 12a-12b are preliminary diagrams showing initial values of periodic input wavelengths obtained by automatically recognizing a three-dimensional velocity-spectral image of a vertical acceleration vibration on a Beijing-Tianjin line according to an embodiment of the present invention
  • FIG. 13 is a schematic flowchart of another method for analyzing acceleration response of a rail vehicle according to an embodiment of the present invention.
  • FIG. 14a- FIG. 14c are respectively a three-dimensional velocity-transfer diagram of the vertical transfer of the wheel-to-frame according to the embodiment of the present invention, and a local peak identification of the three-dimensional velocity-transfer diagram of the vertical transfer of the wheel-to-frame vertical transfer without speed change.
  • 15a-c are respectively a three-dimensional velocity-transfer diagram of the lateral nominal transfer of the wheel-to-frame provided by the present embodiment, and a local peak recognition result of the three-dimensional velocity-transfer diagram of the laterally-transferred transfer of the wheel-to-frame, respectively Data nominal transfer diagram;
  • 16a-FIG. 16c are diagrams showing local peak identification results of a three-dimensional velocity-spectrum image of a wheelset, a frame, and a vehicle body vertical vibration that do not vary with speed, respectively, according to an embodiment of the present invention
  • FIG. 17a- FIG. 17c are diagrams showing local peak identification results of the wheel-to-frame, frame-vehicle body, wheelset-vehicle body vertical vibration three-dimensional velocity-transmission diagram without speed change according to the embodiment;
  • 18a to 18c are diagrams showing local peak identification results of the three-dimensional velocity-spectral image of the lateral vibration of the wheelset, the frame and the vehicle body, which are not changed with the speed, respectively, provided by the embodiment;
  • 19a-FIG. 19c are diagrams showing local peak identification results of the wheel-to-frame, frame-vehicle body, wheelset-vehicle body lateral vibration three-dimensional velocity-transfer diagram without speed change according to the embodiment;
  • FIG. 20a-20c are the snake line identification and evaluation diagrams provided by the embodiment;
  • FIG. 21 is a three-dimensional velocity-spectrum diagram of the lateral acceleration of the low-frequency wheel pair provided by the embodiment;
  • FIG. 22a-22c are respectively a three-dimensional velocity-spectrum, three-dimensional velocity-maximum spectrum, and three-dimensionality of the vehicle body vertical acceleration response provided by the embodiment. Automatic identification of local peaks in the velocity-spectrum map that do not vary with speed;
  • FIG. 23 is a schematic structural diagram of an apparatus for analyzing acceleration response of a rail vehicle according to an embodiment of the present invention.
  • FIG. 24 is a schematic structural diagram of an identification unit according to an embodiment of the present disclosure.
  • FIG. 25 is a schematic structural diagram of another identification unit according to an embodiment of the present disclosure.
  • FIG. 26 is a schematic structural diagram of a wavelength interval setting module according to an embodiment of the present invention
  • FIG. 27 is a schematic structural diagram of another apparatus for analyzing acceleration response of a rail vehicle according to an embodiment of the present invention
  • FIG. 28 is a schematic structural diagram of a nominal transmission unit according to an embodiment of the present invention.
  • the acceleration response of the rail vehicle is determined simultaneously by the characteristics of the rail vehicle and the input of the rail vehicle, and the object of the embodiment of the present invention is the acceleration of the rail vehicle.
  • the characteristics of the rail vehicle and the rail vehicle input are separated as much as possible, and the rail vehicle characteristics and the rail vehicle input are quantified. It is very difficult to unconditionally separate the separation.
  • the inventor considers that the actual rail vehicles are running at different speeds, and the influence of the running speed on the acceleration response of the rail vehicle actually shows the influence of speed on the vehicle characteristics and the vehicle input; Vehicle characteristics, vehicle input vary with speed The statistical variation law, and these laws can be quantified, then the joint characteristics of the speed-frequency domain can be used to better extract the characteristics of the rail vehicle and its input.
  • the input of the rail vehicle mainly includes two kinds of track input and wheel set input.
  • the track input may include random track irregularity and periodic track irregularity.
  • the input of the wheel pair can be equivalent to a special periodic track. Smooth.
  • Orbital irregularity is generally represented by a statistically significant orbital spectrum. From the perspective of a rail vehicle, the study and application of the orbital spectrum focuses on the conversion of the spatial frequency domain to the spatial domain, such as the conversion from the spatial frequency domain to the spatial domain.
  • As an input to the dynamic simulation of the rail vehicle as a target signal for random excitation in the laboratory, etc.
  • the research and application of the transition relationship between the orbit spectrum from the spatial frequency domain to the time-frequency domain is very low. However, this is just one of the theoretical foundations for solving the characteristics of the rail vehicle and the input characteristics of the vehicle through the acceleration response of the rail vehicle.
  • the numerical laws of the acceleration input power spectrum and the acceleration input amplitude spectrum are mainly utilized, especially in the automatic identification of related parameters.
  • the embodiment of the present invention only needs to utilize any one of an acceleration input power spectrum and an acceleration input amplitude spectrum, which may be determined according to an operator's habit.
  • the following is the data law when using the acceleration input power spectrum and the acceleration input amplitude spectrum:
  • the time frequency is proportional to the speed, the corresponding frequency interval is simultaneously scaled, and the acceleration input power spectrum of the rail vehicle is proportional to the speed 3 powers;
  • the time frequency is also proportional to the speed, the corresponding frequency interval is not scaled, and the acceleration input power spectrum of the rail vehicle is proportional to the speed of the fourth power;
  • the actual input can be seen as a combination of the above two cases.
  • the time frequency is proportional to the speed
  • the corresponding frequency interval is simultaneously scaled
  • the acceleration input amplitude spectrum of the rail vehicle is proportional to the speed of 1. 5 power
  • the time frequency is also proportional to the speed, the corresponding frequency interval is not scaled, and the acceleration input amplitude spectrum of the rail vehicle is proportional to the speed 2 power;
  • the actual input can be seen as a combination of the above two cases.
  • the law of rail vehicle characteristics changing with speed can be understood from two levels: Only the above part of the rail vehicle wheelset is considered. Generally, since the dynamic parameters of the above part of the rail vehicle wheelset are not significantly related to the speed, it can be considered as the speed. Irrelevant; wheelset is the most distinctive component of railway rail vehicles. Due to the particularity of wheel-rail restraint, self-excited vibration occurs after the running speed is greater than a certain value. This value is generally related to the wheel-rail profile, contact conditions, The upper part of the rail vehicle is associated with the constraint of the wheel pair.
  • the characteristics of the rail vehicle do not change much when the rail vehicle is not approaching the meandering speed; when the snake speed is exceeded, self-excited vibration will occur, which is manifested as a sharp change in the characteristics of the rail vehicle, but at this time, the characteristics of the rail vehicle wheel pair will still be Stable, snakes can be seen as a special external input to some extent.
  • the acceleration response of the rail vehicle is determined by the characteristics of the rail vehicle and the vehicle input. From the characteristics of the rail vehicle obtained above and the law of the change of the input of the rail vehicle with the speed, the response of the rail vehicle can be obtained as a speed. Characteristics of degree changes:
  • the acceleration response of the rail vehicle generally has a local peak that varies linearly with the speed frequency, corresponding to the input of the period length, such as the roundness of the wheel, the eccentricity, the length of the feature of the track structure, etc.;
  • the acceleration response of the rail vehicle generally has a local peak that does not change with the speed, corresponding to the natural frequency of the rail vehicle system (including the natural frequency of the structure);
  • Embodiments of the present invention are based on the above-described characteristics of rail vehicle characteristics, vehicle input, and vehicle acceleration response as a function of speed.
  • Embodiment 1 is based on the above-described characteristics of rail vehicle characteristics, vehicle input, and vehicle acceleration response as a function of speed.
  • FIG. 1 is a schematic flow chart of a method for analyzing acceleration response of a rail vehicle according to an embodiment of the present invention.
  • the embodiment of the present invention is a comprehensive vertical direction for a high speed train with a speed class of 350 km/h on the Beijing-Tianjin line. , horizontal characteristics and input analysis, this analysis is based on the speed-frequency domain, the method includes the following steps:
  • Step a Obtain the acceleration response signal of the rail vehicle and its corresponding speed information.
  • an acceleration sensor and a speed sensor may be installed on the rail vehicle.
  • the same Vertical and lateral acceleration sensors and speed sensors are installed at the appropriate positions of the wheelset, frame and body of the bogie, and the six acceleration responses and corresponding speeds of the train on the Beijing-Tianjin line and the normal operation of the downlink are obtained.
  • Signal, sampling frequency is 1000Hz.
  • Step b grouping the acceleration signals according to speed.
  • the acceleration signals are grouped by intervals, and the acceleration signals can be grouped by different speed intervals. Obviously, it is also necessary to comply with the general principles of statistical data processing, such as eliminating the speed grades with fewer samples, so that the results of the evaluation are reliable.
  • the acceleration signals are grouped at a speed interval of 10 km/h, and the minimum number of samples is 8.
  • Step c Obtain a power spectrum or a power equivalent amplitude spectrum of the acceleration response signal corresponding to each speed level according to the periodogram method.
  • the power spectrum of each speed-level acceleration response signal is obtained according to the periodogram method, or the corresponding amplitude spectrum is obtained according to the principle of power equivalence, and the sample length corresponding to the periodogram is 4 s. It should be noted that both the power spectrum and the amplitude spectrum can be used to implement the method of the embodiment of the present invention. However, since the power spectrum changes relatively violently, and the result is intuitive and conforms to the habit of rail vehicle research, the embodiment of the present invention The legends are all represented by amplitude spectra.
  • Step d The result of step c is represented by a three-dimensional velocity-spectral map.
  • the X-axis represents the analysis frequency (here, for the sake of display, the upper limit of the analysis frequency is 80 Hz)
  • the y-axis represents the velocity
  • the z-axis represents the spectrum
  • the three-dimensional velocity-spectrum is used.
  • the picture shows the characteristics of the rail vehicle system and the analysis of the inputs provides an intuitive, comprehensive view of the most information possible with the least amount of data.
  • the following three-dimensional velocity-spectral map can be obtained: a three-dimensional velocity-spectral map of the vertical vibration of the wheelset, a three-dimensional velocity-spectral map of the vertical motion of the frame, and a three-dimensional velocity-spectrum of the vertical vibration of the vehicle body.
  • a three-dimensional velocity-spectral map of the vertical vibration of the wheelset a three-dimensional velocity-spectral map of the vertical motion of the frame
  • a three-dimensional velocity-spectrum of the vertical vibration of the vehicle body Fig. 2D, 3D, 3D, 3a .
  • Step e Identify local peaks in the three-dimensional velocity-spectral map that do not vary with velocity and/or local peaks that vary linearly with velocity.
  • the local peaks that do not vary with speed in the three-dimensional velocity-spectral map and the local peaks that vary linearly with speed can reflect the characteristics of the rail vehicle and its input.
  • the local peaks that do not vary with speed generally correspond to the vehicle.
  • Characteristics, such as: vehicle moving head natural frequency, vehicle nod The natural frequency, the natural vibration of the bogie, etc., and the local peaks that vary linearly with speed generally correspond to inputs with constant wavelengths, such as wheel pair not round, wheelset eccentricity, fixed input brought by track splicing structure, and the like.
  • Fig. 8 is a comparison diagram of the displacement input power spectrum obtained by the displacement signal of a spatial domain at the same time sampling frequency at a time difference of the operating speed.
  • the frequency is proportional to the speed, the running speed is doubled, and the input frequency is doubled;
  • the random input power spectrum is inversely proportional to the speed, the speed is doubled, and the corresponding power spectrum is reduced by half; the periodic excitation power spectrum constant.
  • the time frequency is proportional to the speed, the corresponding frequency interval is simultaneously scaled, and the acceleration input power spectrum of the rail vehicle is proportional to the speed 3 powers;
  • the time frequency is also proportional to the speed, the corresponding frequency interval is not scaled, and the acceleration input power spectrum of the rail vehicle is proportional to the speed 4th power;
  • the actual input can be seen as a combination of the above two cases.
  • the time frequency is proportional to the speed
  • the corresponding frequency interval is simultaneously scaled
  • the acceleration input amplitude spectrum of the rail vehicle is proportional to the speed of 1. 5 power
  • the time frequency is also proportional to the speed, the corresponding frequency interval is not scaled, and the acceleration input amplitude spectrum of the rail vehicle is proportional to the speed 2 power;
  • the actual input can be seen as a combination of the above two cases.
  • the vehicle acceleration response can be normalized by using various vehicle input and speed characteristics to highlight and extract vehicle characteristics and fixed waves. Long input, etc.
  • Step el normalize the power spectrum of the acceleration response signal corresponding to each speed grade in step c by dividing the power of the third power or the power equivalent amplitude spectrum by the speed of 1.5.
  • the processing is performed to obtain the normalized data.
  • the amplitude spectrum is used in the embodiment. Therefore, the embodiment of the present invention performs the normalization process by dividing the amplitude spectrum by the speed of 1.5.
  • Step e3 obtaining an intermediate value of the normalized data corresponding to each analysis frequency, and obtaining an analysis frequency-intermediate value curve;
  • Step e5 Perform zero-phase shift filtering on the analysis frequency-intermediate curve to identify a local peak in the three-dimensional velocity-spectral map that does not vary with speed.
  • the above-mentioned steps el, e3, e5 can exclude the interference of the local peaks which vary with the speed, and only highlight the local peaks which do not change with the speed, and the automatic identification results of the local peaks which do not vary with the speed obtained according to the steps el, e3, e5
  • the curve is the filtered analysis frequency-intermediate value curve, and the corresponding three-dimensional velocity-spectral maps are respectively Fig. 2a - Fig. 7a.
  • the local peak that can be manually identified from Fig. 2a that does not vary with speed is about 44. 4 Hz
  • the local peak that is automatically recognized by Fig. b that does not change with speed is 44.43 Hz
  • the results are basically the same.
  • Fig. 2a and Fig. 2b are respectively a three-dimensional spectrogram of the vertical vibration of the wheelset and its corresponding automatic recognition result without speed change
  • the local peak which does not vary with the speed may correspond to the bending vibration of the wheelset or the wheelset and the track. Coupled vibration.
  • the vertical vibration acceleration of the vehicle body has a very obvious local peak which does not change with the speed
  • automatic The identification can be obtained with a large peak at 0.98, 27.1, 46.9, and 59.6 Hz.
  • the peak value of about 0.98Hz may correspond to the nodding or ups and downs of the car body; the peak at 59.6Hz may be related to the vibration of the vehicle electrical equipment.
  • Step e2 setting a wavelength interval corresponding to a local peak that varies linearly with speed.
  • the initial wavelength interval is set to [6m 8m], [2.51m 3.14m] (corresponding to the wheel pair diameter of 0.88). To 1 meter);
  • Step e4 At each speed level, obtain a frequency corresponding to the maximum power spectrum in the frequency interval corresponding to the wavelength interval, and obtain a speed-frequency curve.
  • the frequency interval corresponding to the wavelength interval at a certain speed level is [speed/wavelength]
  • Speed/wavelength small value such as the frequency interval corresponding to the wavelength interval [6m 8m] when the speed is 100m/s (360km/h) is [100/8Hz 100/6Hz];
  • Step e6 At each speed level, obtain a power spectrum integral value in a fixed frequency band near a frequency corresponding to a maximum power spectrum in the frequency interval corresponding to the wavelength interval, obtain a speed-integral value curve, or obtain a maximum corresponding to the power spectrum
  • the fixed frequency band is taken as [-lHz 1 ⁇ ]
  • the fixed frequency band is mainly set for To overcome the power leakage caused by the velocity distribution and the periodogram algorithm at the same speed level, the evaluation result is stable;
  • Step e8 performing a fitting on the velocity-frequency curve to obtain a slope of a local peak that varies linearly with speed to obtain a fixed wavelength
  • Step elO Four times fitting the velocity-integral value curve or quadratic fitting the velocity-total amplitude curve to obtain a geometric amplitude of a local peak that varies linearly with speed.
  • the vertical acceleration of the wheelset can be considered to be consistent with the vertical input of the wheelset in a certain frequency band, the local peak analysis of the three-dimensional velocity-spectral image of the vertical acceleration of the wheelset can be accurately obtained. Fixed wavelengths and accurate geometric amplitudes.
  • the cycle input result obtained by automatically recognizing the three-dimensional spectrum of the vertical acceleration vibration of the Beijing-Tianjin line is provided in the embodiment of the present invention, and it can be seen that: a characteristic wavelength is about 6.488 m.
  • the cycle input, the amplitude is about 0. 2430mm, corresponding to the impact of the Beijing-Tianjin line track plate; the other has a 2.887m (round pair diameter of 0. 919m) cycle input, the amplitude is about 0.
  • the slope interval in step e2 can be manually set or automatically set, wherein the automatic setting includes the following steps:
  • Step ell normalize the power spectrum at different speed levels in step c by the fourth power of the corresponding speed to obtain a normalized power spectrum, or divide the power equivalent amplitude spectrum by the corresponding speed.
  • the normalization of the second power is performed to obtain a normalized amplitude spectrum;
  • Step e l2 normalizing the normalized frequency by dividing the frequency corresponding to the power spectrum or the power equivalent amplitude spectrum at different speed levels in step c by the speed;
  • Step e l3 performing frequency interpolation refinement according to the normalized frequency and the normalized power spectrum at each speed level, or according to the normalized frequency and the normalized amplitude spectrum ;
  • Step e l4 obtaining an intermediate value of a normalized power spectrum or a normalized amplitude spectrum corresponding to each refinement frequency in the frequency interpolation refinement process, to obtain a refinement frequency-intermediate value curve;
  • Step el5 Identifying the peak value in the refinement frequency-intermediate value curve, and obtaining an estimated value of the wavelength interval corresponding to the local peak that varies linearly with the velocity.
  • the initial value of the periodic input wavelength obtained by automatically recognizing the three-dimensional spectrogram of the vertical acceleration vibration according to the upward rotation of the Beijing-Tianjin line is as shown in FIG. 10a to FIG. 10b, wherein the speed is normalized to lm / s,
  • the spatial frequency at which the peak is obtained is 0. 155m - 0. 348 m 1
  • the initial wavelength range set in the middle is [6m 8m], [2. 51m 3. 14m], and it is feasible to include the initial value of the automatically identified wavelength of 6.452m, 2. 8 74m.
  • Fig. 1 la - Fig. 1 Id is a cycle input result obtained by automatically recognizing the three-dimensional spectrogram of the vertical acceleration vibration according to the descending wheel of the Beijing-Tianjin line in the present embodiment, and it can be seen that: 6.
  • 12a-12b are preliminary results of the periodic input wavelength obtained by automatically identifying the three-dimensional spectrogram of the vertical acceleration vibration according to the descending wheel of the Beijing-Tianjin line in the present embodiment, wherein the velocity is normalized to lm/s, and the obtained wavelength is obtained.
  • FIG. 13 is a schematic flowchart of a method for analyzing acceleration response of a rail vehicle according to an embodiment of the present invention.
  • the embodiment of the present invention is also an analysis of a high-speed train with a speed class of 350 km/h on the Beijing-Tianjin line.
  • Step a-step e is similar to that in the above embodiment, and will not be described again here.
  • the embodiment further includes a step f_step i, and the steps are performed.
  • the premise is that at least two acceleration signals are simultaneously acquired in the step a and there is an input-output relationship between the acceleration signals, and the step f_ i i specifically includes:
  • Step f at each speed level, obtain the nominal transfer between the acceleration signals according to the power spectrum or amplitude spectrum obtained in step c, the algorithm of step f and the classical single input-single output system transfer function
  • the amplitude estimates are the same but the vehicle-track system is a multi-input system, in order to reflect the difference between the patents used in the nominal transfer;
  • Step g The result of step f is represented by a three-dimensional velocity-pass graph
  • Step h obtaining the intermediate value of the nominal transfer obtained by the step f at each analysis frequency, and obtaining an analysis frequency-nominal transfer intermediate value curve;
  • Step i The analysis frequency obtained by the step h - the nominal transfer intermediate value curve is subjected to zero phase shift filtering to identify the local peak of the nominal transfer.
  • the speed-frequency domain analysis of the independent response of the rail vehicle provides an intuitive and comprehensive view of the characteristics of the rail vehicle and the analysis of the input. Since the acceleration response depends on the characteristics and input of the rail vehicle at the same time, the characteristics of the rail vehicle estimated by the independent acceleration response are inevitable. It will be affected by the input. When multiple acceleration responses with a clear correlation are obtained, the nominal transfer function can be used to refine the analysis of the rail vehicle characteristics.
  • the calculation of the nominal transfer function between the above acceleration signals is performed.
  • the nominal transfer function is directly obtained by dividing the amplitude of the nominal output spectrum by the amplitude of the nominal input spectrum. If it is a power spectrum, the result needs to be processed.
  • the nominal transfer estimate is directly performed based on the global data for comparison of the results.
  • 14a-14c are three-dimensional speeds of vertical nominal transfer of wheelset-framework in an embodiment of the present invention, respectively.
  • 15a-c are respectively a three-dimensional velocity-transfer diagram of the lateral nominal transfer of the wheelset-frame in the present embodiment, and a local peak identification result and global data in the three-dimensional velocity-transfer diagram of the wheel-to-frame lateral nominal transfer without speed change, respectively.
  • the nominal transfer diagram shows that: There are few components that change with speed; the nominal transfer function is basically unchanged at each speed level, which is consistent with the theoretical analysis of the method; as can be seen from Figure 15a - Figure 15c, the three analysis The results are very close.
  • the velocity-frequency domain analysis of the acceleration response of the independent rail vehicle can obtain local peaks that do not change with speed (these local peaks generally correspond to the characteristics of the rail vehicle), and the frequency domain characteristics of the characteristics of the rail vehicle can also be obtained from the nominal transfer function analysis.
  • the relationship between the characteristics of the rail vehicles obtained by the two methods is discussed below.
  • 16a to 16c are diagrams showing local peak identification results of a wheel-to-beam, a frame, and a vehicle body vertical vibration in a three-dimensional spectrum diagram, which do not change with speed, respectively.
  • Fig. 17a - Fig. 17c are diagrams showing the results of local peak identification which does not vary with speed in the wheel-to-frame, frame-body, wheelset-body vertical speed-transfer diagram of the present embodiment.
  • Fig. 16b is the peak at 21. 2, 25. 4, 30. 8, 44. 7, 74. 7Hz and 20. 8 , 24. 7 , 31 in Fig. 17a. 0, 46. 1, 73.
  • the peak at 7 Hz agrees well;
  • the peak at 8. 5 Hz in Fig. 16b, 6.8 Hz in Fig. 16c, and the peak at 7.1 and 4.64 Hz in Fig. 17a have a certain value.
  • 0 ⁇ In Figure 16c, the peak at 0. 98, 13. 7, 27. 1, 38. 3, 59. 6 Hz and Figure 7b 0. 7 3, 11. 7, 27. 3, 40. 0, 59.
  • the peak value at 6 Hz is close; in general, the local peak that does not change with speed is identified by the vertical independent response speed-frequency domain method and the vertical nominal The peak consistency of the transmission recognition is better, and the resolution of the peak obtained by the vertical nominal transmission is relatively high.
  • 18a to 18c are respectively a partial peak recognition result in the three-dimensional spectrogram of the wheel pair, the frame, and the vehicle body in the lateral vibration of the present embodiment, which does not change with the speed.
  • 19a to 19c are diagrams showing local peak identification results of the wheel-to-frame, frame-body, wheelset-vehicle lateral velocity-transfer diagrams, respectively, which do not vary with speed in the present embodiment.
  • the local peak of the independent acceleration response-frequency domain method does not change with the speed and the peak consistency obtained by the nominal transmission is better; independent response
  • the speed-frequency domain method has a relatively low frequency resolution; it must also be aware that nominal transmission is not a classical transfer, and its peak value needs to be confirmed in conjunction with actual conditions.
  • the local peak which does not change with the speed can be used to assess the degree of snake behavior of the vehicle.
  • FIG. 20a-c is a snake line identification and evaluation diagram in the embodiment, and the object is a wheelset lateral acceleration Degree, the analysis frequency range is below 10 Hz
  • Figure 20a is the analysis result according to the full velocity domain data, that is, the low frequency part of Fig. 5, the amplified wheelset lateral acceleration three-dimensional velocity-spectrum is as shown in Fig. 21, one can be seen
  • a local peak that does not change with speed, the frequency is about 5.37 Hz, which corresponds to an inherent characteristic of the vehicle
  • Fig. 20b shows the result of local peak analysis that does not vary with speed based on the wheel-to-direction lateral acceleration data with a speed greater than 320 km/h. There are two local peaks that do not change with speed.
  • the frequency is about 5.37Hz and 6.84Hz. Observe Figure 21. Obviously, there is no periodic input and system resonance in this area; therefore, there is an 'additional input' at high speed, according to the previous
  • the analysis may be a meandering, the corresponding frequency is about 6.84 Hz;
  • Figure 20c compares the maximum amplitude in the 1 Hz bandpass around the two center frequencies of the full speed stage 5. 37 Hz, 6.84 Hz (one of the 100% percentile amplitude spectrum) Estimate), you can see that the maximum amplitude rule near 5.37Hz rises, and the maximum amplitude near 6.8411 ⁇ 2 suddenly increases at 2901 ⁇ /1, and then maintains a relatively large value. She gave birth to a degree of hunting motion.
  • This embodiment is the reason for the vertical abnormal vibration of a subway rail vehicle and proposes maintenance and improvement directions, including the following steps:
  • Step 1 Correctly obtain the digital signal of the acceleration response of the rail vehicle and its corresponding speed information.
  • the vertical, horizontal and vertical three-direction accelerations near the traction pin of a metro rail vehicle were measured by a few-W4 type rail vehicle running quality meter.
  • the speed, the operating conditions and other related information were matched according to the train monitoring data.
  • the sampling frequency was 256 Hz.
  • the original signal filtering frequency is 20Hz.
  • Step 2 The acceleration signals are grouped at equal speed intervals. According to the periodogram method, the power amplitude equivalent spectrum and the maximum value of the amplitude spectrum of each speed grade acceleration signal are obtained.
  • the speed interval is taken as 5 km/h, and the analysis length is 4 s.
  • Step 3 The result of step 2 is represented by a three-dimensional velocity-spectral map, identifying local peaks that do not vary with velocity and identifying local peaks that vary linearly with velocity.
  • Step 4 Resonance identification and evaluation.
  • the possible resonance velocity-frequency region is determined and the degree of resonance is evaluated in conjunction with the velocity-percentage spectrum.
  • the velocity-spectrum map shows that there are two sets of local peaks that do not change with speed in the vertical direction of the car body, which are about 1.25 and 13.25 Hz, corresponding to the lines D and E respectively, which correspond to the vertical characteristics of the car body.
  • the vertical vibration anomaly corresponds to the zero position of the handle, that is, the idle condition of the train, so that the conclusion can be drawn:
  • the vertical abnormal vibration of the subway rail vehicle is under certain working conditions, the wheel alignment Inputs (eccentricity, out-of-round, etc.) are transmitted by anomalies and are caused by vertical resonance with the rail vehicle system.
  • the input of the abnormal vibration mainly comes from the wheel pair, and the wheel pair can be inspected and repaired first;
  • FIG. 23 is a schematic structural diagram of an apparatus for analyzing acceleration response of a rail vehicle according to an embodiment of the present invention, where the apparatus includes: an acceleration and speed acquiring unit 110, a grouping unit 120, and spectrum calculation.
  • the unit 130, the three-dimensional display unit 140, and the identification unit 150 are schematic structural diagrams of an apparatus for analyzing acceleration response of a rail vehicle according to an embodiment of the present invention, where the apparatus includes: an acceleration and speed acquiring unit 110, a grouping unit 120, and spectrum calculation.
  • the unit 130, the three-dimensional display unit 140, and the identification unit 150 includes: an acceleration and speed acquiring unit 110, a grouping unit 120, and spectrum calculation.
  • the acceleration and velocity acquisition unit 110 is configured to acquire a rail vehicle acceleration response signal and its corresponding speed information.
  • the unit can be implemented by using an acceleration sensor and a speed sensor.
  • the grouping unit 120 is configured to group the acceleration signals according to speed. In this embodiment, the grouping unit 120 generally selects equal speed intervals to group the acceleration signals, and may also select different speed intervals to group the acceleration signals.
  • the spectrum calculation unit 130 is configured to obtain a power spectrum or a power equivalent amplitude spectrum of the acceleration response signal corresponding to each speed level according to the periodogram method.
  • the three-dimensional display unit 140 is configured to represent the calculation result of the spectrum calculation unit by a three-dimensional velocity-spectral map.
  • the identification unit 150 is configured to identify local peaks that do not vary with velocity in the three-dimensional velocity-spectral map and/or local peaks that vary linearly with velocity.
  • the local peaks that do not vary with speed in the three-dimensional velocity-spectral map and the local peaks that vary linearly with speed can reflect the characteristics of the rail vehicle and its fixed wavelength input.
  • the local peaks that do not vary with speed generally correspond.
  • the characteristics of the vehicle (including structural self-vibration), and the local peaks that vary linearly with speed generally correspond to constant wavelength inputs, such as wheel pair not round, wheelset eccentricity, fixed input from track structure, and so on.
  • the identification unit 150 is used to automatically recognize the three-dimensional velocity - Local peaks in the spectrogram that do not vary with velocity and/or local peaks that vary linearly with velocity.
  • the further embodiment may include the following parts: a normalization module 151, an intermediate value acquisition module 153 and an identification module 155 (shown in FIG. 24), wherein: the normalization module 151 is configured to divide the power spectrum of the acceleration response signal corresponding to each speed level by The normalized data is obtained by dividing the velocity of the third power or the power equivalent amplitude spectrum by the speed of 1.5 to the normalization process;
  • the intermediate value obtaining module 153 is configured to obtain an intermediate value of the normalized data at each analysis frequency to obtain an analysis frequency-intermediate value curve;
  • the identification module 155 is configured to perform zero phase shift filtering on the analysis frequency-intermediate curve to identify a local peak that does not vary with velocity in the three dimensional velocity-spectral map.
  • the further portion may include the following: a wavelength interval setting module 152, a first curve acquiring module 154, The second curve acquisition module 156, the first fitting module 158, and the second fitting module 160 (shown in FIG. 25), wherein:
  • the wavelength interval setting module 152 is configured to set a wavelength interval corresponding to a local peak that varies linearly with speed;
  • the first curve obtaining module 154 is configured to obtain, at each speed level, a frequency corresponding to a maximum power spectrum in the frequency interval corresponding to the wavelength interval, to obtain a speed-frequency curve;
  • the second curve obtaining module 156 is configured to obtain, at each speed level, a power spectrum integral value in a fixed frequency band near a frequency corresponding to a maximum power spectrum in the frequency interval corresponding to the wavelength interval, obtain a speed-integral value curve, or obtain the power.
  • the maximum value of the spectrum corresponds to a velocity-total amplitude curve corresponding to the total amplitude of the power equivalent in a fixed frequency band near the frequency;
  • the first fitting module 158 is configured to perform a fitting on the velocity-frequency curve to obtain a slope of a local peak that varies linearly with speed to obtain a fixed wavelength;
  • the second fitting module 160 is configured to perform four fittings on the velocity-integrated value curve or quadratic fitting the velocity-total amplitude curve to obtain a geometric amplitude of a local peak that varies linearly with speed.
  • the wavelength interval setting module 152 can be manually set or automatically set when the wavelength interval is set.
  • the wavelength interval setting module 152 can further include: the first normalization sub-module 1521 a second normalization sub-module 1522, a refinement processing sub-module 1 523, an intermediate value acquisition sub-module 1524, and a wavelength interval acquisition sub-module 1525 (shown in FIG. 26).
  • the first normalization sub-module 1521 a second normalization sub-module 1522
  • a refinement processing sub-module 1 523 an intermediate value acquisition sub-module 1524
  • a wavelength interval acquisition sub-module 1525 shown in FIG. 26.
  • the first normalization sub-module 1521 is configured to normalize the power spectrum at different speed levels by the fourth power of the corresponding speed to obtain a normalized power spectrum, or divide the power equivalent amplitude spectrum by the corresponding Normalization of the velocity of the second power to obtain a normalized amplitude spectrum;
  • the second normalization sub-module 1522 is configured to normalize the frequency corresponding to the power spectrum or the power equivalent amplitude spectrum at different speed levels by a normalized frequency to obtain a normalized frequency;
  • the refinement processing sub-module 1523 is configured to perform frequency interpolation according to the normalized frequency and the normalized power spectrum at each speed level, or according to the normalized frequency and the normalized amplitude spectrum Refinement processing;
  • the intermediate value acquisition sub-module 1524 is configured to obtain an intermediate value of a normalized power spectrum or a normalized amplitude spectrum corresponding to each refinement frequency in the frequency interpolation refinement process, to obtain a refinement frequency-intermediate value curve;
  • the interval acquisition sub-module 1525 is configured to identify the peak value in the refinement frequency-intermediate value curve, obtain an initial value of the fixed wavelength input, and obtain a wavelength interval corresponding to the local peak value that changes linearly with the velocity.
  • the rail vehicle acceleration response analyzing apparatus in this embodiment further includes a nominal transfer unit 170 (shown in FIG. 27) for acquiring at least two accelerations simultaneously when the acceleration and speed acquiring unit simultaneously acquires When the signal has a certain input-output relationship between the signals, the calculation of the nominal transfer function is performed according to the speed-spectrum data.
  • the nominal transfer unit 170 may further include: a nominal transfer calculation module 171, a nominal transfer display module 172, and a nominal transfer intermediate The value acquisition module 173, the nominal transfer identification module 174 (shown in Figure 28), wherein:
  • the nominal transfer calculation module 171 is configured to obtain a nominal transfer between the acceleration signals according to a power spectrum or an amplitude spectrum obtained in the spectrum calculation unit 130 at each speed level;
  • the nominal transfer display module 172 is configured to represent the results of the nominal transfer calculation module 171 in a three-dimensional velocity-transfer map
  • the nominal transfer intermediate value acquisition module 173 is configured to obtain the intermediate value of the nominal transfer obtained by the nominal transfer calculation module 171 at each analysis frequency to obtain an analysis frequency-nominal transfer intermediate value curve; the nominal transfer identification module 174 is used to transfer the nominal value to the middle.
  • the nominal intermediate value curve is subjected to zero phase shift filtering to identify local peaks of nominal transfer.
  • the rail vehicle acceleration response analyzing apparatus in the embodiment further includes a resonance recognizing and evaluating unit 180 and a meandering recognition and evaluation unit 190 (shown in FIG. 27), wherein:
  • the resonance identification and evaluation module 180 is configured to determine a possible resonance velocity-frequency region based on the local peaks that are not detected by the velocity and the local peaks that vary linearly with the velocity, and combined with the velocity-percentile spectrum evaluation. The degree of resonance.
  • the meandering recognition and evaluation module 190 is configured to detect a local peak that does not vary with speed according to the speed component in the velocity-spectral map, a local peak that varies linearly with speed, and a non-speed that is obtained from the high-speed, low-frequency segment identification of the velocity-spectrum map.
  • the local peak of the change determine the possible meandering speed-frequency region, and combine the velocity-percentile spectrum to assess the degree of serpentine.
  • the apparatus for analyzing the acceleration response of the rail vehicle can comprehensively, intuitively and effectively obtain the characteristics of the rail vehicle and the vehicle input of the fixed wavelength. As long as the original signal band is sufficiently wide, the analysis result is not affected by the filtering frequency or the like;
  • the embodiment of the invention can be used not only for manual identification and qualitative analysis of acceleration signals, but also for automatic quantitative evaluation of rail vehicle characteristics and inputs, and the algorithm is simple and easy to implement; the embodiment of the invention is applicable to rail vehicles, especially high-speed rail vehicles. Test, monitoring, overhaul and feature optimization have strong engineering application value.
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)

Abstract

L'invention concerne un procédé permettant d'analyser une réponse d'accélération d'un véhicule ferroviaire, le procédé comprenant les étapes suivantes : a) l'obtention de signaux de réponse d'accélération et d'informations de vitesse correspondantes d'un véhicule ferroviaire ; b) les signaux de réponse d'accélération sont regroupés en fonction de la vitesse ; c) un spectre d'énergie ou un spectre d'amplitude équivalent à un spectre d'énergie des signaux de réponse d'accélération correspondant à chaque niveau de vitesse est obtenu selon un procédé donnant des périodogrammes ; d) le spectre d'énergie ou le spectre d'amplitude est présenté sous la forme d'un graphique spectral tridimensionnel de vitesse-fréquence ; et e) les valeurs maximales locales ne variant pas avec la vitesse et/ou les valeurs maximales locales variant de manière linéaire avec la vitesse sont identifiées.
PCT/CN2009/075963 2009-12-24 2009-12-24 Procédé et dispositif permettant d'analyser une réponse d'accélération d'un véhicule ferroviaire WO2011075896A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2009/075963 WO2011075896A1 (fr) 2009-12-24 2009-12-24 Procédé et dispositif permettant d'analyser une réponse d'accélération d'un véhicule ferroviaire

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2009/075963 WO2011075896A1 (fr) 2009-12-24 2009-12-24 Procédé et dispositif permettant d'analyser une réponse d'accélération d'un véhicule ferroviaire

Publications (1)

Publication Number Publication Date
WO2011075896A1 true WO2011075896A1 (fr) 2011-06-30

Family

ID=44194913

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2009/075963 WO2011075896A1 (fr) 2009-12-24 2009-12-24 Procédé et dispositif permettant d'analyser une réponse d'accélération d'un véhicule ferroviaire

Country Status (1)

Country Link
WO (1) WO2011075896A1 (fr)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1274978B1 (fr) * 2000-04-19 2004-06-23 DB Reise & Touristik AG Procede et dispositif pour surveiller les caracteristiques de roulement d'un vehicule sur rails
JP2006189388A (ja) * 2005-01-07 2006-07-20 East Japan Railway Co 振動試験装置、及び振動試験方法

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1274978B1 (fr) * 2000-04-19 2004-06-23 DB Reise & Touristik AG Procede et dispositif pour surveiller les caracteristiques de roulement d'un vehicule sur rails
JP2006189388A (ja) * 2005-01-07 2006-07-20 East Japan Railway Co 振動試験装置、及び振動試験方法

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
CHI, MAORU ET AL.: "Influence of hunting motion on ride quality of railway vehicle", JOURNAL OF VIBRATION ENGINEERING, vol. 21, no. 6, December 2008 (2008-12-01), pages 639 - 643 *
CHI, MAORU ET AL.: "Vibrant response characteristic of railway vehicle", JOURNAL OF TRAFFIC AND RANSPORTATION ENGINEERING, vol. 7, no. 5, October 2007 (2007-10-01), pages 6 - 11 *

Similar Documents

Publication Publication Date Title
Yang et al. Vibration singularity analysis for milling tool condition monitoring
CN108845028B (zh) 一种高速铁路钢轨波磨动态检测方法和装置
Cantero et al. Railway infrastructure damage detection using wavelet transformed acceleration response of traversing vehicle
CN113343928B (zh) 变速路段高速铁路钢轨波磨检测方法及装置、计算机设备
CN108515984B (zh) 一种车轮伤损检测方法及装置
CN103969069B (zh) 机械设备工况监测与故障诊断方法
Haase et al. Damage identification based on ridges and maxima lines of the wavelet transform
JP5384166B2 (ja) 構造体変状検知システム
JP6255292B2 (ja) 橋梁のモニタリングシステム、モニタリング方法、及びプログラム
CN113705412B (zh) 基于深度学习的多源数据融合的轨道状态检测方法及装置
CN106092015B (zh) 一种轨道表面凹陷长度检测方法
CN102980756A (zh) 动刚度的测试方法及系统
CN101738497A (zh) 轨道车辆加速度响应分析的方法及装置
CN103884482A (zh) 一种基于压缩机的振动测试方法和系统
CN115758289B (zh) 一种基于多任务学习神经网络的钢轨波磨识别方法
CN113776760B (zh) 基于整轴振动分析的列车轮对失圆故障监测方法和系统
CN104215454A (zh) 一种汽车后桥主减速器噪音性能测试方法
CN112162034A (zh) 一种应用结构噪声的钢-混凝土组合梁损伤识别方法
CN112556627A (zh) 钢轨波磨检测系统及方法
JP2018004469A (ja) 構造体変状検知システム、構造体変状検知方法、及びプログラム
Morichika et al. Estimation of displacement response in steel plate girder bridge using a single MEMS accelerometer
CN105372080B (zh) 一种有轨电车及其嵌入式轨道耦合动力学测试装置及方法
CN116252820B (zh) 改进频域积分法驱动的高速列车车轮多边形定量检测方法
CN110143217B (zh) 轨道状态测量方法、系统及装置
KR100797055B1 (ko) 철도궤도의 효율적 유지보수를 위한 3차원 데이터 형식의철도 궤도틀림도 측정방법

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 09852446

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 09852446

Country of ref document: EP

Kind code of ref document: A1