A method for dynamic stress analysis of a frog

Abstract

The application discloses a kind of frog dynamic stress analysis methods, and the sensor dynamic data of real car is collected in frog area when passing, data content includes strain and acceleration, the position relationship between measuring point, then a method is used to decompose time domain signal into several parts, then carry out Hilbert-Huang (HHT) transformation, convert time domain signal to time-frequency domain, and the relationship between time, transient frequency and transient energy is solved.Transient frequency corresponds to the inherent frequency of each component, the transient energy of each frequency band can be analyzed by this, and the application state of each component is analyzed.The size of instantaneous energy can be used to judge the use state of the vibration component, and through comparison of multiple working conditions, it can be used to guide maintenance;The position of maximum instantaneous energy can be used to judge whether the transient energy borne by the weak position is reasonable, whether the transient energy is too large;Using Hilbert-Huang transformation method, the signal can be decomposed into several frequency bands, and the number of decomposition frequency bands is not limited.

Description

Technical Field

[0001] This invention relates to the field of railway turnouts, and in particular to a method for dynamic stress analysis of turnouts. Background Technology

[0002] With the rapid development of high-speed turnout construction in China in recent years, dynamic performance testing technology has become relatively mature. The joint commissioning and testing of new high-speed railway lines before their opening involves evaluating the dynamic performance of the track structure by collecting data on wheel-rail vertical force, lateral force, and track structure deformation and vibration. However, the traditional shear stress method cannot obtain the wheel-rail vertical force P and wheel-rail lateral force Q within the harmful space of fixed-type turnouts and for non-standard profile track components. Currently, testing and analysis in this area focus on static wheel-rail contact analysis, while dynamic performance testing remains largely unexplored, primarily relying on simulation analysis.

[0003] The known method for collecting dynamic stress data on frogs utilizes sensors deployed on the locomotive and rolling stock, with data from the vehicle providing feedback on frog performance. This method can analyze the frog area as a whole vibrating component. However, the track system under the wheels in the frog area is very complex and highly discrete, making it difficult to analyze the usage of each individual track component.

[0004] Document CN107765610A discloses a GPRS-based remote monitoring system and method for the service status of railway turnouts. The monitoring system includes at least a field acquisition module, a GPRS signal transmission module, and a host computer. The field acquisition modules are respectively deployed in the switch area and frog area of ​​the railway turnout. Each field acquisition module is connected to the host computer via the GPRS signal transmission module. The field acquisition modules deployed in the switch area and frog area synchronously acquire strain and acceleration information of the turnout structure when a train passes through, and wirelessly transmit the information to the host computer via the GPRS signal transmission module. The advantages of this invention are: it achieves the function of on-site monitoring, evaluation, and early warning of the service status of railway turnouts; data acquisition is more efficient and accurate; and it improves the level of information management of railway engineering equipment and the level of scientific decision-making in maintenance and repair.

[0005] None of the aforementioned patents provide an in-depth and detailed analysis of the collected acceleration signals, nor do they explain how the analysis of acceleration signals can be combined with the actual working conditions. Therefore, a dynamic force analysis method for turnouts is needed to solve the above problems. Summary of the Invention

[0006] This invention addresses the problem of existing methods for collecting dynamic stress data on frogs, which utilize sensors deployed on locomotives to provide feedback on frog performance. While this method treats the frog area as a whole vibrating component, the track system under the wheels in the frog area is complex and highly discrete, making it difficult to analyze the usage of each track component individually. This invention provides a dynamic stress analysis method for frogs. It collects dynamic sensor data from the frog area during actual vehicle passage, including strain, acceleration, and the positional relationships between measuring points. The time-domain signal is then decomposed into several parts using a method, followed by a Hilbert-Huang (HHT) transform to convert the time-domain signal to the time-frequency domain, and the relationship between time, transient frequency, and transient energy is determined. The transient frequency corresponds to the natural frequency of each component, allowing analysis of the transient energy in each frequency band and the application status of each component, thus solving the problems of existing technologies.

[0007] This invention provides a method for dynamic force analysis of a turnout, comprising the following steps:

[0008] S1. Select several nodes at the standard cross-section in the fork area, and select several nodes at the non-standard cross-section in the hazardous space of the fork area.

[0009] S2. Install strain sensors and acceleration sensors at nodes in the standard cross-section, and install acceleration sensors at non-standard cross-sections. Record the distance between the sensors. The strain sensors and acceleration sensors detect wheel-rail force and acceleration, and proceed to step S5. Set the sampling frequency for the acceleration sensors and strain sensors, and collect the time-domain waveform data of the acceleration sensors and strain sensors when the actual vehicle passes through the frog area with the sampling frequency as the period, and proceed to step S3.

[0010] S3. Perform time-domain and frequency-domain analysis on the time-domain waveform data to obtain the train speed, impact time, and the relationship between time t, transient energy E, and instantaneous frequency f.

[0011] S4. Determine the natural vibration frequency F of each component in the track. Based on the correspondence between the natural vibration frequency F and the instantaneous frequency f, obtain the turnout structure component corresponding to the transient energy E of each frequency band in the time-frequency domain data of the acceleration signal, and proceed to step S6.

[0012] S5. Calculate the time required for the same wheel to pass through the two sensors using the time information from the two strain sensors, and calculate the actual vehicle speed based on the distance between the two strain sensors, and then proceed to step S6.

[0013] S6. Based on the actual vehicle speed and the time when the maximum transient energy occurs, obtain the location where the maximum transient energy occurs;

[0014] S7. Evaluate the strength of this structural component under impact load based on the transient energy E of each structural component;

[0015] By comparing the changes in transient energy E of each structural component, the severity of the operating conditions and whether maintenance is required can be determined.

[0016] Based on the magnitude of the transient energy E and the location where the maximum value occurs, determine whether the weak point bears the maximum transient energy E, and further determine whether this situation is caused by unreasonable structural design.

[0017] Time-frequency domain analysis is performed on the time-domain signal, decomposing the signal into various frequency bands and calculating the instantaneous frequency and instantaneous energy of each band. The natural vibration frequencies of each component are correlated with the instantaneous frequencies of the measured signal, and the instantaneous energy in the acceleration signal is extracted and correlated one-to-one with the natural vibration frequencies.

[0018] The dynamic force analysis method for a turnout described in this invention, as a preferred embodiment, involves the following specific method for collecting time-domain waveform data from acceleration and strain sensors when a real vehicle passes through the turnout area in step S2:

[0019] Time-domain analysis was performed on the waveform data of the strain measurement points to obtain the time corresponding to the peak strain under impact.

[0020] Based on the distance between the acceleration measuring point and the strain measuring point, determine the approximate time when each wheel passes through the acceleration measuring point;

[0021] Centered on the time of the acceleration measurement point, acceleration measurement data of duration T are taken to obtain the impact time-domain waveform data a of each wheel passing through the acceleration measurement point;

[0022] VMD variational mode decomposition is performed on the impact time-domain waveform data a to obtain discrete sub-signal data in the impact time domain. Hilbert-Huang transform is then performed on the discrete sub-signal data in the impact time domain to convert the discrete sub-signal data in the impact time domain to the time-frequency domain, thereby obtaining the correspondence between the acceleration waveform data a at time t, various instantaneous frequencies f, and instantaneous energy E.

[0023] As is well known, an impact signal is a broadband signal. When an impact force acts on a structure, the vibration signal frequency contains multiple natural frequencies of that structure. Therefore, the impact signal generated when a train passes a frog contains the natural frequencies of multiple systems, including the rail components, wheel load system, turnout sleepers, track bed, and fasteners, among others.

[0024] This technical solution employs a time-frequency transformation analysis method to divide the measured track component acceleration signal into multiple frequency bands, calculating the instantaneous frequency and instantaneous energy of each band. Since the natural frequencies of the vibrating components of the frog can be measured individually or derived from previous experimental results, the instantaneous energy corresponding to each natural frequency of the vibrating components is summed to determine the energy magnitude. The instantaneous energy magnitude can be used to evaluate the stress condition of each component.

[0025] This technical solution uses VMD variational mode decomposition, but other decomposition methods can also be used to decompose the time-domain signal into several parts. Therefore, no limitation is made on the time-domain signal decomposition method.

[0026] The Hilbert spectrum of the impact signal on the rail includes the impact energy from the train wheels, the energy from the turnout sleepers, and the energy from the fastening system. However, the natural frequencies of each component differ, and the installation state also affects the natural frequencies. For example, the acceleration on the turnout sleepers can be used to analyze the natural frequency range of the sleepers currently in use, and sensors installed on other components can also be used to analyze their natural frequencies. Simulation analysis can also be used to determine the natural frequencies of each component. Once the natural frequency range is clear, the frequency components of the rail impact can be roughly determined, and the sum of the transient energy of the rail impact can be calculated.

[0027] In summary, by inputting acceleration signals into the analysis system, the transient energy of each component is obtained.

[0028] Further data analysis can be used to compare the transient energy of rail impact, the transient energy of turnout sleepers and other components; and to compare the differences between the same measuring point on different frogs and the differences between different positions on the same frog.

[0029] The dynamic stress analysis method for a turnout described in this invention, as a preferred embodiment, specifically includes step S3 as follows:

[0030] S31. The natural vibration frequency F of each component of the track is obtained through field tests or based on theoretical modal analysis;

[0031] S32. By comparing the natural vibration frequency F of the structural component with the instantaneous frequency f, determine the structural component corresponding to each instantaneous frequency f of the acceleration waveform data a.

[0032] S33. Obtain the transient energy E of the structural component based on the instantaneous frequency f corresponding to each structural component.

[0033] The dynamic force analysis method for a turnout described in this invention, as a preferred embodiment, involves setting up acceleration sensors, including unidirectional acceleration sensors and tridirectional acceleration sensors, at non-standard cross-sections.

[0034] The dynamic stress analysis method for a turnout described in this invention, as a preferred embodiment, specifically includes step S7 as follows:

[0035] Using the wheel-rail force and transient energy at the selected standard section measured on-site as a reference benchmark, the strength of this structural component under impact load is evaluated based on the transient energy E of each structural component.

[0036] Calculate the change in transient energy E of each structural component and the difference ΔE between it and the historical data E'. Divide ΔE into several segments from the theoretical minimum to the theoretical maximum value, and select a threshold E between the maximum and minimum values. TH The severity of the operating conditions is determined based on the value range of ΔE, and the relationship between ΔE and E... TH The size relationship determines whether repair is needed;

[0037] Based on the magnitude and location of the transient energy E, compare it with the location of the weakest component of the frog to determine whether the two location information overlap. If they do, combine the overall information of the frog to determine whether the structural design is unreasonable. Otherwise, output the result of the relationship between the magnitude and location of the transient energy E.

[0038] The beneficial effects of this invention are as follows:

[0039] (1) This method determines the time of impact load on a component based on the time when the instantaneous maximum energy of each component occurs; the velocity can be calculated from the strain data and the measurement point location, and the impact location can be inferred.

[0040] (2) In the context of measurable wheel-rail force in standard cross-section, the wheel-rail force at other locations (non-standard cross-section of rail component) can be obtained by comparing the instantaneous energy magnitude with that of the standard cross-section.

[0041] (3) The instantaneous energy level can be used to determine the operating status of the vibrating components. By comparing multiple operating conditions, it can be used to guide maintenance and repair.

[0042] (4) The location of the maximum instantaneous energy can be used to judge whether the transient energy borne by the weak point is reasonable and whether the transient energy is too large;

[0043] (5) The Hilbert-Huang transform method can decompose the signal into several frequency bands, and the number of frequency bands is unlimited. Attached Figure Description

[0044] Figure 1 This is a schematic diagram of a dynamic stress analysis method for a fork. Detailed Implementation

[0045] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0046] Example 1

[0047] like Figure 1 As shown, a dynamic force analysis method for a turnout includes the following steps:

[0048] S1. Select several nodes at the standard cross-section in the fork area, and select several nodes at the non-standard cross-section in the hazardous space of the fork area.

[0049] S2. Install strain sensors and acceleration sensors at nodes in the standard cross-section, and install acceleration sensors at non-standard cross-sections. Record the distance between the sensors. The strain sensors and acceleration sensors detect wheel-rail force and acceleration, and proceed to step S5. Set the sampling frequency for the acceleration sensors and strain sensors, and collect the time-domain waveform data of the acceleration sensors and strain sensors when the actual vehicle passes through the frog area with the sampling frequency as the period, and proceed to step S3.

[0050] S3. Perform time-domain and frequency-domain analysis on the time-domain waveform data to obtain the train speed, impact time, and the relationship between time t, transient energy E, and instantaneous frequency f.

[0051] S4. Determine the natural vibration frequency F of each component in the track. Based on the correspondence between the natural vibration frequency F and the instantaneous frequency f, obtain the turnout structure component corresponding to the transient energy E of each frequency band in the time-frequency domain data of the acceleration signal, and proceed to step S6.

[0052] S5. Calculate the time required for the same wheel to pass through the two sensors using the time information from the two strain sensors, and calculate the actual vehicle speed based on the distance between the two strain sensors, and then proceed to step S6.

[0053] S6. Based on the actual vehicle speed and the time when the maximum transient energy occurs, obtain the location where the maximum transient energy occurs;

[0054] S7. Evaluate the strength of this structural component under impact load based on the transient energy E of each structural component;

[0055] By comparing the changes in transient energy E of each structural component, the severity of the operating conditions and whether maintenance is required can be determined.

[0056] Based on the magnitude of the transient energy E and the location where the maximum value occurs, determine whether the weak point bears the maximum transient energy E, and further determine whether this situation is caused by unreasonable structural design.

[0057] Specifically, step S8 includes:

[0058] Using the wheel-rail force and transient energy at the selected standard section measured on-site as a reference benchmark, the strength of this structural component under impact load is evaluated based on the transient energy E of each structural component.

[0059] Calculate the change in transient energy E of each structural component and the difference ΔE between it and the historical data E'. Divide ΔE into several segments from the theoretical minimum to the theoretical maximum value, and select a threshold E between the maximum and minimum values. TH The severity of the operating conditions is determined based on the value range of ΔE, and the relationship between ΔE and E... TH The size relationship determines whether repair is needed;

[0060] Based on the magnitude and location of the transient energy E, compare it with the location of the weakest component of the frog to determine whether the two location information overlap. If they do, combine the overall information of the frog to determine whether the structural design is unreasonable. Otherwise, output the result of the relationship between the magnitude and location of the transient energy E.

[0061] The information collected in steps S1 and S2 of this technical solution includes the following:

[0062] 1. Collect the vertical stress and acceleration of the wing rail or fork rail (standard section) as the basis for comparing the acceleration of non-standard section rails;

[0063] 2. Collect the vibration acceleration of the center rail and wing rail within the wheel-rail force transition range (the impact position of the center rail or wing rail is measured by a triaxial accelerometer);

[0064] 3. Collect the vibration acceleration of the center rail and wing rail within the force transition range of the turnout wheel-rail;

[0065] 4. Collect the vibration acceleration of the main affected turnout sleepers within the wheel-rail force transition range of the turnout;

[0066] 5. The distance between each measuring point.

[0067] Furthermore, in this embodiment, the sampling frequency of all measuring points is above 5kHz, the rail acceleration measurement range is 2000g, and the turnout sleeper is 500g.

[0068] The specific method for collecting time-domain waveform data from the acceleration and strain sensors when the actual vehicle passes through the turnout area in step S2 is as follows:

[0069] Time-domain analysis was performed on the waveform data of the strain measurement points to obtain the time corresponding to the peak strain under impact.

[0070] Based on the distance between the acceleration measuring point and the strain measuring point, determine the approximate time when each wheel passes through the acceleration measuring point;

[0071] Centered on the time of the acceleration measurement point, acceleration measurement data of duration T are taken to obtain the impact time-domain waveform data a of each wheel passing through the acceleration measurement point;

[0072] VMD variational mode decomposition is performed on the impact time-domain waveform data a to obtain discrete sub-signal data in the impact time domain. Hilbert-Huang transform is then performed on the discrete sub-signal data in the impact time domain to convert the discrete sub-signal data in the impact time domain to the time-frequency domain, thereby obtaining the correspondence between the acceleration waveform data a at time t, various instantaneous frequencies f, and instantaneous energy E.

[0073] Step S3 specifically includes:

[0074] S31. The natural vibration frequency F of each component of the track is obtained through field tests or based on theoretical modal analysis;

[0075] S32. By comparing the natural vibration frequency F of the structural component with the instantaneous frequency f, determine the structural component corresponding to each instantaneous frequency f of the acceleration waveform data a.

[0076] S33. Obtain the transient energy E of the structural component based on the instantaneous frequency f corresponding to each structural component.

[0077] Accelerometers, including unidirectional and tridirectional accelerometers, are installed at non-standard cross-sections.

[0078] In this embodiment, the specific principle of energy comparison is as follows: There is a standard profile rail section inside the frog. The wheel-rail force can be obtained by the traditional shear stress method. The instantaneous energy of the wheel-rail force can be compared with that of the non-standard section rail component in the harmful space of the fixed frog. This can be used to analyze the strength of the load on the frog.

[0079] The specific principle for determining the impact point location of a frog is as follows: A three-dimensional acceleration sensor is installed on the impact-bearing rail component of the frog. By calculating the magnitude of the transient energy of the vertical acceleration at the installation location of the measuring point, the time it takes for the wheel to pass through the three-dimensional acceleration sensor is determined. By calculating the magnitude of the transient energy of the acceleration in another direction, the moment when the transient energy of the impact load in that direction reaches its maximum value is determined, which is the time when the load occurs from zero. If the velocity is measurable, the distance corresponding to the two times can be calculated. Thus, the impact point location can be determined, which can be used to analyze the distance between the impact force on the frog and the measured point.

[0080] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for dynamic force analysis of a turnout, characterized in that: Includes the following steps: S1. Select several nodes at the standard cross-section of the fork area and select several nodes at the non-standard cross-section of the hazardous space in the fork area. S2. Strain sensors and acceleration sensors are installed at nodes in the standard cross-section, and acceleration sensors are installed in non-standard cross-sections. The distance between the sensors is recorded. The strain sensors and acceleration sensors detect wheel-rail force and acceleration, and step S5 is performed. The sampling frequency of the acceleration sensors and strain sensors is set, and the time-domain waveform data of the acceleration sensors and strain sensors are collected periodically at the sampling frequency when the actual vehicle passes through the frog area, and step S3 is performed. S3. Perform time-domain and frequency-domain analysis on the time-domain waveform data to obtain the train speed, impact time, and the relationship between time t, transient energy E, and instantaneous frequency f. S4. Determine the natural vibration frequency F of each component in the track, and obtain the turnout structure component corresponding to the transient energy E of each frequency band in the time-frequency domain data of the acceleration signal based on the correspondence between the natural vibration frequency F and the instantaneous frequency f, and proceed to step S6. S5. Calculate the time required for the same wheel to pass through the two strain sensors using the time information from the two strain sensors, and calculate the actual vehicle speed based on the distance between the two strain sensors, and then proceed to step S6. S6. Based on the actual vehicle speed and the time when the maximum transient energy occurs, obtain the location where the maximum transient energy occurs; S7. Evaluate the strength of this structural component under impact load based on the transient energy E of each structural component; By comparing the changes in transient energy E of each structural component, the severity of the operating conditions and whether maintenance is required can be determined. Based on the magnitude of the transient energy E and the location where the maximum value occurs, determine whether the weak point bears the maximum transient energy E, and further determine whether this situation is caused by unreasonable structural design.

2. The dynamic force analysis method for a turnout according to claim 1, characterized in that: The specific method for collecting the time-domain waveform data of the acceleration sensor and the strain sensor when the actual vehicle passes through the turnout area in step S2 is as follows: Time-domain analysis was performed on the waveform data of the strain measurement points to obtain the time corresponding to the peak strain under impact. Based on the distance between the acceleration measuring point and the strain measuring point, determine the approximate time when each wheel passes through the acceleration measuring point; Centered on the time of the acceleration measurement point, the acceleration measurement point data with a duration of T is taken to obtain the impact time-domain waveform data a of each wheel passing through the acceleration measurement point; The impact time-domain waveform data a is subjected to VMD variational mode decomposition to obtain discrete sub-signal data in the impact time domain. The discrete sub-signal data in the impact time domain is then subjected to Hilbert-Huang transform to convert the discrete sub-signal data in the impact time domain to the time-frequency domain, thereby obtaining the correspondence between the acceleration waveform data a at time t, each instantaneous frequency f, and instantaneous energy E.

3. The dynamic force analysis method for a turnout according to claim 1, characterized in that: Step S3 specifically includes: S31. The natural vibration frequency F of each component of the track is obtained through field tests or based on theoretical modal analysis; S32. By comparing the natural vibration frequency F of the structural component with the instantaneous frequency f, determine the structural component corresponding to each instantaneous frequency f of the acceleration waveform data a. S33. Obtain the transient energy E of the structural component based on the instantaneous frequency f corresponding to each structural component.

4. The dynamic force analysis method for a turnout according to claim 1, characterized in that: The acceleration sensors installed at the non-standard cross-section include unidirectional acceleration sensors and tridirectional acceleration sensors.

5. The dynamic force analysis method for a turnout according to claim 1, characterized in that: Step S7 specifically involves: Using the wheel-rail force and transient energy at the selected standard section measured on-site as a reference benchmark, the strength of this structural component under impact load is evaluated based on the transient energy E of each structural component. Calculate the change in transient energy E of each structural component and the difference ΔE between it and the historical data E'. Divide ΔE into several segments from the theoretical minimum to the theoretical maximum value, and select a threshold E between the maximum and minimum values. TH The severity of the operating conditions is determined based on the value range of ΔE, and the relationship between ΔE and E... TH The size relationship determines whether repair is needed; Based on the magnitude and location of the transient energy E, compare it with the location of the weakest component of the frog to determine whether the two location information overlap. If they do, combine the overall information of the frog to determine whether the structural design is unreasonable. Otherwise, output the result of the relationship between the magnitude and location of the transient energy E.

Images