JP5378093B2 - Rail axial force measuring device and rail axial force measuring method - Google Patents

Description

  The present invention relates to a rail axial force measuring device and a rail axial force measuring method.

In rails such as railways, rail overhang and buckling may occur due to an increase in axial force accompanying temperature rise. If the rails overhang or buckle, the rail operation will be hindered and dangerous. In order to prevent this, the axial force of the rail is measured. Conventionally, the measurement of the axial force of the rail has been performed by measuring the stress directly between the rail play at a predetermined temperature or indirectly measuring the stroke at the expansion joint. In the case of a long rail, the stress and distortion are directly measured by cutting the middle of the rail.
In addition to the direct strain measurement method that has been put into practical use, various methods are currently being studied, including an acoustoelastic method, a permeability measurement method, an acoustic emission method, an X-ray method, and a VERSE method. Among these, Patent Document 1 shows an example of a method for measuring rail axial force using the magnetic permeability of the rail.

JP-A-7-280669

  However, only the direct distortion method is widely used at present. In the direct strain measurement method, it is necessary to unfasten the rail or to cut the rail in the middle when measuring the axial force, and it takes time for the measurement work, and there is a possibility that it may hinder the operation of the railway etc. is there.

  The present invention has been made in view of the above-described problems, and it is possible to measure the axial force of a rail without removing the rail of the actual track in a natural state or cutting the rail in the field. An object of the present invention is to provide a rail axial force measuring device and the like that can be used.

According to a first aspect of the invention for achieving the above object, there is provided a vibrator for vibrating a rail at a predetermined frequency, a receiver for detecting a vibration state of the rail, and the rail while changing the frequency by the vibrator. An analyzer for obtaining natural frequencies for a predetermined vibration mode of the rail, based on vibration state data at the time of vibration at each frequency output from the receiver at a plurality of positions of the rail when vibrated; And a calculation device for calculating the axial force of the rail from the natural frequency obtained by the analyzer based on a correspondence relationship between the axial force of the rail obtained in advance and the natural frequency for the predetermined vibration mode of the rail. And a rail axial force measuring device.

  With the above configuration, the axial force of the rail is measured based on the correspondence obtained in advance from the natural frequency obtained by vibrating the rail on the actual track on site. Accordingly, there is provided a rail axial force measuring apparatus capable of measuring the axial force of the rail without removing the rail of the actual track in the natural state or cutting the rail.

  The correspondence relationship is obtained by performing analysis on a state in which rail axial force is applied to an analysis model simulating the rail.

  The analysis of the state in which the rail axial force is applied to the analysis model is performed using Floquet transformation.

  With the above configuration, it is possible to perform pre-analysis in line with the rails of the actual track in the natural state. Is possible.

  A second invention for achieving the above-described object is a rail axial force measuring method for measuring the axial force of a rail in a state where the rail is fixed in the middle of the sleeper, and the predetermined position of the rail is determined while changing the frequency. Detecting a vibration state at the time of vibration at each frequency of the rail at another plurality of positions of the rail, and a predetermined vibration mode of the rail based on the vibration state detected in the detection step And an analysis step for obtaining a natural frequency for the rail, and a correspondence relationship between the axial force of the rail obtained in advance and the natural frequency for a predetermined vibration mode of the rail, based on the natural frequency obtained in the analysis step. A rail axial force measuring method comprising: a calculating step of calculating an axial force of the rail.

  With the above configuration, the axial force of the rail is measured based on the correspondence obtained in advance from the natural frequency obtained by vibrating the rail on the actual track on site. Therefore, there is provided a rail axial force measurement method capable of measuring the axial force of the rail without removing the rail of the actual track in the natural state or cutting the rail.

  The correspondence relationship is obtained by performing analysis on a state in which rail axial force is applied to an analysis model simulating the rail.

  The analysis of the state in which the rail axial force is applied to the analysis model is performed using Floquet transformation.

  With the above configuration, it is possible to perform pre-analysis in line with the rails of the actual track in the natural state. Is possible.

  According to the present invention, it is possible to provide a rail axial force measuring device and the like that can measure the axial force of the rail without removing the rail of the actual track in a natural state or cutting the rail in the field. it can.

The figure which shows the state of a rail and a rail measuring device Configuration diagram showing the configuration of analysis / calculation device and receiver Flow chart showing the flow of rail axial force measurement method Diagram explaining the rail model Diagram showing the relationship between natural frequency, wave number, and rail axial force Diagram showing the relationship between rail axial force and natural frequency Diagram explaining rail axial force measurement

  Hereinafter, embodiments of a rail axial force measuring device and the like of the present invention will be described with reference to the drawings.

  First, the rail axial force measuring apparatus of this embodiment is demonstrated, referring FIG. 1, FIG. FIG. 1 is a diagram for explaining a state of a rail and a rail axial force measuring device. FIG. 1 (a) is a perspective view showing a state in which the rail axial force measuring device is attached to the rail, and FIG. It is sectional drawing which shows a sleeper. FIG. 2 is a configuration diagram showing the configurations of the analysis / calculation device and the receiver.

  As shown in FIG. 1A, the rail axial force measuring device 1 of this embodiment includes a vibrator 3, a receiver 5, and an analysis / calculation device 6. The receiver 5 and the analysis / calculation device 6 are connected by a cable 8.

  Further, sleepers 9 are provided at intervals L in the rail axial direction at the lower portion of the rail 7. As shown in FIG. 1 (b), the rail 7 and the sleeper 9 are connected via a track pad 11. The sleeper 9 is attached to the upper part of the road bed 15 via the vibration-proof pad 13. The track pad 11 and the vibration-proof pad 13 are made of rubber or the like, for example, and have an effect of alleviating shaking of the rail 7 and the sleeper 9.

  The vibration exciter 3 is attached to one side of the rail 7 so as to be in contact with the side surface, and applies vibration in the vertical direction to the rail 7 at a predetermined frequency.

  The vibration exciter 3 includes a signal generator (not shown) that generates a signal for applying vibration to the rail at an arbitrary frequency, an excitation device (not shown) that applies vibration to the rail based on the signal, and the like. The vibration exciter 3 may be one that loads an impact on a rail by an impulse hammer.

  The receiver 5 is attached to another side surface of the rail 7 so as to be in contact with the vibrator 3 at a different position, detects the vibration state of the rail 7, and transmits the vibration data to the analysis / calculation device 6 via the cable 8. Output. Note that vibration data may be output to the analysis / calculation device 6 by performing wireless communication instead of the cable 8. Moreover, you may make it attach the receiver 5 to the same side as the side where the vibrator 3 of the rail 7 was attached.

As shown in FIG. 2, the receiver 5 includes an accelerometer 31, an amplifier 33, an A / D converter 35, a communication interface 37, and the like. The accelerometer 31 detects the vibration state of the rail 7 as an acceleration signal. The amplifier 33 amplifies the acceleration signal detected by the accelerometer 31. The A / D converter 35 quantizes the amplified acceleration signal and converts it into digital data. The communication interface 37 outputs the acceleration data converted into digital data to the analysis / calculation device 6 via the cable 8.
The receiver 5 may be one that detects the vibration state of the rail 7 by a speed signal or a displacement signal, for example, using a speedometer or a displacement meter instead of the accelerometer 31.

  The analysis / calculation device 6 analyzes the natural frequency based on the vibration data input from the receiver 5 via the cable 8 and calculates the rail axial force based on the natural frequency.

  As shown in FIG. 2, the analysis / calculation device 6 includes a CPU 17, a memory 19, a display unit 21, a media input / output unit 22, a storage unit 23, a communication unit 24, an input unit 25, an output unit 27, and the like. Are connected by a system bus 29.

  The CPU 17 executes a program stored in the memory 19 or the storage unit 23 and executes various processes of the analysis / calculation device 6 described later. The memory 19 includes a RAM and a ROM, and stores various data and programs. The display unit 21 is a display and can display various types of information in accordance with the processing of the analysis / calculation device 6 described later. The media input / output unit 22 controls input / output using a CD-ROM, a DVD-ROM, or the like. The storage unit 23 is a hard disk or the like, and stores data and programs necessary for various processes of the analysis / calculation apparatus 6 described later. The communication unit 24 includes communication interfaces such as RS-232C input / output, wireless communication, and a modem, and controls communication with the outside. The input unit 25 includes an input device such as a keyboard or a mouse, and the output unit 27 includes a printer or the like.

  In the rail axial force measurement method of this embodiment, a program used for vibration data acquisition, natural frequency analysis, or rail axial force calculation is input in advance from the media input / output unit 22 from a medium such as a CD-ROM. It is stored in the storage unit 23 and executed by the CPU 17 or the like as necessary.

  Next, the rail axial force measuring method of the present embodiment will be described with reference to FIGS. In the rail axial force measuring method of this embodiment, the rail axial force of the rail 7 shown in FIG. 3 is a flowchart showing the flow of the rail axial force measuring method of the present embodiment, FIG. 4 is a diagram for explaining the rail model, FIG. 5 is a diagram showing the relationship between the natural frequency, the wave number, and the rail axial force, FIG. FIG. 7 is a diagram illustrating the relationship between rail axial force and natural frequency, and FIG. 7 is a diagram illustrating measurement of rail axial force.

  As shown in FIG. 3, in the rail axial force measurement method of the present embodiment, first, the CPU 17 of the analysis / calculation device 6 executes a model analysis program stored in advance in the storage unit 23 to obtain the actual trajectory shown in FIG. 1. The rail model 10 which is an analysis model simulating the cross-sectional structure and the connection state between the rail 7 and the sleeper 9 and the like of the rail 7 of the actual track according to various conditions such as the rail set based on the rail 7 is finite. It creates using the element method (step 101).

  The rail model will be described with reference to FIG. As shown in FIG. 4A, in this embodiment, based on the actual track of the rail 7 as shown in FIG. Modeling an infinite periodic structure including the anti-seismic pad 13 and the road bed 15 and the sleepers 9 and the like arranged in the axial direction of the rail 7 with the periodic length L. The vibration mode analysis of the rail model shown in FIG. 4A is reduced to the vibration mode analysis in one unit (rail model 10) of the rail model shown in FIG. The rail model 10 having only one unit can be created as an analysis model using the finite element method.

By the finite element method, the rail 7 has its cross section and axial direction discretized into a plurality of finite elements that are coupled to each other at the nodes. At this time, the discretization method such as the number and shape of the finite elements can be set in step 101. Since the rail 7 is represented by a discrete finite element, a minute deformation can be taken into account, so that a more accurate analysis result can be obtained. In Step 101, when the rail model 10 is created, the mass density ρ (kg / m ³ ), Young's modulus E (Pa), Poisson's ratio σ, shear elastic modulus G, and cross-sectional area A (m ² ) And a value such as a cross-sectional secondary moment I (m ⁴ ) is also set. The rail 7 is modeled as an Euler beam that takes into account bending deformation associated with vibration, but may be modeled as a Timoshenko beam that takes into account the effects of shear deformation and rotational inertia. .
The sleeper 9 is modeled as a mass point of mass m (kg). The track pad 11 is modeled as a spring having a spring constant k _r (N / m) connecting the rail 7 and the sleeper 9. The anti-vibration pad 13 is modeled as a spring having a spring constant k _s (N / m) connecting the sleeper 9 and the road bed 15. The road bed 15 is modeled as a rigid body uniform in the axial direction of the rail 7. In the rail model 10, the track pad 11, the sleeper 9, and the vibration isolating pad 13 are arranged at the center in the axial direction of the rail 7 having a length L, and act on the finite element of the rail 7 corresponding to the position. These values of m, k _r , and k _s are also set when the rail model 10 is created in step 101. The sleeper interval L and the like are also set in step 101. The rail model 10 is created as described above.

  Subsequently, the CPU 17 applies a predetermined rail axial force in the axial direction of the rail 7 of the rail model 10 according to the setting of the predetermined rail axial force, and vibrates the rail model 10 in this state using Floquet transformation. A mode analysis is performed to obtain a natural frequency for a predetermined vibration mode (step 102).

  In the vibration mode analysis, the frequency, wave number, axial force is applied to the finite element equation obtained by discretization and setting of the axial force for the one-unit rail model 10 shown in FIG. We derive the eigenvalue problem with By solving this, the vibration mode is analyzed. This will be described below.

  In a linear structure, the axial force and the natural frequency are related, and if a tensile force is applied to both ends, the natural frequency increases. Conversely, if a compressive force is applied, the natural frequency decreases. Therefore, the axial force of the rail can be obtained from the natural frequency (in a predetermined vibration mode) for the rail. For example, when assuming an unsupported rail in a free space or an infinite length rail with a uniform elastic support state, the natural frequency can be obtained by analyzing the vibration mode using the Fourier transform. .

  However, with respect to practical use, there is a drawback in performing an analysis on an unsupported rail or an infinite length rail with a uniform elastic support as described above. This is because the actual rail is supported and fixed by sleepers along the way. In the Fourier analysis, it is difficult to formulate a state where the support is fixed on the way in this way. Therefore, in this embodiment, the Floquet transformation is applied instead of the Fourier transformation, and the vibration mode is analyzed by the Floquet wave number expression based on this.

ここで、ｘ_ｌとκはそれぞれ（−Ｌ／２，Ｌ／２）及び（−π／Ｌ，π／Ｌ）の区間内の実数である。式（２）より、ｕ^‡はκとｘ_ｌについてそれぞれ次のような周期性を持つ。
ｕ^‡（ｘ_ｌ，κ＋２π／Ｌ）＝ｕ^‡（ｘ_ｌ，κ）…（３）
ｕ^‡（ｘ_ｌ＋Ｌ，κ）＝ｕ^‡（ｘ_ｌ，κ）ｅ^−ｉκＬ…（４）
式（３）は第１種周期性、式（４）は第２種周期性と呼ばれる。また、κは通常の波数に対応するもので、Ｆｌｏｑｕｅｔ波数と呼ばれる。 The above-described Floquet transformation is known but will be briefly described. First, with respect to the rail 7 of the periodic length L in the one-unit rail model 10 of FIG. 4B having the periodic length L, the displacement response u (for the sake of simplicity) under the circular frequency ω is targeted for the stationary problem. Suppose one dimension is given by the following equation.
u (x, t) = u ^† (x) e ^iωt (1)
Here, x represents the position along the axial direction of the rail 7, and in FIG. 4B, the central portion (arrangement position of the sleeper 9 etc.) of the rail 7 is set to 0, and in FIG. The right direction is positive. t is a time variable and i is an imaginary unit. The Floquet transform u ^‡ of the steady solution u ^† is given by the following equation.

Here, _xl and κ are real numbers in the sections (−L / 2, L / 2) and (−π / L, π / L), respectively. From equation (2), u ^‡ has the following periodicity for κ and _xl , respectively.
u ^‡ (x ₁ , κ + 2π / L) = u ^‡ (x ₁ , κ) (3)
u ^‡ (x ₁ + L, κ) = u ^‡ (x ₁ , κ) e ^−iκL (4)
Formula (3) is called 1st type periodicity, and Formula (4) is called 2nd type periodicity. Κ corresponds to a normal wave number and is called a Floquet wave number.

ここで、［Ｗ^†］は離散化された各有限要素の仮想節点変位ベクトル、［Ｗ^†＊］はその共役である。［Ｋ］は離散化された各有限要素の剛性を示すレールの剛性行列、［Ｍ］は離散化された各有限要素の質量を示す質量行列、｛Ｕ^†｝は離散化された各有限要素の節点における変位を示す変位ベクトルである。Ｎ_ｉ、Ｎ_ｊはたわみｕ^†の補間関数であり、３次元Ｈｅｒｍｉｔｅ多項式により与えられる。なお、［Ｋ］、［Ｍ］はステップ１０１で設定された各値に基づいて求められる。 By the way, from the flexural vibration of the rail 7, which is a beam that receives an axial force, the following equation of motion can be obtained for u ^† .
EId ⁴ u ^† / dx ⁴ + Nd ² u ^† / dx ² −ω ² ρAu ^† = 0 (5)
Here, EI is the bending rigidity of the rail, N is the rail axial force, ω is the circular frequency, ρ is the mass density of the rail, and A is the cross-sectional area of the rail. A virtual work equation is derived from Equation (5), and the following finite element equation is obtained corresponding to the discretization described above.
^{[W † *] T [K} -NC-ω 2 M] {U †} = 0 ... (6)

Here, [W ^† ] is a virtual nodal displacement vector of each finite element that has been discretized, and [W ^{† *} ] is its conjugate. [K] is a rail stiffness matrix indicating the stiffness of each discretized finite element, [M] is a mass matrix indicating the mass of each discretized finite element, and {U ^† } is each discretized finite element It is a displacement vector which shows the displacement in a nodal point. N _i and N _j are interpolation functions of the deflection u ^† and are given by a three-dimensional Hermitian polynomial. [K] and [M] are obtained based on the values set in step 101.

According to the Floquet principle relating to one unit of the periodic structure as shown in FIG. 4B, the steady solutions u ^† and w ^{† *} have the following second type periodicity. This is due to the fact that u ^† also has periodicity as shown in equation (4) according to equation (2).
u _{L / 2} ^† = u _{−L / 2} ^† e− ^iκL , w _{L / 2} ^{† *} = w _{−L / 2} ^{† *} e ^iκL (8)
Here, u _{−L / 2} ^† and u _{L / 2} ^† are the nodal displacement vectors of the left and right rail ends of one unit, and w _{−L / 2} ^{† *} and w _{L / 2} ^{† *} are the left and right rail ends of one unit. Is a conjugate of the virtual nodal displacement vector. Equation (8) is applied to Equation (6), and u _{L / 2} and w _{L / 2} ^* are eliminated to obtain the following equation.
[K′−NC′−ω ² M ′] {U ′} = 0 (9)
Here, 'represents that the condition of the equation (8) is imposed and the determinant is arranged so that it does not match the original matrix. Since [K], [C], and [M] in Equation (6) are real symmetric matrices, the coefficient matrix in Equation (9) is a Hermite matrix. Further, due to the positive definiteness of each matrix, the values of the natural frequency ω and the axial force N are real values. In step 102, the eigenvalue problem of equation (9) is solved to obtain the Floquet wave number κ and the natural frequency ω for a predetermined axial force N.

An example of the result is shown in FIG. FIG. 5 is a schematic diagram showing the relationship of κ−f when the frequency (vibration) number f is taken on the vertical axis and the Floquet wave number κ is taken on the horizontal axis, and the axial force N ₁ is larger than the axial force N ₂ . It is. The axial force N is positive in the rail compression direction, and there is a relationship of f = ω / 2π between f and ω.

In the case of the axial force N ₁ , vibrations are not damped on the rail with these frequencies as natural frequencies between f ₀ (Hz) and f ₁ (Hz) and between f ₂ (Hz) and f ₃ (Hz). There are modes that propagate to. These are called passband frequencies. On the other hand, there is no mode in which vibration propagates without damping the rail between f ₁ (Hz) and f ₂ (Hz). These are called stopband frequencies. In the case of the axial force N ₂ , each frequency (band) decreases as indicated by f ₀ ′ to f ₃ ′.

Here, the vibration mode that vibrates with f ₂ (f ₂ ′) (Hz) as a natural frequency is a vibration mode that vibrates at a wavelength of 2 L with the spring position as a node. This is a vibration mode called pinned-pinned resonance, and is represented in FIG. In this mode, the sensitivity, that is, the amount of change in the natural frequency with respect to the change in the axial force N is large, and in addition, vibration occurs with the position of the sleeper 9 as a node. There is an advantage that it does not depend on the physical property values of the vibration-proof pad 13 or the like. Therefore, it is preferable to estimate the axial force N by examining the natural frequency of the pinned-pinned resonance vibration mode. That is, in step 102, the natural frequency in the pinned-pinned resonance vibration mode when a predetermined axial force N is applied is obtained. However, the natural frequency in another vibration mode may be obtained and used for subsequent axial force estimation. Further, since the range of the stop band or the pass band also changes in accordance with the change in the axial force N, these ranges, the upper limit value and the lower limit value thereof may be obtained and used for subsequent axial force estimation.

  By using the Floquet transform, it is possible to perform an analysis in conformity with the rail 7 of the real track in the natural state. It becomes possible to take correspondence. The details of the Floquet transformation are described in Kazuhisa Abe, Takumi Furuya, Kazuhiro Benkuro “Analysis of Wave Propagation of Infinite Length Rails Supported by Sleepers”, Applied Mechanics Papers, Vol. 10, p. 1029-1036, 20077.8.

  The CPU 17 of the analysis / calculation device 6 performs the vibration mode analysis of step 102 for a plurality of rail axial forces set in advance, and the natural frequency for a predetermined (pinned-pinned resonance) vibration mode in the rail model 10. And the rail axial force are obtained (step 103). Note that the value of the rail axial force to be applied to the rail model 10 in step 103 is set in consideration of the measurement accuracy of the target rail axial force. For the measurement of the rail axial force, it is sufficient if it can be practically measured in units of 5 to 10 MPa. For example, the interval of the rail axial force set in consideration of this can be determined.

  FIG. 6B is a schematic diagram showing an example in which the natural frequency for the vibration mode of the pinned-pinned mode is obtained for various axial forces N and the relationship with the axial force N is shown. The vertical axis in FIG. 6B is the rail axial force N, and the rail compression direction is positive. The horizontal axis is the natural frequency. FIG. 6B shows that the rail axial force N and the natural frequency have a one-to-one relationship, and that the natural frequency decreases as the rail axial force in the compression direction increases.

  The CPU 17 of the analysis / calculation device 6 causes the storage unit 23 to store data of the correspondence relationship between the natural frequency value and the rail axial force value for the predetermined (pinned-pinned resonance) vibration mode (step) 104).

  The above analysis is performed before the rail axial force measurement in the field. In this embodiment, the CPU 17 of the analysis / calculation device 6 executes the model analysis program stored in the storage unit 23 and performs the above analysis. However, the present invention is not limited to this, and another analysis device such as a computer is used. The above analysis may be performed. In this case, the data on the correspondence between the natural frequency and the rail axial force for the pinned-pinned resonance vibration mode, which is the analysis result, is stored in a storage medium such as a CD-ROM, and the rail axial force measurement is performed on site. Prior to this, the storage medium 23 is input to the analysis / calculation device 6 from the media input / output unit 22 and stored in the storage unit 23.

  On the other hand, as shown in FIG. 1 and FIG. 7A, the vibrator 3 is attached to one side surface of the rail 7 of the actual track in the natural state in which the sleepers 9 are provided in the lower part. Further, the receiver 5 is attached to another side surface of the rail 7 at a position different from the attachment position of the vibrator 3 (step 201).

  Subsequently, vertical vibration is applied to the rail 7 on one side surface of the rail 7 at a predetermined frequency (vibration) for which vibration data is scheduled to be acquired using the vibrator 3 (step 202).

  The receiver 5 detects the vibration in the vertical direction of the rail 7 at the mounting position as a time change of acceleration (step 203, detection process). The detected vibration is output to the analysis / calculation device 6 as vibration data. The CPU 17 of the analysis / calculation device 6 acquires the vibration data, stores it in the storage unit 23 in association with the position information of the vibration data detection position (attachment position of the receiver 5) (step 204). At the time of vibration data acquisition, the vibration data acquisition program stored in the storage unit 23 is executed by the CPU 17 of the analysis / calculation device 6 and the above processing is performed. Here, the vibration data detection position is input in advance as a horizontal distance between the vibration position and the vibration position by the vibration exciter 3, for example, and stored in the memory 19 or the storage unit 23 as the vibration data acquisition program is executed.

  If data has not been acquired for all frequencies for which vibration data is scheduled to be acquired and there is a frequency for which vibration data has not been acquired (No in step 205), the frequency for exciting the rail 7 is changed (step 206), and step 202 is performed. Repeat the process in step 204. When data acquisition is performed for all frequencies for which vibration data acquisition is planned (Yes in step 205), the process proceeds to step 207. Note that the frequency at which vibration data is acquired can be appropriately determined in consideration of the measurement accuracy of the target rail axial force and the like.

  If data is not acquired at all the positions where vibration data is scheduled to be acquired, and there is a position where vibration data has not been acquired (No in step 207), the receiver 5 is moved by a predetermined distance to another side surface of the rail 7. (Step 208), the following steps 202 to 206 are repeated to obtain vibration data at the mounting position of the receiver 5 at the time of rail vibration at each frequency. It is also possible to obtain vibration data by exciting the rail 7 in the lateral direction with the vibrator 3.

  When the above steps are repeated and vibration data at the time of rail excitation at each frequency is acquired at all positions where vibration data is scheduled to be acquired (Yes in step 207), the process proceeds to step 209. The vibration data detection position is variously determined according to the vibration mode to be analyzed and the measurement situation.

  In step 209, the CPU 17 of the analysis / calculation device 6 executes the vibration mode analysis program stored in the storage unit 23, and performs a predetermined (pinned-pinned resonance) vibration mode from the vibration data stored in the storage unit 23. The natural frequency is obtained by analysis and stored in the storage unit 23. (Step 209, analysis process).

  For example, positions A, B, C, and D shown in FIGS. 1A and 7A are positions where vibration data is acquired along the rail axis direction. Here, it is assumed that each position interval is equal to L / 3, and sleepers 9 are arranged at an intermediate position between B and C. For example, when the rail 7 is vibrated at a certain low frequency, vibration data (acceleration u ″ (t)) of the positions A, B, C, and D in the axial direction of the rail 7 shown in FIG. The vibration mode at this time is considered to be the pinned-pinned resonance vibration mode shown in Fig. 6A, and the frequency at that time is set to the pinned-pinned resonance. Stored as the natural frequency for the vibration mode.

  The actual vibration mode analysis program reads, for example, vibration data stored in the storage unit 23 and executes a high-speed FFT or the like to obtain a Fourier amplitude spectrum, or a cross spectrum or transfer function between vibration data acquired at different positions. The natural frequency is obtained for a predetermined (pinned-pinned resonance) vibration mode by performing analysis along with vibration data detection position information, and a known one can be used.

  Subsequently, the CPU 17 of the analysis / calculation device 6 executes a rail axial force calculation program stored in the storage unit 23, and obtains a predetermined (pinned-pinned resonance) vibration obtained in step 209 and stored in the storage unit 23. Correlation between natural frequency and rail axial force for a predetermined (pinned-pinned resonance) vibration mode obtained by model analysis in steps 101 to 103 and stored in the storage unit 23 in step 104 based on the natural frequency of the mode Is used to calculate the rail axial force of the rail 7 on the actual track in the natural state of the site (step 210, calculation step). As described above, the rail axial force of the rail 7 on the actual track in the natural state of the site is measured.

For example, it is assumed that the natural frequency obtained in step 209 corresponds to 630 (Hz). The CPU 17 executes a rail axial force calculation program, and, for example, from the data on the correspondence relationship between the natural frequency and the rail axial force with respect to the pinned-pinned resonance vibration mode obtained by model analysis shown in FIG. A rail axial force corresponding to the natural frequency (630 (Hz)) obtained in step 209 is obtained, and the rail axial force at this time is calculated to be about 40 (N / mm ² ).

  In step 210, correspondence data obtained by model analysis is output in advance in the form of a graph as shown in FIG. 6B, and the graph is obtained from the natural frequency obtained by the measurer in step 209. It is also possible to calculate the rail axial force by reading.

  As described above, according to the rail axial force measuring device and the like of this embodiment, the natural frequency for a predetermined vibration mode is analyzed from the vibration data obtained by vibrating the rail of the actual track in the natural state at the site. By calculating the rail axial force based on the relationship between the natural frequency and the rail axial force obtained by prior analysis, the rail axial force of the actual track in the natural state can be measured on site. It can be carried out. When measuring the rail axial force, it is only necessary to vibrate the rail, so it is not necessary to unfasten or cut off the rail of the actual track in the natural state.

The preferred embodiments of the rail axial force measuring device and the like according to the present invention have been described above with reference to the accompanying drawings, but the present invention is not limited to such examples. It will be apparent to those skilled in the art that various changes or modifications can be conceived within the scope of the technical idea disclosed in the present application, and these naturally belong to the technical scope of the present invention. Understood.
For example, in the present embodiment, vibrations are detected at a plurality of positions by sequentially changing the mounting position of the receiver 5, but vibrations can also be detected at a plurality of positions at a time using a plurality of receivers 5. .

DESCRIPTION OF SYMBOLS 1 ......... Rail axial force measuring device 3 ... …… Excitation device 5 ... …… Receiver 6 ... …… Analysis and calculation device 7 ... …… Rail 8 ……… Cable 9 ……… Sleeve 11 ……… Orbit pad 13 …… Vibration isolation pad 15 ………… Road bed

Claims (6)

A vibrator that vibrates the rail at a predetermined frequency;
A receiver for detecting a vibration state of the rail;
When the rail is vibrated while changing the frequency by the exciter, the rail is determined based on vibration state data at the time of vibration at each frequency output from the receiver at a plurality of positions of the rail. An analyzer for determining the natural frequency for the vibration mode of
A calculation device that calculates the axial force of the rail from the natural frequency obtained by the analysis device based on a correspondence relationship between the axial force of the rail obtained in advance and the natural frequency for the predetermined vibration mode of the rail; ,
A rail axial force measuring device comprising:   The rail axial force measuring device according to claim 1, wherein the correspondence relationship is obtained by performing analysis on a state in which rail axial force is applied to an analysis model simulating the rail.   The rail axial force measuring device according to claim 2, wherein the analysis with respect to a state in which a rail axial force is applied to the analysis model is performed using a Floquet transform. A rail axial force measurement method for measuring the axial force of a rail that is fixed in the middle of a sleeper,
A detection step of vibrating a predetermined position of the rail while changing the frequency, and detecting a vibration state at the time of vibration at each frequency of the rail at a plurality of other positions of the rail;
An analysis step for obtaining a natural frequency for a predetermined vibration mode of the rail based on the vibration state detected in the detection step;
A calculation step of calculating the axial force of the rail from the natural frequency obtained in the analysis step, based on a correspondence relationship between the axial force of the rail obtained in advance and the natural frequency for the predetermined vibration mode of the rail; ,
A rail axial force measuring method comprising:   5. The rail axial force measuring method according to claim 4, wherein the correspondence relationship is obtained by performing an analysis on a state in which a rail axial force is applied to an analysis model simulating the rail.   6. The rail axial force measuring method according to claim 5, wherein the analysis with respect to a state in which rail axial force is applied to the analysis model is performed using Floquet transformation.

Images