Wheel load estimation method, and track monitoring method using this method
By integrating a vibration accelerometer and additional sensors to measure vertical and horizontal deformations, the method addresses the limitations of conventional wheel load estimation, providing accurate and continuous monitoring of wheel loads and track conditions.
Patent Information
- Application Number
- JP2023070562
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-04-24
- Publication Date
- 2024-11-06
AI Technical Summary
Conventional wheel load estimation methods using PQ monitoring trolleys are unable to accurately measure instantaneous and local changes in wheel loads due to limitations in sensor placement and the inability to account for external forces, leading to incomplete data acquisition.
Installation of a vibration accelerometer on the axle box of the PQ monitoring cart, combined with sensors to measure vertical and horizontal deformations, allows for continuous data collection and accurate estimation of wheel loads, including instantaneous fluctuations.
Enables highly accurate wheel load estimation and trajectory monitoring by accounting for vertical and horizontal accelerations, deformations, and lateral forces, facilitating real-time safety assessments and early detection of track abnormalities.
Smart Images

Figure 2024156253000001_ABST
Abstract
Description
[Technical field]
[0001] The present invention relates to a method for estimating wheel load based on data obtained from a commercial vehicle, a track monitoring method using this method, and a commercial vehicle or the like for carrying out this method. [Background technology]
[0002] The derailment coefficient is used as an index to measure the running safety of railway vehicles. The derailment coefficient is the ratio of the wheel load P, which is the vertical load acting between the wheels of the railway vehicle and the rail, to the lateral force Q, which is the horizontal load acting between the wheels of the railway vehicle and the rail, and is expressed as Q / P, which is the value obtained by dividing Q by P.
[0003] The derailment coefficient can be thought of as a value that indicates the increased risk of derailment when the lateral force Q acting in the direction that causes the wheels to deviate increases relative to the wheel load P that presses the wheels against the rails. The derailment coefficient is an indicator of the running safety of the vehicle itself, and is also used to detect track abnormalities, since any displacement of the rails for any reason can be detected as a change in the derailment coefficient.
[0004] It has been proposed to make it possible to measure the derailment coefficient on a daily basis by using a PQ monitoring bogie equipped with sensors that can estimate the wheel load P and lateral force Q with a certain degree of accuracy (see, for example, Patent Document 1, Patent Document 2, and Non-Patent Document 1).
[0005] The derailment coefficient is generally measured using PQ wheelsets, which are made by processing the wheels and attaching strain gauges to them. The strain gauges are attached to holes in the wheels to detect the longitudinal deformation of the wheels caused by the wheel load, as well as the bending deformation of the wheels caused by lateral pressure. However, the measurement of the derailment coefficient using PQ wheelsets is only carried out on very limited occasions, such as when a new line is opened or a new type of vehicle is introduced, and only one-off or intermittent measurements can be performed, so the frequency of measurement is extremely low. This is because holes must be drilled in the wheelsets to attach the strain gauges, and wheelsets with holes cannot be used on commercial lines for long periods due to strength issues, and because it is necessary to avoid the effects of heat and brakes cannot be applied, they could not be used on vehicles in commercial operation.
[0006] In contrast, with a PQ monitoring bogie, strain gauges and various sensors are attached to the static system, such as the bogie frame, rather than to the wheels, making it possible to estimate the wheel load P and lateral force Q with a certain degree of accuracy.By installing it on a commercial vehicle, it is possible to continuously collect large amounts of data over the long term and constantly observe the derailment coefficient Q / P.
[0007] In order to estimate the wheel load P in the PQ monitoring trolley, it is necessary to consider that the measurement value is as unaffected as possible by various external forces, such as lateral pressure and forces due to acceleration and deceleration. Generally, in the PQ monitoring trolley, in order to eliminate the effects of these external forces as much as possible, a signal corresponding to the wheel load P is detected at a position as close to the wheel as possible. Specifically, as described in Non-Patent Document 1, a method is used in which a sensor (magnetostrictive displacement meter) is attached to the axle spring and the strain and deflection generated in the axle spring is measured.
[0008] In addition, in order to estimate the lateral force Q in a PQ monitoring bogie, for example, as shown in Patent Document 1, a sensor for measuring lateral force (non-contact displacement meter) is attached to the bottom of the bogie axle box using a jig to detect the bending deformation of the wheel. [Prior art documents] [Patent documents]
[0009] [Patent Document 1] JP 2006-88967 A [Patent Document 2] JP 2015-51674 A [Non-patent literature]
[0010] [Non-Patent Document 1] "Method of measuring wheel / rail contact force without using PQ wheelset", Transactions of the Japan Society of Mechanical Engineers (Series C), Vol. 77, No. 774 (2011-2), pp. 147-155 Summary of the Invention [Problem to be solved by the invention]
[0011] As described above, by using the PQ monitoring trolley, the wheel load P can be estimated based on daily measurement data from commercial vehicles.
[0012] However, the conventional wheel load estimation method using the PQ monitoring bogie uses a magnetostrictive displacement meter to measure the displacement of the axle spring, and is therefore unable to follow momentary, shock-induced wheel load fluctuations.As a result, the conventional wheel load estimation method does not measure momentary wheel load fluctuations, and is therefore unable to respond to localized changes in the track.
[0013] The present invention has been made in consideration of the above circumstances, and has an object to provide a wheel load estimation method that is highly accurate by making it possible to obtain data continuously and over the long term from commercial vehicles and to follow momentary wheel load fluctuations, and a track monitoring method using this method. [Means for solving the problem]
[0014] In order to achieve the above object, the inventors have conducted extensive research and discovered that by attaching a vibration accelerometer to the axle box of a PQ monitoring bogie and performing calculations taking into account the data obtained from this vibration accelerometer, it is possible to accurately estimate the wheel load, including the instantaneous load in the vertical direction to the rail surface. The present invention was completed based on the findings of the inventors.
[0015] The wheel load estimation method of the present invention comprises the steps of: A method for estimating a wheel load of a commercial vehicle while the vehicle is running, the method comprising: running a commercial vehicle equipped with a first sensor that measures the acceleration of vibration in the vertical direction of an axle box on an inner rail side and an outer rail side; and a second sensor that measures the deflection amount of an axle spring on an inner rail side and an outer rail side; a first step of multiplying an output value of the inner rail side and an output value of the outer rail side by a first sensor by a required coefficient and adding them together to obtain vertical accelerations at the wheel positions for the inner rail side and the outer rail side, respectively; The method has a second step of estimating the wheel loads on the inner rail side and the outer rail side based on a gravity term obtained by multiplying the weight of the axle, the gravitational acceleration, and a required coefficient, an axle spring force term obtained by multiplying the output value of the inner rail side and the output value of the outer rail side from the second sensor by a required coefficient and then adding them together, and an inertia force term obtained by multiplying 1 / 2 the mass of the axle and the output value of the first step.
[0016] The present invention further includes a third sensor for measuring the amount of horizontal deformation of the wheel plate portion due to bending on the inner rail side and the outer rail side, In addition to the second step, a third step may be included in which the wheel loads of the inner rail side and the outer rail side are estimated by subtracting the output value of the inner rail side from the output value of the outer rail side by the third sensor and multiplying the result by a required coefficient to add or subtract a lateral force term.
[0017] In the present invention, the second step may include at least a step of performing calculations expressed by the following formulas (1) and (2).
number
number
[0018] In the present invention, the second and third steps may include at least the steps of performing the calculations expressed by the following equations (3) and (4).
number
number
[0019] The present invention may include at least a process of performing the calculations expressed by the following formulas (5) and (6) in the first step. Note that the present invention may include a process of performing the calculations expressed by the following formulas (5) and (6) in the first step, whether or not the third step that takes into account the lateral force term is included.
number
number
[0020] In the present invention, the third step may include a process of applying a notch filter for reducing a signal caused by a wheelset surface wobble when determining the lateral force term.
[0021] The commercial vehicle of the present invention is a bogie having an axle box vertical vibration signal receiving unit that receives an output signal from a first sensor, an axle spring vertical displacement signal receiving unit that receives an output signal from a second sensor, and a wheel displacement signal receiving unit that receives an output signal from a third sensor; The vehicle includes a vehicle body and an underfloor. A control unit is provided under the floor which inputs the speed signal and train information transmitted from the car body unit, and is connected to the axle box up and down vibration signal receiving unit, the axle spring up and down displacement signal receiving unit and the wheel displacement signal receiving unit via an amplifier unit, and transmits data to an information processing device that can implement any of the wheel load estimation methods described above.
[0022] In the present invention, a commercial vehicle capable of measuring the state of a track may be provided with the above-mentioned information processing device in its vehicle body. In this case, the information processing device executes the first and second steps described above.
[0023] The method for monitoring a trajectory of the present invention includes the steps of: A commercial vehicle is run, and the inner rail side lateral force and the outer rail side lateral force are respectively determined based on the output value of the third sensor. These values are then divided by the inner rail side wheel load and the outer rail side wheel load respectively determined by any of the wheel load estimation methods described above to calculate derailment coefficients for the inner rail side and the outer rail side, and changes in the derailment coefficients in the section in which the commercial vehicle is running are monitored. Effect of the Invention
[0024] The present invention is a method for estimating wheel load based on data obtained from a commercial vehicle. Like conventional PQ monitoring bogies, it uses data obtained from a second sensor that measures the deflection of the axle spring on the inner rail side and the outer rail side. In addition, as a new configuration that did not exist previously, a first sensor (e.g., a vibration accelerometer) is provided that measures the acceleration of the vertical vibration of the axle box on the inner rail side and the outer rail side, and calculations are performed using the data obtained from the first sensor.
[0025] Therefore, the present invention has a first step in which the output value of the inner rail side and the output value of the outer rail side from the first sensor are multiplied by a required coefficient and then added together to obtain the vertical acceleration at the wheel position for each of the inner rail side and the outer rail side.
[0026] In addition, the present invention has a second step of estimating the wheel loads on the inner rail side and the outer rail side based on a gravity term obtained by multiplying the weight of the axle, the gravitational acceleration, and a required coefficient, an axle spring force term obtained by multiplying the output value of the inner rail side and the output value of the outer rail side from the second sensor by a required coefficient and then adding them together, and an inertia force term obtained by multiplying 1 / 2 the mass of the axle and the output value of the first step.
[0027] According to the present invention, in addition to being able to obtain data on the track on which commercial vehicles run continuously and over the long term, by performing calculations that take into account the vertical acceleration at the wheel position for each of the inner and outer rail sides, it is possible to follow even momentary wheel load fluctuations, thereby making it possible to accurately estimate the wheel load.
[0028] In the wheel load estimation method of the present invention, in order to further improve the accuracy of the estimated wheel load, it is preferable to provide a third sensor on the commercial vehicle to measure the amount of horizontal deformation due to bending of the wheel plate portion on the inner rail side and the outer rail side, and to include a third step of adding or subtracting a lateral force term obtained by subtracting the output value of the inner rail side from the output value of the outer rail side by the third sensor and multiplying the result by a required coefficient. [Brief description of the drawings]
[0029] [Figure 1] FIG. 1 is an explanatory diagram showing the meanings of parameters used in an estimation equation of a wheel load estimation method according to an embodiment. [Diagram 2] FIG. 2 is a block diagram showing the relationship between parameters and input / output of functions used in the wheel load estimation method according to the embodiment. [Diagram 3] FIG. 3 is a diagram showing the mounting positions of a first sensor (a vibration acceleration meter) and a second sensor (a magnetostrictive displacement meter) in the PQ monitoring cart according to the embodiment. [Figure 4] FIG. 4 is a diagram showing the mounting position of a third sensor (non-contact displacement meter) in the PQ monitoring cart according to the embodiment. [Diagram 5] FIG. 5 is a system block diagram showing the configuration of a commercial vehicle that implements the wheel load estimation method according to the embodiment. [Figure 6] FIG. 6(a) is a graph showing the results of estimating wheel load in a section where a commercial vehicle was travelling, using the wheel load estimation method according to the embodiment, and FIG. 6(b) is a graph showing the results of estimating wheel load in the same section, using a conventional wheel load estimation method that does not use the first sensor. [Figure 7A] FIG. 7A is a graph showing the results of estimating wheel load in a section in which a commercial vehicle traveled, using a wheel load estimation method according to an embodiment including a lateral force term. [Figure 7B] FIG. 7B is a graph showing quasi-static wheel load changes obtained by removing the value based on the inertial force term from the data in FIG. 7A. [Figure 8A] FIG. 8A is a graph showing the results of estimating wheel load in a section in which a commercial vehicle traveled, using a wheel load estimation method according to an embodiment that does not include a lateral force term. [Figure 8B] FIG. 8B is a graph showing quasi-static wheel load changes obtained by removing the value based on the inertial force term from the data in FIG. 8A. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0030] Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings. Note that the following embodiment is merely an example of the present invention, and is not intended to limit the scope of the present invention, its applications, or its uses.
[0031] The wheel load estimation method of this embodiment uses a PQ monitoring bogie equipped with a first sensor, a second sensor, and a third sensor. This embodiment is a highly accurate wheel load estimation method that aims to be able to acquire track data continuously and over the long term, as well as to follow momentary wheel load fluctuations by performing calculations that take into account the vertical acceleration at the wheel positions on the inner rail side and the outer rail side.
[0032] [Explanation of each parameter] The PQ monitoring cart 10 used to implement the wheel load estimation method of this embodiment is equipped with a left wheel 11 and a right wheel 12, as shown in Fig. 1. The PQ monitoring cart 10 is a cart that can be run as a commercial vehicle. The track consisting of a left rail 13 and a right rail 14 is a curved section, and a cant is provided so that the left wheel 11 side is higher and the right wheel 12 side is lower.
[0033] In the example shown in Fig. 1, the left wheel 11 is the outer rail side wheel, and the right wheel 12 is the inner rail side wheel. There are cases where the cant is provided so that the left wheel 11 side is lower and the right wheel 12 side is higher, so in this specification, the terms outer rail side wheel and inner rail side wheel are used as necessary. In the example shown in Fig. 1, the left wheel 11 is the outer rail side wheel, and the right wheel 12 is the inner rail side wheel. The PQ monitoring cart 10 is equipped with axle springs 15, 16 that support the left wheel 11 and right wheel 12, respectively.
[0034] Hereinafter, the meanings of the parameters used in the estimation equation used in the wheel load estimation method of this embodiment will be described. First, the parameters used in this embodiment will be listed.
[0035] P l :Estimated wheel load of the left wheel (estimated value) P r : Estimated wheel load of the right wheel (estimated value) Z r : Vertical displacement of the right shaft spring measured by a magnetostrictive displacement meter (measured value) Z l : Vertical displacement of the left axis spring measured by a magnetostrictive displacement meter (measured value) Q r : Lateral force on the right wheel (measured value) Q l : Lateral force on the left wheel (measured value) a sr : Output value (measured value) from the right axle box vibration accelerometer a sl : Output value (measured value) from the left axle box vibration accelerometer b l : Distance from the center of the wheelset to the contact point between the left wheel and the rail (approximate value) b r : Distance from the center of the wheelset to the contact point between the right wheel and the rail (approximate value)
[0036] where b l and b r Regarding this, an example of approximation using a linear model is shown in the following equations (7) and (8).
number
number
[0037] b0: Distance from the center of the wheelset to the left and right wheel-rail contact points when the wheelset is in neutral on a straight track with a constant gauge and no track irregularity (design value) y: Amount of left and right movement of the axle (approximate value) δ: Flange gap size (design value) s: Slack size (design value)
[0038] Here, y is defined as in the following formula (9) depending on whether the section is a straight line section, a left curve section, or a right curve section.
number
[0039] m: Mass of the wheel set (design value) g: gravitational acceleration (known) b z : Half the distance between the shaft and spring (design value) k: Upper and lower spring constants of the axle springs (design value) b s : Distance from the center of the wheel set to the installation position of the left and right axle box acceleration sensors (design value) r: Wheel diameter (design value) a zl : Vertical acceleration at the left wheel position a zr : Vertical acceleration at the right wheel position
[0040] The meanings of the above parameters will be further explained with reference to FIG.
[0041] The estimated wheel load of the left wheel 11 is P l The estimated wheel load of the right wheel 12 is P r Let us assume that l and P r is the final estimated value obtained by the calculation according to the wheel load estimation method of this embodiment, and is the target value of the present invention.
[0042] The vertical displacement of the right shaft spring 16 is Z r The vertical displacement of the left shaft spring 15 is Z l Let us assume that Z r and Z l is a value measured by the magnetostrictive displacement meter, and is a measured value. The magnetostrictive displacement meter corresponds to the second sensor.
[0043] The lateral force acting on the right wheel 12 is Q r The lateral force acting on the left wheel 11 is Q l Let us assume that. Q r and Q l is a value measured by the non-contact displacement meter, which corresponds to the third sensor.
[0044] The output value of the vibration accelerometer for the left axle box is a sl The output value of the vibration accelerometer for the right axle box is asr Let us assume that sl and a sr is a value measured by the vibration accelerometer, which corresponds to the first sensor.
[0045] The vertical acceleration at the position of the left wheel 11 is a zl The vertical acceleration at the position of the right wheel 12 is a zr Let us assume that zl and a zr is a value that can be calculated by inputting the output value from the vibration accelerometer and performing the first step of calculation described below.
[0046] The distance from the center of the wheel set to the installation position of the vibration accelerometers on the left and right axle boxes is b s In this embodiment, b s Use the design value.
[0047] The distance from the center of the wheel set to the contact point between the left wheel 11 and the rail is b l The distance from the center of the wheel set to the contact point between the right wheel 12 and the rail is b r In this embodiment, b l and b r The approximate value is used. To calculate the approximate value, a linear model is used using the left and right movement amount y of the wheelset, the flange clearance δ, and the slack amount s, as described above.
[0048] The weight of the wheel set is m. In this embodiment, m is a design value.
[0049] The acceleration due to gravity is g. g is a known value, specifically, 9.80665 m / s 2 It is.
[0050] Half the distance between the axle springs 15 and 16 is b z Let us assume that b z Use the design value.
[0051] The upper and lower spring constants of the shaft springs 15 and 16 are denoted by k. In this embodiment, k is a design value.
[0052] The wheel diameter is r. In this embodiment, r is a design value. However, r is changed every time the wheel is turned.
[0053] [Explanation of the estimation formula] Next, an example of an estimation formula for the wheel load estimation method of this embodiment will be described. Fig. 2 shows, as a block diagram, the relationship between the parameters input to function A and the parameters output by function A, and the relationship between the parameters input to function B and the parameters output by function B. In Fig. 2, the calculation process performed by function A corresponds to the first step. The calculation process performed by function B corresponds to the second and third steps.
[0054] In the first step, the function A expressed by the following equations (5) and (6) is calculated. Specifically, in the first step, the output value a sl and the output value a from the vibration accelerometer of the right axle box sr and the distance b from the center of the wheel set to the installation position of the left and right axle box acceleration sensors s and the distance b from the center of the wheel set to the contact point between the left wheel 11 and the rail 13. l and the distance b from the center of the wheel set to the contact point between the right wheel 12 and the rail 14. r The input values are as follows. Then, the calculations shown in the following expressions (5) and (6) are performed.
number
number
[0055] That is, in the first step, the output value of the first sensor on the inner rail side and the output value of the outer rail side are multiplied by a coefficient based on the left and right movement amount of the wheel set and then added to obtain the vertical acceleration at the wheel position on the inner rail side and the outer rail side. The output value obtained as a result of the calculation in the first step is the vertical acceleration a zl and the vertical acceleration a at the position of the right wheel 12 zr It is.
[0056] In the second and third steps, the function B expressed by the following formulas (3) and (4) is calculated. Specifically, in the second step, the vertical displacement Z of the right shaft spring 16 measured by the magnetostrictive displacement meter is calculated. r and the vertical displacement Z of the left shaft spring 15 measured by a magnetostrictive displacement meter. l and the vertical acceleration a at the position of the left wheel 11 zl and the vertical acceleration a at the position of the right wheel 12 zr and the distance b from the center of the wheel set to the contact point between the left wheel 11 and the rail 13. l and the distance b from the center of the wheel set to the contact point between the right wheel 12 and the rail 14. r The mass m of the wheel set, the gravitational acceleration g, and half the distance b between the axle springs 15 and 16. z The input values are the vertical spring constant k of the axle springs 15 and 16, and the wheel diameter r. zl and the vertical acceleration at the right wheel position a zr is the output value of the first step. In the third step, the lateral force Q applied to the right wheel 12 is r and the lateral force Q acting on the left wheel 11 l and the distance b from the center of the wheel set to the contact point between the left wheel 11 and the rail 13. l and the distance b from the center of the wheel set to the contact point between the right wheel 12 and the rail 14. r and the wheel diameter r are input values. In the second and third steps, the calculations shown in the following equations (3) and (4) are performed.
number
number
[0057] The above estimation formula calculates, from left to right, the "gravity term" + "right axle spring force term" + "left axle spring force term" + "inertia force term" + "lateral force term."
[0058] That is, in the second step, the "gravity term" is calculated by multiplying the weight of the axle, the acceleration of gravity, and a coefficient based on the left-right movement of the axle, the "axle spring force term" is calculated by multiplying the output value of the inner rail side and the output value of the outer rail side by a coefficient based on the left-right movement of the axle and the upper and lower spring constants of the axle springs and adding them together, and the "inertia force term" is calculated by multiplying 1 / 2 the mass of the axle and the output value of the first step. Furthermore, in the third step, the wheel loads of the inner rail side and the outer rail side are estimated by adding or subtracting the "lateral force term" calculated by subtracting the output value of the inner rail side from the output value of the outer rail side by the third sensor and multiplying it by a coefficient based on the left-right movement of the axle. The output values obtained in the second and third steps are the estimated wheel load P l and the estimated wheel load of the right wheel P r It is.
[0059] [Sensor description] The PQ monitoring bogie 10, which implements the wheel load estimation method of this embodiment, has three types of sensors attached to both the inner rail side and the outer rail side. Specifically, as shown in FIG. 3, the PQ monitoring bogie 10 is attached with a vibration accelerometer 1 that measures the acceleration of vibration in the vertical direction for the axle box on the right wheel 12 side and the axle box on the left wheel 11 side. The vibration accelerometer 1 is attached to the base with a cable tie 1a. The vibration accelerometer 1 corresponds to a first sensor that measures the acceleration of vibration in the vertical direction of the axle box on the inner rail side and the outer rail side. In addition, the PQ monitoring bogie 10 is attached with a magnetostrictive displacement meter 2 that measures the deflection amount of each axle spring on the right wheel 12 side and the left wheel 11 side. The magnetostrictive displacement meter 2 corresponds to a second sensor that measures the deflection amount of the axle spring on the inner rail side and the outer rail side.
[0060] 4, non-contact displacement gauges 3a, 3b, and 3c are attached to the PQ monitoring cart 10 to measure the amount of horizontal deformation caused by bending of each wheel plate portion on the right wheel 12 side and the left wheel 11 side. Specifically, the non-contact displacement gauge 3a is disposed via a first displacement gauge mounting seat 19 on an axle box 18 that rotatably supports the wheel set via a bearing 17.
[0061] The non-contact displacement meter 3a measures the amount of deformation of the rim portion 20a or plate portion 20b of the wheel 20 while the commercial vehicle is running. The non-contact displacement meter 3b is attached to the axle box 18 via a second displacement meter mounting seat 21, measures the displacement of the boss portion 20c of the wheel 20, and calculates the amount of deformation of the rim portion 20a or plate portion 20b of the wheel 20, which corresponds to the lateral pressure, from the difference. The non-contact displacement meter 3c is placed symmetrically to the non-contact displacement meter 3b with respect to the axle, and corrects the inclination of the axle box 18. These non-contact displacement meter 3a, 3b, 3c correspond to a third sensor that measures the amount of horizontal deformation due to bending of the wheel plate portion on the inner rail side and the outer rail side.
[0062] Generally, to detect wheel deformation due to lateral force, the detected displacement must have an accuracy of at least 0.005 mm. This accuracy is possible with a displacement gauge with a resolution of 0.002 mm, but requires correction taking into account errors and noise caused by sensor installation, etc. In this embodiment, a known correction method, for example that disclosed in Patent Document 1 mentioned above, is used. Specifically, this embodiment uses the three non-contact displacement gauges 3a, 3b, 3c mentioned above to perform correction taking into account the relative displacement between the axle box and the wheel, thereby improving the accuracy of the lateral force estimate. In this embodiment, first, the amount δ of movement of the wheelset in the thrust direction relative to the axle box due to thrust clearance of the bearing, etc. is calculated. s Next, calculate the amount of inclination δ of the axle box relative to the wheel set caused by the lateral force acting on the bogie frame. T After calculating the above, the corrected plate deformation amount δ is calculated. Specifically, the corrected plate deformation amount δ is expressed by the following equation using the displacement amounts detected by the non-contact displacement meters 3a, 3b, and 3c.
[0063] δ s =(δ1+δ2) / 2 δ T = B tan θ=(B / A)(δ2-δ1) δ = δ3 - (δ s +δ T ) =δ3- (δ1+δ2) / 2-(B / A)(δ2-δ1) In the above formula, δ1 is the displacement of the upper part of the wheel hub, δ2 is the displacement of the lower part of the wheel hub, δ3 is the displacement of the wheel plate, A is the distance between the upper and lower sensors installed on the boss, B is the distance from the center of the axle to the sensors installed on the wheel plate, and θ is the relative inclination angle between the axle box and the wheel (see Figure 2 of Patent Document 1).
[0064] [Explanation of commercial vehicles and PQ data wireless transmission equipment] As shown in Fig. 5, the commercial vehicle used in this embodiment includes a vehicle body 50, an underfloor 60, and a bogie 70. A PQ data wireless transmission device 80 including an information processing device 81, a ground-to-vehicle communication function device 82, and an antenna 83 is installed in the vehicle body 50. The PQ data wireless transmission device 80 is a device capable of performing data communication with devices outside the commercial vehicle by using wireless communication.
[0065] The bogie 70 uses the PQ monitoring bogie 10. The bogie 70 is equipped with at least an axle box vertical vibration signal receiver 71 that receives an output signal from the vibration accelerometer 1, an axle spring vertical displacement signal receiver 72 that receives an output signal from the magnetostrictive displacement gauge 2, and a wheel displacement signal receiver 73 that receives output signals from the non-contact displacement gauges 3a, 3b, and 3c. The bogie 70 may further include a monolink load signal receiver.
[0066] A PQ monitor controller 61 is installed under the floor 60 and is capable of receiving at least the speed signal 51 and train information 52 transmitted from the car body 50. The train information 52 includes the vehicle number identifying the commercial vehicle, as well as position data and time data of the commercial vehicle. The PQ monitor controller 61 and the information processing device 81 receive power from a DC 100V power supply 53 provided in the car body 50.
[0067] The PQ monitor controller 61 is connected to each of the axle box vertical vibration signal receiver 71, the axle spring vertical displacement signal receiver 72, and the wheel displacement signal receiver 73 via a connector base 62 and an amplifier unit 63. The amplifier unit 63 performs processing such as signal amplification. The data is then sent to an information processing device 81 for arithmetic processing. The PQ monitor controller 61 corresponds to a control unit that transmits to the information processing device 81 data required to implement the wheel load estimation method of this embodiment.
[0068] In this embodiment, an information processing device 81 is provided in the vehicle body 50. The information processing device 81 includes a central processing unit (CPU), a main memory, and a secondary storage device (HDD, SSD, etc.), and can load an application program recorded in the secondary storage device into the main memory and perform required calculations in the central processing unit. The information processing device 81 executes a first step of multiplying the output value of the inner rail side and the output value of the outer rail side by a coefficient based on the left-right movement amount of the wheel set and adding them together to obtain the vertical acceleration at the wheel position for each of the inner rail side and the outer rail side, and a second step of estimating the wheel load for each of the inner rail side and the outer rail side based on a "gravity term" obtained by multiplying the weight of the wheel set, the gravitational acceleration, and a coefficient based on the left-right movement amount of the wheel set, an "axle spring force term" obtained by multiplying the output value of the inner rail side and the output value of the outer rail side by a coefficient based on the left-right movement amount of the wheel set and the up-down spring constant of the axle spring and adding them together, and an "inertia force term" obtained by multiplying the mass of the wheel set and 1 / 2 the output value of the first step. The commercial vehicle of this embodiment is a commercial vehicle capable of measuring the state of the track.
[0069] [Driving test 1] In order to confirm the effect of the wheel load estimation method of this embodiment, a running test 1 was carried out by running a commercial vehicle equipped with the PQ monitoring bogie of the above configuration on a subway commercial line. In the running test 1, a conventional wheel load estimation method using only the output value of a magnetostrictive displacement meter without using the output value of a vibration accelerometer was also carried out at the same time. The running test 1 was carried out during normal business hours in the section between Kita-Senju Station and Machiya Station on the Tokyo Metro Chiyoda Line.
[0070] The results of the wheel load estimation method of this embodiment are shown in Fig. 6(a), and the results of the conventional estimation method are shown in Fig. 6(b). In each graph, the vertical axis represents the estimated wheel load P l The horizontal axis indicates the mileage of commercial vehicles.
[0071] Comparing the two graphs, it was confirmed that the wheel load estimation method of this embodiment accurately observed the vibrational wheel load in the curved sections at around 3.5 km, 4 km before, around 4.3 km, and around 5.3 km, compared to the conventional method. In addition, the impact at the point where the turnout was located at around 3.75 km and the impact at the joint of the fixed-length rail at 4.5 to 5 km were also detected with high accuracy when using the wheel load estimation method of this embodiment. Note that, although FIG. 6 shows only the wheel load of the left wheel, it was confirmed that the same result was obtained for the wheel load of the right wheel.
[0072] [First Modification of the Embodiment] In the wheel load estimation method of this modified example, in the block diagram of FIG. 2, the calculation process performed by function A is the same as the first step using the above-mentioned equations (5) and (6), but the calculation process performed by function B does not take into account the lateral force term, and only the second step using the following equations (1) and (2) is carried out.
[0073] Specifically, in the second step of this modified example, the vertical displacement Z of the right shaft spring 16 measured by the magnetostrictive displacement meter is r and the vertical displacement Z of the left shaft spring 15 measured by a magnetostrictive displacement meter. l and the vertical acceleration a at the position of the left wheel 11 zl and the vertical acceleration a at the position of the right wheel 12 zr and the distance b from the center of the wheel set to the contact point between the left wheel 11 and the rail 13. l and the distance b from the center of the wheel set to the contact point between the right wheel 12 and the rail 14. r The mass m of the wheel set, the gravitational acceleration g, and half the distance b between the axle springs 15 and 16. zThe input values are the vertical spring constant k of the axle springs 15 and 16, and the wheel diameter r. zl and the vertical acceleration at the right wheel position a zr is the output value of the first step.
number
number
[0074] The above estimation formula calculates, from the left, the "gravity term" + "right axle spring force term" + "left axle spring force term" + "inertia force term."
[0075] In other words, only the second step is performed, and the following are added: a "gravity term" obtained by multiplying the weight of the axle, the acceleration of gravity, and a coefficient based on the left and right movement of the axle; an "axle spring force term" obtained by multiplying the output value of the inner rail side and the output value of the outer rail side by a coefficient based on the left and right movement of the axle and the upper and lower spring constants of the axle springs, respectively, from the second sensor; and an "inertia force term" obtained by multiplying 1 / 2 the mass of the axle and the output value of the first step. In this modified example, the output value obtained by the second step is the estimated wheel load P l and the estimated wheel load of the right wheel P r It is.
[0076] [Driving test 2] In order to compare the difference in effectiveness between the wheel load estimation method of the above-mentioned embodiment, which performs the calculations of formulas (3) and (4) that also take into account the lateral force term, and the wheel load estimation method of this modified example, which does not take the lateral force term into account, running test 2 was carried out by running a commercial vehicle equipped with a PQ monitoring bogie on a commercial subway line. Running test 2 was carried out during normal business hours in the section between Sendagi Station and Nezu Station on the Tokyo Metro Chiyoda Line.
[0077] Fig. 7A shows the wheel load change according to the method of the above-mentioned embodiment, which performs calculations taking into account the lateral force term, and Fig. 7B shows the quasi-static wheel load change obtained by removing the value based on the inertial force term (first step) from that data. Fig. 8A shows the wheel load change according to the method of this modified example, which does not consider the lateral force term, and Fig. 8B shows the quasi-static wheel load change obtained by removing the value based on the inertial force term (first step) from that data. In each graph, the vertical axis represents the estimated wheel load of the right wheel at the front of the vehicle in the traveling direction, P r The horizontal axis indicates the mileage of commercial vehicles.
[0078] Comparing the graph in Fig. 7A with the graph in Fig. 8A, which shows the results of this modified example without considering the lateral force term, larger values are detected in the curve in the section of the traveled distance of 8.8 to 8.95 km in the graph in Fig. 7A. The same tendency can be confirmed by comparing the graphs in Fig. 7B and Fig. 8B, which exclude the inertial force term (first step). It can be said that this difference is the effect of performing calculations that take the lateral force term into account.
[0079] Therefore, in the present invention, if it is desired to further improve the accuracy of the estimated wheel load, it is more preferable to provide a third sensor (as an example, non-contact displacement meters 3a, 3b, 3c) that measures the amount of horizontal deformation due to bending of the wheel plate portion on the inner rail side and the outer rail side, and to include a third step of subtracting the output value of the inner rail side from the output value of the outer rail side by the third sensor and multiplying the result by a coefficient based on the lateral force acting on the wheel set, and then adding or subtracting the lateral force term obtained.
[0080] However, in the present invention, the effect of performing a calculation that takes into account the output value of the first sensor (the vibration accelerometer 1 as an example) is significantly large, and a comparison of the results of Figs. 7A and 7B shows that the impacts caused by the gaps (joints) at the connections between the rails and the unevenness caused by wear on the rail surface are estimated with high accuracy, whereas the effect of performing a calculation that further takes into account the lateral force term is relatively small in the overall effect.
[0081] Therefore, the wheel load estimation method of this modified example, which does not take into account the lateral force term, also has the advantage of being able to reduce equipment costs and maintenance efforts while maintaining a certain level of accuracy.
[0082] [Second Modification of the Embodiment] In the above embodiment, when the third step is carried out to perform a calculation including the lateral force term, the output data from the non-contact displacement gauges 3a, 3b, and 3c also includes data on displacement caused by "surface wobble" in which the wheels and axles are not fitted at right angles to each other and wobble occurs. Conventionally, surface wobble was dealt with by averaging, but there was a problem that some data was lost in the averaging process, making it difficult to measure instantaneous lateral force.
[0083] Therefore, in this modified example, a notch filter is applied to reduce the signal caused by the wheelset surface wobble, thereby enabling highly accurate estimation of the lateral force. That is, in the wheel load estimation method of this modified example, the third step includes a process of applying a notch filter that reduces the signal caused by the wheelset surface wobble when calculating the lateral force term.
[0084] Specifically, when determining the lateral force term based on the output data from the non-contact displacement gauges 3a, 3b, and 3c, the amplitude of the component believed to be caused by surface wobble can be effectively reduced by converting the data into data relative to the travel distance and then applying the notch filter shown in the following equation 10. Furthermore, unlike conventional averaging processing, the method using this notch filter makes it easy to detect instantaneous increases in lateral force.
number
[0085] [Third Modification of the Embodiment] It is important to improve the running safety of railway vehicles to define indicators for derailment of railway vehicles, evaluate the indicators, and take measures according to the evaluation results. It is known that the derailment coefficient (Q / P), which is the ratio of the wheel load P, which is the vertical load acting between the wheels of a railway vehicle and the rail, to the lateral force Q, which is the horizontal load acting between the wheels of a railway vehicle and the rail, varies greatly depending on the track irregularity, the support condition of the rail, and the contact condition between the wheels and the rail.
[0086] Therefore, it is believed possible to estimate the condition of the rails by determining the derailment coefficient calculated from the estimated values of wheel load P and lateral force Q. Meanwhile, in the past, track irregularity values were measured using rail inspection cars and other instruments to measure the condition of the rails, but such measurements can only be made intermittently and are unable to capture short-term changes.
[0087] Therefore, in the track monitoring method of this modified example, in order to grasp changes in the track condition in a timely manner, the wheel load estimation method of the above-mentioned embodiment is used to estimate the wheel load P with higher accuracy than before, and the derailment coefficient (Q / P) on operating lines is constantly observed.
[0088] In this modified example, for example, a commercial vehicle equipped with the PQ monitoring bogie 10 is driven, and the inner rail side lateral force and the outer rail side lateral force (Q r , Q l ) are calculated, and these values are used to calculate the inner rail side wheel load and the outer rail side wheel load (P r、 P l ) to obtain the derailment coefficients (Q r / P r , Q l / P l ) and monitors changes in the derailment coefficient for the section in which commercial vehicles are running.
[0089] In order to make it easier to visually grasp the results of the track monitoring method according to this modified example, a graph may be created with the running distance of the commercial vehicle on the horizontal axis and the derailment coefficient on the vertical axis. Note that since the derailment coefficient is generally calculated for the outer rail wheels of the front axle in the traveling direction of a specific bogie when the rail vehicle runs on a curved section of the track, the outer rail derailment coefficient may be calculated in the same manner in this modified example.
[0090] The track monitoring method of this modification may be less accurate than a method that uses a rail inspection car or the like to measure track irregularity values, but it fully meets the required accuracy for the purpose of early detection of abnormal values on commercial lines. Also, compared to the conventional monitoring method using a PQ monitoring bogie, it has the advantage of being able to detect sudden wheel load P with high accuracy, so track abnormalities such as rail joint drop, rail breakage, and rail peeling can be identified early and accurately.
[0091] [Other variations] In the above-mentioned embodiment, the information processing device 81 may be located outside the business vehicle. In this case, while the business vehicle is traveling, the output data of the PQ monitor controller 61 is only recorded in a secondary storage device, and the recorded data is read later using an external information processing device to perform the calculations of the above-mentioned first, second, and third steps. [Industrial Applicability]
[0092] The present invention can be applied to systems that constantly monitor changes in wheel load and derailment coefficient using commercial vehicles, confirming running safety in real time while detecting changes in track conditions and abnormal values at an early stage. [Explanation of symbols]
[0093] 1. Vibration accelerometer (first sensor) 2 Magnetostrictive displacement gauge (second sensor) 3a,3b,3c Non-contact displacement meter (3rd sensor) 11 Left wheel 12 Right Wheel 13 Left rail 14 Right rail 15 Left shaft spring 16 Right shaft spring 17 Bearings 18 Axle box 50 Body 51 Speed signal 52 Train information 60 Underfloor 61 PQ Monitor Controller 62 Connector stand 63 Amplifier section 70 Carts 71 Axle box vertical vibration signal receiver 72 Vertical displacement signal receiver 73 Wheel displacement signal receiver 80 PQ data wireless transmission device 81 Information processing equipment
Claims
1. A method for estimating a wheel load of a commercial vehicle while the vehicle is traveling, the method comprising: running a commercial vehicle equipped with a first sensor that measures the acceleration of vibration in a vertical direction of an axle box on an inner rail side and an outer rail side; and a second sensor that measures the deflection amount of an axle spring on an inner rail side and an outer rail side; a first step of multiplying an output value of the inner rail side and an output value of the outer rail side by a required coefficient and adding them together to obtain vertical accelerations at the wheel positions for the inner rail side and the outer rail side, respectively; a second step of estimating the wheel load for each of the inner rail side and the outer rail side based on a gravity term obtained by multiplying the weight of the axle, the gravitational acceleration, and a required coefficient, an axle spring force term obtained by multiplying the output value of the inner rail side and the output value of the outer rail side by the second sensor, each multiplied by a required coefficient, and then adding them together, and an inertia force term obtained by multiplying 1 / 2 the mass of the axle and the output value of the first step.
2. A third sensor is further provided for measuring the amount of horizontal deformation caused by bending of the wheel plate portion on the inner rail side and the outer rail side, 2. The wheel load estimation method according to claim 1, further comprising, together with the second step, a third step of estimating the wheel loads for each of the inner rail side and the outer rail side by adding or subtracting a lateral force term obtained by subtracting an output value of the inner rail side from an output value of the outer rail side by the third sensor and multiplying the result by a required coefficient.
3. 2. The wheel load estimation method according to claim 1, wherein the second step includes at least a step of performing calculations expressed by the following equations (1) and (2). [0010] [0025] (where: P l : Estimated wheel load of outer wheel P r : Estimated wheel load of inner rail wheels Z r : Vertical displacement of the inner rail axle spring measured with a magnetostrictive displacement gauge Z l : Vertical displacement of the outer axle spring measured with a magnetostrictive displacement gauge a zl :Vertical acceleration at the outer wheel position a zr :Vertical acceleration at the inner wheel position b l : Distance from the center of the wheel set to the contact point between the outer wheel and the rail b r : Distance from the center of the wheelset to the contact point between the inner wheel and the rail m: Mass of the axle b z : Half the distance between the shaft and spring k: Upper and lower spring constant of the axle spring It is.)
4. 3. The wheel load estimation method according to claim 2, wherein the second step and the third step include at least a step of performing calculations expressed by the following equations (3) and (4). [0030] [0045] (where: P l : Estimated wheel load of outer wheel P r : Estimated wheel load of inner rail wheels Z r : Vertical displacement of the inner rail axle spring measured with a magnetostrictive displacement gauge Z l : Vertical displacement of the outer axle spring measured with a magnetostrictive displacement gauge Q r : Lateral force acting on the inner wheel Q l : Lateral force acting on outer wheel a zl :Vertical acceleration at the outer wheel position a zr :Vertical acceleration at the inner wheel position b l : Distance from the center of the wheel set to the contact point between the outer wheel and the rail b r : Distance from the center of the wheelset to the contact point between the inner wheel and the rail m: Mass of the axle g: gravitational acceleration b z : Half the distance between the shaft and spring k: Upper and lower spring constant of the axle spring r: Wheel diameter It is.)
5. 4. The wheel load estimation method according to claim 3, wherein the first step includes at least a step of performing calculations expressed by the following equations (5) and (6). [0050] [006] (where: a zl :Vertical acceleration at the outer wheel position a zr :Vertical acceleration at the inner wheel position a sl :Outside wheel axle box acceleration sensor output value a sr :Output value from inner rail axle box acceleration sensor b s : Distance from the center of the wheel set to the location of the inner and outer axle box acceleration sensors b l : Distance from the center of the wheel set to the contact point between the outer wheel and the rail b r : Distance from the center of the wheelset to the contact point between the inner wheel and the rail It is.)
6. 5. The wheel load estimation method according to claim 4, wherein the first step includes at least a step of performing calculations expressed by the following equations (5) and (6). [0050] [006] (where: a zl :Vertical acceleration at the outer wheel position a zr :Vertical acceleration at the inner wheel position a sl :Outside wheel axle box acceleration sensor output value a sr :Output value from inner rail axle box acceleration sensor b s : Distance from the center of the wheel set to the location of the inner and outer axle box acceleration sensors b l : Distance from the center of the wheel set to the contact point between the outer wheel and the rail b r : Distance from the center of the wheelset to the contact point between the inner wheel and the rail It is.)
7. 7. The method for estimating a wheel load according to claim 2, wherein the third step includes a process of applying a notch filter to reduce a signal caused by a wheelset surface wobble when determining the lateral force term.
8. It is a commercial vehicle, a bogie having an axle box vertical vibration signal receiving unit that receives an output signal of the first sensor, an axle spring vertical displacement signal receiving unit that receives an output signal of the second sensor, and a wheel displacement signal receiving unit that receives an output signal of the third sensor; The vehicle includes a vehicle body and an underfloor. A control unit is provided under the floor of a commercial vehicle which receives a speed signal and train information transmitted from the car body unit, is connected to the axle box up and down vibration signal receiving unit, the axle spring up and down displacement signal receiving unit, and the wheel displacement signal receiving unit via an amplifier unit, and transmits data enabling the wheel load estimation method according to any one of claims 1 to 6 to an information processing device.
9. A commercial vehicle capable of measuring track conditions, The vehicle body section is provided with the information processing device, The information processing device includes: a first step of multiplying an output value of the inner rail side and an output value of the outer rail side by a required coefficient and adding them together to obtain vertical accelerations at the wheel positions for the inner rail side and the outer rail side, respectively; 9. A commercial vehicle as claimed in claim 8, further comprising: a second step of estimating the wheel loads on the inner rail side and the outer rail side based on a gravity term obtained by multiplying the weight of the axle, the gravitational acceleration, and a required coefficient; an axle spring force term obtained by multiplying the output value of the inner rail side and the output value of the outer rail side by the second sensor, each multiplied by a required coefficient, and then adding them together; and an inertia force term obtained by multiplying 1 / 2 the mass of the axle and the output value of the first step.
10. A track monitoring method comprising the steps of: running the commercial vehicle while determining the inner rail side lateral force and the outer rail side lateral force based on the output value of the third sensor; dividing these values by the inner rail side wheel load and the outer rail side wheel load determined by the wheel load estimation method of any one of claims 2, 4 or 6 to calculate derailment coefficients for the inner rail side and the outer rail side; and monitoring changes in the derailment coefficients in the section in which the commercial vehicle is running.
Citation Information
Patent Citations
Horizontal pressure measurement method and truck for railway vehicle
JP2006088967A
Railroad state-measuring method, and business vehicle capable of measuring railroad state
JP2015051674A
Cited By
Information processing device, information processing method, and program
JP7948733B1
