Lateral pressure measuring apparatus and method for measuring lateral pressure

Abstract

To provide a lateral pressure measuring apparatus and a method for measuring lateral pressure capable of improving the measurement accuracy of lateral pressure by reducing the influence of hysteresis caused by frictional force on the output of a bridge circuit connected to a strain gauge.SOLUTION: A lateral pressure measuring apparatus 10 includes: (A) a plurality of bridge circuits including strain gauges (a plate-surface strain gauge 60 and a hole strain gauge 70) for measuring strain of a wheel 40; (B) a coupling coefficient determination unit 120 configured to determine linear combination coefficients for outputs from the plurality of bridge circuits, the outputs including hysteresis that differs between an increase and a decrease in lateral pressure; and (C) a lateral pressure estimation unit 130 configured to calculate the estimated value of a lateral pressure Q by computing the weighted sum of the outputs from the plurality of bridge circuits using the linear combination coefficients determined by the coupling coefficient determination unit 120, (D) the plurality of bridge circuits being combined in such a manner that the lateral pressure Q and a frictional force can be separated.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a lateral force measuring device and a lateral force measuring method for measuring lateral forces acting on wheels of a railway vehicle. [Background technology]

[0002] In recent years, efforts have been made to increase the speed of Shinkansen trains, but at the same time, there is a demand for braking distances to be comparable to those of conventional trains.To improve deceleration during braking, therefore, centrally fastened brake discs have been developed, which are less susceptible to thermal deformation of the brake disc than conventional inner-circumferentially fastened brake discs (see Non-Patent Document 1).

[0003] Meanwhile, a PQ measurement method is known for evaluating the running safety of railway vehicles. This method measures the contact force between the wheel and the rail using a special wheelset called a PQ wheelset, which has a large number of strain gauges attached to the wheel. When this PQ measurement method is applied to a centrally fastened brake disc, it is said that, particularly in measuring lateral force, the amount of strain cannot be properly converted to lateral force due to the effect of so-called hysteresis, which generates a difference in the amount of strain measured when loaded and unloaded (see Non-Patent Document 2). Furthermore, Non-Patent Document 2 proposes a method for measuring shear strain at the wheel rim and converting it into lateral force, in order to solve the problem of not being able to properly convert the amount of strain into lateral force.

[0004] However, Non-Patent Document 3 shows that the method proposed in Non-Patent Document 2 can be strongly affected by the wheel load action depending on the contact position in the lateral direction, and the accuracy of measuring lateral force can be significantly reduced. [Prior art documents] [Patent documents]

[0005] [Non-Patent Document 1] Arai, Kato, and Asano. "Development of a foundation brake device for high-speed operation," JR East Technical Review No. 22, pp. 7-10, 2008 [Non-patent document 2] Ishida, Endo. "Lateral Force Measurement Method for Wheels with Centrally Fastened Brake Discs," Meisei University Faculty of Science and Engineering Research Bulletin, Vol. 53, pp. 39-46, 2017 [Non-patent document 3] Hondo, et al.: "Lateral force measurement method using shear strain at wheel load measurement points of PQ wheelset," Proceedings of the Transportation and Logistics Conference, DOI: 10.1299 / jsmetld.2020.29.1201, 2020 Summary of the Invention [Problem to be solved by the invention]

[0006] The present invention has been made in consideration of the above circumstances, and aims to provide a lateral force measuring device and a lateral force measuring method that can improve the accuracy of lateral force measurement by reducing the effect of hysteresis caused by frictional force on the output of a bridge circuit connected to a strain gauge in a PQ measurement method using a wheel equipped with a centrally fastened brake disc. [Means for solving the problem]

[0007] [1] In order to solve the above-mentioned problems, according to a first aspect of the present invention, there is provided a lateral force measuring device for performing PQ measurements on a wheel equipped with a centrally fastened brake disc, the device comprising: (A) a plurality of bridge circuits including strain gauges attached to the wheel to measure strain on the wheel; (B) a coupling coefficient determination unit that determines a linear coupling coefficient of outputs from each of the plurality of bridge circuits, the outputs including hysteresis that differs when the lateral force increases and when the lateral force decreases; and (C) a lateral force estimation unit that calculates an estimated value of lateral force by calculating a weighted sum of outputs from the plurality of bridge circuits using the linear coupling coefficient determined by the coupling coefficient determination unit; and (D) the plurality of bridge circuits are combined in such a way that lateral force and frictional force can be separated.

[0008] [2] Furthermore, in order to solve the above-mentioned problems, according to a second aspect of the present invention, there is provided a lateral force measurement method for a lateral force measurement device for performing PQ measurements on a wheel equipped with a centrally fastened brake disc, the method comprising: a determination step of determining a linear combination coefficient of outputs from multiple bridge circuits, each bridge circuit including a strain gauge that measures the strain of the wheel attached to the wheel, the outputs including hysteresis that differs when the lateral force increases and when the lateral force decreases; and an estimation step of calculating an estimated value of lateral force by calculating a weighted sum of the outputs from the multiple bridge circuits using the determined linear combination coefficient, wherein the multiple bridge circuits are combined in such a way that the lateral force and frictional force can be separated. [Effects of the Invention]

[0009] According to the present invention, it is possible to provide a lateral force measuring device and a lateral force measuring method that can improve the accuracy of lateral force measurement by reducing the effect of hysteresis caused by frictional force on the output of a bridge circuit connected to a strain gauge in a PQ measurement method using a wheel equipped with a centrally fastened brake disc. [Brief explanation of the drawings]

[0010] [Figure 1] 1 is a diagram showing a schematic configuration of a lateral force measuring device according to an embodiment of the present invention; [Figure 2] FIG. 2 is a perspective view showing a PQ wheelset used to measure a derailment coefficient in the lateral force measuring device shown in FIG. 1, and showing wheel load P and lateral force Q. [Figure 3] FIG. 3 is a plan view showing the wheel of the PQ wheelset shown in FIG. 2, and showing the strain gauges that make up the lateral force measuring device attached to the PQ wheelset, with the brake disc removed from the wheel. [Figure 4] FIG. 3 is a cross-sectional view showing the wheel of FIG. 2 cut along line II. [Figure 5]FIG. 3 is a plan view showing the shape of a brake disc attached to the PQ wheel set shown in FIG. 2. [Figure 6] FIG. 3 is a diagram showing a bridge circuit for measuring lateral force in the PQ wheelset shown in FIG. 2. [Figure 7] In the PQ wheelset shown in Fig. 2, this is a bridge circuit that corresponds to the lateral force measurement method using shear strain. [Figure 8] 10 is a graph showing loading and unloading states when analyzing the influence of hysteresis in the present embodiment. [Figure 9] 9 is a graph showing the outputs of the bridge circuits in the analysis shown in FIG. 8, where (a) shows the output of the bridge circuit Qb1 and (b) shows the output of the bridge circuit Qs1. [Figure 10] 10 is a graph showing the relationship between the loaded lateral pressure Q and the estimated lateral pressure ^Q, applying equation (13) to the same data as in the analysis results shown in FIG. 9. DETAILED DESCRIPTION OF THE INVENTION

[0011] A lateral force measuring device 10 and a lateral force measuring method according to an embodiment of the present invention will now be described with reference to the drawings.

[0012] [1. Overall configuration of the lateral force measuring device 10] 1 is a diagram showing the schematic configuration of a lateral force measuring device 10 according to this embodiment. The lateral force measuring device 10 of this embodiment comprises a PQ wheelset 20, a slip ring device 100, a strain signal processing unit 110, a coupling coefficient determination unit 120, and a lateral force estimation unit 130.

[0013] <1-1. About PQ Wheelset 20> Fig. 2 is a perspective view showing the PQ wheelset 20 used to measure the derailment coefficient of the lateral force measuring device 10 shown in Fig. 1, and showing the wheel load P and lateral force Q. Fig. 3 is a plan view showing the wheel of the PQ wheelset 20 shown in Fig. 2, with the strain gauges that make up the lateral force measuring device 10 attached to the PQ wheelset 20, and shows the state with the brake disc 80 removed from the wheel 40. Fig. 4 is a cross-sectional view showing the wheel 40 of Fig. 2 cut along line II.

[0014] Here, the PQ wheelset 20 is a wheelset for measuring the wheel load P and lateral force Q acting between the wheel 40 and the rail R using a strain gauge 60 attached to the wheel 40, and the wheelset itself can be considered as a load cell. The PQ wheelset 20 is equipped with a wheelset 30 and strain gauges (plate strain gauge 60 and hole strain gauge 70).

[0015] As shown in Fig. 1, the wheel load P is a radial force of the wheel 40 that the wheel 40 receives from the rail R, and the lateral force Q is a force in the axial direction of the axle 50 that the wheel 40 receives from the rail R. When the wheel 40 is not tilted in the up-down direction, the wheel load P is a force in the up-down direction, and the lateral force Q is a force in the left-right direction connecting the pair of rails R. The derailment coefficient is the value (Q / P) obtained by dividing the lateral force Q by the wheel load P.

[0016] As shown in Fig. 1, the wheel set 30 is configured by press-fitting the vicinity of both ends of the axle 50 onto a pair of left and right wheels 40 and then fixing them. In the configuration shown in Fig. 1, brake discs 80 are also attached to the anti-flange side (outside of the pair of rails R) and flange side (inside of the pair of rails R) of the wheel 40, respectively.

[0017] As shown in Figure 2, the wheel 40 is a wheel in which a hub portion 41, a rim portion 42, a tread surface 43, a flange 44, and a plate portion 45 are integrated together. Of these, the hub portion 41 is a portion provided at the radial center of the wheel 40, and is the portion to which the axle 50 is fixed after being press-fitted. The hub portion 41 is formed so that it is thicker in the direction of the rotation axis (the axial direction of the axle 50) than the plate portion 45, and a hub hole 41a is formed in the center thereof for the axle 50 to be press-fitted into.

[0018] The rim portion 42 is a ring-shaped portion provided on the outer periphery of the wheel 40, and constitutes a so-called tire portion that rolls in contact with the rail R. For this reason, the rim portion 42 is provided with a larger axial dimension than the plate portion 45.

[0019] The tread 43 is the portion present on the outer peripheral surface of the rim portion 42 that comes into contact with the upper surface of the rail R. The tread 43 is formed to have a predetermined tread gradient that becomes smaller toward the center in the axial direction of the axle 50, in order to allow a difference in the turning radius between the left and right wheels when passing through a curve, allowing the vehicle to pass through the curve smoothly.

[0020] The flange 44 is a portion that protrudes radially outward from the vicinity of the inner end of the tread 43. The flange 44 is a portion that limits the axial movement of the wheel 40 relative to the rail R in order to prevent the wheel 40 from coming off the rail R.

[0021] The plate portion 45 is a substantially flat portion that connects the outer peripheral surface of the hub portion 41 and the inner peripheral surface of the rim portion 42. This plate portion 45 is provided with a plurality of (12 in FIG. 2 ) bolt holes 46. The bolt holes 46 are holes for inserting bolts for fastening the brake disc 80, and are arranged on the anti-flange side (outside of the pair of rails R) and flange side (inside of the pair of rails R) of the wheel 40, respectively. The plurality of bolt holes 46 are arranged at predetermined intervals (at equal intervals) on an arrangement circumference that is located a predetermined distance from the radial center of the wheel 40.

[0022] In addition to the plurality of bolt holes 46, the plate portion 45 is also provided with a plurality of (six in FIG. 3) measurement holes 47. The plurality of measurement holes 47 are holes in which strain gauges (hole strain gauges 60) for measuring the wheel load P are placed.

[0023] In the following description, the measurement holes 47 are assigned sub-numbers in clockwise order, and the hole strain gauges 60, described below, are also assigned sub-numbers corresponding to these sub-numbers. Specifically, the measurement hole 47 located at a position (345 degrees) between 11 o'clock (330 degrees) and 12 o'clock (0 degrees) on the clock basis is referred to as measurement hole 47-1. Furthermore, the measurement hole 47 located at a position (45 degrees) between 1 o'clock (30 degrees) and 2 o'clock (60 degrees) is referred to as measurement hole 47-2. Similarly, the measurement holes 47 located at the position between 3 o'clock (90 degrees) and 4 o'clock (120 degrees) (105 degrees), the position between 5 o'clock (150 degrees) and 6 o'clock (180 degrees) (165 degrees), the position between 7 o'clock (210 degrees) and 8 o'clock (240 degrees) (225 degrees), and the position between 9 o'clock (270 degrees) and 10 o'clock (300 degrees) (285 degrees) are designated as measurement holes 47-3, 47-4, 47-5, and 47-6, respectively.

[0024] The axle 50 is a shaft-shaped member having both ends inserted into the hub hole 41a.

[0025] The numbers of the bolt holes 46 and measurement holes 47 are not limited to the above-mentioned 12 and 6, but can be changed as appropriate.

[0026] <1-2. About strain gauges> Next, the strain gauges will be described. In this embodiment, the strain gauges include a plate surface strain gauge 60 and a hole strain gauge 70. The plate surface strain gauge 60 corresponds to the first strain gauge, and the hole strain gauge 70 corresponds to the second strain gauge.

[0027] The plate surface strain gauge 60 is a strain gauge attached to a predetermined position (corresponding to the first position) of the plate portion 45, and is used to measure lateral pressure due to wheel bending. A single-axis strain gauge can be used as this plate surface strain gauge 60.

[0028] Furthermore, the hole strain gauge 70 is a strain gauge placed in the measurement hole 47. In this embodiment, the hole strain gauge 70 is a triaxial strain gauge, and of the triaxial strain gauges, two orthogonal strain gauges are used to measure shear strain and calculate lateral pressure. Of the triaxial strain gauges, the central strain gauge that is not orthogonal is used to measure wheel load and is not used to convert lateral pressure. Therefore, if there is no need to use it to measure wheel load, a biaxial strain gauge can be used as the hole strain gauge 70 instead of a triaxial strain gauge.

[0029] Here, the above-mentioned plate surface strain gauge 60 is attached on the circumference of the surface of plate portion 45 on the side opposite the flange (referred to as the plate surface) and the surface on the flange side (referred to as the plate back surface) located a predetermined distance from the radial center of wheel 40. This circumference is set to pass through a location midway between measurement hole 47 and hub hole 41a. Furthermore, plate surface strain gauge 60 is disposed at a predetermined position (first position) on a line connecting the centers of measurement hole 47 and hub hole 41a in the radial direction.

[0030] In the following description, the plate surface strain gauge 60 arranged on the line connecting the measurement hole 47-1 and the center of the hub hole 41a on the plate surface of the plate part 45 will be referred to as plate surface strain gauge 61a. Similarly, the plate surface strain gauges 60 arranged on the line connecting the measurement holes 47-2, 47-3, 47-4, 47-5, 47-6 and the center of the hub hole 41a on the plate surface of the plate part 45 will be referred to as plate surface strain gauges 62a, 63a, 64a, 65a, 66a.

[0031] Furthermore, on the rear surface of plate portion 45, plate surface strain gauge 60 arranged on the line connecting measurement hole 47-1 and the center of hub hole 41a is referred to as strain gauge 61b. Similarly, on the rear surface of plate portion 45, plate surface strain gauges 60 arranged on the lines connecting measurement holes 47-2, 47-3, 47-4, 47-5, 47-6 and the center of hub hole 41a are referred to as plate surface strain gauges 62b, 63b, 64b, 65b, 66b.

[0032] 3, a pair of hole strain gauges 70 are attached to each measurement hole 47. Here, as shown in FIG. 3(a), they are arranged so that the plate portion 45 on the opposite flange side faces outward. In this state, of the pair of hole strain gauges 70 attached to each measurement hole 47, the one attached on the counterclockwise side is marked with the letter "A," and the one attached on the clockwise side is marked with the letter "B."

[0033] Of the hole strain gauges 70 attached to the inner circumferential surfaces (corresponding to the second positions) of measurement holes 47-1 to 47-6, those attached counterclockwise and in a manner that points toward the plate surface as they move toward the outer diameter are referred to as hole strain gauges 71A1, 72A1, 73A1, 74A1, 75A1, and 76A1. Also, of the hole strain gauges 70, those attached counterclockwise and in a manner that points toward the plate back surface as they move toward the outer diameter are referred to as hole strain gauges 71A3, 72A3, 73A3, 74A3, 75A3, and 76A3.

[0034] Similarly, of the hole strain gauges 70 attached to the inner surfaces of measurement holes 47-1 to 47-6, those attached clockwise and in a manner that points toward the plate surface as they move toward the outer diameter are referred to as hole strain gauges 71B1, 72B1, 73B1, 74B1, 75B1, and 76B1. Also, of the hole strain gauges 70, those attached clockwise and in a manner that points toward the plate back surface as they move toward the outer diameter are referred to as hole strain gauges 71B3, 72B3, 73B3, 74B3, 75B3, and 76B3.

[0035] <1-3. About Brake Disc 80> Figure 5 is a plan view showing the shape of the brake disc 80 attached to the PQ wheelset 20 shown in Figure 2. Note that the brake disc 80 shown in Figure 5 is a schematic illustration and is not a faithful model of an actual center-fastened brake disc. This illustration is also simplified to allow for use as the brake disc 80 for analysis using the finite element method, which will be described later. Therefore, the illustration does not include the numerous heat dissipation fins and other features that are arranged on an actual brake disc.

[0036] The brake disc 80 shown in FIG. 5 is fixed to the plate portion 45 on the opposite side to the flange of the wheel 40 and to the plate portion 45 on the flange side via bolts.

[0037] The brake disc 80 is provided in a ring shape, and is provided with bolt fastening holes 81 and ribs 82. The bolt fastening holes 81 are holes that pass through the brake disc 80, and are portions into which bolts are inserted while aligned with the bolt holes 46. For this reason, the same number of bolt fastening holes 81 as the number of bolt holes 46 are provided.

[0038] The rib 82 is a ring-shaped protruding portion around the bolt fastening hole 81, and protrudes from the surface of the brake disc 80 facing the plate portion 45. In Fig. 4, the bolt fastening holes 81 are numbered sequentially in a counterclockwise direction.

[0039] Specifically, the bolt fastening hole 81 located at 12 o'clock (0 degrees) on a clock basis is designated as bolt fastening hole 81a. Similarly, the bolt fastening holes 81 located at 11 o'clock (330 degrees), 10 o'clock (300 degrees), 9 o'clock (270 degrees), 8 o'clock (240 degrees), 7 o'clock (210 degrees), 6 o'clock (180 degrees), 5 o'clock (150 degrees), 4 o'clock (120 degrees), 3 o'clock (90 degrees), 2 o'clock (60 degrees), and 1 o'clock (30 degrees) are designated as bolt fastening holes 81b, 81c, 81d, 81e, 81f, 81g, 81h, 81i, 81j, 81k, and 81l, respectively.

[0040] <1-4. About bridge circuit Qb and Qs> Fig. 6 is a diagram showing bridge circuits Qb, Qs for measuring lateral force in the PQ wheelset 20. Fig. 7 is a bridge circuit corresponding to the lateral force measurement method using shear strain in the PQ wheelset 20. Two types of bridge circuits Qb, Qs are provided: a bridge circuit Qb corresponding to the lateral force measurement method using wheel bending as shown in Fig. 6, and a bridge circuit Qs corresponding to lateral force measurement using shear strain as shown in Fig. 7.

[0041] Furthermore, in this embodiment, the bridge circuit Qb shown in Fig. 6 includes three systems: bridge circuits Qb1, Qb2, and Qb3. Also, the bridge circuit Qs shown in Fig. 7 includes three systems: bridge circuits Qs1, Qs2, and Qs3. In this embodiment, the bridge circuits Qb1, Qb2, and Qb3 are referred to collectively or simply as bridge circuit Qb when there is no need to distinguish them from one another. Similarly, the bridge circuits Qs1, Qs2, and Qs3 are referred to collectively or simply as bridge circuit Qs when there is no need to distinguish them from one another.

[0042] In this way, the effects of hysteresis can be further reduced when calculating the lateral force Q using three systems each for the bridge circuit Qb and the bridge circuit Qs. However, the effects of hysteresis can also be sufficiently reduced when calculating the lateral force Q using one system each for the bridge circuit Qb and the bridge circuit Qs.

[0043] In the bridge circuit Qb1, the bridge circuit Qb2, and the bridge circuit Qb3, the plate surface strain gauges 60 included therein are arranged so that the angular positions in the rotation direction of the axle 50 are shifted by 60 degrees from one another.

[0044] The bridge circuit Qb1 is configured by sequentially connecting plate surface strain gauges 61a, 64a, 64b, and 61b in a circular configuration. The bridge circuit Qb2 is configured by sequentially connecting plate surface strain gauges 62a, 65a, 65b, and 62b in a circular configuration. The bridge circuit Qb3 is configured by sequentially connecting plate surface strain gauges 63a, 66a, 66b, and 63b in a circular configuration.

[0045] In addition, in the bridge circuit Qb1, an input voltage Vin is applied between the plate surface strain gauge 61a and the plate surface strain gauge 64a and between the plate surface strain gauge 61b and the plate surface strain gauge 64b, and the voltage between the plate surface strain gauge 64a and the plate surface strain gauge 64b and between the plate surface strain gauge 61a and the plate surface strain gauge 61b is set as an output voltage Vout.

[0046] Similarly, in bridge circuit Qb2, an input voltage Vin is applied between plate surface strain gauge 62a and plate surface strain gauge 65a and between plate surface strain gauge 62b and plate surface strain gauge 65b, and the voltage between plate surface strain gauge 65a and plate surface strain gauge 65b and between plate surface strain gauge 62a and plate surface strain gauge 62b is set as output voltage Vout.

[0047] Similarly, in bridge circuit Qb3, an input voltage Vin is applied between plate surface strain gauge 63a and plate surface strain gauge 66a and between plate surface strain gauge 63b and plate surface strain gauge 66b, and the voltage between plate surface strain gauge 66a and plate surface strain gauge 66b and between plate surface strain gauge 63a and plate surface strain gauge 63b is set as output voltage Vout.

[0048] Furthermore, in the bridge circuits Qs1, Qs2, and Qs3 shown in FIG. 7, the plate surface strain gauges 60 included therein are arranged so that the angular positions in the rotation direction of the axle 50 are shifted by 60 degrees from one another.

[0049] The bridge circuit Qs1 is configured by sequentially connecting hole strain gauges 71A1, 71B1, 74A1, 74B1, 74B3, 74A3, 71B3, and 71A3 in a ring. The bridge circuit Qs2 is configured by sequentially connecting hole strain gauges 72A1, 72B1, 75A1, 75B1, 75B3, 75A3, 72B3, and 72A3 in a ring. The bridge circuit Qs3 is configured by sequentially connecting hole strain gauges 73A1, 73B1, 76A1, 76B1, 76B3, 76A3, 73B3, and 73A3 in a ring.

[0050] In addition, in the bridge circuit Qs1, an input voltage Vin is applied between the hole strain gauge 71B1 and the hole strain gauge 74A1, and between the hole strain gauge 71B3 and the hole strain gauge 74A3, and the voltage between the hole strain gauge 71A1 and the hole strain gauge 71A3, and between the hole strain gauge 74B1 and the hole strain gauge 74B3 is the output voltage Vout.

[0051] Similarly, in bridge circuit Qs2, an input voltage Vin is applied between hole strain gauge 72B1 and hole strain gauge 75A1, and between hole strain gauge 72B3 and hole strain gauge 75A3, and the voltages between hole strain gauge 72A1 and hole strain gauge 72A3 and between hole strain gauge 75B1 and hole strain gauge 75B3 are set as output voltage Vout.

[0052] Similarly, in bridge circuit Qs3, input voltage Vin is applied between hole strain gauge 73B1 and hole strain gauge 76A1 and between hole strain gauge 73B3 and hole strain gauge 76A3, and the voltage between hole strain gauge 73A1 and hole strain gauge 73A3 and between hole strain gauge 76B1 and hole strain gauge 76B3 is set as output voltage Vout.

[0053] <1-5. Regarding the slip ring device 100 and the strain signal processing unit 110> The outputs of the bridge circuits Qb1 to Qb3 and the outputs of the bridge circuits Qs1 to Qs3 are input to the slip ring device 100. The slip ring device 100 outputs a signal to the strain signal processing unit 110.

[0054] The outputs of the bridge circuits Qb1 to Qb3 and the outputs of the bridge circuits Qs1 to Qs3 are input to the distortion signal processing unit 110. The distortion signal processing unit 110 amplifies the input signals and performs signal processing such as AD conversion, and then transmits a predetermined signal to the coupling coefficient determination unit 120.

[0055] <1-6. Coupling coefficient determination unit 120 and lateral force estimation unit 130> Next, the coupling coefficient determination unit 120 and the lateral force estimation unit 130 will be described. For example, in a computer configured with a CPU, memory (ROM, RAM, non-volatile memory, etc.), and other elements, the coupling coefficient determination unit 120 is functionally realized by the cooperation of predetermined programs and data in the memory.

[0056] The coupling coefficient determiner 120 determines (calculates) linear coupling coefficients (described later) based on the outputs from the multiple bridge circuits Qb and Qs, i.e., the outputs from bridge circuits Qb1-Qb3 and bridge circuits Qs1-Qs3, which are input via the strain signal processor 110. The outputs from the multiple bridge circuits Qb and Qs include hysteresis that differs depending on whether the lateral force is increasing or decreasing. The strain signal processor 110 determines the linear coupling coefficients based on the outputs including this hysteresis. The coupling coefficient determiner 120 also determines the coupling coefficients based on the relationship between applied lateral force and strain obtained in a preliminary (i.e., pre-travel) verification test (static load test).

[0057] The coupling coefficient determination unit 120 determines the linear coupling coefficients before a road test based on the outputs (corresponding to the certification test data) from the bridge circuits Qb and Qs of the multiple systems. Therefore, the coupling coefficient determination unit 120 does not need to be performed every time a road test is performed. The linear coupling coefficients determined by the coupling coefficient determination unit 120 can be used in the road test. Details of determining the coupling coefficients will be described later.

[0058] Furthermore, for example, in a computer configured with a CPU, memory (ROM, RAM, non-volatile memory, etc.), and other elements, the lateral force estimation unit 130 is functionally realized by the cooperation of predetermined programs and data in the memory.

[0059] The lateral force estimation unit 130 is a part that, during a vehicle running test, calculates an estimated lateral force ^Q, which is an estimated value of the lateral force Q, by calculating a weighted sum (Equation (11) or Equation (13) described later) of the outputs from the multiple bridge circuits Qb and Qs using the linear combination coefficients determined by the combination coefficient determination unit 120. The lateral force estimation unit 130 may be mounted on the vehicle on which the running test is performed, or may be installed in a location other than the vehicle.

[0060] [2. Method for improving the measurement accuracy of lateral force Q by utilizing multiple bridge circuit outputs] Next, we will explain the details of how the coupling coefficient is determined by the coupling coefficient determiner 120 and how the measurement accuracy of the lateral force Q is improved. When the frictional force between the lateral force Q and the brake disc 80 acts on the wheel 40, the output ε (referring to the symbol on the left side of "Equation 1"; the same applies hereinafter) of the bridge circuits Qb and Qs can be expressed as in the following equation (1).

number

[0061] The output components of the bridge circuits Qb and Qs due to frictional force can be expressed in the form of, for example, the following equation.

number

[0062] Alternatively, since the heat dissipation fins where the brake disc 80 is in contact with the wheel 40 are arranged discretely, the sensitivity a f If we approximate that (p) is constant, we can also express it discretely as in equation (3) below.

number

[0063] From this equation (3), we can say that F fs It has a strong nonlinearity with respect to the lateral pressure Q, and it is difficult to model this in detail, but the force F fsThe strain (bridge output) that occurs when acts on the sth contact area is expressed as a proportionality coefficient if it is within the range of linear elasticity. fs The force F should be proportional to the friction force. fs and coefficient a fs In theory, if the force F is known, the hysteresis due to friction can be removed by signal processing. fs It is difficult to measure the coefficient a fs It is also difficult to directly identify the

[0064] So, the force F fs and coefficient a fs Rather than directly determining εf, we consider determining εf indirectly from the outputs of multiple bridge circuits Qb and Qs and subtracting this. As a simple example, consider the case where the friction force is a concentrated load (i.e., when N = 1). Alternatively, even if the load is distributed, it can be treated in the same way as a concentrated load if it falls under the following (a) or (b). (a) When the same friction force acts on all contact areas (b) The magnitude of the frictional force acting on each contact area is different, but the spatial ratio is constant.

[0065] The above (b) is, for example, the frictional force, which is the constant F f and the distribution function d(p), and d(p) a f In either case, the constant corresponding to the sensitivity is a f This can be expressed as the following equation (4).

number

[0066] Now, suppose that outputs ε1 and ε2 are obtained from the two bridge circuits Qb and Qs as outputs of the bridge circuits Qb and Qs, and can be expressed as the following equations (5) and (6), respectively.

number

[0067] Here, we consider canceling the friction force term by linearly combining ε1 and ε2. That is, consider the calculation of the following equation (7) with c1 and c2 as appropriate constants, and then consider determining the constants c1 and c2 in equation (7) so that the friction force term is further cancelled out.

number

[0068] Further rearranging equation (7) gives equation (8) below. TIFF2025149553000008.tif11170

[0069] Therefore, if the constants c1 and c2 are determined so as to satisfy the following equations (9) and (10), the friction force term can be canceled out by linear combination.

number

[0070] However, as mentioned above, the sensitivity to friction force a f1 and a f2 Since it is difficult to directly grasp the above, the simultaneous equations (9) and (10) cannot be directly used. Therefore, as an alternative method, we can use the above equation (8) by adding c1a1+c2a2=1 in the above equation (9) and c1a in the above equation (10). f1 +c2a f2 = 0 and consider the following equation (11).

number

[0071] In this case, there are two unknowns, c1 and c2, but only one equation can be established, so it is necessary to use at least two test results under different loading conditions. Furthermore, as a prerequisite for the above discussion to be valid, the simultaneous equations, equations (9) and (10), must be linearly independent. In other words, it is not possible to determine the constants c1 and c2 without considering the combination of two bridge circuits Qb and Qs that have different ratios of sensitivity to lateral force Q and sensitivity to friction force.

[0072] In cases where the frictional force cannot be eliminated using two bridge circuits Qb and Qs, one effective method is to increase the number of bridge circuits Qb and Qs used in the lateral force calculation. Below, the previous discussion of linear combination will be generalized to cases where three or more bridge circuits Qb and Qs are used.

[0073] The number of bridge circuits is M, and R M Let vector ε belong to the vector that lists the outputs of all bridge circuits Qb and Qs (R is a real number). M The coefficient vector c can be determined so as to satisfy the following equation (12), where c is a vector in which coefficients belonging to the above equation are arranged. Note that equation (12) corresponds to calculation equation 1.

number

[0074] The above equation (12) is equivalent to equation (11) when dealing only with two bridge circuits Qb and Qs, and cannot be solved as is because the number of equations is insufficient compared to the number of unknowns. On the other hand, in a static load test, multiple pairs of ε and Q are obtained during the process, and each pair is distinguished by a subscript, such as ε1, Q1, ε2, Q2, .... Using a matrix arranging these, a simultaneous equation such as equation (13) below can be constructed, and by solving this, the coefficient vector c can be determined.

number

[0075] As described above, by determining the coefficient vector c in the static load test, the friction force term can be canceled, and the influence of hysteresis caused by friction force can be reduced from the outputs of the multiple bridge circuits Qb and Qs, thereby improving the measurement accuracy of the lateral force Q.

[0076] [3. Example of analysis results] Next, the extent of the hysteresis effect was analyzed using a specified EFM method. In this analysis, a load equivalent to lateral force Q was applied to the tread 43 directly above bolt fastening hole 81a (No. 1) at 0kN, 0.5kN, 1kN, up to 50kN, and then unloaded in 0.5kN increments, from 50kN to 49.5kN, up to 0kN. Figure 8 shows the loading and unloading conditions.

[0077] The bolt preload was set to 25 kN, and the friction coefficient was set to 0.25. It was also assumed that the deformation of the wheel 40 and the brake disc 80 was minimal.

[0078] The contact force and sliding displacement at each contact point are evaluated, and the strain of the wheel 40 is evaluated at the positions and directions shown in Fig. 3. Note that this analytical model of the wheel 40 is a modified version of the PQ wheelset 20 model for conventional lines, with hole positions and diameters changed, and a total of 18 holes are provided, including 12 bolt holes 46 and 6 measurement holes 47, as shown in Fig. 3.

[0079] Using the obtained estimated strain amount, the output values ​​are evaluated for the bridge circuit Qb alone shown in Figure 6 and the bridge circuit Qs alone shown in Figure 7. Note that the bridge circuit Qb shown in Figure 6 corresponds to the lateral force measurement method using wheel bending, and the bridge circuit Qs shown in Figure 7 corresponds to the lateral force measurement method using shear strain. An example of this analysis result is shown in Figure 9. In Figure 9, (a) shows the output of the bridge circuit Qb, and (b) shows the output of the bridge circuit Qs. Note that in Figures 9(a) and (b), the solid lines show the output of the bridge circuits Qb and Qs when the lateral force Q increases, and the dashed lines show the output of the bridge circuits Qb and Qs when the lateral force Q decreases.

[0080] In this evaluation, the bridge circuit Qb1 shown in FIG. 6(a) and the bridge circuit Qs1 shown in FIG. 7(a) are evaluated.

[0081] As shown in Figure 9, if hysteresis were not occurring, the solid line and dashed line should match when the lateral force Q increases and when it decreases, but they do not match for either bridge circuit Qb1 or Qs1, which indicates that hysteresis is occurring. In this analysis, the outputs of bridge circuits Qb1 and Qs1 when a lateral force Q of 50 kN is applied are used as the reference (matched) in order to evaluate the relative strength of the effect of hysteresis.

[0082] The ratio of the maximum absolute value of the hysteresis width was calculated to be 3.9% for the bridge circuit Qb1 alone and 56.9% for the bridge circuit Qs1 alone.

[0083] Meanwhile, we attempted to calculate the lateral force Q from the outputs of the two bridge circuits Qb1 and Qs1 by applying the above equation (13) to the same data as shown in Figure 9. In this case, the linear combination coefficients were determined to be c1 = 0.0412 for bridge circuit Qb1 and c2 = 0.0314 for bridge circuit Qs1. Using these linear combination coefficients, we calculated the estimated lateral force ^Q as ^Q = c1εb + c2εs (see equation (11)), where εb and εs are the outputs of bridge circuits Qb1 and Qs1, respectively.

[0084] Based on the above, the relationship between the loaded lateral pressure Q and the estimated lateral pressure ^Q is shown in Figure 10. In Figure 10, the horizontal axis represents the loaded lateral pressure Q, and the vertical axis represents the estimated lateral pressure obtained by the linear combination of bridge circuit Qb1 and bridge circuit Qs1 (i.e., the above ^Q = c1εb + c2εs).

[0085] As shown in Figure 12, it can be seen that the influence of relative hysteresis is reduced by combining the outputs of the two bridge circuits Qb1 and Qs1. Specifically, as with the evaluation of the influence of hysteresis shown in Figure 9, the output of the bridge circuits Qb1 and Qs1 when a lateral force Q was applied of 50 kN was used as the reference for evaluation. The ratio of the maximum absolute value of the hysteresis width of the estimated lateral force ^Q was evaluated and found to be 0.51%.

[0086] Therefore, the results of this evaluation show that the effect of hysteresis caused by friction with the brake disc 80 is reduced by estimating the lateral pressure Q from the output values ​​of multiple bridge circuits (in this example, two bridge circuits Qb1, Qs1) rather than estimating the lateral pressure Q from the output value of a single bridge circuit (bridge circuit Qb or bridge circuit Qs).

[0087] [Note] The contents of the above-described embodiments can be understood, for example, as follows. The above-described lateral force measuring device 10 can produce the following effects. [1] That is, in this embodiment, the lateral force measuring device 10 is for measuring PQ on a wheel equipped with a centrally fastened brake disc 80, and includes: (A) multiple bridge circuits Qb, Qs each including strain gauges (plate strain gauges 60, hole strain gauges 70) attached to the wheel 40 to measure strain on the wheel 40; (B) a coupling coefficient determination unit 120 that determines a linear coupling coefficient of the outputs from each of the multiple bridge circuits Qb, Qs, the outputs including hysteresis that differs when the lateral force increases and when the lateral force decreases; and (C) a lateral force estimation unit 130 that calculates an estimate of the lateral force Q by calculating a weighted sum of the outputs from the multiple bridge circuits Qb, Qs using the linear coupling coefficient determined by the coupling coefficient determination unit 120; and (D) the multiple bridge circuits Qb, Qs are combined in such a way that the lateral force Q and frictional force can be separated.

[0088] In this configuration, the multiple bridge circuits Qb, Qs are combined in a way that separates the lateral force Q from the frictional force, thereby canceling (removing) the influence of the frictional force from the output results of the bridge circuits Qb, Qs. This reduces the influence of hysteresis caused by slippage between the wheel 40 and the brake disc 80, making it possible to improve the measurement accuracy of the lateral force Q.

[0089] [2] In the above-described embodiment, in the above item [1], the lateral force estimation unit 130 may be configured as a lateral force measurement device characterized by calculating an estimated value of the lateral force based on the following (calculation formula 1). Q = ε T c (calculation formula 1; formula (12)) In the above formula 1, Q is the estimated lateral force, ε is a vector that lists the outputs of all bridge circuits Qb and Qs of multiple systems, and ε T is the transpose of the matrix indicated by ε. c is a coefficient vector in which the linear coupling coefficients are arranged in the same number as the number of systems of the bridge circuit of multiple systems, and the coupling coefficient determination unit 120 determines the transposed matrix ε of the lateral load Q applied during the preliminary verification test (static load test) and the bridge output obtained at that time. T This was calculated from the value of

[0090] Based on the above equation (12), by calculating the transpose matrix of the vector arranging the outputs of all bridge circuits Qb and Qs of multiple systems and further using the linear combination coefficient c obtained in a preliminary certification test, it is possible to reduce the effect of hysteresis caused by sliding displacement between the wheel 40 and the brake disc 80 and improve the measurement accuracy of the lateral force Q.

[0091] [3] Furthermore, in the above-described embodiments, in the above items [1] or [2], the multiple bridge circuits Qb, Qs may be a bridge circuit Qb consisting of a first strain gauge (plate surface strain gauge 60) for measuring lateral force attached to a first position of the wheel 40 (a predetermined position of the plate portion 45), and a bridge circuit Qs including a second strain gauge (hole strain gauge 70) for measuring lateral force due to shear strain of the wheel 40 attached to a second position (the inner surface of the measurement holes 47-1 to 47-6) different from the first position (a predetermined position of the plate portion 45).

[0092] In this configuration, plate surface strain gauges 60 and hole strain gauges 70 are attached to the plate portions 45 of the wheel 40, which have completely different surface directions, and to the inner circumferential surfaces of the measurement holes 47-1 to 47-6, respectively. This makes it possible to improve the measurement accuracy of the lateral force Q.

[0093] [4] In addition, the lateral force measurement method of the lateral force measurement device 10 for measuring PQ on a wheel 40 equipped with a centrally fastened brake disc 80 in this embodiment includes a determination step of determining a linear combination coefficient of the outputs of each bridge circuit Qb, Qs, which includes multiple strain gauges (plate strain gauges 60, hole strain gauges 70) attached to the wheel 40 and measure the strain of the wheel 40, and which outputs include hysteresis that differs when the lateral force increases and when the lateral force decreases, and an estimation step of using the determined linear combination coefficient to calculate a weighted sum of the outputs from the multiple bridge circuits Qb, Qs, and calculating an estimated lateral force value Q, where the multiple bridge circuits Qb, Qs are combined in such a way that lateral force and friction force can be separated.

[0094] In this lateral force measurement method, the multiple bridge circuits Qb, Qs are combined in a way that separates the lateral force Q from the frictional force (i.e., linearly independent), so that the influence of the frictional force can be canceled (removed) from the output results of the bridge circuits Qb, Qs. This reduces the influence of hysteresis caused by slippage between the wheel 40 and the brake disc 80, making it possible to improve the measurement accuracy of the lateral force Q.

[0095] <Modification> Although the embodiments of the present invention have been described above, the present invention can be modified in various other ways, which will be described below.

[0096] In the above embodiment, the coefficient vector c (linear combination coefficients) is determined based on the bridge circuits Qb and Qs. However, when multiple bridge circuits Qs exist, the coefficient vector c (linear combination coefficients) may be determined from the multiple bridge circuits Qs. [Explanation of symbols]

[0097] 10...Lateral force measuring device, 20...PQ wheelset, 30...wheelset, 40...wheel, 41...hub portion, 41a...hub hole, 42...rim portion, 43...tread, 44...flange, 45...plate portion, 46...bolt hole, 47, 47-1 to 47-6...measurement hole, 50...axle, 60, 61a to 66a, 61b to 66b...plate surface strain gauge, 70, 71A1 to 76A1, 71A3 to 76A3, 71B 1 to 76B1, 71B3 to 76B3...hole strain gauge, 80...brake disc, 81, 81a to 81l...bolt fastening hole, 82...rib, 100...slip ring device, 110...signal processing unit, 120...coupling coefficient determination unit, 130...lateral force estimation unit, P...wheel load, Q...lateral force, Qb, Qb1 to Qb3...bridge circuit, Qs, Qs1 to Qs3...bridge circuit, R...rail

Claims

1. A lateral force measuring device for performing PQ measurements on wheels equipped with centrally fastened brake discs, comprising: (A) a plurality of bridge circuits including strain gauges attached to the wheels for measuring strain of the wheels; (B) a coupling coefficient determination unit that determines a linear coupling coefficient of outputs from the bridge circuits of the plurality of systems, the outputs including hysteresis that differs when the lateral pressure increases and when the lateral pressure decreases; (C) a lateral force estimation unit that calculates an estimated value of lateral force by calculating a weighted sum of outputs from the bridge circuits of the plurality of systems using the linear combination coefficients determined by the combination coefficient determination unit; and Equipped with (D) The bridge circuits of the plurality of systems are combined so as to separate the lateral pressure and the frictional force. A lateral force measuring device characterized by:

2. The lateral force measuring device according to claim 1, The lateral force measuring device is characterized in that the lateral force estimation unit calculates an estimated value of the lateral force based on the following (calculation formula 1). Q = ε T c (Calculation formula 1) Q is the estimated lateral force, ε is a vector obtained by arranging the outputs of all the bridge circuits of the multiple systems, and ε T is the transposed matrix of the matrix indicated by ε. c is a coefficient vector in which the linear combination coefficients are arranged in the same number as the number of systems of the bridge circuits of the plurality of systems, and the coupling coefficient determination unit determines a transposed matrix ε of the lateral load Q applied during a preliminary verification test and the bridge output obtained at that time. T This was calculated from the value of

3. The lateral force measuring device according to claim 1, The plurality of bridge circuits are bridge circuits each including a first strain gauge attached to a first position of the wheel for measuring lateral force, and a second strain gauge attached to a second position different from the first position for measuring lateral force due to shear strain of the wheel. A lateral force measuring device characterized by:

4. A lateral force measurement method for a lateral force measurement device for performing PQ measurement on a wheel equipped with a centrally fastened brake disc, comprising: a determining step of determining a linear combination coefficient of outputs of each bridge circuit, which includes a plurality of strain gauges attached to the wheels and measure the strain of the wheels, the outputs including hysteresis that differs depending on whether the lateral pressure is increasing or decreasing; an estimation step of calculating an estimated value of lateral force by calculating a weighted sum of outputs from the bridge circuits of the plurality of systems using the determined linear combination coefficients; and The bridge circuits of the plurality of systems are combined in such a way that lateral pressure and frictional force can be separated. A lateral force measuring method characterized by:

Images