Track measuring station

Abstract

A fixed track measuring point (101) and a wheel contact point determination method are disclosed. The fixed track measuring point (101) is configured to be attached to a web (21) of a rail (11) of a track (1) of a predetermined track type.The stationary track measuring station (101) is provided with a plurality of interconnected strain gauges, which are to be arranged in different orientations on the web (21), wherein strain gauges arranged and interconnected on an inner side (21a) of the web (21) correspond to a mirror image of the strain gauges arranged and interconnected on an outer side (21b) of the web, a data acquisition unit configured to evaluate measurement signals obtained from the strain gauges and to determine from them the forces applied by each individual wheel of a rail vehicle traveling over the track (1) in a longitudinal direction (x), a transverse direction (y) and a vertical direction (z) of the rail, and a calculation unit which has a memory in which signals from simulated rail vehicle crossings are stored via a numerical model of the track (1).By comparing the measurement signals with stored data of the simulated rail vehicle crossings, among other things a wheel contact point (e) in the transverse direction of the rail (11) can be determined.

Description

The disclosure concerns a fixed track measuring station and a wheel contact point determination method. In addition to information concerning forces acting on a rail during the passage of a rail vehicle, information concerning the current position of a wheel contact point during the passage of the rail vehicle is important in order to be able to detect damage to the rail alone and also to the entire track caused by loads during operation in a timely manner. The track described in the application consists of a structure made up of two rails connected by several sleepers or other supports, such as Vignoles, broad-footed rails, or grooved rails, fastening devices for connecting each rail to a sleeper, and the associated substructure. The sleepers can be made of materials such as wood, concrete, steel, or plastic. Fastening devices can include, for example, so-called Epsilon clamps or other small metal fasteners, possibly with additional intermediate layers. The substructure can be made of ballast, preferably compacted ballast, concrete, or a steel supporting structure of a bridge. It should be noted that the examples mentioned here are not exhaustive. Alternatives known to those skilled in the art are also included, particularly with regard to the fasteners and the substructure. A track type therefore corresponds to a specific combination of rails, sleepers, fasteners, and substructure. Elastic intermediate layers and sleeper pads, if present, are also preferably included in the track type. The railway vehicle addressed in the application is any traction vehicle, such as a railcar, a steam, electric or diesel locomotive, a draisine, an electric or diesel multiple unit, passenger cars (train coaches) or freight wagons with any number of axles, special vehicles, and any combination thereof. As above, the examples given here are not exhaustive. Alternatives known to those skilled in the art also fall under the term "rail vehicle," such as hybrid locomotives, battery-powered locomotives, or hybrid, battery-powered, or fuel cell multiple units. EP 1 904 356 B1 discloses a device for detecting the risk of derailment of railway vehicles. This device involves attaching strain gauges to the rail, which measure longitudinal stresses in the rail and detect horizontal movement of the wheel contact point. For this purpose, the rail foot is supported by a measuring element that detects horizontal and vertical forces at a horizontally moving wheel contact point. Strain gauges are designed on the outer edge of the rail head (BFR) and on both outer edges of the rail foot (BFR) to detect horizontal displacements. The measuring element supporting the rail foot has at least one beam-shaped force application section that can be connected to the rail foot and at least one beam-shaped force dissipation section that can be connected to a sleeper.These parts are arranged parallel and horizontally one above the other, and at least one pair of measuring countersinks is arranged between the force input and force output parts on at least one deformation body. EP 1 521 072 A1 describes a method for detecting the elastic deformation of railway tracks, wherein strain gauges are provided on the rail web and on the rail foot. From JP 2002 - 236 065 A a method for measuring the stresses in a railway rail has become known, wherein the strain gauges used are arranged in the form of Wheatstone bridges. Finally, CN 107 014 528 A shows a system for monitoring the forces occurring between the wheel and the rail. Here too, strain gauges are attached to the rail. The aforementioned devices and methods, according to the state of the art, require a considerable amount of effort to attach the necessary components, such as strain gauges, to suitable locations on a rail. Firstly, attachment to the rail head or foot is difficult due to the curvature of the rail profile in these areas. Secondly, preparing an application point, particularly at the foot, can be complex. For example, it may be necessary to remove track ballast at the application point on the rail foot, attach the strain gauge to the underside of the rail, which is difficult to inspect, and solder it overhead to the required wires. There is a need for a stationary track measuring station and a wheel contact point determination method, the components of which can be easily and reliably attached to a rail, in particular exclusively to the web of the rail. The invention relates to a stationary track measuring point configured to be attached to a web of a rail of a track of a predetermined track type. The stationary track measuring station has a plurality of interconnected strain gauges, which are to be arranged in different orientations on the web, wherein strain gauges arranged and interconnected on an inside of the web (which are also referred to as strain gauges) correspond to a mirror image of the strain gauges arranged and interconnected on an outside of the web; a data acquisition unit which is configured to evaluate measurement signals obtained from the strain gauges and from this to determine the forces applied by each individual wheel of a rail vehicle traveling over the track in a longitudinal direction, a transverse direction and a vertical direction of the rail; and a calculation unit which has a memory in which signals from simulated rail vehicle crossings are stored via a numerical model of the track. The calculation unit is configured to compare measurement signals obtained from the interconnected strain gauges with the stored signals from the simulated rail vehicle crossings, to assign the obtained measurement signals to one of the simulated rail vehicle crossings whose simulated measurement signals correspond to the obtained measurement signals, and thereby to determine a wheel contact point in the transverse direction of the rail. Since the number of different track types is manageable despite the various possible combinations of rails, sleepers, or other supports, these different track types can be simulated using finite element analysis and stored in databases. Similarly, a large number of crossings by a rail vehicle can be simulated with relatively little computational effort using just one wheel of the rail vehicle. The differences between the individual simulated crossings are essentially based on a change in the wheel's point of contact. This is because, particularly on straight tracks, the conical wheel of the rail vehicle contacts the rounded head of the rail at a point known as the wheel contact point. The resulting track of successive wheel contact points meanders (sinusoidally) on the rail head, as the conical running surfaces of the wheels provide self-centering guidance. Consequently, when the same rail vehicle passes over the same point on the track, different forces can be applied, depending primarily on the position of the wheel contact point. By comparing the measurement signals obtained from the strain gauges with the stored signals from the simulated rail crossings, it is possible to assign the obtained measurement signals to a specific simulated rail crossing. This numerical model and the entirety of the simulated rail crossings are also referred to as a digital twin. Since the simulated wheel contact point is also known, the actual wheel contact point can be determined. This is done by identifying the known wheel contact point of the simulated rail vehicle crossing whose simulated forces in the longitudinal, lateral, and vertical directions correspond to the forces actually determined from the obtained measurement signals. Furthermore, it is possible to correct the measurement signals when the axle spacing in a bogie is very small. With a very small distance between the axles in the bogie, e.g., < 2 m, measurement signals generated by one axle can overlap with those generated by the other axle. Furthermore, the dynamics of the forces at the moment the wheel disc passes over the track measuring point can be derived by comparison. Advantageously, during the determination of the forces in the longitudinal, transverse and vertical direction of the rail, a conversion factor determined in advance by applying known forces in the longitudinal, transverse and vertical direction of the rail and comparing them with the forces determined by the data acquisition unit can be applied for each of the directions from longitudinal, transverse and vertical direction. Advantageously, the strain gauges can be connected in at least three full bridges. Preferably, the number of full bridges can be four. Alternatively or additionally, some of the strain gauges can be connected to form quarter or half bridges. It is advantageous for each full bridge to have at least four strain gauges. Advantageously, the position and orientation of the strain gauges on the rail web and to each other can be determined by a numerical optimization algorithm based on the simulated rail vehicle crossings via the numerical model of the predetermined track type. This makes it possible to select an area of ​​the bridge for the attachment and alignment of the strain gauges where the greatest strains and compressions occur during train crossings. This ensures that sufficiently strong measurement signals can be generated. An inventive method for determining the wheel contact point at the head of a rail of a track of a predetermined track type comprises the following steps: Arranging a plurality of interconnected strain gauges in different orientations on the web, wherein strain gauges arranged and interconnected on an inner side of the web correspond to a mirror image of the strain gauges arranged and interconnected on an outer side of the web; evaluating the measurement signals obtained from the strain gauges with a data acquisition unit; and determining the forces applied by each individual wheel of a rail vehicle traveling over the track in a longitudinal direction, a transverse direction, and a vertical direction of the rail from the evaluated measurement signals; and comparing the measurement signals obtained from the interconnected strain gauges with pre-stored signals from simulated rail vehicle crossings.- Assigning the measurement signals obtained from the interconnected strain gauges to one of the simulated rail vehicle crossings whose simulated measurement signals correspond to the obtained measurement signals, and - thereby determining a wheel contact point in the transverse direction of the rail. Advantageously, during the determination of the forces in the longitudinal, transverse and vertical direction of the rail, a conversion factor k, determined in advance by applying known forces in the longitudinal, transverse and vertical direction of the rail and comparing them with the forces determined by the data acquisition unit, can be applied for each of the directions from longitudinal, transverse and vertical direction. Advantageously, the position and orientation of the strain gauges on the rail web and to each other can be determined in advance by a numerical optimization algorithm based on the simulated rail vehicle crossings via the numerical model of the predetermined track type. Further advantages become apparent from the figures, which schematically illustrate currently preferred embodiments of the invention. In the figures, Fig. 1 shows a section of a track with rails and sleepers; Fig. 2 a detail of a finite element (FE) model of a track; Fig. 3 a cross-sectional view of a Vignoles rail profile; Fig. 4 a cross-sectional view of a grooved rail profile. Figure 1 schematically shows a track 1. The track 1 discussed here consists of two rails 11 connected by sleepers 13, which are fastened to the sleepers 13 by known means such that the distance between the rails corresponds to a predetermined track gauge. An example of the known means is shown in Figure 2, which depicts a clamping plate 15, a screw 17, and an elastic epsilon clamp 19. An elastic intermediate layer 14 is located between the foot 23 of the rail 11 and the sleeper 13. The Epsilon clamp 19 is pre-tensioned against the foot 23 of the rail 11 by screwing the screw 17 into the sleeper 13. Due to the elasticity of the Epsilon clamp 19, a secure connection is ensured at all times, even if the components change length due to, for example, temperature fluctuations. The plate 15 serves to distribute forces acting directly on the threshold 13 from the epsilon clamp 19, so that the material of the threshold 13 is not excessively damaged. The schematic representation in Fig. 1 shows only a very short section of track with three sleepers. In reality, however, the track discussed here is considerably longer, ranging from a few kilometers to several hundred kilometers. A track type according to the invention refers to the combination of the track discussed above with the track bed. In most cases, this consists of ballast. Alternatively, a solid roadway with a base layer of concrete or asphalt can also form the bed, or the rails can be attached directly to a stable, flat substructure without rail supports, using screws or spring clips. Furthermore, the term "track type" encompasses not only continuous welded track but also various types of rail joints, such as fishplate joints. To reduce the required computing power, sleepers and ballast can be modeled using anisotropic materials in order to achieve corresponding background stiffnesses. According to the invention, FE models of the various track types are prepared in sufficient length to also take into account the effects of a force applied by a rail vehicle on adjacent areas of the measuring point. For example, it has proven useful to create an FE model of a track with, say, thirteen sleeper bays. In such a model, if a force is applied in the middle of the rail, i.e., in the seventh sleeper bay, no influence of the load is discernible at the end sections. Furthermore, crossings with different types of rail vehicles are simulated on each individual FE model of the various track types. A predetermined number of crossings are simulated with one and the same rail vehicle, each time with different sinusoidal wheel trajectories and thus different track contact points. This results in a large number of data sets for each track type. To reduce computing power, the simulation is performed using only one wheel. From the measurement results of these simulated rail vehicle crossings, the most suitable locations for attaching strain gauges and their ideal alignment on a real rail to be measured 11 are determined using an optimization algorithm. The strain gauges are preferably attached to the web 21 of the rail 11 in the form of full bridges, thus forming a fixed track measuring point 101. An arrangement of the strain gauges on an outer surface 21a is analogous to a mirrored arrangement of the strain gauges on an inner surface 21b of the web 21. According to the invention, the strain gauges only need to be attached to the rail web, since the optimization algorithm was able to determine, using the simulated track types, at which points on the web sufficiently high measurement values ​​can be obtained. Thus, with the track measuring point 101 according to the invention, it is unnecessary to attach strain gauges to the head 25 of the rail 11 or to the foot 23 of the rail 11. It should be noted that the optimal attachment point for strain gauges to determine the vertical force is actually on the underside of the foot 23, but this is practically difficult to implement. It would be too time-consuming to remove the ballast or other type of roadbed to create sufficient space on the underside of the foot 23 of the rail 11 for attaching strain gauges. This would require applying the strain gauge(s) overhead and soldering them to electrical wires, which is very complex. To increase the accuracy of the forces determined at track measuring point 101 from the recorded signals, calibration is preferably carried out after the strain gauges are attached. In this process, known forces are applied to the head 25 of the rail 11 in the longitudinal direction x, transverse direction y and vertical direction z, and a conversion factor k between the applied force and the force measured by the strain gauges is determined. The forces are determined by a data acquisition unit not shown in the figures. This unit evaluates the measurement signals obtained from the strain gauges and calculates the forces exerted by each individual wheel of a rail vehicle traveling over track 1 in the longitudinal direction x, the transverse direction y and the vertical direction z of the rail 11. A processing unit, also not shown in the figures, has a memory in which the data from the simulated rail vehicle crossings are stored. Within the processing unit, the signals determined by the data acquisition unit are compared with the data from the simulated rail vehicle crossings. The data set from the simulated rail vehicle crossings whose values ​​are closest to those determined by the data acquisition unit is identified as the reference data set. From this comparison and the assignment of the reference data set to the measurement results, parameters that cannot be directly measured with the strain gauges can be determined. These are primarily the wheel contact point e, which is shown schematically in Figures 3 and 4 for a Vignoles rail or a grooved rail. In addition to the wheel contact point, the measurement signals can also be corrected in the event of a rail vehicle with a very small axle spacing in the bogie passing over the track. If the axle spacing in the bogie is very small, e.g., less than 2 m, the measurement signals of one axle overlap with those of the other. This leads to systematic measurement errors, which can be corrected by comparison with the data from simulated rail vehicle crossings. The dynamics of the forces at the moment the wheel disc passes over the track measuring point can also be derived by comparison with the simulated measured values.

Claims

A stationary track measuring point (101) configured to be attached to a web (21) of a rail (11) of a track (1) of a predetermined track type, comprising a plurality of interconnected strain gauges arranged in different orientations on the web (21), wherein strain gauges arranged and interconnected on an inner side (21a) of the web (21) correspond to a mirror image of strain gauges arranged and interconnected on an outer side (21b) of the web, a data acquisition unit configured to evaluate measurement signals obtained from the strain gauges and to determine from them the forces applied by each individual wheel of a rail vehicle traveling over the track (1) in a longitudinal direction (x), a transverse direction (y) and a vertical direction (z) of the rail, and a computation unit comprising a memory.in which signals from simulated rail vehicle crossings are stored via a numerical model of the track (1), and which is configured, - to compare measurement signals obtained from the interconnected strain gauges with the stored signals from the simulated rail vehicle crossings, - to assign the obtained measurement signals to one of the simulated rail vehicle crossings whose simulated measurement signals correspond to the obtained measurement signals, and - thereby to determine a wheel contact point (e) in the transverse direction (y) of the rail (11). A fixed track measuring station (101) according to claim 1, wherein, during the determination of the forces in the longitudinal direction (x), the transverse direction (y) and the vertical direction (z) of the rail, a conversion factor (k) determined in advance by applying known forces in the longitudinal direction (x), the transverse direction (y) and the vertical direction (z) of the rail and comparing them with the forces determined by the data acquisition unit is applied for each of the directions from longitudinal direction (x), transverse direction (y) and vertical direction (z). Fixed track measuring station (101) according to claim 1 or 2, wherein the strain gauges are connected to form at least three full bridges. Fixed track measuring station (101) according to claim 3, wherein the number of full bridges is four. Fixed track measuring station (101) according to one of claims 1 to 4, wherein a part of the strain gauges is connected to form quarter or half bridges. Fixed track measuring station (101) according to claim 3 or 4, wherein each full bridge has at least four strain gauges. Fixed track measuring point (101) according to one of claims 1 to 6, wherein the position and orientation of the strain gauges on the web (21) of the rail (11) and to each other is determined by a numerical optimization algorithm based on the simulated rail vehicle crossings via the numerical model of the predetermined track type. Wheel contact point determination method for determining a wheel contact point at the head of a rail (11) of a track (1) of a predetermined track type, comprising arranging a plurality of interconnected strain gauges in different orientations on the web (21), wherein strain gauges arranged and interconnected on an inner side (21a) of the web (21) correspond to a mirror image of strain gauges arranged and interconnected on an outer side (21b) of the web, evaluating the measurement signals obtained from the strain gauges with a data acquisition unit, and determining the forces applied by each individual wheel of a rail vehicle traveling over the track (1) in a longitudinal direction (x), a transverse direction (y) and a vertical direction (z) of the rail from the evaluated measurement signals, and comparing the measurement signals obtained from the interconnected strain gauges with pre-stored signals from simulated rail vehicle crossings.- Assigning the measurement signals obtained from the interconnected strain gauges to one of the simulated rail vehicle crossings whose simulated measurement signals correspond to the obtained measurement signals, and - thereby determining a wheel contact point (e) in the transverse direction (y) of the rail (11)., Wheel contact point determination method according to claim 8, wherein during the determination of the forces in the longitudinal direction (x), the transverse direction (y) and the vertical direction (z) of the rail, a conversion factor (k) determined in advance by applying known forces in the longitudinal direction (x), the transverse direction (y) and the vertical direction (z) of the rail and comparing them with the forces determined by the data acquisition unit is applied for each of the directions from longitudinal direction (x), transverse direction (y) and vertical direction (z). Wheel contact point determination method according to claim 8 or 9, wherein the position and orientation of the strain gauges on the web (21) of the rail (11) and to each other are determined in advance by a numerical optimization algorithm based on the simulated rail vehicle crossings via the numerical model of the predetermined track type.

Images