EP0123774B1 - Method of identification of waggons in a fast running train and device for carrying out the method - Google Patents

Abstract

1. Method for the identifying of wagons running in a train formation, in particular a rapidly travelling one, with the use of an arrangement for the measuring of the wheel loads with at least one measuring point, wherein also the speed of the train formation is determined, characterized thereby, that strain gauges, which are fastened at the web of a rail at the level of the neutral fibre and oriented orthogonally to the neutral fibre, are used as measuring point, that - on a wheel running over the measuring point - the then arising measurement peak of the measurement signal is triggered at a threshold valuehat the time interval between two measurement peaks on the running of two successive wheels over the measuring point is measured each time on the threshold value being fallen below or exceeded and this measured time interval is added to half the difference of the measured time intervals between the threshold value being fallen below and being exceeded at each of both the successive measurement peaks, that the result is processed together with the speed into an axle spacing and that the wagon concerned is identified by this axle spacing or by several successive axle spacing.

Description

The invention relates to a method for identifying wagons running in a particularly fast moving train set using an arrangement for measuring the wheel loads with at least one measuring point, the speed of the train set being determined, and a device for carrying out this method.

The increasing automation in connection with an increase in the driving speed and the payload of railway vehicles requires the simplest possible control of the respective loading condition of a railway wagon. Existing safety regulations restrict the wheel loads of an axle, for example of freight wagons. There are also certain regulations for the ratio of the wheel loads of an axle. Another limitation in loading applies to the relationship between the axles and the bogies. Exceeding these restrictions or incorrect loading of the wagons can lead to a safety risk, because overloaded or skewed wagons no longer show their natural driving behavior.

In view of the growing demands for short transport times, the previously common methods for determining the wheel loads of wagons are no longer economical. The weighbridges used for this, on which the wagons could be weighed individually, are time-consuming and personnel-intensive. The weighed wagons must also be identified. It has therefore started to use dynamic scales that make it possible to measure the wheel loads when trains are moving. A proposal which is not yet part of the prior art (EP-A-0 099 001) describes an arrangement for measuring the wheel loads of fast-moving rail vehicles with at least one measuring point and at least one strain gauge attached to the web of a rail at the level of the neutral fiber, the main direction of action of which is orthogonal to neutral fiber, ie is arranged vertically. When a wheel overflows through such a measuring point, pronounced measuring peaks are obtained, from which the respective wheel load can be determined with great accuracy.

There is still no sufficiently reliable method for identifying wagons running in a fast-moving train group. - Although it is known to provide wagons with automatically readable markings, this requires special facilities on the wagons and in the area of the tracks. However, the information received is unreliable because it can be falsified by weather conditions. In addition, this information has no relation to the respective wheel loads.

In a generic method (DD-A-151 220), the speed of the train set is determined in addition to the wheel loads. Additional position sensors are used for this purpose, which are arranged along the rail at a fixed distance from one another and independent of the measurement for the wheel loads. This means that only center distances can be recorded that are defined by the specified distances of the position sensors. Other center distances cannot be recorded. This is also not possible with other known arrangements for determining the position of a wheel (US-A-38 61 203) or for determining wheel loads with the aid of several measuring points arranged one behind the other, taking into account the so-called sine run (FR-A-23 44 004) . To determine the wheel loads themselves, there are weighing rails (DE-B-15 49 244) or the measuring points already mentioned with vertically aligned strain gauges attached to the web of a rail (GB-A-864 193).

The object of the invention is to identify different wagons in moving, in particular fast-moving train sets.

This object is achieved in a method of the type described above in that strain gauges are used as the measuring point, which are attached to the web of a rail at the level of the neutral fiber and are aligned orthogonally to the neutral fiber, that when a wheel overflows over the measuring point, the resultant Measurement peak of the measurement signal is triggered at a threshold value, that the time period between two measurement peaks when two successive wheels overrun the measuring point is measured in each case when the threshold value is undershot or exceeded, and this measured time period is half the difference in the measured time periods between undershooting and exceeding the threshold value it is added to each of the two successive measurement peaks that the result is processed with the speed to form an axis distance and that the relevant wagon is identified with this axis distance or with several successive axis distances.

In the method according to the invention, the wagon is identified by its center distance. An analysis of the wagons available in the area of the German Federal Railways and their respective center distances has shown that under certain conditions it is possible to identify a certain wagon on the basis of its center distance and thereby distinguish it from other wagons. However, this presupposes that the measurement of the center distance is carried out with sufficient accuracy. Studies have shown that a accurate identification of a specific wagon is possible if the measured center distance can be determined with an error of less than 3.5%. Such error limits can, however, be observed with the described method, because this method is based on the measurement of time periods which are defined by intersections between a constant threshold value and relatively steeply increasing or decreasing measured values. The measurement values themselves are not affected by this, in particular the measurement peaks, as stated at the beginning, can be used to determine the wheel loads. This then also gives the possibility of assigning the determined wheel loads to the identified wagon.

The speed of the train set can also be detected using the arrangement of measuring points with strain gauges, namely by measuring the time between overflows of a wheel over two adjacent measuring points.

bestimmt werden, wobei

• S_A (k) = eine Maßzahl (mit Berücksichtigung der Geschwindigkeit) für den Achsabstand,
• i (k) - die Anzahl der Zählimpulse zwischen zwei Meßpeaks beim Überlauf zweier aufeinanderfolgender Räder über die Meßstelle jeweils bei Unter- oder Überschreitung des Schwellenwertes,
• j (k) = die Anzahl der Zählimpulse zwischen Unter- und Überschreitung des Schwellenwertes an einem Meßpeak,
• k - Index für aufeinanderfolgende Meßpeaks.

Otherwise, it is expedient to digitize the measured values and process them digitally. Then the times can also be measured digitally with a counter and the center distance according to the formula

be determined, whereby

• S _A (k) = a measure (taking into account the speed) for the center distance,
• i (k) - the number of counting pulses between two measuring peaks when two successive wheels overflow through the measuring point, each time the threshold value is exceeded or not reached,
• j (k) = the number of counts between falling below and exceeding the threshold value at a measuring peak,
• k - index for successive measurement peaks.

The difference to be added to the number i (k) takes into account that the wheel loads of successive wheels can be very different and that accordingly the assigned measurement peaks turn out differently. The difference takes these differences into account, so that the measure S _A (k) practically indicates counting pulses between the centers of two successive peaks. The center distance is obtained by dividing this measure S _A (k) by the measured speed.

Since there are not only biaxial, but also multiaxial wagons, one will measure the center distances between several successive peaks and then to identify the wagons in question, compare the determined axle spacing (s) with corresponding values of an identification matrix. Such an identification matrix contains, for example, the center distances of all railway wagons in traffic.

The device according to the invention for carrying out the method described has a measuring amplifier for each measuring point, an analog-digital converter connected downstream of the measuring amplifier, a trigger which can be set to a threshold value, a counter and a memory for the counting pulses, and a processor for performing the necessary mathematical operations. among others must also be carried out to determine the wheel loads. According to the invention, this device is characterized by a memory with an identification matrix for center distances of different wagons, so that in the course of determining the wheel loads it is also possible to identify the respective wagon and thus to assign measured wheel loads to the wagon in question.

• Fig. 1 in schematischer Darstellung die Seitenansicht eines Schienenabschnittes mit einer Meßstelle,
• Fig. 2 in schematischer Darstellung die Wiedergabe eines typischen Meßsignals an dieser Meßstelle,
• Fig. 3 in schematischer Darstellung ein Blockschaltbild einer Vorrichtung zur Bestimmung der Radlasten und zur Identifizierung des betreffenden Waggons aus dem Meßsignal,
• Fig. 4 in schematischer Darstellung das Meßsignal beim Überlauf von zwei aufeinanderfolgenden Rädern über die Meßstelle,
• Fig. 5 in schematischer Darstellung die Seitenansicht eines zweiachsigen und eines dreiachsigen Waggons mit typischen Abmessungen in Millimetern.

In the following an embodiment of the invention shown in the drawing is explained; show it:

• 1 is a schematic representation of the side view of a rail section with a measuring point,
• 2 shows a schematic representation of the reproduction of a typical measuring signal at this measuring point,
• 3 shows a schematic representation of a block diagram of a device for determining the wheel loads and for identifying the relevant wagon from the measurement signal,
• 4 shows a schematic representation of the measuring signal when two successive wheels overflow via the measuring point,
• Fig. 5 shows a schematic representation of the side view of a two-axle and a three-axle wagon with typical dimensions in millimeters.

The arrangement shown in Figure 1 is initially used to measure and determine the wheel loads of high-speed rail vehicles. On a rail 1 with rail head 2, rail web 3 and rail foot 4, a temperature-independent strain gauge 6 is attached on both sides of the web 3 at the level of the neutral fiber 5 so that its main direction of action is perpendicular to the neutral fiber 5, i.e. vertical, extends. The strain gauges 6 are otherwise attached so that they are arranged substantially symmetrically to the neutral fiber 5 with respect to their meandering turns. It is not shown that the strain gauges 6 and the measuring point defined thereby are located directly above a threshold.

If the measuring point of the rail 1 is overrun by a wheel 7 with an assigned wheel load 8, then the rail 1 is deformed in the area of the measuring point, as well as at other points, when overflowed by the wheel 7, as is exaggerated in FIG. 1. In other words, the rail 1 and in particular its web 3 are compressed in the vertical direction. The strains that occur are detected by the strain gauges 6 and converted into measurement signals.

A typical course of a measuring signal 9 when a wheel 7 overflows through a measuring point is shown in FIG. One recognizes a function superimposed by high-frequency interference, which according to theoretical considerations is a product of an exponential function and the sum of circular functions. Only the maximum of this function is decisive for determining the desired wheel load.

In order to be able to reliably determine the maximum of the measurement signal 9 and thus the desired wheel load 8, the measurement signal 9 is first broken down into two signal components, namely a constant signal component and a variable signal component. For this purpose, a device is used, as is shown schematically in FIG. 3. First, the measurement signal emitted by the strain gauge 6 is fed to a measuring amplifier 10, which may also contain the usual bridge circuit for several strain gauges 6. The amplified measurement signal is then digitized in an analog-digital converter 11. A trigger 12, which can be set to a predetermined threshold value 13, is also connected into the signal flow. This trigger 12 consequently separates the measurement signal into a constant signal component defined by the threshold value 13 and a variable signal component which remains in the window 14 indicated by dash-dotted lines in FIG. It goes without saying that the trigger 12 can also be connected upstream of the analog-digital converter 11, so that only the signal components remaining in the window 14 are fed to the converter.

When the threshold value 13 is exceeded and / or undershot, the trigger 12 also acts on a counter 15, the counting pulses of which are assigned to the respective digital measured values.

In order to simultaneously identify the wagon just passing over the measuring point, the speed of the train set must first be determined. To do this, it is sufficient if the time between wheel overflows is determined via two adjacent measuring points. Otherwise, the identification of the wagons passing over the measuring point is based on the counting pulses emitted by the counter 15.

If one assumes that at a constant speed of the train set, the time period between the overflow of two successive wheels via a measuring point is proportional to the center distance of these two wheels, then the center distance can be determined from the time between the overflow of two successive wheels via a measuring point passes.

The actual conditions are shown in FIG. The measurement signal 9 can be seen with two successive measurement peaks 17, 18. The threshold value 13 is also shown. Since it would be technically very difficult and too imprecise to determine the center of the measurement peaks 17 and 18, on the one hand, the number i (k) of the counting pulses between two measuring peaks 17, 18 when two successive wheels overrun the measuring point in each case when the threshold 13 is undershot, and moreover the number j (k) of the counting pulses between exceeding and undershooting the threshold 13 at each of the measuring peaks 17, 18 .

wobei k ein Index für aufeinanderfolgende Meßwerte ist. Diese Formel berücksichtigt, daß die Meßpeaks 17, 18 unterschiedlich ausgebildet sein können, weil aufeinanderfolgende Räder unterschiedlich belastet sein können. Jedenfalls kann aus der Maßzahl S_A (k) unter Berücksichtigung der ermittelten Geschwindigkeit der Achsabstand zweier aufeinanderfolgender, die Meßstelle überlaufender Räder mit großer Genauigkeit, d.h. mit einem Fehler, der kleiner ist als 3,5 %, bestimmt werden.A dimension number S _A (k) for the center distance of the two wheels overflowing the measuring point then results from the following formula:

where k is an index for successive measured values. This formula takes into account that the measurement peaks 17, 18 can be designed differently because successive wheels can be loaded differently. In any event, the center distance of two successive, the measuring point on running wheels with high accuracy, ie, determined with an error which is less than 3.5% from the measured value S _A (k) taking into account the determined speed.

The processing of the measured values and the counting impulses takes place in a processor 16, which also includes a memory 19 with an identification matrix for center distances of different wagons. This memory contains the center distances for different wagons, each without buffer dimensions, the wagons being appropriately arranged according to the number of the respective axles.

The distances from wagon to wagon, which result from the addition of two successive buffer dimensions, are outside the possible center distances if the error limits of less than 3.5% are observed. You therefore do not need to be identified.

FIG. 5 schematically shows the particularly critical case of a two-axle wagon in comparison with a three-axle wagon, with axis spacings of 8000 millimeters occurring in the cases. Since, when a two-axle wagon is coupled to a three-axle wagon, - with a maximum buffer dimension between the two wagons, there is an axle spacing of 7460 millimeters, the two wagons can be identified in any case if the error of the measured axle spacing is less than 3.5 % is.

Claims (6)

1. Method for the identifying of wagons running in a train formation, in particular a rapidly travelling one, with the use of an arrangement for the measuring of the wheel loads with at least one measuring point, wherein also the speed of the train formation is determined, characterised thereby, that strain gauges, which are fastened at the web of a rail at the level of the neutral fibre and oriented orthogonally to the neutral fibre, are used as measuring point, that - on a wheel running over the measuring point - the then arising measurement peak of the measurement signal is triggered at a threshold valuehat the time interval between two measurement peaks on the running of two successive wheels over the measuring point is measured each time on the threshold value being fallen below or exceeded and this measured time interval is added to half the difference of the measured time intervals between the threshold value being fallen below and being exceeded at each of both the successive measurement peaks, that the result is processed together with the speed into an axle spacing and that the wagon concerned is identified by this axle spacing or by several successive axle spacings.

2. Method according to claim 1, characterised thereby, that the speed of the train formation is determined by measurement of the time between two neighbouring measuring points being overrun by a wheel.

3. Method according to claim 1 or 2, characterised thereby, that the times are measured digitally by a counter and the axle spacing is determined according to the formula

S_A (k) = a measurement number (taking the speed into consideration) for the axle spacing,

i (k) = the number of the counting pulses between two measurement peaks on the running of two successive wheels over the measuring point each time on the threshold value being fallen below or exceeded,

j (k) = the number of the counting pulses between the threshold value being fallen below and exceeded at a measurement peak and index for successive measurement peaks.

4. Method according to one of the claims 1 to 3, characterised thereby, that for identification of the wagon concerned, the determined axle spacing is or spacings are compared with the corresponding values of an identification matrix.

5. Apparatus for the performance of the method according to one of the claims 1 to 4 with a measurement amplifier for each measuring point, an analog-to-digital converter connected behind the amplifier, a trigger device settable to a threshold value counter and a store for the counting pulses, and with a processor, characterised by a store (19) with an identification matrix for axle spacings of different wagons.

Images