AU2022209303A1 - Method of counting axles with computer-aided evaluation - Google Patents

Abstract

Method for counting axles with computer-assisted evaluation The subject matter of the disclosure is a method for counting axles, in which an axle counting sensor installed on a track is passed by a wheel, the axle counting sensor generates a measurement signal (Ul... U2) and the waveform (VL1... VL2) of the measurement signal (Ul... U2) is evaluated on a computer assisted basis and thus the wheel is identified. During the evaluation of the measurement signal (Ul... U2), at least one maximum (Ml... M4) of the signal amplitude is searched for in the waveform (VL1... VL2) of the measurement signal (Ul... U2). The amplitude of the measurement signal (Ul... U2) is normalized during an amplitude normalization in such a manner that the maximum (Ml... M4) is identical to a predefined target value (ZW). A dynamic time normalization is performed before and after the maximum (Ml... M4) for the waveform (VL1... VL2) of the measurement signal (Ul... U2). The waveform (NV1... NV2), which has been normalized by amplitude normalization and time normalization, of the measurement signal (Ul... U2) is compared with patterns (Ml... M2) both of at least one waveform (VL1... VL2) for the measurement signal (Ul... U2) when a wheel (RD) passes and of at least one waveform (VL1... VL2) for the measurement signal (Ul... U2) when an error occurs. The disclosure further comprises a computer program product and a provision apparatus for the computer program product. Fig 2 LULII uuuU 2/2 FIGL ZV U2 M2i M4 ----------- j----a------------------------------- --------- IM3 VL2 ul i t2 --------------- \---------------------- ---- VL1 * M3~ ZVR N '-"N N Z51 ZF2 r5Z3 zw zw zw x NVl V NV3 Ml M2 , 2 X

Description

LULII uuuU

2/2 FIGL ZV

U2 M2i M4 ----------- j----a------------------------------- ---------

IM3

VL2

ul i t2 --------------- \---------------------- ----

VL1 * M3~

ZVR

Z51 N ZF2 r5Z3 '-"N N

zw zw zw x

NVl V NV3

Ml M2 ,

2 X

Description

Method for counting axles with computer-assisted evaluation

The disclosure relates to a method for counting axles, in

which an axle counting sensor installed on a track is passed

by a wheel, the axle counting sensor generates a measurement

signal, the waveform of the measurement signal is evaluated on

a computer-assisted basis, wherein the wheel is identified.

Additionally, the disclosure relates to a computer program

product as well as to a provision apparatus for said computer

program product, wherein the computer program product is

equipped with program commands for performing this method.

During axle counting by way of axle counters, it is known that

a wide range of interference occurs, from simple noise or

environmental effects to sagging cables on trains or what are

known as sideways running effects for trucks in tight curves.

It is therefore desired in particular to suppress interference

which is similar to signals of wheels or trucks, in order to

be able to reliably recognize the wheel signals.

A related problem has to be solved, for example, during the

computer-assisted recognition of handwriting. A suitable

method for recognizing handwriting is described by Claus

Bahlmann et al. in IEEE TRANSACTIONS ON PATTERN ANALYSIS AND

MACHINE INTELLIGENCE, VOL. 26, NO. 3, MARCH 2004 in the

article "The Writer Independent Online Handwriting Recognition

System frog on hand and Cluster Generative Statistical Dynamic

Time Warping". This involves recognizing letters despite the

differences resulting from different handwritings. This cannot

be readily transferred to axle counters, however, as in axle

counters it is necessary to make a distinction between useful

signals, which indicate wheel passes, and interference signals.

In this context, a satisfactory level of safety additionally

has to be achieved. It should be taken into consideration that

the interference signals may be of an extent that they are

misinterpreted as a wheel pass. For this purpose, the

following example is to be indicated, without limiting the

generality.

When viewed in terms of measurement technology, a truck

consists of two successive wheels, i.e. two maximums of the

signal amplitude of the measurement signal with a certain

plateau therebetween. In this context, a measurement error may

occur, which is referred to as sideways running. In what is

known as sideways running, said plateau may be raised such

that a third wheel is incorrectly recognized.

It is an object of the present invention to substantially

overcome, or at least ameliorate, one or more of disadvantages

of existing arrangements, or provide a useful alternative.

Some embodiments of the invention are intended to specify a

method for counting axles, which has a comparatively high

level of safety in relation to incorrect recognition of wheel

passes. Additionally, other embodiments of the invention are

intended to specify a computer program product as well as a

provision apparatus for said computer program product, with

which the aforementioned method can be performed.

One aspect provides a method for counting axles, in which * an axle counting sensor installed on a track is passed by

a wheel, * the axle counting sensor generates a measurement signal,

* the waveform of the measurement signal is evaluated on a

computer-assisted basis, wherein the wheel is identified, wherein during the evaluation of the measurement signal, * at least one maximum of the signal amplitude is searched for in the waveform of the measurement signal, * the amplitude of the measurement signal is normalized during an amplitude normalization in such a manner that the maximum is identical to a predefined target value, * a dynamic time normalization is performed before and after the maximum for the waveform of the measurement signal, wherein the waveform, which has been normalized by amplitude normalization and time normalization, of the measurement signal is compared with patterns * both of at least one waveform for the measurement signal when a wheel passes * and of at least one waveform for the measurement signal when an error occurs.

Another aspect provides a computer program product with program commands for performing the method of the above aspect.

A further aspect provides a provision apparatus for the computer program product of the above aspect, wherein the provision apparatus stores and/or provides the computer program product.

Some embodiments of the present invention provide that, during the evaluation of the measurement signal, at least one maximum of the signal amplitude is searched for in the waveform of the measurement signal, the amplitude of the measurement signal is normalized during an amplitude normalization in such a manner that the maximum is identical to a predefined target value, a dynamic time normalization is performed before and after the maximum for the waveform of the measurement signal, wherein the waveform, which has been normalized by amplitude normalization and time normalization, of the measurement signal is compared with patterns both of at least one waveform for the measurement signal when a wheel passes and of at least one waveform for the measurement signal when an error occurs.

The measurement signal is a temporal waveform of the measured

measurement variable, preferably the signal voltage, which has

respective maximums caused by the passing of the wheel of an

axle, but also by interference effects. This means that, by

way of the computer-assisted evaluation of the measurement

signal, it is possible to recognize the event to be detected,

in that a wheel has passed the axle counting sensor, but it is

also possible for interference signals to be incorrectly

recognized as such a wheel pass.

According to the disclosure, both an amplitude normalization

and a dynamic time normalization, also referred to as dynamic

time warping (in the following DTW for short) are applied to

the measurement result. This has the advantage that the

measurement signal is normalized both with regard to the

amplitude of its maximum and with regard to the length of its

waveform in terms of time. This then facilitates the

comparison of the waveform of the measurement signal to be

evaluated, with predefined time duration and predefined

maximum amplitude, with patterns of various waveforms (more on

this in the following). This improves the reliability of a

pattern recognition for wheel passes and minimizes the

probability of the occurrence of incorrect evaluation results.

The amplitude normalization proceeds in such a manner that the

maximum of the considered waveform of the measurement signal after normalization is identical to a predefined target value. Preferably, it is possible to normalize to 1, i.e. the target value is equal to 1. This is not necessarily required, however. What is important is that the predefined target value of the maximum matches the maximums that are contained in the patterns, with which the relevant waveform of the measurement signal is to be compared.

The DTW is performed in order to identify, in a temporally limited section of the entire waveform of the measurement signal which extends before and after the maximum, which is to be compared with patterns for the purpose of recognizing wheel passes or occurring errors. Since a wheel pass generates a signal waveform which rises up to a maximum and subsequently falls again, a maximum is contained in the waveforms identified by the DTW in each case.

A time normalization takes place for the purpose of being able to perform a comparison of the relevant waveform of the measurement signal with the patterns. In this context, it should be taken into consideration in particular that the waveform of the measurement signal in particular depends upon the speed of the vehicle passing the axle counting sensor. A higher speed generates a steeper, shorter rise up to the maximum (and subsequently a corresponding fall). By comparison, a lower speed generates a flatter, longer rise up to the maximum (and subsequently a corresponding fall).

The principle of DTW is known, for example, from voice recognition (the recognition of speech features when dictating): here, individual words should be recognized from a spoken text through comparison with stored voice patterns. One problem consists in the words often being pronounced differently. Above all, vowels are often voiced longer or shorter. For successful pattern comparison, the word should therefore be stretched or compressed accordingly; not evenly, but rather primarily at the vowels that have been voiced longer or shorter. The dynamic time warping algorithm achieves this adaptive time normalization. Another application case is the recognition of handwriting. Here, pattern recognition of individual letters takes place, wherein the aim is to recognize the letters in different handwritings.

The disclosure makes use of the knowledge that, in comparison with a character recognition or voice recognition, the measurement signals of an axle counter have a comparatively low complexity. On the other hand, however, there are errors which may be mistaken for a wheel pass and therefore lead to incorrect results during the evaluation. Despite the comparatively low complexity of the patterns, these have to be recognized in a reliable manner. Some embodiments of the disclosure begin here by defining patterns not only for the events to be recognized of a wheel pass of various vehicles, but also for typically occurring errors, which then can be recognized as such and cannot be mistaken for a wheel pass.

In other words, the disclosure aims to recognize not only the events which are desired to occur and are to be counted, but also to deliberately also recognize the events which are not supposed to occur, accordingly are not supposed to be counted, but could be incorrectly recognized as an event to be counted. If these events are recognized as errors with certainty, these can be excluded as counting events, even if their assessment as event of a wheel pass to be counted were to be uncertain. This represents the added value according to the disclosure caused by increasing the recognition reliability.

"Computer-assisted" or "computer-implemented", in the context of the disclosure, may be understood to mean an implementation of the method in which at least one computer or processor carries out at least one method step of the method.

The expression "computer" covers all electronic devices with

data processing properties. Computers may be, for example,

personal computers, servers, handheld computers, mobile radio

devices and other communication devices that process data in a

computer-assisted manner, processors and other electronic

devices for data processing, which preferably may also be

combined to form a network.

A "processor", in the context of the disclosure, may be

understood to mean, for example, a converter, a sensor for

generating measurement signals or an electronic circuit. A

processor may involve, in particular, a central processing

unit (CPU), a microprocessor, a microcontroller, or a digital

signal processor, possibly in combination with a memory unit

for storing program commands, etc. A processor may also be

understood to mean a virtualized processor or a soft CPU.

A "memory unit", in the context of the disclosure, may be

understood to mean, for example, a computer-readable memory in

the form of random access memory (RAM) or data storage (hard

drive or data carrier).

"Interfaces" may be implemented as hardware, for example in a

cable-bound manner or as a wireless connection, and/or as

software, for example as an interaction between individual

program modules or program parts of one or more computer

programs.

"Program modules" are to be understood to mean individual

functional units that enable a program sequence of method steps according to the disclosure. These functional units may be implemented in a single computer program or in a plurality of computer programs that communicate with one another. The interfaces realized here may be implemented as software within a single processor or as hardware, if a plurality of processors are used.

In accordance with one embodiment of the disclosure, it is provided that the waveform, which has been normalized by amplitude normalization and time normalization, of the measurement signal is compared with patterns both of at least one waveform for the measurement signal when an individual wheel passes and of at least one waveform for the measurement signal when two wheels of a truck pass.

This embodiment of the disclosure makes use of the knowledge that the double axle of a truck, i.e. the two wheels that pass the axle counting sensor in this case, results in a characteristic pattern with two maximums. If these two maximums are identified as belonging to the truck by way of the DTW, then a normalization may take place with reference to said double event. Subsequently, this may be compared with the associated pattern. This results in the achievement of a further increase in the reliability. A recognized truck thus counts twice with regard to an axle counting, since it possesses two axles.

Should a truck not be recognized, then the two wheels may also be identified and assessed as individual wheels. In this context, this results in the same counting results, with the prerequisite that both wheels are identified. This shows that, with the definition of patterns that belong to a truck, an additional identification possibility is created, with the effect that the recognition reliability is improved. The reasoning for this is that the pattern of a truck makes more characteristic assessment criteria available and thus can be identified in a simpler manner. Should it not be recognized as a truck, however, there is still the fallback option of recognizing the individual wheels.

In accordance with one embodiment of the disclosure, it is provided that a comparison with patterns of a waveform for the measurement signal when two wheels of a truck pass is only performed when the temporal offset of the maximums in the waveform of the measurement signal does not exceed a limit value predefined as a function of the speed of a vehicle passing the axle counting sensor.

This measure is based on the knowledge that, when a truck passes, the axle counting sensor records two maximums, one after the other, in quick succession. In other words, it can be excluded that a truck is involved if the maximums are not measured within a speed-dependent time interval that is characteristic for trucks.

In order to be able to predefine the limit value, the speed of the vehicle crossing the axle counting sensor has to be known. There are different options for doing so. The speed can be ascertained by means of another sensor, for example, and fed into the method as an input variable. For example, the speed can be measured in the vehicle and transferred via radio to a computer, which performs the calculations of the method according to the disclosure.

Another option consists in estimating the speed from the relationship of a pattern of maximums (according to the axle counting pulses). Trucks are usually installed in vehicles of a certain length, meaning that trucks generate maximums that lie close together in each case and then a longer pause

(passing of the vehicle center) or a shorter pause (between

two coupled vehicles) occurs. From the ratio of the pauses, it

is possible to estimate the speed and thus also to determine

the speed-dependent limit value.

A further option consists in using what are known as double

axle counters, in which two axle counting sensors are

installed in close succession. As the distance between the

axle counting sensors is known, by determining the time offset

of the maximums generated by the same wheel in the two axle

counting sensors it is possible to infer the speed (more on

this in the following).

In accordance with one embodiment of the disclosure, it is

provided that the waveform, which has been normalized by

amplitude normalization and time normalization, of the

measurement signal is compared with at least one pattern of a

waveform for the measurement signal on occurrence of a

sideways running that occurs when trucks are negotiating a

curve.

As already mentioned, sideways running involves a measurable

excess signal increase of the measurement signal of an axle

counting sensor, which generates a maximum between the two

maximums of the wheel passes of a truck. Sideways running

preferably occurs when the axle counting sensor is installed

in a curve and the measurement takes place while the truck is

negotiating the curve.

If possible sideways running effects are defined as a pattern

of occurring errors, then when sideways running occurs during

the measurement by an axle counting sensor, as part of DTW it

is possible to generate a waveform which can be assigned to said error following comparison with the pattern available for the sideways running. If this assignment is unambiguous, the relevant waveform of the measurement signal can be excluded from an assignment of the event of a wheel pass. In particular, this is advantageous if an assignment to a wheel pass were a borderline case, and in case of doubt a non present axle would be incorrectly counted.

In other words, there are cases in which the method according

to the disclosure can be used with a higher certainty when

counting axles belonging to a truck. The occurrence of

incorrectly counted axles can therefore be excluded or the

probability of such an event can at least be reduced.

In accordance with one embodiment of the disclosure, it is

provided that an axle counter is used which, arranged one

behind the other in the direction of travel, has a first axle

counting sensor and a second axle counting sensor, wherein the

method is run through in succession for the first axle

counting sensor and the second axle counting sensor.

This involves what are known as double axle counters, the use

of which is widespread. The two installed axle counting

sensors, i.e. the first axle counting sensor and the second

axle counting sensor, therefore in each case generate the same

maximums in quick succession in the waveform of the

measurement signal over time, at least if no interference is

present. In this case, the maximums correspond to the counted

wheels. Otherwise, it is likewise possible to detect

interference signals that lead to maximums.

The use of two axle counting sensors does not change the

functional principle of the axle counter. The first axle

counting sensor and the second axle counting sensor function in the same way as the axle counting sensor of an axle counter, in which only a single axle counting sensor is installed. The statements made in the context of this disclosure therefore equally apply to the axle counting sensor or the first axle counting sensor as well as the second axle counting sensor, unless described otherwise.

The use of a first axle counting sensor and a second axle

counting sensor has the advantage that the axle counter has a

higher safety in relation to failure. In addition, as long as

the first axle counting sensor and the second axle counting

sensor are in operation, the sensor signals can also be used

to ascertain the speed of the vehicle passing the axle

counter. In this context, the maximums generated by one and

the same wheel are examined with regard to their time offset

in the first axle counting sensor and in the second axle

counting sensor and the speed is determined while taking into

consideration the known distance between the first axle

counting sensor and the second axle counting sensor.

In accordance with one embodiment of the disclosure, it is

provided that the maximums in the first waveform detected by

the first axle counting sensor and the second waveform

detected by the second axle counting sensor are compared, and

of the maximums in the first waveform and second waveform,

which have been normalized by amplitude normalization and time

normalization, of the measurement signal, only those which are

present both in the first waveform and in the second waveform

are compared with patterns.

This embodiment of the disclosure makes use of the knowledge

that the event of a wheel passing the axle counting sensor is

reliably recognized as a maximum in the waveform of the

measurement signal. For this reason, these maximums also have to appear in the two measured waveforms of the measurement signals. If a maximum only appears in one of the two waveforms of the measurement signals, then it can be reliably inferred that this involves an interference signal, which should not be counted per se. For this reason, it is advantageous to exclude this maximum from an assessment with regard to the presence of a wheel pass from the outset, whereby incorrect recognition decreases and thus the safety in relation to errors is advantageously enhanced during the recognition of wheels.

In accordance with one embodiment of the disclosure, it is provided that the maximums in the first waveform detected by the first axle counting sensor and the second waveform detected by the second axle counting sensor are compared and the waveform of the measurement signal before and after a maximum, which is to be taken into consideration during the dynamic time normalization, is determined while taking into consideration a time offset between a comparable maximum of the first waveform and the second waveform.

If, in the first waveform and the second waveform, maximums are found which correspond to one another, then the time offset that can be determined therefrom can advantageously be used to obtain a speed-dependent measure for the time limits of the waveform to be taken into consideration during the dynamic time normalization. This advantageously ensures that the waveform during the dynamic time normalization has a satisfactory range in order to contain the characteristics to be assessed for a later comparison with the patterns.

Furthermore, a computer program product with program commands for performing the stated method according to the disclosure and/or exemplary embodiments thereof is claimed, wherein the method according to the disclosure and/or exemplary embodiments thereof in each case can be performed by means of the computer program product.

Moreover, a provision apparatus for storing and/or providing

the computer program product is claimed. The provision

apparatus is, for example, a memory unit which stores and/or

provides the computer program product. As an alternative

and/or in addition, the provision apparatus is, for example, a

network service, a computer system, a server system, in

particular a distributed, for example cloud-based computer

system and/or virtual computer system, which preferably stores

and/or provides the computer program product in the form of a

data stream.

The provision takes place, for example, in the form of a

program data block as a file, in particular as a download

file, or as a data stream, in particular as a download data

stream, of the computer program product. This provision may

also, however, take place as a partial download, for example,

which consists of a plurality of parts. Such a computer

program product is read into a system, for example using the

provision apparatus, so that the method according to the

disclosure is made to be carried out on a computer.

Further details of some embodiments of the invention are

explained below, making reference to the drawing. Identical or

corresponding drawing elements are provided with the same

reference characters in each case and are explained multiple

times only insofar as differences arise between the individual

figures.

The exemplary embodiments set out in the following involve

preferred embodiments of the invention. The components of the

embodiments as described in the exemplary embodiments each represent individual features of an embodiment of the invention that are to be regarded as independent of one another and each also develop the embodiment of the invention independently of one another and are thus also to be considered individually, or in a different combination from that shown, as a constituent part of the embodiment of the invention. Furthermore, the components described can also be combined with the previously described features of embodiments of the invention.

In the drawings:

Figure 1 shows an exemplary embodiment of the apparatus according to an embodiment of the invention with its cause and-effect relationships in schematic form with a computer infrastructure as a block circuit diagram, wherein the individual functional units contain program modules, which in each case are able to run in one or more processors, and the interfaces accordingly may be designed as software or hardware,

Figure 2 shows an exemplary embodiment of the method according to an embodiment of the invention, wherein the individual method steps can be implemented individually or in groups by way of program modules and wherein the interfaces are indicated by way of example in accordance with Figure 1.

In Figure 1, a vehicle FZ is shown which is in transit in a direction of travel FR on a track GL. The vehicle FZ has trucks DG, which are provided with two axles in each case. These are indicated by wheels RD in Figure 1.

As soon as the wheels RD pass over an axle counter AZL, having a first axle counting sensor AZ1 and a second axle counting sensor AZ2, a pulse is generated in the waveform of the measurement signal Ul, U2 (cf. Figure 2) (more on this in the following).

The axle counter AZL is connected to an evaluation unit AE, which has a first computer CPl. This computer CP1 is connected both to the first axle counting sensor AZ1 as well as to the second axle counting sensor AZ2 via a sixth interface S6. Instead of two axle counting sensors, it is also possible for an individual axle counting sensor AZ to be used, for this reason one of the two axle counting sensors is referred to both with the reference character AZ and with the reference character AZ1.

Additionally accommodated in the evaluation unit AE is a first memory facility SE1, which is connected to the first computer CP1 via a fifth interface S5. This contains, for example, a program for performing the method according to an embodiment of the invention as well as a library with various patterns Ml, M2 (cf. Figure 2), which are used for certain waveforms VL1, VL2 to be measured, represented by normalized waveforms NV1, NV2, NV3 (cf. Figure 2).

Furthermore, the first computer CP1 is connected to a second computer CP2 in a control center LZ via a third interface S3. The second computer CP2 is moreover connected to a second memory facility SE2 via a fourth interface S4. The control center acts as a representative for a trackside facility, such as an interlocking or an automatic train control system.

The vehicle FZ as well as the control center LZ have antennas AT, so that these are able to communicate with one another via a second interface S2. Additionally, the vehicle FZ is able to communicate with a satellite STL via a first interface Sl. In this manner, for example, it is possible to track the vehicle

FZ, wherein the satellite STL involves a navigation satellite.

The method according to the disclosure has program modules,

which optionally are able to run in the first computer CP1 or

in the second computer CP2. This depends upon how "smart" the

arrangement of axle counters formed by the axle counter AZL

and the evaluation unit AE is embodied to be.

In Figure 2, the sequence of the method according to the

disclosure is shown on the basis of a flow diagram. In this

context, schematic representations of the signal waveforms are

chosen, in order to explain the individual method steps. In

the upper part of Figure 2, the waveforms VL1 of the first

axle counting sensor AZ1 and VL2 of the second axle counting

sensor AZ2 are shown. To this end, a diagram is chosen in

which the measurement signal Ul, U2 is shown in the form of an

output voltage over time t. In the lower part of Figure 2, the

subsequent method steps of a normalization with the result of

normalized waveforms NV1, NV2, NV3 as well as a comparison

with patterns Ml, M2 are shown.

In the methods shown in Figure 2, as already mentioned, the

axle counter AZL in accordance with Figure 1 with a first axle

counting sensor AZ1 and a second axle counting sensor AZ2 is

used. In the same way, it is conceivable to use an axle

counter with only one axle counting sensor AZ, wherein Figure

2 would appear similar, i.e. the diagram of the waveform VL1

as well as the measures associated therewith, indicated by

arrows, would be omitted.

On the basis of the waveform VL1 and the waveform VL2, it can

first be recognized that the axle counting sensors AZ1, AZ2

are installed in the track GL with a lateral offset in the direction of travel. This leads to a time offset ZVM of comparable maximums. This is indicated in Figure 2 by the first maximum, which is generated on the basis of the first wheel RD of the truck DG passing, being selected in the first waveform VL1 and in the second waveform VL2.

Furthermore, it can be recognized in the waveforms VL1, VL2 that it involves two wheels (axles) of a truck crossing. This can be recognized since, in the waveforms VL1, VL2, in addition to the time-offset first maximum Ml a further second maximum M2 can be recognized, which likewise is shifted by the time offset ZVM and has great similarity to the first maximum Ml. The first maximum Ml and the second maximum M2 are in each case spaced apart from one another by a time offset ZVR between each of the wheel passes. This time offset ZVR corresponds particularly to the time difference, which lies between the wheel pass of the first wheel of the truck DG and the second wheel RD of the truck DG.

In order to generate the normalized waveforms NV1, NV2, NV3, in accordance with the method according to the disclosure a normalization N is performed in a manner not shown in further detail. This normalization contains an amplitude normalization of the measurement signal Ul, U2 to a target value ZW, which lies at 1 in the exemplary embodiment in accordance with Figure 2. Additionally, a dynamic time normalization is performed, wherein the first waveform VL1 or the second waveform VL2 is in each case considered before and after the identified maximum Ml, M2, M3, to the extent that the waveform associated with the maximum Ml, M2, M3 can be characterized (and can be compared with patterns Ml, M2, more on this in the following). This produces the normalized waveforms NV1, NV2, NV3, effectively in time windows ZF1, ZF2, ZF3, which correspond to the patterns Ml, M2 in their temporal extent.

As can be recognized, the evaluation of the first maximum Ml leads to the generation of the first normalized waveform NV1 and the evaluation of the second maximum M2 leads to a generation of the third normalized waveform NV3. Furthermore, a third maximum M3 can be recognized both in the first waveform VL1 and in the second waveform VL2, which leads to the generation of a second normalized waveform NV2. A fourth maximum M4 can only be established in the second waveform VL2, and for this reason is discarded for normalization (indicated by an X). The reason for this may be that the fourth maximum M4 may not involve a wheel pass, since it would have to be able to be recognized both in the first waveform VL1 and in the second waveform VL2.

In the final step, a pattern comparison of the normalized waveforms NV1, NV2, NV3 takes place. This results in the first normalized waveform NV1 and the third normalized waveform NV3 in each case matching the first pattern Ml, which represents a wheel pass. This leads to a counting result of 2. The second normalized waveform NV2 is identified by means of the second pattern M2, which represents sideways running. For this reason, the normalized waveform NV2 is excluded from counting (indicated by an X).

Figure 2 indicates that the pattern Ml and the second pattern M2 have a trust region characterized by the hatched area, which permits certain fluctuations with regard to the normalized waveforms NV1, NV2, NV3. This allows for the fact that the measured waveforms VL1, VL2 are subject to certain tolerance fluctuations. In addition to a measurement tolerance, it should also be taken into consideration that different vehicles generate different measurement signals, which for example depend upon circumstances such as wheel wear of the vehicle.

List of reference characters

LZ Control center

FZ Vehicle

DG Truck

RD Wheel

FR Direction of travel

GL Track

AT Antenna

STL Satellite

AZL Axle counter

AZ, AZ1, AZ2 Axle counting sensor

AE Evaluation unit

CP1 ... CP2 Computer

SE1... SE2 Memory facility

S1... S5 Interface

VL1... VL2 Waveform

Ml... M4 Maximum

NV1... NV3 Normalized waveform

ZF1... ZF3 Time window

Ul... U2 Measurement signal

Ml... M2 Pattern

N Normalization

ZW Target value

ZVM Time offset between comparable maximums

ZVR Time offset between wheel passes

2 Counting result

X Exclusion

Claims (9)

Claims

1. A method for counting axles, in which

* an axle counting sensor installed on a track is passed by

a wheel, * the axle counting sensor generates a measurement signal,

* the waveform of the measurement signal is evaluated on a

computer-assisted basis, wherein the wheel is identified,

wherein

during the evaluation of the measurement signal,

* at least one maximum of the signal amplitude is searched

for in the waveform of the measurement signal, * the amplitude of the measurement signal is normalized

during an amplitude normalization in such a manner that

the maximum is identical to a predefined target value, * a dynamic time normalization is performed before and

after the maximum for the waveform of the measurement

signal,

wherein the waveform, which has been normalized by amplitude

normalization and time normalization, of the measurement

signal is compared with patterns * both of at least one waveform for the measurement signal

when a wheel passes * and of at least one waveform for the measurement signal

when an error occurs.

2. The method as claimed in claim 1,

wherein

the waveform, which has been normalized by amplitude

normalization and time normalization, of the measurement

signal is compared with patterns

* both of at least one waveform for the measurement signal

when an individual wheel passes

* and of at least one waveform for the measurement signal

when two wheels of a truck pass.

3. The method as claimed in claim 2,

wherein

a comparison with patterns of a waveform for the measurement

signal when two wheels of a truck pass is only performed when

the temporal offset of the maximums in the waveform of the

measurement signal does not exceed a limit value predefined as

a function of the speed of a vehicle passing the axle counting

sensor.

4. The method as claimed in any one of the preceding claims,

wherein

the waveform, which has been normalized by amplitude

normalization and time normalization, of the measurement

signal is compared with at least one pattern of a waveform for

the measurement signal on occurrence of a sideways running

that occurs when trucks are negotiating a curve.

5. The method as claimed in any one of the preceding claims,

wherein

an axle counter is used which, arranged one behind the other

in the direction of travel, has a first axle counting sensor

and a second axle counting sensor, wherein the method is run

through in succession for the first axle counting sensor and

the second axle counting sensor.

6. The method as claimed in claim 5,

wherein

the maximums in the first waveform detected by the first axle

counting sensor and the second waveform detected by the second

axle counting sensor are compared, and of the maximums in the

first waveform and second waveform, which have been normalized by amplitude normalization and time normalization, of the measurement signal, only those which are present both in the first waveform and in the second waveform are compared with patterns.

7. The method as claimed in claim 5 or 6,

wherein

the maximums in the first waveform detected by the first axle

counting sensor and the second waveform (VL1) detected by the

second axle counting sensor (VL2) are compared and the

waveform of the measurement signal before and after a maximum,

which is to be taken into consideration during the dynamic

time normalization, is determined while taking into

consideration a time offset between a comparable maximum of

the first waveform and the second waveform.

8. A computer program product with program commands for

performing the method as claimed in any one of claims 1 - 7.

9. A provision apparatus for the computer program product as

claimed in claim 8, wherein the provision apparatus stores

and/or provides the computer program product.

Siemens Mobility GmbH Patent Attorneys for the Applicant/Nominated Person SPRUSON&FERGUSON